Datasets:

LLMDH

/

OpenScience

like 0

Follow

LLM - Digital Humanities 7

https://openalex.org/W3162327270

https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0245311&type=printable

English

Effect of an inverted seated position with upper arm blood flow restriction on measures of elbow flexors neuromuscular performance

PloS one

cc-by

Results Inversion induced significant pre-exercise intervention decreases in elbow flexors MVC (21.1%, Z2 p = 0.48, p = 0.02) and resting evoked twitch forces (29.4%, Z2 p = 0.34, p = 0.03). The 30-s MVC induced significantly greater pre- to post-test decreases in potentiated twitch force (Z2 p = 0.61, p = 0.0009) during inversion (#75%) than upright (#65.3%) conditions. Over- all, BFR decreased MVC force 4.8% (Z2 p = 0.37, p = 0.05). For upright position, BFR induced 21.0% reductions in M-wave amplitude (Z2 p = 0.44, p = 0.04). There were no significant differ- ences for electromyographic activity or voluntary activation as measured with the interpo- lated twitch technique. For all conditions, there was a significant increase in pain scale between the 40–60 s intervals and post-30-s MVC (upright<inversion, and without BFR<BFR). Copyright: © 2021 Ahmadi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All data is available from the Open Science Framework (DOI: 10. 17605/OSF.IO/TZM4B). Funding: This study was funded by the Natural Science and Engineering Research Council of Canada (RGPIN-2017-03728) to DGB. Purpose The objective of the investigation was to determine the concomitant effects of upper arm blood flow restriction (BFR) and inversion on elbow flexors neuromuscular responses. Hamid Ahmadi1, Nehara Herat1, Shahab Alizadeh1, Duane C. Button1, Urs GranacherID2, David G. BehmID1* 1 School of Human Kinetics and Recreation, Memorial University of Newfoundland, St. John’s, 1 School of Human Kinetics and Recreation, Memorial University of Newfoundland, St. John’s, Newfoundland and Labrador, Canada, 2 Division of Training and Movement Science, University of Potsdam, Potsdam, Germany Newfoundland and Labrador, Canada, 2 Division of Training and Movement Science, University of Potsdam, Potsdam, Germany * [email protected] * [email protected] a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Ahmadi H, Herat N, Alizadeh S, Button DC, Granacher U, Behm DG (2021) Effect of an inverted seated position with upper arm blood flow restriction on measures of elbow flexors neuromuscular performance. PLoS ONE 16(5): e0245311. https://doi.org/10.1371/journal. pone.0245311 PLOS ONE PLOS ONE RESEARCH ARTICLE Methods Randomly allocated, 13 volunteers performed four conditions in a within-subject design: rest (control, 1-min upright position without BFR), control (1-min upright with BFR), 1-min inverted (without BFR), and 1-min inverted with BFR. Evoked and voluntary contractile prop- erties, before, during and after a 30-s maximum voluntary contraction (MVC) exercise inter- vention were examined as well as pain scale. Introduction Individuals can experience an involuntary inverted posture such as with an overturned vehicle or voluntary inversion with aerial maneuvers and sports (e.g. gymnastics). A decrease in force or power output, with an increased perceived difficulty in some situations can be life threaten- ing or affect performance. For example, in an overturned vehicle (e.g. car, helicopter) the vic- tim must have sufficient force to release the seat belt and shoulder harness while constrained by their body mass and blood flow can be restricted by both the harness and near maximal muscle contractions. Although inhibited neuromuscular function (i.e., force, rate of force development) has been reported when shifting from an upright to an inverted position [1–5], the underlying mechanisms are not clearly elucidated. Altered sympathetic nervous system activity during inversion (i.e., higher hydrostatic pressure increases vagal inputs), has been suggested as one primary mechanism that may influence changes in neuromuscular functions with inversion [1–4]. Inversion-induced hydrostatic pressure has also been suggested to con- tribute to neuromuscular impairments in animals [6–9] and humans [10]. Changes in perfusion pressure to a target muscle can be found during contractions at mod- erate-to-high intensities [11], low-intensity contractions combined with blood flow restriction (BFR) [12], and when the position of a working muscle changes in respect to the level of heart (i.e., above the heart induced lower perfusion pressure) [13]. Decrements in neural function and perceived exertion can be exacerbated by changes in perfusion pressure and reduced oxy- gen-induced peripheral fatigue [14], which for example can occur if an individual is trapped in a position with load exerted upon an immobilized limb. A squeezed or compressed limb can lead to ischaemia, increasing metabolic by-product accumulation, thereby activating pain afferents (group III and IV) [15–17], contributing to central nervous system inhibition [15]. Decreased force production and muscle performance were observed with changes in perfusion pressure (i.e. an arm lifted above the heart level) [18] and graded ischaemia of the lower limb [10]. Hobbs and McCloskey [19] indicated that with ischemia, there was greater muscle activ- ity (electromyography: EMG) to keep the force output at the requisite level. Partial occlusion (increased pressure and ischaemia) concomitant with prolonged exercise could influence the perception of effort and sense of pain [20]. Conclusion The concomitant application of inversion with elbow flexors BFR only amplified neuro- muscular performance impairments to a small degree. Individuals who execute forceful Competing interests: The authors have declared that no competing interests exist. 1 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance contractions when inverted or with BFR should be cognizant that force output may be impaired. contractions when inverted or with BFR should be cognizant that force output may be impaired. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 Participants Based on the force data from previous studies on similar research topics [1,2], a statistical a pri- ori power analysis (GPower 3.1, Dusseldorf, Germany) indicated that a minimum of six par- ticipants would be needed to attain an alpha of 0.05 (α error) with an actual power of 0.8 (1-β error). We were able to recruit a convenience sample of 13 (males, n = 7, age: 24.7 ± 4.9, height: 178.3 ± 8.3 cm, and mass: 79.9 ± 8.6 kg and females, n = 6, age: 24.5 ± 4.8, height: 162.0 ± 3.6 cm, and mass: 73.5 ± 14.3 kg) healthy physically active university students. Participants per- formed structured physical activity 3–4 days per week, which included resistance training on a regular (1–3 times per week) basis for the prior six months. They had no previous history of cerebral, hypertensive, or visual health problems or injuries. The participants were given an overview of all procedures (i.e., orientation and testing sessions) before data collection. If will- ing to participate, participants signed the consent form and completed a ‘Physical Activity Readiness Questionnaire for Everyone’ (PAR-Q+: Canadian Society for Exercise Physiology, approved September 12, 2011 version). The Interdisciplinary Committee on Ethics in Human Research, Memorial University of Newfoundland (ICEHR Approval #: 20192154-HK) approved this study and the study was conducted in accordance with the latest version of the Declaration of Helsinki. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 Introduction Although both inversion [1–4] and blood flow restriction [10,14,18,19] can alter force output and muscle activation, the combination of inversion (increased hydrostatic pressure) with blood flow restriction (BFR: increased perfu- sion pressure), on neuromuscular performance has not been previously investigated. It is unknown whether additive effects of hydrostatic (whole body effects) and perfusion (local muscle effects) pressure occur (inversion and BFR respectively), exacerbating neuromuscular performance decrements. With these contexts, the objective of this study was to investigate the potential effects of a one-minute inverted position with upper arm BFR on measures of elbow flexors isometric maximum voluntary contraction (MVC) force production, biceps and triceps brachii electro- myographic (EMG) activity, perceived pain, and performance fatiguability (extent of force reduction with a 30-s MVC). Based on prior inversion studies [1–4], it was hypothesized, that inversion and BFR would decrease both voluntary and evoked force output, decrease voluntary activation, and increase fatigue during the 30-s MVC, and the addition of BFR to inversion would amplify these impairments to neuromuscular function. 2 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance Experimental design https://doi.org/10.1371/journal.pone.0245311.g001 https://doi.org/10.1371/journal.pone.0245311.g001 MVC), followed by another evoked twitch and MVC with ITT (5- and 10-s after 30-s MVC respectively) (Fig 1). It is noted that there was a 5-s transition from both upright to supine, and supine to inversion positions. In total, a participant was placed at each condition for 150s, from positioning at the desired posture to post-30s MVC measures (i.e., post-fatigue evoked twitch, MVC ITT, and potentiated twitch force). MVC), followed by another evoked twitch and MVC with ITT (5- and 10-s after 30-s MVC respectively) (Fig 1). It is noted that there was a 5-s transition from both upright to supine, and supine to inversion positions. In total, a participant was placed at each condition for 150s, from positioning at the desired posture to post-30s MVC measures (i.e., post-fatigue evoked twitch, MVC ITT, and potentiated twitch force). Experimental design Based on the recommendation of the Canadian Society for Exercise Physiology [21], partici- pants were advised to not smoke, drink alcohol or partake in intensive physical activity six hours prior to testing and to not eat food or ingest caffeine two hours before participating in the testing procedure. Participants attended an orientation session, at least a week before data collection, where they were familiarized with both upright and inversion postures; and also became familiar with BFR, the interpolated twitch technique (ITT), EMG electrode placement, and isometric maximal voluntary contractions (MVC) techniques. Following the orientation session, all participants were capable of achieving near maximal (83.5%-95.9%) voluntary acti- vation before commencing the experimental conditions. Using random allocation (generated by Microsoft Excel) participants performed four, ~60 minutes, experimental conditions in a within subject design: 1) Control (1-min upright position without BFR), 2) Control with BFR (1-min upright position with BFR), 3) 1-min inversion (without BFR), and 4) Combined: 1-min inversion with BFR. Testing included 1) initial testing (upright seated position pre- fatigue), 2) pre-testing (one of the four conditions pre-fatigue) and post-testing after the 30-s MVC (one of the four conditions after the 30-s MVC). There was approximately 48 h between each experimental condition. To decrease diurnal rhythms effects, all of the test procedures were completed at approximately the same time of day. The temperature of the laboratory was maintained at 20C. Maximal evoked muscle twitch (i.e., resting position) and voluntary contractile properties (MVC with ITT) were initially tested from an upright seated position. The participant then performed a warm-up, which consisted of arm cycling (cycle ergometer: Monark Ltd. Swe- den), at 70 rpm with 1 kp for 5-min; followed by a specific warm-up of five 5-s isometric elbow flexion (~50% of perceived maximum) contractions. Following the warm-up and a 5-min rest period, pre-fatigue BFR testing procedures (relative to the posture) with the participant posi- tioned in the inversion chair were assessed. The subsequent condition/position (without/with an upper right arm BFR) also included an evoked twitch (10-s after achieving desired position or condition), MVC with ITT (5-s after the evoked twitch), a 30-s MVC protocol (5-s after PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 3 / 19 PLOS ONE Inversion, blood flow, and neuromuscular performance Fig 1. Experimental design. https://doi.org/10.1371/journal.pone.0245311.g001 Fig 1. Experimental design. Evoked contractile properties To assess evoked twitch forces and muscle compound action potentials (M-wave), evoked con- tractile properties were measured. To stimulate the musculocutaneous nerve, stimulating elec- trodes (electrode width was 5 cm) were placed at the intersection of the biceps brachii and deltoid (anode) and antecubital space (cathode), respectively [22]. Furthermore, the placement of electrodes was marked with ink from test to test to maintain the correct position of elec- trodes during each session. Stimulating electrodes were connected to a stimulator (Digitimer Stimulator, Model DS7AH, Hertfordshire, UK) with a maximum of one ampere (A) and 400 volts (V). Then, both amperage (10 mA– 1 A) and voltage (100–400 V) of the 200-μs duration square wave pulse were increased sequentially by 10mA increments until a plateau in the twitch torque and M-wave was attained. The electrode leads were taped to the participants’ arm to reduce movement and stress on the lines. While the initial resting twitch involved a single stimulus, the superimposed and subse- quent potentiated twitches had a 10 ms inter-pulse interval between two maximal twitches during biceps brachii nerve stimulation [23]. The ITT using doublet stimulation has been reported to be a valid and reliable measure of voluntary muscle activation (VMA) [23]. Previous studies [23–26] have shown that there is a possibility to activate all muscle fibres via a superimposed twitch on a voluntary contraction (ITT). For consistency, each session commenced with an initial evoked twitch. Since evoked single twitches are sensitive to prior contractions resulting in a potentiated response, the twitches were evoked both prior to (rest- ing twitch) and after (potentiated twitch) the MVC testing [27,28]. Superimposed twitches were delivered during 2–3 MVCs (4-s duration with 2-min rest between each MVC), before the BFR and fatigue intervention, and once following the BFR and 30-s MVC fatiguing muscle contraction (i.e., MVC during elbow flexion). The first supramaximal (120% of maximum) electrical stimulation was delivered at the 3-s point of the MVC (all participants could achieve maximal force within this duration), and the second twitch (as a potentiated twitch) was evoked at a 3-s interval after the MVC (subject was instructed to relax). This procedure was repeated if the MVC force of the second MVC was 5% higher than the first MVC [1,22– 25,28,29]. To estimate VMA, the amplitudes of the superimposed and post-contraction stimu- lation were compared [voluntary activation = [1 –(superimposed twitch/potentiated twitch)]  100] [23]. Force measures For voluntary and evoked force measures, the participants were seated in an inversion chair (initially in an upright position), which was designed and constructed by Technical Services of Memorial University of Newfoundland [1,3,4]. The chair can rotate through a 360-degree range. Straps secured the participant at the head, torso, shoulder, hip, and thighs. Hips and knees were positioned at 900 during data collection. A Wheatstone bridge configuration strain gauge (Omega Engineering Inc., Don Mills, Ont.) via a high-tension wire cable was attached to a reinforced strap around the right wrist to assess force output, and forces were collected and amplified via analog to digital data collection hardware and software (i.e., Biopac System Inc. DA 100, A/D convertor MP100WSW; Holliston, MA). Due to the physical layout of the labo- ratory equipment, the right elbow flexors were tested. 4 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance Voluntary contractile properties With the same set-up as the evoked contractile properties, elbow flexors MVC isometric force was measured while seated and secured in the inversion chair with a supinated forearm at a 90o elbow flexion angle with shoulders at 0o (mid-frontal plane), with a reinforced strap around the right wrist. An instruction (“as hard and as fast as possible”), with verbal encour- agement (“go go”) was provided by the researcher during the entire 4-s isometric MVC. The forces detected by the strain gauge, were used to analyze the peak isometric MVC. Evoked contractile properties A fatiguing protocol consisting of a 30-s MVC of the elbow flexors, was performed using an isometric elbow flexion MVC (right arm). The researcher provided consistent verbal encour- agement in terms of wording and timing (e.g., “keep it up” every 10 s, starting at 10 s point). Electromyography (EMG) Surface EMG was monitored during evoked twitches (muscle action potential: M-wave), 4-s MVC and 30-s MVC. First, the skin was prepared before electrode placement with shaving (removal hair), abrading (to remove possible dead epithelial cells), and cleaning the area with an alcohol swab to remove oils. Then, two pairs of bipolar electrodes (Kendall Medi-trace 100 5 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance series, Chikopee, Mass.) were positioned (established by SENIAM) [30] at the mid-belly of the biceps brachii (i.e., at 50% on the line from medial acromion to the fossa cubit) as an agonist muscle (corresponding to the muscle fibres), and the lateral head of the triceps brachii (i.e., halfway from the posterior crista of the acromion to the olecranon as an antagonist muscle). Bipolar electrodes were placed collar to collar (2 cm apart).The reference electrode was posi- tioned on the ulnar styloid process. The EMG signal was collected at 2000 Hz, band-pass filter 10–500 Hz and amplified 1000x (Biopac System MEC 100 amplifier, Santa Barbara, Calif; input impedance = 2M, common mode rejection ratio > 100 dB minimum [50/60 Hz]). The collected data (via the A/D con- verter, Biopac MP150) was stored on a personal computer for post-processing analysis. The final data (i.e., raw EMG) was rectified and integrated over 500 ms following an MVC [2–4]. Universal pain assessment tool The researcher utilized a universal pain assessment tool (scale with numbers and cartoon facial Figs) [33] to assess the degree of overall discomfort (pain perception) during inversion. Also, to investigate whether there was any interaction between the effects of inversion and BFR (dur- ing two control sessions), the sense of discomfort with and without BFR was examined. Blood flow restriction (BFR) To decrease potential harmful biological effects, which can occur following an extremely high dose of BFR [31], previous evidence [32] using Doppler ultrasonograms showed that a moder- ate BFR and partial occlusion of the brachial artery can lead to 100% venous restriction and partial arterial BFR, respectively. In order to reduce hydrostatic pressure effects, and perfusion to the right elbow flexors muscles, BFR was implemented. At the resting position, a pressure cuff (A+ Med 7–62 pressure cuff; Toronto, Canada) was placed around the upper right arm, while positioned at the level of the heart. Then, the hand bulb was squeezed to manually find a pulse elimination pressure (100% BFR). The individualized BFR in the present study was set relative to brachial systolic blood pressure. To create partial BFR for upright/inverted position, the hand bulb was squeezed to 50% of complete BFR. Furthermore, with a maximal isometric contraction, the blood flow would be further restricted, obviating the need for maximal BFR with a cuff. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 Reliability With the exception of moderate internal consistency (0.7) for biceps brachii’s M-wave, and acceptable (0.7  α < 0.8) reliability for resting twitch force, ICC reliability scores were excel- lent (0.82  α < 0.94) for MVC forces, potentiated twitch forces, biceps and triceps brachii EMG (Table 2). Statistical analysis The SPSS software (version 23.0, SPSS, Inc. Chicago, IL) was used for statistical analysis. Nor- mality of the data was assessed and confirmed using a histogram chart to illustrate skewness and kurtosis with a Shapiro-Wilks test. The value of Greenhouse-Geisser were reported if the assumption of sphericity was not met. Initially, sex effects were considered as a factor, but with a lack of any significant differences, the sample population was integrated for analysis. To determine the effect of inversion with BFR on fatigue index of the 30-s MVC, a two-way repeated measures ANOVA (2 seated positions ⤬2 blood flow conditions), while three-way repeated measures ANOVA (2 seated positions ⤬2 blood flow conditions ⤬2 times) was con- ducted for EMG median frequency. For the evoked twitches, MVCs, ITT, potentiation twitches, muscle action potential (M-) waves, EMG, a three-way ANOVA (2 seated positions and 2 blood flow conditions and three times [initial upright resting position, pre-, and post- 30-s MVC]) was applied. Further, a repeated measures ANOVA was conducted to analyze the pain scale. Differences were considered significant when a minimum value of p = 0.05 was reached. Planned pairwise comparisons; Bonferroni adjustment, was selected to compare main effects. Additionally, the calculated partial eta squared (Z2 p) by SPSS was reported as a magnitude of outcomes (effect sizes); which is classified as small (0.00  Z2 p  0.24), medium (0.25  Z2 p  0.39), and large (Z2 p  0.40) [39]. Day to day reliability of measures (for initial upright resting position test) was assessed with Cronbach’s alpha intraclass correlation coeffi- cient (ICC). Results Within the results text and Figs, significant interactions are reported. Main effects are listed and described in Table 1. All data is available from the Open Science Framework: DOI 10. 17605/OSF.IO/TZM4B. Representative raw data sample tracings are provided in Fig 2. Data analysis All data analyses were conducted using the AcqKnowledge software program (Biopac System Inc. Holliston, MA). With the single electrical stimulation, peak twitch force (Newtons), time to peak twitch force (ms), half relaxation time (ms) and M wave (millivolts: mV) were mea- sured for resting and potentiated twitches. Peak MVC forces [34] were analyzed. The resting force output in the upright position was zeroed and used as the baseline when compensating for the suspended arm mass in the inverted position. The mass of the arm was recorded in the inverted position and this value subtracted from the peak force in order to counterbalance the force of gravity negatively affecting force output in the upright position. A force fatigue index was calculated from the 30-s MVC, which involved dividing the mean force of the last 5-sec- onds of the fatigue protocol into the mean force output of the first 5-seconds. The peak amplitude of integrated EMGs, from the biceps and triceps brachii, were analyzed from a one second period of the 4-s MVC (before the superimposed twitch) [29]. Furthermore, a Fast Fourier transform was used to report the EMG median frequency following the fatigue protocol, as it is a reliable indicator for signal conduction velocity (following a fatigued- PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 6 / 19 PLOS ONE Inversion, blood flow, and neuromuscular performance exercise) and generally considered a more sensitive indicator of fatigue than a raw EMG signal [35–38]. PLOS ONE Table 1. Main effects (means ± standard deviation). Main Effects for Time Initial Test Pre-Fatigue Post-Fatigue Initial to Pre- Fatigue Pre to Post-Fatigue MVC Force (N) F(2,16) = 52.58 309.32 ± 84.55 269.18 ± 75.16 226.30 ± 63.38 p<0.001, Z2 p = 0.82, Δ14.9% p<0.001, Z2 p = 0.84 Δ18.9% F100 (N) (F(2,18) = 31.81 73.28 ± 36.47 52.14 ± 25.64 34.32 ± 18.17 p = 0.003, Z2 p = 0.63 Δ40.5% p<0.001, Z2 p = 0.80 Δ51.9% BB EMG (mV/s) F(1,12) = 22.41 0.30 ± 0.16 0.29 ± 0.15 0.24 ± 0.12 p>0.05 p<0.001, Z2 p = 0.65 Δ22.0% TB EMG (mV/s) F(1,8) = 7.34 0.07 ± 0.02 0.06 ± 0.02 0.05 ± 0.01 p = 0.02, Z2 p = 0.47 Δ16.6% p = 0.003, Z2 p = 0.67 Δ20.0% BB M-wave (mV) F(1,7) = 3.89 9.05 ± 4.06 8.96 ± 3.22 8.49 ± 2.95 p>0.05 p = 0.08, Z2 p = 0.35 Δ7.6% PTF (N) F(2,10) = 36.3 66.48 ± 19.11 63.82 ± 18.38 28.86 ± 14.07 p>0.05 p<0.001, Z2 p = 0.87 Δ121.1% VMA (%) F(1,4) = 6.62 91.71 ± 5.36 92.47 ± 3.44 82.53 ± 11.35 p>0.05 p = 0.06, Z2 p = 0.62 Δ12.0% Main Effects for BFR no BFR Conditions BFR Conditions no-BFR to BFR MVC Force (N) F(1,8) = 4.84 274.61 ± 76.60 261.93 ± 72.12 p = 0.05, Z2 p = 0.37 Δ4.8% BB M-Wave (mV) F(1,7) = 7.28 9.67 ± 2.11 7.99 ± 2.44 P = 0.03, Z2 p = 0.51 Δ21.0% Perceived Pain Scale F(1,12) = 7.85, 2.5 ± 2.59 3.19 ± 2.27 p = 0.01, Z2 p = 0.39 15.6% Main Effects for Seated Position Upright Seated Positions Inverted Seated Positions Upright vs. Voluntary contractile properties Elbow flexors MVC. A significant seated position x time interaction (F(2,16) = 5.07, p = 0.02, (Z2 p = 0.48) (Fig 3) was observed for elbow flexor MVC force. The interaction revealed 9.1% and 21.1% MVC force decrements with the initial measures exceeding pre-fatigue mea- sures for upright and inverted positions, respectively. Furthermore, there were 21.4% and 14.2% force impairments from the initial and pre-test to post-test respectively for the upright position. Similarly, there were 33.1% and 17.7% decrements from the initial and pre-test to post-test respectively for the inverted position. A non-significant, medium effect size, seated position x BFR interaction (F(1,8) = 3.79, p = 0.08, Z2 p = 0.32); demonstrated 9.1% lower MVC forces with BFR compared to without BFR for inverted seated positions. 7 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 Inversion, blood flow, and neuromuscular performance PLOS ONE Inverted F100 (N) F(1,9) = 3.42 58.27 ± 28.01 48.22 ± 25.51 p = 0.09, Z2 p = 0.27 Δ8.3% Perceived Pain Scale F(1,12) = 14.56, 2.26 ± 1.98 3.42 ± 2.8 p = 0.002, Z2 p = 0.54 Δ26.5% Main Time Effects for Fatigue–Force, and Fatigue -EMG Relationships 0-5s interval (no-BFR and BFR) 25-30s interval (no-BFR and BFR) 0-5s to 25-30s intervals Force (N) F(1,11) = 29.96 278.42 ± 85.72 242.86 ± 74.45 p<0.001, Z2 p = 0.73 Δ14.6% BB EMG Median Frequency (Hz) F(1,10) = 84.65 57.67 ± 10.61 41.32 ± 5.59 p<0.001, Z2 p = 0.89 Δ39.5% TB EMG Median Frequency (Hz) F(1,12) = 86.69 59.25 ± 12.53 44.26 ± 9.25 p<0.001, Z2 p = 0.87 Δ33.8% Table 1. Main effects (means ± standard deviation). BB = Biceps Brachii, EMG: Electromyography, F100: Force output within the first 100 ms of the MVC, MVC: Isometric maximal voluntary contraction, M-wave: Muscle action potential, PTF = Potentiated Twitch Force, RTF = Resting Twitch Force, TB = Triceps Brachii, VMA = Voluntary Muscle Activation. Voluntary muscle activation (VMA %). A non-significant, but large eta2 magnitude, BFR x time interaction revealed 7.9% and 16.4% decrements post-30-s MVC (F(1,4) = 5.48, p = 0.07, Z2 p = 0.57) for without BFR and BFR conditions, respectively (Fig 4). Evoked twitch contractile properties. A significant seated position x time interaction (F(1,11) = 5.80, p = 0.03, Z2 p = 0.34), for resting pre-MVC twitch force revealed that initial values exceeded pre-30-s MVC, by 29.4% for inverted seated positions (Fig 5). No significant effects or interaction were found for baseline time to peak twitch force. Meanwhile, a non-significant, large magnitude effect size effect was observed (F(1,6) = 4.81, p = 0.07, Z2 p = 0.44), in terms of seated position and BFR, when comparing upright values (without BFR<BFR, 3.9%") with inverted seated positions (without BFR>BFR, 5.6%#). No significant effects were observed for half relaxation time-twitch force. Biceps and triceps brachii M-wave. A significant seated position x BFR interaction (F(1,7) = 5.69, p = 0.04, Z2 p = 0.44), revealed that the without BFR condition exceeded BFR condition biceps brachii’s M-wave by 30.4% and 12.5% for upright, and inverted seated positions, respec- tively (Fig 6). Biceps and triceps brachii M-wave. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE A significant seated position x BFR interaction (F(1,7) = 5.69, p = 0.04, Z2 p = 0.44), revealed that the without BFR condition exceeded BFR condition biceps brachii’s M-wave by 30.4% and 12.5% for upright, and inverted seated positions, respec- tively (Fig 6). 8 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance Fig 2. Representative 1) evoked resting twitch, 2) MVC force, 3) interpolated twitch technique (ITT), 4) biceps brachii EMG, 5) triceps brachii EMG and 6) potentiated twitch force (PTF) tracings presented under different conditions/posture. Top channel represents elbow flexor MVC force, second channel illustrates biceps brachii EMG and third channel shows triceps brachii EMG. Fig 2. Representative 1) evoked resting twitch, 2) MVC force, 3) interpolated twitch technique (ITT), 4) biceps brachii EMG, 5) triceps brachii EMG and 6) potentiated twitch force (PTF) tracings presented under different conditions/posture. Top channel represents elbow flexor MVC force, second channel illustrates biceps brachii EMG and third channel shows triceps brachii EMG. https://doi.org/10.1371/journal.pone.0245311.g002 Evoked potentiated twitch contractile properties. Significant seated position x time interaction (F(2,10) = 7.91, p = 0.009, Z2 p = 0.61) was observed for potentiated twitch force (PTF). The pre- and post-30-s MVC results showed 65.3% and 75% PTF impairments for upright and inverted positions from pre-to post 30-s MVC, respectively (Fig 7). Fatigue–force relation. When comparing force output values, at 0–5 s and 25–30 s inter- vals of the 30-s MVC task in the without BFR conditions versus 0–5 s and 25–30 s of BFR con- ditions, there was 4.1% force decrements for BFR effects (F(1,11) = 5.54, p = 0.03, Z2 p = 0.33). Daily intraclass correlation coefficient reliability (initial resting position test performed on separate days). Table 2. Daily intraclass correlation coefficient reliability (initial resting position test performed on separate days). MVC F100 BB EMG TB EMG RTF BB M-wave TB M-wave PTF ICC 0.93 0.82 0.94 0.83 0.77 0.66 0.6 0.87 CV 0.28 0.51 0.54 0.48 0.41 0.34 0.40 0.43 TEM 35.17 32.68 0.07 0.03 15.77 2.05 1.7 10.12 BB = Biceps Brachii, CV: Coefficient of variance, EMG: Electromyography, F100: Force output within the first 100 ms of the MVC, MVC: Isometric maximal voluntary contraction, M-wave: Muscle action potential, PTF = Potentiated Twitch Force, RTF = Resting Twitch Force, TB = Triceps Brachii, TEM = typical error of measurement. BB = Biceps Brachii, CV: Coefficient of variance, EMG: Electromyography, F100: Force output within the first 100 ms of the MVC, MVC: Isometric maximal voluntary contraction, M-wave: Muscle action potential, PTF = Potentiated Twitch Force, RTF = Resting Twitch Force, TB = Triceps Brachii, TEM = typical error of measurement. https://doi.org/10.1371/journal.pone.0245311.t002 PLOS ONE BB = Biceps Brachii, CV: Coefficient of variance, EMG: Electromyography, F100: Force output within the first 100 ms of the MVC, MVC: Isometric maximal voluntary contraction, M-wave: Muscle action potential, PTF = Potentiated Twitch Force, RTF = Resting Twitch Force, TB = Triceps Brachii, TEM = typical error of measurement PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 9 / 19 PLOS ONE Inversion, blood flow, and neuromuscular performance Fig 3. Isometric maximal voluntary contraction force (MVC) interaction effects for seated position and time. Star () symbol represents that significant MVC force decreases between initial and pre-fatigue tests, for upright and inverted seated positions. Means and standard deviations are illustrated. There was no statistically significant BFR interaction. Fig 3. Isometric maximal voluntary contraction force (MVC) interaction effects for seated position and time. Star () symbol represents that significant MVC force decreases between initial and pre-fatigue tests, for upright and inverted seated positions. Means and standard deviations are illustrated. There was no statistically significant BFR interaction. https://doi.org/10.1371/journal.pone.0245311.g003 Fatigue—EMG relation. Biceps brachii. Furthermore, contrasts revealed a non-signifi- cant, medium magnitude effect size, seated position x time interaction (F(1,10) = 3.84, p = 0.07, Z2 p = 0.27); with 45.6% and 38.6% decreases in biceps brachii EMG median frequency during upright without BFR and BFR conditions respectively, versus similar 32.8% and 41.4% median frequency decrements during the inverted seated position for without BFR and BFR condi- tions, respectively (Table 2). Perceived pain—pain scale. Seated position x BFR interaction revealed significant differences (F(1,12) = 6.55, p = 0.02, Z2 p = 0.35) between upright (without BFR<BFR, 0.64 to 1.53, 58.1%") and inversion (without BFR>BFR, 2.70 to 2.43, -11.1%) positions. For the interaction of BFR x time Fig 4. Voluntary muscle activation (VMA) interaction effects for BFR and time. There was a non-significant effect for percentage of VMA (p = 0.07), with 7.9% and 16.4% decreases post-fatigue for without BFR and BFR conditions. Means and standard deviations are illustrated. There was no statistically significant interaction between seated positions (inverted versus upright). https://doi.org/10.1371/journal.pone.0245311.g004 Fig 4. Voluntary muscle activation (VMA) interaction effects for BFR and time. There was a non-significant effect for percentage of VMA (p = 0.07), with 7.9% and 16.4% decreases post-fatigue for without BFR and BFR conditions. Means and standard deviations are illustrated. There was no statistically significant interaction between seated positions (inverted versus upright). PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE https://doi.org/10.1371/journal.pone.0245311.g004 https://doi.org/10.1371/journal.pone.0245311.g004 10 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance Fig 5. Resting twitch force interaction effects for seated position and time. Star () symbol represents significant decreases between initial and pre-fatigue tests, for upright and inverted seated positions. Means and standard deviations are illustrated. There was no statistically significant BFR interaction. https://doi.org/10.1371/journal.pone.0245311.g005 Fig 5. Resting twitch force interaction effects for seated position and time. Star () symbol represents significant decreases between initial and pre-fatigue tests, for upright and inverted seated positions. Means and standard deviations are illustrated. There was no statistically significant BFR interaction. https://doi.org/10.1371/journal.pone.0245311.g005 (F(2.05,24.62) = 4.68, p = 0.01, Z2 p = 0.28), a 4.6% pain scale decrease between 20–40 s and 40-60s intervals for the without BFR condition, contrasted with an overall (both without BFR and BFR) increase from initial (upright/inverted) to post-30-s MVC measures. Furthermore, a seated posi- tion x BFR x time interaction (F(2.87,34.47) = 3.18, p = 0.03, Z2 p = 0.29) showed significant differences when comparing without BFR and BFR, upright versus inverted conditions between 0–20 s and Fig 6. Biceps brachii M-wave interaction effects for seated position and BFR. Star () symbol represents that significant decreases in amplitude of M-wave for Biceps Brachii, between without BFR and BFR, with 30.4% and 12.5% for upright and inverted seated positions, respectively. Means and standard deviations are illustrated. https://doi.org/10.1371/journal.pone.0245311.g006 Fig 6. Biceps brachii M-wave interaction effects for seated position and BFR. Star () symbol represents that significant decreases in amplitude of M-wave for Biceps Brachii, between without BFR and BFR, with 30.4% and 12.5% for upright and inverted seated positions, respectively. Means and standard deviations are illustrated. https://doi.org/10.1371/journal.pone.0245311.g006 11 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance Fig 7. Potentiation Twitch Force (PTF) interaction effects for seated position and time. Star () symbol represents that significant Potentiation Twitch Force decreases between pre- and post-fatigue tests, for upright and inverted seated positions. The hashtag or number symbol (#) indicates that PTF tested with inversion post-fatigue was significantly lower than all other times and conditions. Means and standard deviations are illustrated. There was no statistically significant BFR interaction. htt //d i /10 1371/j l 0245311 007 Fig 7. Potentiation Twitch Force (PTF) interaction effects for seated position and time. https://doi.org/10.1371/journal.pone.0245311.g007 20–40 s intervals (-32.6% (0.61 to 0.46) without BFR and 13.6%" (1.46 to 1.69) for BFR upright positions, versus 7.8%" (2.69 to 2.92) and 3.2%" (2.38 to 2.46) for inverted without BFR and BFR conditions), and between 40–60 s and post-30-s MVC (77.5%" (0.38 to 1.69)) and 35.2%" (1.84 to 2.84) for upright without BFR and BFR versus 13.9%" (2.84 to 3.30) and 26%" (2.61 to 3.53) for inverted without BFR and BFR conditions). A main effect for time (F(1.37, 16.49) = 11.39, p = 0.002, Z2 p = 0.48) showed 34.6% (Z2 p = 0.45, initial<pre-30-s MVC), 14% (Z2 p = 0.26, pre-30-s MVC<0–20 s interval), and 32.3% (Z2 p = 0.49, 40–60 s interval<post-30-s MVC) pain scale increases. Main effects for seated position and BFR showed 26.5% (upright<inversion) (F(1,12) = 14.56, p = 0.002, Z2 p = 0.54) and 15.6% increases (without BFR<BFR) (F(1,12) = 7.85, p = 0.01, Z2 p = 0.39) for pain scale. PLOS ONE Star () symbol represents that significant Potentiation Twitch Force decreases between pre- and post-fatigue tests, for upright and inverted seated positions. The hashtag or number symbol (#) indicates that PTF tested with inversion post-fatigue was significantly lower than all other times and conditions. Means and standard deviations are illustrated. There was no statistically significant BFR interaction. https://doi.org/10.1371/journal.pone.0245311.g007 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 Inversion effects Evoked contractile properties. Significant reductions in pre-test evoked twitch force with inversion are partially in accord with a previous inversion study that showed non-significant, but moderate magnitude decreases (18.6%, effect size [ES] = 0.74) for elbow flexors resting twitch force from upright to the inverted position contrasting with non-significant moderate magnitude increases (17.5%, ES = 0.76) for knee extensors [3]. Similarly, greater 30-s MVC- induced PTF decreases with inversion versus upright, in this research, also conforms with Neary et al. [3], who showed non-significant (p = 0.06), small magnitude, decreases for elbow flexor PTF (11.8%, ES = 0.44); whereas they reported large magnitude increases (p = 0.03, ES = 1.27, 27.3%) for leg extensor PTF forces during seated inversion. With inversion, the position of the arm and leg induces higher and lower hydrostatic pres- sure respectively, hence contributing to the contrasting arm and leg responses in the Neary et al. [3] study. With high gravitational pressure, previous animal studies have shown an altered function of acetylcholine receptors [8], relative decrease in psoas single muscle fibre isometric force [6], decreased tetanic force for extensor digitorum longus muscle [9], and decreased Ca++ influx into the nerve terminal [7]. In humans, Sundberg and Kaisjer [10] showed that a graded occluded lower limb (up to 50 mmHg to increase pressure) under upright conditions caused a progressive decrease in muscle function. These results suggest higher hydrostatic or perfusion pressures on the arm during inversion can significantly reduce evoked forces. In contrast, Paddock and Behm [4] reported no significant alterations to evoked twitch properties with seated inversion. In the Paddock and Behm study, only male subjects were recruited, and they spent 30-s rather than 1-min in the inverted position. In addition, after each contraction, the participants were returned to an upright position for 2-minutes before adopting the inverted position again for another series of voluntary and evoked contractions. Hence, the differences in duration and recovery from inversion may have attenuated changes in hydrostatic pressure while sex differences may have contributed to the differences in evoked twitch property results. Voluntary contractile properties. The 21.1% inversion-induced MVC force decrease in the present study aligns with the 18.9–21.1%, 10.4%, and 6.1% impairments reported by Johar et al. [2], Hearn et al. [1] and Paddock and Behm [4], respectively. It might be argued that the relatively less substantial force decreases experienced by the participants in the Hearn et al. Discussion Prior studies have reported upon the deficits related to global (whole body) effects of inver- sion-induced changes in hydrostatic pressure [1–5]. Similarly, there is a body of literature exhibiting impairments associated with local (muscle group) BFR-induced increases in perfu- sion pressure [10,14,18,19]. The present study is the first to investigate the separate and com- bined effects of inversion (increased hydrostatic pressure) and BFR (increased perfusion pressure) on voluntary and evoked contractile properties. Major findings of this study were that inversion induced significantly greater decreases in resting twitch and elbow flexors MVC forces before the 30-s MVC task. Following the 30-s MVC task, inversion induced greater decreases in PTF. BFR led to overall (inversion and upright) detriments in MVC force, as well as greater decreases than without BFR while in the inverted position. Furthermore, there was a greater decrease in M-wave amplitude for the upright versus inverted position. In addition, the concomitant application of inversion of the participant with BFR of the elbow flexors only amplified neuromuscular performance impairments to a small degree. Representative traces of the measures are presented in Fig 2. 12 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance Inversion effects [1] and Paddock and Behm [4] studies may be attributed to shorter durations of inversion (30-s or less in each study) with 2-min recovery periods to an upright position between each MVC. However, Johar and colleagues had very similar force decreases as the present study and had participants exert MVCs for 6-s following rapid rotations of 1- or 3-s to an inverted position. Thus, MVCs were competed in less than 10-s in the Johar et al. [2] study. Neary et al. [3] reported no significant change in elbow flexor MVC force, which might be attributed to the highly trained track and field (athletics) athletes recruited in their study, who were permitted 1-min recovery periods between contractions. Hence, the relatively lower or non-significant force decrements in the Hearn et al. [1], Paddock and Behm [4] and Neary et al. [3] studies respectively may be more related to the greater recovery periods between inversion and return to upright positions. The decreases in elbow flexors MVC force with inversion can be related to several factors. The decreased force output can be related to a muscle stiffening mechanism, associated with a threat of instability [40], that the participants could have perceived when inverted with only straps holding them in the chair. Adkin et al. [41] found that the stiffening strategy can nega- tively influence voluntary movement. Increased co-contractile activity with inversion has been PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 13 / 19 PLOS ONE Inversion, blood flow, and neuromuscular performance previously reported as a factor counteracting target force output [1,4], however there were no significant increases in triceps brachii EMG in the present study. It is possible that an increased focus on stabilizing functions of the shoulders, and trunk muscles [42], could negatively impact the force output of the elbow flexors [43,44]. A shift from mobilizing to stabilizing strategies of the neuromuscular system has been reported to contribute to force reduction [43,45]. There were no significant inversion-induced decrements in biceps brachii EMG, which contrasts with 21.7%-35.9%, 26.6%, 47.9% decreases with Johar et al. [2], Paddock and Behm [4] and Hearn et al. [1] studies, respectively. As mentioned previously, these contrasts may be attributed to the greater inversion and contraction recovery periods or recruited population (i.e., track and field athletes) in the cited studies. Inversion effects Hence, while these other studies postulated inversion-induced neural inhibition due to increases in cerebral blood pooling and intracra- nial pressure [46–48], or attenuated sympathetic drive [18,46,49–51], the lack of EMG and ITT changes in the present study suggests that neural influences did not play a major role in MVC force reductions. Voluntary activation (ITT) was decreased following the 30-s MVC for both BFR and no-BFR but EMG did not significantly change. The curvilinear nature of the EMG- force relationship often resulting in a plateau at high voluntary forces diminishes the EMG sensitivity to force changes at high contraction intensities [25]. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 Blood Flow Restriction (BFR) The modest but significant decrements in MVC force (<5%) and only near significant (p = 0.07) changes in EMG activity with blood flow occlusion from a blood pressure cuff PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 14 / 19 PLOS ONE Inversion, blood flow, and neuromuscular performance might be attributed to the occlusion effects from the MVC. Since a MVC also contributes to BFR, the additional effect of the blood pressure cuff on a maximal contraction might not have been as prominent as it might be with submaximal contractions. Perceived pain. Previous seated inversion studies [2–4] have mentioned that there was a sense of discomfort (i.e., distinct swelling around the head region) during inversion but did not directly measure or report the degree of discomfort. Overall, pain perception tended to be higher with BFR (15.6%) as well as when inverted (26.5%). Increased pain perception with inversion would be related to the increased hydrostatic pressure around the head region (Pad- dock and Behm 2009). BFR would induce a hypoxic muscle environment resulting in a greater reliance on anaerobic metabolism, higher accumulation of metabolites and with the BFR, an inability to dispose these metabolites [10,52]. A compressed limb can lead to ischaemia accu- mulating metabolites and with increased local acidity would activate group III and IV pain afferents promoting a sensation of increased discomfort and pain [17,20,60]. p g p Inversion and BFR interactions. The data tended to indicate that BFR with inversion induced small (~5–12.5%) multiplicative effects upon voluntary and evoked muscle force and activation. A non-significant (p = 0.07) but large partial eta2 magnitude BFR x inversion inter- action showed that evoked twitch forces had a BFR-induced greater increase (5.6%) than with- out BFR in the inverted position. In contrast, MVC forces with BFR showed greater decreases (-9.1%) than without BFR for the inverted seated position. The M-wave amplitude was more impaired by BFR (-12.5%) than without BFR in the inverted position. Finally, pain perception increased with BFR (11.1%) versus without BFR with the inverted position. Thus, the combi- nation of inversion and BFR provided some amplification of the deficits compared to inver- sion alone (i.e., decreased elbow flexor resting twitch and MVC forces). The BFR-induced hypoxic environment attenuates blood substrates leading to a greater accumulation of meta- bolic by-products that could negatively impact force production. On the other hand, BFR may increase neuromuscular excitation (i.e., increased M-wave amplitudes). Blood Flow Restriction (BFR) Neural compression and blood flow impairment have elevated EMG with submaximal contractions [61,62]. Copithorne et al. [53] reported that blood flow occlusion led to a more rapid and greater increase in motoneuron excitability but had no effect on motor cortical excitability suggesting that group III/IV afferent feedback was the primary cause of the enhanced motoneuron excit- ability. Thus, BFR can induce both peripheral impairments and central excitation. These con- trasting effects in combination with inversion-induced (i.e., increased hydrostatic pressure) alterations provided small but not major amplification of the deficits. There were some minor limitations to the research. Due to the research procedure, data could only be obtained from the right limb. While the partial BFR was set relative to the posi- tion (upright/inverted), it could not exclusively indicate 50% BFR upright was exactly the same as inverted. Furthermore, female participants were not screened for birth control or menstrual cycle phase, which could have affected the performance of some of the women. As the present study recruited active adults who resistance trained on a consistent basis, we would hypothe- size that the observed impairments might be more substantial with an untrained population with less experience coping with partial or full occlusion (e.g., moderate to high intensity con- tractions while resistance training) and increases in hydrostatic pressures (e.g., Valsalva maneuvers while resistance training). Blood Flow Restriction (BFR) Voluntary and evoked contractile properties. Overall, BFR induced 4.8% lower MVC force values than without BFR as well as 9.1% lower MVC forces with BFR in an inverted posi- tion versus without BFR. This relative impairment is similar to the BFR-induced deficits for time to peak twitch. Without BFR, time to peak twitch increased 3.9% versus BFR when upright versus a 5.6% decrease with BFR vs. without BFR when inverted. Overall, BFR also impaired M-wave amplitudes by 21% (30.4% and 12.5% deficits for upright, and inverted seated positions respectively) compared to without BFR. These findings are similar to the force [10] and neuromuscular efficiency deficits reported by Hobbs and McCloskey [19] with graded ischaemia. Ischaemia and the associated pain can contribute to central nervous system inhibition [15,16] adversely affecting force output. Simi- lar to inversion, the occlusion of venous blood flow return would increase hydrostatic / perfu- sion pressure at the muscle resulting in similar force deficits seen in animals [6,9] and humans [10]. Hogan et al. [52] reported force decrements with diminished oxygen delivery to working muscle. Copithorne et al. [53] showed that BFR induced more significant decreases (~80%) in time to task failure during a sustained isometric fatiguing task, in comparison with normal blood flow condition. Furthermore, impairments in the evoked time to peak twitch and M- wave could be partly attributed to an alteration in muscle fibre conduction velocity [54] with nerve compression [55] as well as impaired sarcoplasmic reticulum Ca2+ uptake [56]. With median nerve compression in the carpal tunnel, Lunderborg et al. [57] observed a change in endoneural microcirculation. While it would be logical to suspect peripheral alterations with BFR, central responses can also be influenced. The greater BFR-induced decrease (p = 0.07, 16.4%) in biceps brachii EMG versus without BFR (7.9%) may reflect a greater hypoxia placing a greater emphasis on anaerobic metabolism, resulting in an earlier higher threshold motor unit recruitment accelerating the onset of force decrements [58,59]. Occluding arterial flow in the arm, during the post-exercise period, Big- land-Ritchie et al. [15] observed an impairment of neuromuscular transmission. Project administration: Hamid Ahmadi. Project administration: Hamid Ahmadi. Supervision: Duane C. Button, David G. Behm. Writing – original draft: Hamid Ahmadi. Writing – original draft: Hamid Ahmadi. Writing – review & editing: Nehara Herat, Shahab Alizadeh, Duane C. Button, Urs Grana- cher, David G. Behm. Author Contributions Conceptualization: Hamid Ahmadi, Urs Granacher, David G. Behm. Conceptualization: Hamid Ahmadi, Urs Granacher, David G. Behm. Data curation: Hamid Ahmadi, Nehara Herat, Shahab Alizadeh. Formal analysis: Hamid Ahmadi, Shahab Alizadeh, David G. Behm. Investigation: Hamid Ahmadi. Methodology: Hamid Ahmadi, Shahab Alizadeh, Duane C. Button, Urs Granacher, David G. Behm. Methodology: Hamid Ahmadi, Shahab Alizadeh, Duane C. Button, Urs Granacher, David G. Behm. Conclusions Hydrostatic pressure-induced peripheral muscle impairments with the inverted position, may have contributed to decreases in elbow flexor twitch, PTF, and MVC forces before the fatigue task. BFR did decrease voluntary force and M-wave amplitudes with only a small additional PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 15 / 19 PLOS ONE Inversion, blood flow, and neuromuscular performance increase to the inversion-induced impediments. A decrease in force or power output, with an increased perceived difficulty in stressful and life-threatening conditions such as accidents involving inversion or blood flow restriction (i.e., overturned vehicles, aerial maneuvers) can be life threatening. The results provide insights into the separate and combined effects of global (inversion) and local (BFR) increases in hydrostatic/perfusion pressure on voluntary and evoked contractile properties. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 References 1. Hearn J, Cahill F, Behm DG. An inverted seated posture decreases elbow flexion force and muscle acti- vation. Eur J Appl Physiol 2009; 106:139–147. https://doi.org/10.1007/s00421-009-0999-4 PMID: 19214555 2. Johar P, Grove V, DiSanto MC, Button DC, Behm DG. A rapid rotation to an inverted seated posture inhibits muscle force, activation, heart rate and blood pressure. Eur J Appl Physiol 2013; 113(8):2632– 9. https://doi.org/10.1007/s00421-013-2632-9 PMID: 23546453 3. Neary JP, Salmon DM, Dahlstrom BK, Casey EJ, Behm DG. Effect of an inverted seated position on single and sustained isometric contractions and cardiovascular parameters of trained individuals. Human Mov Sci 2015; 40:119–133. https://doi.org/10.1016/j.humov.2014.12.008 PMID: 25553559 4. Paddock N, Behm DG. The effect of an inverted body position on lower limb muscle force and activa- tion. Appl Physiol Nutr Metab 2009; 34:673–680. https://doi.org/10.1139/H09-056 PMID: 19767803 5. Smith DM, McAuliffe J, Johnson MJ, Button DC, Behm DG. Seated inversion adversely affects vigilance tasks and suppresses heart rate and blood pressure. Occup Ergo 2013; 11:153–163. https://doi.org/ 10.3233/OER-140215 6. Geeves MA, Ranatunga KW. Tension responses to increased hydrostatic pressure in glycerinated rab- bit psoas muscle fibres. Proc R Soc Lond B Biol Sci 1987; 232:217–226. https://doi.org/10.1098/rspb. 1987.0070 PMID: 2892206 7. Grossman Y, Kendig JJ. Evidence for reduced presynaptic Ca2+ entry in a lobster neuromuscular junc- tion at high pressure. J Physiol 1990; 420:355–364. https://doi.org/10.1113/jphysiol.1990.sp017917 PMID: 1969963 8. Heinemann SH, Stuhmer W, Conti F. Single acetylcholine receptor channel currents recorded at high hydrostatic pressure. Proc Natl Acad Sci USA 1987; 84(10):3229–33. https://doi.org/10.1073/pnas.84. 10.3229 PMID: 2437577 9. Ranatunga KW, Geeves MA. Changes produced by increased hydrostatic pressure in isometric con- traction of rat fast muscle. J Physiol 1991; 441:423–431. https://doi.org/10.1113/jphysiol.1991. sp018759 PMID: 1816380 16 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance 10. Sundberg CJ, Kaisjer L. Effects of graded restriction of perfusion on circulation and metabolism in the working leg; quantification of a human ischaemia-model. Acta Physiol Scand 1992; 146:1–9. https:// doi.org/10.1111/j.1748-1716.1992.tb09386.x PMID: 1442118 11. Sadamoto T, Bonde-Petersen F, Suzuki Y. Skeletal muscle tension, flow, pressure, and EMG during sustained isometric contraction in humans. Eur J Appl Physiol 1983; 51:395–408. https://doi.org/10. 1007/BF00429076 PMID: 6685038 12. Yasuda T, Brechue WF, Fujita T, Shirakawa J, Sato Y, Abe T. Muscle activation during low-intensity muscle contraction with restricted blood flow. J Sports Sci 2009; 27(5):479–489. https://doi.org/10. 1080/02640410802626567 PMID: 19253083 13. Wright JR, McCloskey DI, Fitzpatrick RC. References Effects of muscle perfusion pressure on fatigue and systemic arterial pressure in human subjects. J Appl Physiol 1999; 86:845–851. https://doi.org/10.1152/jappl. 1999.86.3.845 PMID: 10066695 14. Amann M, Calbet JAL. Convective oxygen transport and fatigue. J Appl Physiol 2008; 104:861–870. https://doi.org/10.1152/japplphysiol.01008.2007 PMID: 17962570 15. Bigland-Ritchie B, Dawson NJ, Johansson RS, Lippold OCJ. Reflex origin for the slowing of motoneu- ron firing rates in fatigue of human voluntary contraction. J Physiol 1986; 379:451–459. https://doi.org/ 10.1113/jphysiol.1986.sp016263 PMID: 3560001 16. Garland SJ. Role of small diameter afferents in reflex inhibition during human muscle fatigue J Physiol 1991; 435:547–558. https://doi.org/10.1113/jphysiol.1991.sp018524 PMID: 1770449 17. Leonard CT, Kane J, Perdaems J, Frank C, Graetzer DG, Moritani T. Neural modulation of muscle con- tractile properties during fatigue: afferent feedback dependence. Electroenceph Clin Neurophysiol 1994; 93:209–217. https://doi.org/10.1016/0168-5597(94)90042-6 PMID: 7515797 18. Fitzpatrick R, Taylor JL, McCloskey DI. Effect of arterial perfusion pressure on face production in work- ing human hand muscles. J Physiol 1996; 495(3):885–891. https://doi.org/10.1113/jphysiol.1996. sp021640 PMID: 8887790 19. Hobbs SF, Mccloskey DJ. Effects of blood pressure on force production in cat and human muscle. J Appl Physiol 1987; 63(2):834–839. https://doi.org/10.1152/jappl.1987.63.2.834 PMID: 3654443 20. Hollander DB, Reeves GV, Clavier JD, Francois MR, Thomas C, Kraemer RR. Partial occlusion during resistance exercise alters effort sense and pain. J Strength Cond Res 2010; 24(1):235–243. https://doi. org/10.1519/JSC.0b013e3181c7badf PMID: 19935100 21. Health Canada. The Canadian physical activity, fitness and lifestyle approach ( 3rd ed.). Canadian Soci- ety for Exercise Physiology, Health Canada Publishers, Ottawa, Ont. 2004. 22. Halperin I, Copithorne D, Behm GD. Unilateral isometric muscle fatigue decreases force production and activation of contralateral knee extensors but not elbow flexor. Appl Physiol Nutr Metab 2014; 39:1338– 44. https://doi.org/10.1139/apnm-2014-0109 PMID: 25291403 23. Behm DG, St-Pierre DMM, Perez D. Muscle inactivation: assessment of interpolated twitch technique. J Appl Physiol 1996; 81:2267–73. https://doi.org/10.1152/jappl.1996.81.5.2267 PMID: 8941554 24. Behm DG, St-Pierre DM. Effect of fatigue duration and muscle type on voluntary and evoked contractile properties. J Appl Physiol 1997a; 82:1654–61. https://doi.org/10.1152/jappl.1997.82.5.1654 PMID: 9134916 25. Behm DG, St-Pierre DMM. The Muscle Activation-Force Relationship is Unaffected by Ischaemic Recovery. Can J Appl Physiol 1997b; 22(5):468–478. 26. Merton PA. Voluntary strength and fatigue. J Physiol Lond 1954; 123:553–564. https://doi.org/10.1113/ jphysiol.1954.sp005070 PMID: 13152698 27. Behm DG. Force maintenance with submaximal fatiguing contraction. Can J Appl Physiol 2004; 29 (3):274–290. https://doi.org/10.1139/h04-019 PMID: 15199227 28. Behm DG, Button DC, Barbour G, Butt JC, Young WB. Conflicting effect of fatigue and potentiation on voluntary force. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 References J Strength Cond Res 2004; 18(2):365–372. https://doi.org/10.1519/R-12982.1 PMID: 15141999 29. Button DC, Behm DG. The effect of stimulus anticipation on the interpolated twitch technique. J Sports Sci Med 2008; 7(4):520–4. PMID: 24149960 30. Hermens HJ, Freriks B, Merletti R, Stegeman D, Blok J, Rau G, et al. European Recommendations for Surface Electromyography: Results of the SENIAM Project. Publisher: Roessingh Research and Development, Netherlands 1999; 24–83. 31. Loenneke JP, Thiebaud RS, Abe T, Bemben MG. Blood flow restriction recommendations: the horm- esis hypothesis. Med Hypotheses 2014; 82:623–626. https://doi.org/10.1016/j.mehy.2014.02.023 PMID: 24636784 17 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance 32. Takarada Y, Takazawa H, Sato Y, Takebayashi S, Tanaka Y, Ishii N. Effects of resistance exercise combined with moderate vascular occlusion on muscular function in human. J Appl Physiol 2000; 88:2097–2106. https://doi.org/10.1152/jappl.2000.88.6.2097 PMID: 10846023 33. Belcheva A, Shindova M. Pain Perception of Pediatric Patients during Cavity Preparation with ER: YAG Laser and Conventional Rotary Instruments. J of IMAB 2014; 20(5):634–637. https://doi.org/10.5272/ jimab.2014205.634 34. Perrine JJ, Edgerton VR. Muscle force-velocity and power-velocity relationships under isokinetic load- ing. Med Sci Sports 1978; 10:159–166. PMID: 723504 35. Agarwal GC, Gottlieb GL. An analysis of the electromyogram by fourier, simulation and experimental techniques. IEEE Trans Biomed En 1975; 22(3):225–9. https://doi.org/10.1109/tbme.1975.324486 PMID: 1116855 36. Daanen HAM, Mazure M, Holewijn M, Van der Velde EA. Reproducibility of the mean power frequency of the surface electromyogram. Eur J App Physiol 1990; 61:274–7. https://doi.org/10.1007/ BF00357612 PMID: 2282913 37. Dimitrova NA, Dimitrov GV. Interpretation of EMG changes with fatigue: facts, pitfalls, and fallacies. J Electromyogr Kinesiol 2003; 13(1):13–36. https://doi.org/10.1016/s1050-6411(02)00083-4 PMID: 12488084 38. Kwatny E, Thomas DH, Kwatny HG. An application of signal processing techniques to the study of myo- electric signals. IEEE Trans Biomed Eng 1970; 17(4):303–313. https://doi.org/10.1109/tbme.1970. 4502758 PMID: 5518826 39. Cohen J. Statistical Power Analysis for the Behavioral Sciences ( 2nd ed.). Hillsdale, NJ: Erlbaum Asso- ciates. 1998. 40. Carpenter MG, Frank JS, Silcher CP, Peysar GW. The influence of postural threat on the control of upright stance. Exp Brain Res 2001; 138(2):210–8. https://doi.org/10.1007/s002210100681 PMID: 11417462 41. Adkin AL, Frank JS, Carpenter MG, Peysar GW. Fear of falling modifies anticipatory postural control. Exp Brain Res 2002; 143(2):160–70. https://doi.org/10.1007/s00221-001-0974-8 PMID: 11880892 42. Arokoski JP, Valta T, Airaksinen O, Kankaanpaa M. Back and abdominal muscle function during stabili- zation exercises. Arch Phys Med Rehabil 2001; 82(8):1089–98. https://doi.org/10.1053/apmr.2001. 23819 PMID: 11494189 43. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 References Anderson KG, Behm DG. Maintenance of EMG activity and loss of force output with instability. J Strength Cond Res 2004; 18(3):637–40. https://doi.org/10.1519/1533-4287(2004)18<637: MOEAAL>2.0.CO;2 PMID: 15320684 44. Drinkwater EJ, Pritchett EJ, Behm DG. Effect of instability and resistance on unintentional squat-lifting kinetics. Int J Sports Physiol Perform 2007; 2(4):400–13. https://doi.org/10.1123/ijspp.2.4.400 PMID: 19171958 45. Behm DG, Anderson KG. The role of instability with resistance training. J Strength Cond Res 2006; 20 (3):716–22. https://doi.org/10.1519/R-18475.1 PMID: 16937988 46. Bosone D, Ozturk V, Roatta S, Cavallini A, Tosi P, Micieli G. Cerebral hemodynamic response to acute intracranial hypertension induced by head-down tilt. Funct Neurol 2004; 19(1):31–35. PMID: 15212114 47. Matsubara S, Sawa Y, Yokoji H, Takamori M. Shy-Drager syndrome. Effect of fludrocortisone and L- threo-3,4-dihydroxyphenylserine on the blood pressure and regional cerebral blood flow. J Neurol Neu- rosurg Psychiatry 1990; 53(11):994–7. https://doi.org/10.1136/jnnp.53.11.994 PMID: 2283531 48. Warkentin S, Passant U, Minthon L, Karlson S, Edvinsson L, Faldt R, et al. Redistribution of blood flow in the cerebral cortex of normal subjects during head-up postural change. Clin Auton Res 1992; 2 (2):119–24. https://doi.org/10.1007/BF01819667 PMID: 1638106 49. Cooke WH, Carter JR, Kuusela TA. Human cerebrovascular and autonomic rhythms during vestibular activation. Am J Physiol Regul Integr Comp Physiol 2004; 286:R838–R843. https://doi.org/10.1152/ ajpregu.00562.2003 PMID: 14715492 50. Cooke WH, Dowlyn MM. Power spectral analysis imperfectly informs changes in sympathetic traffic dur- ing acute simulated microgravity. Aviat Space Environ Med 2000; 71:1232–1238. PMID: 11439723 51. Fu Q, Sugiyama Y, Kamiya A, Mano T. A comparison of autonomic responses in humans induced by two simulation models of weightlessness: lower body positive pressure and 6˚ head-down tilt. J Auton Nerve Syst 2000; 80:101–107. https://doi.org/10.1016/s0165-1838(00)00081-3 PMID: 10742547 52. Hogan MC, Richardson RS, Kurdak SS. Initial fall in skeletal muscle force development during ischemia is related to oxygen availability. J Appl Physiol 1994; 77(5):2380–4. https://doi.org/10.1152/jappl.1994. 77.5.2380 PMID: 7868458 18 / 19 PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 PLOS ONE Inversion, blood flow, and neuromuscular performance 53. Copithorne DB, Rice CL, McNeil CJ. Effect of blood flow occlusion on corticospinal excitability during sustained low-intensity isometric elbow flexion. J Neurophysiol 2020; 123(3):1113–9. https://doi.org/ 10.1152/jn.00644.2019 PMID: 31995434 54. Farina D, Arendt-Nielsen L, Merletti R, Graven-Nielsen T. Effect of experimental muscle pain on motor unit firing rate and conduction velocity. J Neurophysiol 2004; 91(3):1250–9. https://doi.org/10.1152/jn. 00620.2003 PMID: 14614105 55. Ogata K, Shimon S, Owen J, Manske PR. Effect of compression and devascularisation on ulnar nerve function. PLOS ONE | https://doi.org/10.1371/journal.pone.0245311 May 19, 2021 References J Hand Surg Br 1991; 16(1):104–8. https://doi.org/10.1016/0266-7681(91)90143-c PMID: 2007800 56. Place N, Bruton JD, Westerblad H. Mechanisms of fatigue induced by isometric contractions in exercis- ing humans and in mouse isolated single muscle fibres. Clin Exp Pharmacol Physiol 2009; (3):334–9. https://doi.org/10.1111/j.1440-1681.2008.05021.x PMID: 18671711 57. Lunderborg G, Gelberman RH, Minteer-Convery M, Lee YF, Hargens AR. Median nerve compression in the carpal tunnel-functional response to experimentally induced controlled pressure. J Hand Surg Am 1982; 7(3):252–9. https://doi.org/10.1016/s0363-5023(82)80175-5 PMID: 7086092 58. Gerdle B, Fugl-Meyer AR. Is the mean power frequency shift of the EMG a selective indicator of fatigue of the fast twitch motor units? Acta Physiol Scand 1992; 145(2):129–38. https://doi.org/10.1111/j.1748- 1716.1992.tb09348.x PMID: 1636442 59. Moore DR, Burgomaster KA, Schofield LM, Gibala MJ, Sale DG, Phillips SM. Neuromuscular adapta- tions in human muscle following low intensity resistance training with vascular occlusion. Eur J Appl Physiol 2004; 92(4–5):399–406. https://doi.org/10.1007/s00421-004-1072-y PMID: 15205956 60. Marchettini P, Simone DA, Caputi G, Ochoa JL. Pain from excitation of identified muscle nociceptors in humans. Brain Res 1996; 740(1–2):109–16. https://doi.org/10.1016/s0006-8993(96)00851-7 PMID: 8973804 61. Karabulut M, Cramer JT, Abe T, Sato Y, Bemben MG. Neuromuscular fatigue following low-intensity dynamic exercise with externally applied vascular restriction. J Electromyog Kinesio 2010; 20:440–7. https://doi.org/10.1016/j.jelekin.2009.06.005 PMID: 19640732 62. Yasuda T, Fujita T, Miyagi Y, Kubota Y, Sato Y, Nakajima T, Bemben MG, Abe T. Electromyographic responses of arm and chest muscle during bench press exercise with and without KAATSU. Int J Kaatsu Res 2006; 2:15–18. https://doi.org/10.3806/ijktr.2.15 19 / 19

https://openalex.org/W2754849005

https://europepmc.org/articles/pmc5618646?pdf=render

English

Anti-NKG2D mAb: A New Treatment for Crohn’s Disease?

International journal of molecular sciences

cc-by

Review Anti-NKG2D mAb: A New Treatment for Crohn’s Disease? Kasper Vadstrup 1,2 ID and Flemming Bendtsen 1,* Kasper Vadstrup 1,2 ID and Flemming Bendtsen 1,* 1 Gastro unit, Medical Division, Hvidovre University Hospital, Kettegård Alle 30, DK-2650 Hvidovre, Denmark; [email protected] 2 Medical Affairs, Janssen Immunology, Janssen-Cilag A/S, Bregnerødvej 133, DK-3460 Birkerød, Denmark * Correspondence: fl[email protected]; Tel.: +45-3862-3273 1 Gastro unit, Medical Division, Hvidovre University Hospital, Kettegård Alle 30, DK-2650 Hvidovre, Denmark; [email protected] 2 Medical Affairs, Janssen Immunology, Janssen-Cilag A/S, Bregnerødvej 133, DK-3460 Birkerød, Denmark * Correspondence: fl[email protected]; Tel.: +45-3862-3273 2 Medical Affairs, Janssen Immunology, Janssen-Cilag A/S, Bregnerødvej 133, DK-3460 Birkerød, Denm * Correspondence: fl[email protected]; Tel.: +45-3862-3273 Received: 13 July 2017; Accepted: 11 September 2017; Published: 16 September 2017 Abstract: Crohn’s disease (CD) and ulcerative colitis (UC) are immunologically-mediated, debilitating conditions resulting from destructive inflammation of the gastrointestinal tract. The pathogenesis of IBD is incompletely understood, but is considered to be the result of an abnormal immune response with a wide range of cell types and proteins involved. Natural Killer Group 2D (NKG2D) is an activating receptor constitutively expressed on human Natural Killer (NK), γδ T, mucosal-associated invariant T (MAIT), CD56+ T, and CD8+ T cells. Activation of NKG2D triggers cellular proliferation, cytokine production, and target cell killing. Research into the NKG2D mechanism of action has primarily been focused on cancer and viral infections where cytotoxicity evasion is a concern. In human inflammatory bowel disease (IBD) this system is less characterized, but the ligands have been shown to be highly expressed during intestinal inflammation and the following receptor activation may contribute to tissue degeneration. A recent phase II clinical trial showed that an antibody against NKG2D induced clinical remission of CD in some patients, suggesting NKG2D and its ligands to be of importance in the pathogenesis of CD. This review will describe the receptor and its ligands in intestinal tissues and the clinical potential of blocking NKG2D in Crohn’s disease. Keywords: Crohn’s Disease; IBD; NKG2D; MICA; MICB; ULBP Int. J. Mol. Sci. 2017, 18, 1997; doi:10.3390/ijms18091997 www.mdpi.com/journal/ijms International Journal of Molecular Sciences International Journal of Molecular Sciences Keywords: Crohn’s Disease; IBD; NKG2D; MICA; MICB; ULBP 2. The Natural Killer Group 2D (NKG2D) Receptor NKG2D is an activating receptor expressed on human NK, γδ T, mucosal-associated invariant T (MAIT), CD56+ T, and CD8+ T cells [8,15], which can participate in the recognition of inducible “stressed-self” ligands on the surface of target cells. This will include cells acting as intra-epithelial T cells in the intestinal wall [16]. Activation of NKG2D triggers cellular proliferation, cytokine production, and target cell killing [17]. Under normal conditions in the disease-free human system, the receptor is constitutively expressed on all the lymphocytes listed above in both peripheral blood and the intestinal tissues, but also to a low degree on a subset of CD4+ T cells [18]. NKG2D, or CD314, is encoded by the gene KLRK1 (killer cell lectin-like receptor of the subfamily K member 1) and is found on human chromosome 12. The gene sits close to KLRD1 (CD94) and the cluster of KLRC1 (NKG2A), KLRC2 (NKG2C), KLRC3 (NKG2E), and KLRC4 (NKG2F), which are both activating and inhibitory receptors binding to distinct sets of HLA-E residues [19,20]. With just two alleles different by one amino acid, the KLRK1 gene has limited polymorphism and only one isoform encoding a functional protein has been described in humans. All mammals have KLRK1 orthologs, indicating that the gene is highly conserved during evolution and is an important function across species [21]. NKG2D is a type II membrane protein comprising 216 amino acids with a predicted molecular weight of 25,143 kDa. The protein has an N-terminal intracellular region, a transmembrane domain, a membrane-proximal stalk region, and an extracellular region with a single C-type lectin-like domain. KLRK1 is expressed on the cell surface as a disulfide-bonded homodimer with a molecular weight of approximately 42 kDa when analyzed under reducing conditions and approximately 80 kDa under non-reducing conditions. A cysteine residue just outside the transmembrane region forms the disulfide bond joining the two subunits of the homodimer (Figure 1) [22]. In the adaptive immune system, T cells are mostly dependent on the T cell receptor complex as a result of somatic recombination of genes to recognize and respond to antigens. For NK cells in the innate system, the activation or inhibition of cell signals relies on a wide range of surface receptors where some are shared with subsets of T cells. The NKG2 receptor family are shared between cell types underlining a central role in tumor and pathogen recognition [23]. 1. Introduction Crohn’s disease (CD), an inflammatory bowel disease (IBD), is a complex immunologically- mediated chronic illness that is believed to arise due to a dysregulated immune response to microbiota in the gastrointestinal system. CD is characterized by patchy and transmural inflammation that can occur throughout the gastrointestinal tract, with alternating phases of clinical relapse and remission [1]. CD typically arises between 20 and 30 years of age with symptoms such as abdominal pain, fever, diarrhea with or without bleeding, and weight loss [2]. Despite newer and better therapies, Crohn’s disease often presents a heavy everyday burden, sometimes leading to surgery and disability [3,4]. The biologic antibodies in clinical development adopt parallel mechanisms of action with the same or other targets, namely blocking of a variety of cytokines, inhibition of the similar protein-dependent migration mechanism or neutralization of chemokines. However, other mechanisms may be involved in the pathogenesis of IBD [5,6]. The immune activating receptor Natural Killer Group 2D (NKG2D) has been implicated in the pathogenesis of IBD through its presence on intestinal cytotoxic lymphocytes and the increased expression of activating ligands on inflamed tissue [7]. Normally, the NKG2D receptor and its eight different ligands are a part of a system designed to participate in the recognition of “stressed” cells exposed to viruses or tumor growth. NKG2D is then activated on natural killer (NK) or CD8 T cells to eliminate the ligand-bearing cell as an effective defense mechanism as well as producing pro-inflammatory cytokines. For years, this pathway has been a potentially attractive approach to Int. J. Mol. Sci. 2017, 18, 1997; doi:10.3390/ijms18091997 www.mdpi.com/journal/ijms www.mdpi.com/journal/ijms 2 of 18 Int. J. Mol. Sci. 2017, 18, 1997 counter malignancies and infectious diseases by activating the system [8]. In CD, however, blocking the receptor with a biologic antibody might be a viable treatment option for lowering intestinal cell elimination and the inflammatory environment. NKG2D has been implicated in some [9,10], but not all, mouse models of colitis [11]. In human IBD, the expression and function of NKG2D have not been fully characterized [12], but a recent phase II clinical trial showed that a blocking antibody against NKG2D induced rapid clinical remission of CD in some patients, implicating NKG2D and its ligands to be of importance in the pathogenesis of CD [13]. 1. Introduction There has been limited descriptions or a lack of focus in recent reviews of current and future treatments for IBD on this target among the many in clinical pipelines [14]. This review aims to give an overview of the expression and functional role of the NKG2D pathway in the healthy and inflamed intestine, as well as the clinical potential of a blocking antibody against the receptor in CD. 2. The Natural Killer Group 2D (NKG2D) Receptor J , , Figure 1. Protein structures of Natural Killer Group 2D (NKG2D) and its ligands. NKG2D Fi 1 P t i t t f N t l Kill G 2D (NKG2D) d it li d NKG2D Figure 1. Protein structures of Natural Killer Group 2D (NKG2D) and its ligands. NKG2D is a disulfide-linked homodimer, transmembrane protein that can bind to two adapter molecules, DAP10 (DNAX-activating protein of 10 kDa), also a homodimer. NKG2D ligands are eight different MHC I- like molecules. Four of these (MHC class I polypeptide-related sequence (MIC) A/B and UL16 binding protein (ULBP) 4/5) are bound to the cell membrane by transmembrane (TM) domains with a cytoplasmic tail (CYT), while the other four (ULBP1-3 and 6) are glycophosphatidylinositol (GPI)- anchored. Each has two or three α domains. Figure 1. Protein structures of Natural Killer Group 2D (NKG2D) and its ligands. NKG2D is a disulfide-linked homodimer, transmembrane protein that can bind to two adapter molecules, DAP10 (DNAX-activating protein of 10 kDa), also a homodimer. NKG2D ligands are eight different MHC I-like molecules. Four of these (MHC class I polypeptide-related sequence (MIC) A/B and UL16 binding protein (ULBP) 4/5) are bound to the cell membrane by transmembrane (TM) domains with a cytoplasmic tail (CYT), while the other four (ULBP1-3 and 6) are glycophosphatidylinositol (GPI)-anchored. Each has two or three α domains. 2. The Natural Killer Group 2D (NKG2D) Receptor The frequency of NKG2D+ expression is consistently high on NK and CD8+ T cells and consistently low on CD4+ T cells in humans in the steady state, with other subsets showing differentiating expression [7,24]. While the other members of NKG2 family form heterodimers with CD94, the NKG2D receptor forms stable homodimers on the surface when stabilizing non-covalently with its adaptor molecule DAP10 (DNAX-activating protein of 10 kDa) on the inside of the cell membrane. DAP10 is also a disulfide-linked homodimer [18,25] (Figure 1). NKG2D associates with two DAP10 proteins in the transmembrane region via charged residues within the receptor and its adapter subunits and creates a hexameric complex. A charged amino acid residue (aspartic acid) centrally located within the transmembrane region of DAP10, forms a salt bridge with a charged amino acid residue (arginine) in the transmembrane region of NKG2D to stabilize the receptor complex. Upon interaction with its ligands, the complex transmits an activation signal triggering 3 of 18 Int. J. Mol. Sci. 2017, 18, 1997 lymphocyte cytotoxic granule polarization and degranulation, cytokine production, proliferation, and survival [22]. Int. J. Mol. Sci. 2017, 18, 1997 3 of 17 lymphocyte cytotoxic granule polarization and degranulation, cytokine production, proliferation, and survival [22]. Int. J. Mol. Sci. 2017, 18, 1997 3 of 17 J , , Figure 1. Protein structures of Natural Killer Group 2D (NKG2D) and its ligands. NKG2D is a disulfide-linked homodimer, transmembrane protein that can bind to two adapter molecules, DAP10 (DNAX-activating protein of 10 kDa), also a homodimer. NKG2D ligands are eight different MHC I- like molecules. Four of these (MHC class I polypeptide-related sequence (MIC) A/B and UL16 binding protein (ULBP) 4/5) are bound to the cell membrane by transmembrane (TM) domains with a cytoplasmic tail (CYT), while the other four (ULBP1-3 and 6) are glycophosphatidylinositol (GPI)- anchored. Each has two or three α domains. Figure 1. Protein structures of Natural Killer Group 2D (NKG2D) and its ligands. NKG2D is a disulfide-linked homodimer, transmembrane protein that can bind to two adapter molecules, DAP10 (DNAX-activating protein of 10 kDa), also a homodimer. NKG2D ligands are eight different MHC I-like molecules. Four of these (MHC class I polypeptide-related sequence (MIC) A/B and UL16 binding protein (ULBP) 4/5) are bound to the cell membrane by transmembrane (TM) domains with a cytoplasmic tail (CYT), while the other four (ULBP1-3 and 6) are glycophosphatidylinositol (GPI)-anchored. Each has two or three α domains. 3. The NKG2D Ligands 3. The NKG2D Ligands Eight human ligands for NKG2D have been identified: Human Major Histocompatibility Complex (MHC) class I polypeptide-related sequence A (MICA), MICB, and UL16 binding protein (ULBP) 1–6 (also named RAET1E, RAET1G, RAET1H, RAET1I, RAET1L, and RAET1N). All are located on chromosome 6 within the Major Histocompatibility Complex and have various binding affinities for NKG2D ranging from 10−6 to 10−9 M, with different homology between ligands down to Eight human ligands for NKG2D have been identified: Human Major Histocompatibility Complex (MHC) class I polypeptide-related sequence A (MICA), MICB, and UL16 binding protein (ULBP) 1–6 (also named RAET1E, RAET1G, RAET1H, RAET1I, RAET1L, and RAET1N). All are located on chromosome 6 within the Major Histocompatibility Complex and have various binding affinities 4 of 18 Int. J. Mol. Sci. 2017, 18, 1997 for NKG2D ranging from 10−6 to 10−9 M, with different homology between ligands down to 25%, and with extensive allelic polymorphisms [26,27]. MICA encodes 79 protein variants and MICB encodes 26 protein variants, while the ULBPs exhibit a lower level of polymorphism [28,29]. MICA, MICB and ULBP4 and 5 share a structure of having a transmembrane anchor in the cell and an intracellular cytoplasmic tail, while ULBP1-3 and 6 have a glycophosphatidylinositol (GPI)-anchored organization. MICA and MICB carry three extracellular immunoglobulin-like domains (α1, α2, α3,) and ULBP1–6 carry two domains (α1, α2). However, unlike class I, the ligands are not covalently bound to a monomorphic β2 microglobulin, and its peptide binding groove is empty and does not present peptides [30] (Figure 1). ULBP2 has the unique feature to be expressed at the cell surface either as a transmembrane or a GPI-anchored protein [31]. Allelic variants of the NKG2D ligands have been reported to result in large differences in the affinity of binding to NKG2D. For instance, MICA alleles with a single amino acid substitution at position 129 in the α2 domain of methionine (M) or valine (V) have been classified as having strong or weak binding affinity for NKG2D, respectively. These variable affinities have been suggested to affect thresholds of NK cell triggering and T cell modulation and consequently influencing clinical phenotypes in autoimmune disorders and malignancies [32–34]. q y g p yp g NKG2D-ligand complex crystal structures have been described for some of the ligands and the binding of the different proteins seem to be adaptive fit mechanisms in the receptor [35]. 3. The NKG2D Ligands 3. The NKG2D Ligands Although transcripts are present in some healthy cell types [36], the ligand proteins are rarely present on the cell surface of healthy cells and tissues, but are inducible by virus infection, tumorigenesis, or by stimuli, such as DNA damage, oxidative stress, heat shock, toll-like receptor signaling, or cytokine exposure [16,37–39]. While not present under normal conditions, it seems that every cell type has the ability to express one or several of the ligands under one of the conditions listed above, or by the appropriate stimuli, which can be induced with pro-inflammatory signals. The induction of NKG2D ligand expression is attributed to cellular ”stress” and every type of cancer has the same mechanism. Accelerated proliferation other than cancer development can also result in upregulated ligand expression, for instance healthy cells in embryonic tissues and tissues undergoing wound repair [40–42]. Increased MICA expression has been reported in tissues with inflammation and autoimmune diseases such as type 1 diabetes, celiac disease, rheumatoid arthritis [43–45], in atherosclerosis, where it is found on vascular endothelial cells [45–47], and in asthma [11]. MICB has been located on many of the same sites, but the ULBPs are generally underreported in these diseases. The regulation of NKG2D ligand transcripts is a complex process including epigenetics, RNA degradation, microRNA interference, and regulation of the protein is subject to cleavage from the surface by membrane metalloproteases or intracellular degradation by ubiquitin ligases [38]. This regulation is also different between cell types and maturation statuses. 4. Immunological Pathway of NKG2D NKG2D serves as a sensor for recognition of “induced self” for the detection and removal of hyper-proliferative cells, transformed cells, or cells infected by pathogens. As with the ligands, signaling and regulation by the NKG2D receptor in NK cells and T cells is complex and incompletely understood. While NKG2D is expressed constitutively on essentially all resting human NK cells and CD8+ T cells [48], engagement of NKG2D alone is not sufficient to trigger cell-mediated cytotoxicity or cytokine production [49,50]. The simultaneous engagement of NKG2D and other costimulatory receptors, such as CD335 (NKp46) or CD244 can, under the right circumstances, result in cytolytic activity in resting NK cells [49]. NK cells can also be primed by culture in IL-2 or IL-15—resembling some inflammatory sites—and engagement of NKG2D alone will then be sufficient to initiate degranulation and cytokine production. The result can be increased killing of the ligand-expressing cells and/or production of cytokines, such as IFN-γ, TNF-α, and GM-CSF enhancing the inflammatory environment [51–53]. As described, each NKG2D homodimer associates with two DAP10 homodimers to form a hexameric receptor complex (Figure 1) [25]. The related DAP12 adaptor protein was previously 5 of 18 Int. J. Mol. Sci. 2017, 18, 1997 described as another transducer in this mechanism, but the protein, which signals through an ITAM region, does not form stable complexes with human NKG2D, unlike in mouse, thus identifying a species difference, and was reviewed in [54]. The adaptor molecule of NKG2D, DAP10 has a YINM motif, which recruits a p85 PI3 kinase and Vav-1 signaling complex [22,55]. Once DAP10 is activated, and when the complex is binding to a ligand, the stimulation leads to a PI3K-dependent Akt phosphorylation (alternatively through Vav1), which presumably activates cell survival pathways, NKG2D-mediated calcium release, cytokine production, and cytotoxicity [17]. IL-15 seems to be an essential cytokine for the phosphorylation of the YINM motif on DAP10 leading to the activation of PI3K or Vav-1 [23]. The mechanisms can be distinct, with both PI3K and Vav-1 recruiting Grb2 for downstream signaling, but Vav-1 uses the PLCγ2→Ca2+ or SLP-76 pathway for granule release and cytotoxicity, while PI3K signals through MEK→ERK or directly via Ca2+ increases. DAP10 phosphorylation can also result in cytotoxicity through JNK→cJunN. If PI3K signals Akt instead, the result seems to be increased survival of the lymphocyte. The release of pro-inflammatory cytokines is, instead, believed to happen in the DAP10→JAK2→STAT5 pathway [17] (Figure 2). Int. J. 4. Immunological Pathway of NKG2D Mol. Sci. 2017, 18, 1997 5 of 17 NKG2D-mediated calcium release, cytokine production, and cytotoxicity [17]. IL-15 seems to be an essential cytokine for the phosphorylation of the YINM motif on DAP10 leading to the activation of PI3K or Vav-1 [23]. The mechanisms can be distinct, with both PI3K and Vav-1 recruiting Grb2 for downstream signaling, but Vav-1 uses the PLCγ2→Ca2+ or SLP-76 pathway for granule release and cytotoxicity, while PI3K signals through MEK→ERK or directly via Ca2+ increases. DAP10 phosphorylation can also result in cytotoxicity through JNK→cJunN. If PI3K signals Akt instead, the result seems to be increased survival of the lymphocyte. The release of pro-inflammatory cytokines is, instead, believed to happen in the DAP10→JAK2→STAT5 pathway [17] (Figure 2). Figure 2. Diagram of the signaling pathways of the activated NKG2D receptor complex through DAP10. Cytokines are presented in diamond shapes, signaling proteins in ovals, ions in hexagons and cellular effects in rectangles. DAP10 (DNAX-activating protein of 10 kDa); IL-15 (interleukin 15); Grb2 (Growth factor receptor-bound protein 2); P85 PI3K (P85 subunit of PhosphoInositide 3-Kinase); Vav- 1 (vav guanine nucleotide exchange factor 1); MEK (Mitogen-activated protein kinase kinase); ERK (Extracellular signal-Regulated Kinase); JNK (c-Jun N-terminal protein Kinases); cJunN (c-Jun N- terminal kinase-1); SLP-76 (SH2 domain-containing leukocyte phosphoprotein of 76kDa); CA++ (Ca2+ ion); PLCγ2 (Phospholipase C γ-2); Akt (Protein Kinase B); JAK2 (Janus kinase 2); STAT5 (Signal Transducer and Activator of Transcription 5). In addition to the function of being a potent activating receptor, NKG2D is reportedly involved Figure 2. Diagram of the signaling pathways of the activated NKG2D receptor complex through DAP10. Cytokines are presented in diamond shapes, signaling proteins in ovals, ions in hexagons and cellular effects in rectangles. DAP10 (DNAX-activating protein of 10 kDa); IL-15 (interleukin 15); Grb2 (Growth factor receptor-bound protein 2); P85 PI3K (P85 subunit of PhosphoInositide 3-Kinase); Vav-1 (vav guanine nucleotide exchange factor 1); MEK (Mitogen-activated protein kinase kinase); ERK (Extracellular signal-Regulated Kinase); JNK (c-Jun N-terminal protein Kinases); cJunN (c-Jun N-terminal kinase-1); SLP-76 (SH2 domain-containing leukocyte phosphoprotein of 76kDa); CA++ (Ca2+ ion); PLCγ2 (Phospholipase C γ-2); Akt (Protein Kinase B); JAK2 (Janus kinase 2); STAT5 (Signal Transducer and Activator of Transcription 5). Figure 2. Diagram of the signaling pathways of the activated NKG2D receptor complex through DAP10. Cytokines are presented in diamond shapes, signaling proteins in ovals, ions in hexagons and cellular effects in rectangles. 4. Immunological Pathway of NKG2D DAP10 (DNAX-activating protein of 10 kDa); IL-15 (interleukin 15); Grb2 (Growth factor receptor-bound protein 2); P85 PI3K (P85 subunit of PhosphoInositide 3-Kinase); Vav- 1 (vav guanine nucleotide exchange factor 1); MEK (Mitogen-activated protein kinase kinase); ERK (Extracellular signal-Regulated Kinase); JNK (c-Jun N-terminal protein Kinases); cJunN (c-Jun N- terminal kinase-1); SLP-76 (SH2 domain-containing leukocyte phosphoprotein of 76kDa); CA++ (Ca2+ ion); PLCγ2 (Phospholipase C γ-2); Akt (Protein Kinase B); JAK2 (Janus kinase 2); STAT5 (Signal Transducer and Activator of Transcription 5). In addition to the function of being a potent activating receptor, NKG2D is reportedly involved Figure 2. Diagram of the signaling pathways of the activated NKG2D receptor complex through DAP10. Cytokines are presented in diamond shapes, signaling proteins in ovals, ions in hexagons and cellular effects in rectangles. DAP10 (DNAX-activating protein of 10 kDa); IL-15 (interleukin 15); Grb2 (Growth factor receptor-bound protein 2); P85 PI3K (P85 subunit of PhosphoInositide 3-Kinase); Vav-1 (vav guanine nucleotide exchange factor 1); MEK (Mitogen-activated protein kinase kinase); ERK (Extracellular signal-Regulated Kinase); JNK (c-Jun N-terminal protein Kinases); cJunN (c-Jun N-terminal kinase-1); SLP-76 (SH2 domain-containing leukocyte phosphoprotein of 76kDa); CA++ (Ca2+ ion); PLCγ2 (Phospholipase C γ-2); Akt (Protein Kinase B); JAK2 (Janus kinase 2); STAT5 (Signal Transducer and Activator of Transcription 5). Int. J. Mol. Sci. 2017, 18, 1997 6 of 18 In addition to the function of being a potent activating receptor, NKG2D is reportedly involved in the normal development of NK and T cell function, and even in B cells maturation from hemapoietic precursors. These potential regulatory roles are reviewed elsewhere, but should be observed through the development of the blocking antibody against NKG2D [56]. 5. Cancer Evasion and Infection Control Most knowledge about NKG2D and its ligands comes from cancer and infectious disease research. NKG2D ligands are induced in cells infected with intracellular bacteria and viruses and can, for instance, be induced on dendritic cells responding to microbial pathogens [38]. Pathogens can seemingly induce expression of any of the ligands in a range of different cell types without general rules and it might be dependent on other external factors as well. However, viruses have evolved complicated mechanisms to prevent a given cell from expressing ligands or to truncate the protein to make it soluble or captured intracellularly, underlining the importance of the NKG2D receptor activation. Soluble ligand protein can then antagonize NKG2D to avoid detection by T or NK cells [57,58]. The diversity of NKG2D ligand genes and their polymorphism is likely driven by the pathogens’ evolving mechanisms to escape detection by immune cells through NKG2D pathway. NKG2D ligands are expressed on all cell types by cancers [59]. In some cases, ligands may be induced due to the genomic instability of the transformed cells, resulting in activation of DNA repair pathways [60], but the factors causing NKG2D ligand expression depend on the transformed cells, as well as exposure to cytokines and other factors in the microenvironment [38]. Hyper-proliferation of transformed cells might be another reason for induction caused by activation of transcription factors [42]. Like viruses, primary tumors frequently develop mechanisms for avoiding detection and elimination by lymphocytes. These include systemic release of soluble NKG2D ligands [61,62]. Additionally, when NK or T cells encounter cells bearing NKG2D ligands, a mechanism down-modulates the receptor [63,64]. In addition, tumor-derived factors, such as anti-inflammatory TGF-β, can cause down-regulation of the NKG2D receptor on NK cells and T cells [65–67]. These pathways are further reviewed in [59]. 6. NKG2D in Crohn’s Disease NKG2D on lymphocytes and its ligands MICA, MICB, and ULBP1-6 modulate T and NK cell activity and may contribute to IBD pathogenesis. However, only a few studies have addressed the relation of NKG2D to IBD and especially the ligands are poorly characterized [7,12,68–72]. e o o N G o espec y e g s e poo y c c e e [ , ,68 ] NKG2D is expressed to a high degree on NK and CD8+ T cells in CD also, but can also be detected on CD4+ T cells in inflamed CD intestine [7,73]. The expression level on CD4+ T cells have been shown to increase with Crohn’s disease in the lamina propria compared with controls and patients with ulcerative colitis. CD4+NKG2D+ T cells with a Th1 cytokine profile and expressing perforin were increased in the periphery and in the mucosa in CD. CD4+NKG2D+ T cell clones were functionally active through MICA-NKG2D interactions, producing interferon-γ and killing targets expressing MICA [7]. This defines a subset of CD4+ T cells shown to have cytotoxic and inflammatory properties. Other T cell types can show heterogeneous results between patients. The differences in the NKG2D expression with disease stage might be caused by ligand-induced internalization or due to differences in cytokines in the mucosal gut tissue, as several cytokines are known to either increase [74–77] or decrease [65–67,78,79] NKG2D on NK and T cell subsets. CD56+ αβ T and γδ T cells have previously been found to express NKG2D differently [80] and the balance between these two cell types might be a modulating factor between self-tolerance and autoimmunity [81]. Our group has found that NKG2D expression on γδ T cells (CD45+CD3+γδTCR+) correlate negatively with C-reactive Protein (CRP), while CD56+ T cells (CD45+CD3+γδTCR−CD56+) correlate positively. This might be connected to the finding that NKG2D expression on γδ T cells also correlates negatively to the cells production of GM-CSF, but positively to the release of IL-10, both measured on the protein level. The expression level of NKG2D on CD56+ T cells correlate to the two cytokines in the complete opposite way, which implies 7 of 18 Int. J. Mol. Sci. 2017, 18, 1997 that pro- and anti-inflammatory cytokines influence the NKG2D molecular pathway [73]. Notably, γδ T cells constitute ~40% of the intraepithelial lymphocytes (IEL) [82,83], and might have a protective role in IBD [84]. 6.1. Upregulated Ligands in Crohn’s Disease Weak MICA expression has been reported on the cell surface of some healthy cell types including epithelial cells in the gut [90,91]. Increased MICA expression has also been reported in autoimmune diseases, such as type 1 diabetes, celiac disease, rheumatoid arthritis, and atherosclerosis, where it is found on vascular endothelial cells [43–47]. Additionally, associations between MICA alleles and thyroid disease and Addison’s disease have been observed, pointing to these factors of innate immunity contributing to the pathogenesis of autoimmune disorders [92,93]. Increased levels of MICA and/or MICB have also been observed on epithelial cells and monocytes in CD patients, where they may trigger cytokine release and cytolytic activity [7,12,68,94]. Allez et al. described how increased MICA expression could prime NKG2D+ CD4+ T cells to kill the ligand bearing cells while also producing high levels of IFN-γ and IL-15 [7]. Analyses of ULBP1-6 expression in human CD are incomplete, but ULBP1 and ULBP2 expression have been reported on intestinal monocytes from pediatric IBD patients. This study identified an increase of MICA/B+, ULBP1+, and ULBP2+ cells from mucosal infiltrates in tissue sections from active disease only, not when disease was in remission or in normal controls [12]. The fact that MICA is typically more expressed than MICB could be due to gene promotor polymorphisms [95], and whether this is also the case for the ULBPs needs to be addressed. mRNA differences for MICA in CD patients compared to normal controls have previously been reported [96]. Additionally, several reports have linked polymorphisms in MICA and MICB to increased prevalence of IBD [70,97,98], and other polymorphisms in MICA that, on the contrary, protects from ulcerative colitis, have been found [69]. We have recently shown the presence of all eight ligands on intestinal monocytes and B cells in CD patients by flow cytometry, as well as on endothelial and epithelial cells [73]. To be able to show this, we had to produce antibodies for ULBP3–6, which could explain the few studies on these ligands. The level of expression was highly heterogeneous and differed widely between cell types. No difference in the average expression levels was detected between MICA and MICB, and the ULBPs, but at the mRNA level, the ULBPs were generally more produced. The presence on epithelial cells could be part of the pathogenesis of CD, if lymphocytes eliminate the gut mucosal wall because of ligand presence in the inflammatory microenvironment. 6. NKG2D in Crohn’s Disease CD56+ αβ T cells are activated T cells and may participate in the inflammatory response. Unlike conventional naïve T cells, IELs do not need priming and they will immediately release cytokines or mediate killing of infected target cells. CD8+ T and γδ T cells are most common IELs [85]. These activated T cell subsets or all available activated NK cells can release interferon-gamma (IFN-γ) and use cytotoxic killing without prior cell activation, but just by stimulation of FcRs or NK receptors such as CD94 or NKG2D by stressed or infected cells. Inhibitory lymphocyte receptors, like the heavily-studied PD-1, will ensure that self will be preserved under immunosurveillance by these innate subsets and non-self without ligands for inhibitory receptors will be killed if activated through NKG2D [86]. There could be a tendency toward the down-regulation of the NKG2D receptor on lymphocytes in inflamed CD tissue compared to non-inflamed CD tissue and normal controls. Down-regulation of NKG2D protein is not due to a decrease in NKG2D mRNA levels in CD patients versus normal controls. This lower NKG2D expression in inflamed tissue likely results from either the cytokine balance as mentioned or by increased ligand expression in the inflamed mucosal gut tissue, causing ligand-induced down-regulation of NKG2D [87–89]. 6.2. NKG2D Pathway Contribution to CD Pathology Loss of tolerance towards microbiota crossing the epithelial barrier is likely the initiation of CD, and the pathology accelerates and persists due to the increased amount of immune cells and the mix of cytokines and chemokines creating inflammation. The pathologic response could start with the innate immune system and would then be mediated by antigen presenting cells (APCs) to activate adaptive immunity. Increased signaling or ineffective processing and clearance by epithelial cells, monocytes, neutrophils or dendritic cells (DCs) could lead to an exaggerated cytokine release resulting in an overly aggressive acquired T cell environment [99]. The inflammatory environment in turn activates other cells like cytotoxic Innate like lymphocytes (ILCs), NK cells, CD8+ T and CD4+ T cells capable of killing epithelial cells and other APCs and thus increasing inflammation. The cytotoxic cells can recognize the stress-induced ligands for activating receptors. This pathway could lead to degradation of the ligand-bearing epithelium and result in more microbial influx generating even more inflammation [100,101]. Translational research linked the properties of NKG2D-activated lymphocytes to the potential contribution to pathogenesis in Crohn’s disease. CD4+ T cells expressing NKG2D has been observed to be increased in CD. Like innate lymphocytes, these cells are able to kill NKG2D ligand-bearing cells and produce pro-inflammatory cytokines like IL-17, TNF-α and IFN-γ [7]. It has been shown that most of the oligoclonal expansion of mucosal T cells in CD consists of NKG2D-expressing CD4+ T cells [102]. This expansion will increase concentration of pathogenic TNF-α and IL-17 [103], which has been proven to contribute to disease by the registration of a biologic working through the IL-23/Th17 pathway [104,105]. The implication of CD4+NKG2D+ T cells in gut inflammation has been further shown in a murine model of transfer-induced colitis. In these studies, administration of a specific NKG2D blocking antibody decreased NKG2D expression on CD4+ T cells, blocked the receptor on the other lymphocyte subsets and attenuated the development of colitis, highlighting NKG2D as a possible therapeutic target in IBD [9,10]. Additionally, from coeliac disease it was shown how NKG2D modulates the disease through the cytotoxicity of NK and CD8+ T cells, indicating potential for a blockage in autoimmune diseases [47,106]. By blocking the NKG2D receptor with an antibody in human patients, the ligands will be physically unable to bind resulting in a reduced cytotoxic microenvironment and reduced killing of target cells in CD [107]. 6.1. Upregulated Ligands in Crohn’s Disease The expression of some of the ligands was tended to increase with higher concentrations of pro-inflammatory cytokines like IL-1β and TNF-α. A better characterization of the expression pattern and functional role of NKG2D ligands in CD is warranted to improve the understanding of the mechanism-of-action by NKG2D blockade therapy and the pathway itself. Int. J. Mol. Sci. 2017, 18, 1997 8 of 18 7.1. Phase IIa Results Animal experiments suggest a link between NKG2D and IBD, as NKG2D blockade can attenuate disease progression in certain mouse models of colitis [9,10]. In CD patients, a recent phase IIa study using a human blocking antibody against NKG2D showed significantly increased clinical remission after 12 weeks, suggesting that NKG2D-ligand interactions are viable therapeutic targets [13]. In the randomized, double-blind, parallel-group, placebo-controlled, single-dose, phase 2a trial, 78 CD patients were given either placebo or 2 mg/kg anti-NKG2D mAb and followed for 24 weeks. The primary endpoint was clinical response at four weeks. The trial was stopped prematurely primarily due to a slow recruitment rate and a negative futility analysis of the primary endpoint. However, a significant effect was achieved on the secondary endpoints at week 12 for both the Crohn’s disease activity index (CDAI) score and the Harvey Bradshaw (HBI) score (∆CDAI = −55; p < 0.1 and ∆HBI = −2.7; p < 0.1). In patients naïve to biologic treatment, the CDAI difference from placebo was significant from week 1 through week 12 (p < 0.1). An exposure-response analysis for patients with CDAI > 330 was published suggesting that higher doses and repeated dosing may optimize therapeutic benefit. The failure to reach the primary end point could be due to the dose applied was too low in combination with the rather low number of patients included, but occupancy of the receptor did show to be over 80% for the first eight weeks, falling to ~20% by week 12. Higher doses have been applied in rheumatoid arthritis and a dose response study had not been performed in Crohn’s patients before initiation of the study. Furthermore, multiple dosing in the first 4–8 weeks should have been considered similar to dosage regimens for anti-TNF treatment in Crohn’s disease. The drug was well-tolerated with no evidence of immunogenicity. The findings of this small sized randomized study encourage to further randomized studies in Crohn’s disease with larger sample size, optimizing dose regimens and maybe prolongation of primary endpoint evaluation to 6–8 weeks. 6.2. NKG2D Pathway Contribution to CD Pathology Our group also suggests an additional effect, namely a reduced ability of NKG2D-expressing lymphocytes to cross a ligand-expressing endothelial barrier if blocked by anti-NKG2D antibody [73]. Microvascular intestinal endothelial cells expressing MICA covered a transwell with activated human CD8+ T cells above. When attracted through the monolayer by a chemoattractant, blocking of NKG2D resulted in significant inhibition of T cells migrating through. In that way, fewer activated lymphocytes bearing NKG2D could enter the lamina propria of affected intestinal sites. Besides an increased cytotoxic microenvironment and granular release by lymphocytes activated by a NKG2D ligand interaction, the receptor can also be shown to contribute to the inflammatory balance. Using the previously described explant assay method [108], we added anti-NKG2D mAb to CD intestinal biopsies and observed a tendency towards a decreased cytokine production by the mucosal cells when the NKG2D pathway was blocked (Figure 3). 9 of 18 ls when Int. J. Mol. Sci. 2017, 18, 1997 biopsies and observed a the NKG2D pathway wa Figure 3. Inflamed intestinal mucosal biopsies from four CD patients treated in an explant assay over 24 h with either isotype control or anti-NKG2D antibody. The release of five cytokines to the medium was analyzed and quantified as pg/mL normalized to 100 mg of tissue. The average of measurements from four biopsies are represented in each data point, shown as the mean ± SEM, with two-way ANOVA. Figure 3. Inflamed intestinal mucosal biopsies from four CD patients treated in an explant assay over 24 h with either isotype control or anti-NKG2D antibody. The release of five cytokines to the medium was analyzed and quantified as pg/mL normalized to 100 mg of tissue. The average of measurements from four biopsies are represented in each data point, shown as the mean ± SEM, with two-way ANOVA. Figure 3. Inflamed intestinal mucosal biopsies from four CD patients treated in an explant assay over 24 h with either isotype control or anti-NKG2D antibody. The release of five cytokines to the medium was analyzed and quantified as pg/mL normalized to 100 mg of tissue. The average of measurements from four biopsies are represented in each data point, shown as the mean ± SEM, with two-way ANOVA Figure 3. Inflamed intestinal mucosal biopsies from four CD patients treated in an explant assay over 24 h with either isotype control or anti-NKG2D antibody. 6.2. NKG2D Pathway Contribution to CD Pathology The release of five cytokines to the medium was analyzed and quantified as pg/mL normalized to 100 mg of tissue. The average of measurements from four biopsies are represented in each data point, shown as the mean ± SEM, with two-way ANOVA. 8. Future Perspectives The introduction of anti-TNF-α monoclonal antibodies (mAb) (e.g., Infliximab) was a great advance in CD, leading to improved remission and maintenance hereof in patients with insufficient, or lack of, effect of immunomodulators [111]. The drug has low adverse effects, but only 60–70% of patients will respond and, of responders, ~40% will lose effect within the first year [112]. Targeting and blocking NKG2D would be another new mechanism of action for moderate to severe Crohn’s disease patients. Administration of a blocking antibody against NKG2D has been shown to significantly increase clinical remission in CD patients [107], most likely a result of the NKG2D blockade leading to abrogation of lymphocyte cytotoxicity and cytokine production, but it might also influence migration, recruitment, and retention of inflammatory cells into affected tissue [113]. Tissue-specific homing involves tethering, activation, and firm adhesion steps [114], and the activation might be targeted here. Activation of intestine-derived T cells has been shown to increase their migration [115]. In multiple sclerosis, a CD4+ T cell NKG2D-dependent migration mechanism has been observed [116]. It is therefore possible that the NKG2D and NKG2D-ligand interaction may provide an activating signal to the NK and T cells promoting successful migration, as we have proposed as a supporting mechanism [73]. In this way, anti-NKG2D antibody may uniquely interfere with both intestinal inflammation and lymphocyte homing, the two main processes targeted by current biological therapies for CD and ulcerative colitis [117,118]. Both the mechanism of cytotoxicity and the possible effect on migration should be investigated further. It would be of special interest to know which cells in the CD intestine are targeted for elimination through this pathway and which ligands play integrant roles. More research is needed into the mechanism of action of anti-NKG2D and its therapeutic effect, especially from a genetic perspective. Single nucleotide polymorphism (SNP) is implicated in the extensive polymorphism of the ligands and relates to the binding strength of the receptor and handling of pathogens [119,120]. The single SNP at the protein level that exists for KLRK1 can also produce differences in NKG2D function. NKG2D can vary in the intensity of cell surface expression due to genetic polymorphisms. The functional consequences of polymorphisms in NKG2D and NKG2D ligands may be cooperative or counteracting. The interaction of the variants could be highly important for the outcome of NKG2D signaling and disease associations of the NKG2D signaling pathway [119,121]. 8. Future Perspectives This could result in diverse efficacy between patient populations exposed to the antibody. It could be of great interest to further investigate each NKG2D ligand to explore the exact differences between them and how they are regulated. Evolutionarily, eight ligands to one receptor indicate the importance of the interaction or differences in mechanism. The blocking of a specific ligand instead of the receptor could induce different results in CD patients. Patients with moderate to severe Crohn’s disease, who have failed conventional treatment, and maybe some biologics, could be candidates for treatment with a drug with a new mode of action. Given the natural mechanism of NKG2D-ligand interactions in the human system, special attention should be paid to any adverse effects concerning malignancies and infections in the development and use of a blocking antibody for the use in CD. 7.2. Phase IIb Initiation Two new clinical studies with the anti-NKG2D biologic have been initiated; a phase IIb investigating the safety and efficacy of the drug in participants with moderately to severely active 10 of 18 10 of 18 Int. J. Mol. Sci. 2017, 18, 1997 Crohn's disease including 450 patients for 22 weeks with subcutaneous doses up to 400 mg induction and 200 mg every four weeks [109], and a safety study in healthy Japanese and Caucasian male participants also testing tolerability, pharmacokinetics, and pharmacodynamics following subcutaneous drug injections [110]. Crohn's disease including 450 patients for 22 weeks with subcutaneous doses up to 400 mg induction and 200 mg every four weeks [109], and a safety study in healthy Japanese and Caucasian male participants also testing tolerability, pharmacokinetics, and pharmacodynamics following subcutaneous drug injections [110]. 9. Concluding Remarks The NKG2D pathway poses an attractive new treatment option for Crohn’s disease. The receptor prevalence is high, its biology well-described, and it is straightforward to block the activity through a monoclonal antibody–presumably with limited adverse effects. This could potentially 11 of 18 Int. J. Mol. Sci. 2017, 18, 1997 abrogate lymphocyte destruction of the gut tissue and the system’s contribution to the inflammatory pathogenesis. However, surprisingly little is known about the ligand prevalence and biology, especially in autoimmune diseases. Excellent commercial antibodies for all eight ligands are now available and further research should be directed towards the characterization of ligand-expressing cell types in IBD and the mechanism of action through this potent pathway. Acknowledgments: We would like to thank Teis Jensen for the drawing in Figure 1. Author Contributions: Kasper Vadstrup and Flemming Bendtsen contributed to all phases of the manuscript. Conflicts of Interest: The authors declare no conflict of interest. Acknowledgments: We would like to thank Teis Jensen for the drawing in Figure 1. Author Contributions: Kasper Vadstrup and Flemming Bendtsen contributed to all phases of the manuscript. Conflicts of Interest: The authors declare no conflict of interest. Author Contributions: Kasper Vadstrup and Flemming Bendtsen contributed to all phases of the manuscript. Conflicts of Interest: The authors declare no conflict of interest. References 1. Torres, J.; Mehandru, S.; Colombel, J.F.; Peyrin-Biroulet, L. Crohn’s disease. Lancet 2017, 389, 1741–1755. [CrossRef] 1. Torres, J.; Mehandru, S.; Colombel, J.F.; Peyrin-Biroulet, L. Crohn’s disease. Lancet 2017, 389, 1741–1755. [CrossRef] 2. Laass, M.W.; Roggenbuck, D.; Conrad, K. Diagnosis and classification of Crohn’s disease. Autoimmun. Rev. 2014, 13, 467–471. [CrossRef] [PubMed] 2. Laass, M.W.; Roggenbuck, D.; Conrad, K. Diagnosis and classification of Crohn’s disease. Autoimmun. Rev. 2014, 13, 467–471. [CrossRef] [PubMed] 3. Busch, K.; Sonnenberg, A.; Bansback, N. Impact of inflammatory bowel disease on disability. Curr. Gastroenterol. Rep. 2014, 16, 414. [CrossRef] [PubMed] 3. Busch, K.; Sonnenberg, A.; Bansback, N. Impact of inflammatory bowel disease on disability. Curr. Gastroenterol. Rep. 2014, 16, 414. [CrossRef] [PubMed] Cohen, L.B.; Nanau, R.M.; Delzor, F.; Neuman, M.G. Biologic therapies in inflammatory bowel disease Transl. Res. J. Lab. Clin. Med. 2014, 163, 533–556. [CrossRef] [PubMed] m, D.H.; Cheon, J.H. Pathogenesis of inflammatory bowel disease and recent advances in biologic therapies. mune Netw. 2017, 17, 25–40. [CrossRef] [PubMed] 6. Katsanos, K.H.; Papadakis, K.A. Inflammatory bowel disease: Updates on molecular targets for biologics. Gut Liver 2017, 11, 455–463. [CrossRef] [PubMed] 6. Katsanos, K.H.; Papadakis, K.A. Inflammatory bowel disease: Updates on molecular targets for biologics. Gut Liver 2017, 11, 455–463. [CrossRef] [PubMed] 7. Allez, M.; Tieng, V.; Nakazawa, A.; Treton, X.; Pacault, V.; Dulphy, N.; Caillat-Zucman, S.; Paul, P.; Gornet, J.M.; Douay, C.; et al. Cd4+NKG2D+ T cells in Crohn’s disease mediate inflammatory and cytotoxic responses through mica interactions. Gastroenterology 2007, 132, 2346–2358. [CrossRef] [PubMed] 7. Allez, M.; Tieng, V.; Nakazawa, A.; Treton, X.; Pacault, V.; Dulphy, N.; Caillat-Zucman, S.; Paul, P.; Gornet, J.M.; Douay, C.; et al. Cd4+NKG2D+ T cells in Crohn’s disease mediate inflammatory and cytotoxic responses through mica interactions. Gastroenterology 2007, 132, 2346–2358. [CrossRef] [PubMed] 8. Lanier, L.L. NKG2D receptor and its ligands in host defense. Cancer Immunol. Res. 2015, 3, 575–582. [CrossRef] [PubMed] 9. Ito, Y.; Kanai, T.; Totsuka, T.; Okamoto, R.; Tsuchiya, K.; Nemoto, Y.; Yoshioka, A.; Tomita, T.; Nagaishi, T.; Sakamoto, N.; et al. Blockade of NKG2D signaling prevents the development of murine CD4+ T cell-mediated colitis. Am. J. Physiol. Gastrointest. Liver Physiol. 2008, 294, G199–G207. [CrossRef] [PubMed] 10. Kjellev, S.; Haase, C.; Lundsgaard, D.; Urso, B.; Tornehave, D.; Markholst, H. Inhibition of NKG2D receptor function by antibody therapy attenuates transfer-induced colitis in scid mice. Eur. J. Immunol. 2007, 37, 1397–1406. [CrossRef] [PubMed] 11. Abbreviations Abbreviations Akt Protein kinase B ANOVA Analysis of variance CD Crohn’s disease cJunN c-Jun N-terminal kinase-1 CRP C-reactive protein DAP10 DNAX-activating protein of 10 kDa ERK Extracellular signal-Regulated Kinase Grb2 Growth factor receptor-bound protein 2 GM-CSF Granulocyte macrophage colony-stimulating factor GPI Glycosylphosphatidylinositol IBD Inflammatory bowel disease IEL Intraepithelial lymphocyte IFN Interferon IL Interleukin ILC Innate like lymphocyte JAK2 Janus Kinase 2 JNK c-Jun N-terminal protein Kinases mAB Monoclonal antibody MHC Major histocompatibility complex MIC MHC class I polypeptide-related sequence MEK Mitogen-activated protein kinase kinase NK Natural killer NKG2D Natural killer group 2D PLCγ2 Phospholipase C γ-2 P85 PI3K P85 subunit of phosphoInositide 3-kinase RNA Ribonucleic acid SEM Standard error of the mean SLP-76 SH2 domain-containing leukocyte phosphoprotein of 76 kDa STAT5 Signal transducer and activator of transcription 5 TCR T cell receptor Th T helper cell TM Transmembrane TNF Tumor necrosis factor ULBP UL16 binding protein Vav-1 Vav guanine nucleotide exchange factor 1 12 of 18 12 of 18 Int. J. Mol. Sci. 2017, 18, 1997 References Verneris, M.R.; Karimi, M.; Baker, J.; Jayaswal, A.; Negrin, R.S. Role of NKG2D signaling in the cytotoxicity of activated and expanded CD8+ T cells. Blood 2004, 103, 3065–3072. [CrossRef] [PubMed] 25. Garrity, D.; Call, M.E.; Feng, J.; Wucherpfennig, K.W. The activating NKG2D receptor assembles in the membrane with two signaling dimers into a hexameric structure. Proc. Natl. Acad. Sci. USA 2005, 102, 7641–7646. [CrossRef] [PubMed] 26. Cerwenka, A.; Lanier, L.L. Ligands for natural killer cell receptors: Redundancy or specificity. Immunol. Rev. 2001, 181, 158–169. [CrossRef] [PubMed] 27. O’Callaghan, C.A.; Cerwenka, A.; Willcox, B.E.; Lanier, L.L.; Bjorkman, P.J. Molecular competition for NKG2D: H60 and rae1 compete unequally for NKG2D with dominance of h60. Immunity 2001, 15, 201–211. [CrossRef] 28. Robinson, J.; Waller, M.J.; Parham, P.; Bodmer, J.G.; Marsh, S.G. IMGT/HLA database—A sequence database for the human major histocompatibility complex. Nucleic Acids Res. 2001, 29, 210–213. [CrossRef] [PubMed] 29. Radosavljevic, M.; Cuillerier, B.; Wilson, M.J.; Clement, O.; Wicker, S.; Gilfillan, S.; Beck, S.; Trowsdale, J.; Bahram, S. A cluster of ten novel MHC class I related genes on human chromosome 6q24.2-q25.3. Genomics 28. Robinson, J.; Waller, M.J.; Parham, P.; Bodmer, J.G.; Marsh, S.G. IMGT/HLA database—A sequence database for the human major histocompatibility complex. Nucleic Acids Res. 2001, 29, 210–213. [CrossRef] [PubMed] 29. Radosavljevic, M.; Cuillerier, B.; Wilson, M.J.; Clement, O.; Wicker, S.; Gilfillan, S.; Beck, S.; Trowsdale, J.; Bahram, S. A cluster of ten novel MHC class I related genes on human chromosome 6q24.2-q25.3. Genomics 2002, 79, 114–123. [CrossRef] [PubMed] 30. Ohashi, M.; Eagle, R.A.; Trowsdale, J. Post-translational modification of the NKG2D ligand raet1g leads to cell surface expression of a glycosylphosphatidylinositol-linked isoform. J. Biol. Chem. 2010, 285, 16408–16415. [CrossRef] [PubMed] 31. Fernandez-Messina, L.; Ashiru, O.; Aguera-Gonzalez, S.; Reyburn, H.T.; Vales-Gomez, M. The human NKG2D ligand ULBP2 can be expressed at the cell surface with or without a GPI anchor and both forms can activate nk cells. J. Cell Sci. 2011, 124, 321–327. [CrossRef] [PubMed] 32. Amroun, H.; Djoudi, H.; Busson, M.; Allat, R.; El Sherbini, S.M.; Sloma, I.; Ramasawmy, R.; Brun, M.; Dulphy, N.; Krishnamoorthy, R.; et al. Early-onset ankylosing spondylitis is associated with a functional mica polymorphism. Hum. Immunol. 2005, 66, 1057–1061. [CrossRef] [PubMed] 33. Steinle, A.; Li, P.; Morris, D.L.; Groh, V.; Lanier, L.L.; Strong, R.K.; Spies, T. Interactions of human NKG2D with its ligands MICA, MICB, and homologs of the mouse RAE-1 protein family. Immunogenetics 2001, 53, 279–287. References [CrossRef] [PubMed] 34. Douik, H.; Ben Chaaben, A.; Attia Romdhane, N.; Romdhane, H.B.; Mamoghli, T.; Fortier, C.; Boukouaci, W.; Harzallah, L.; Ghanem, A.; Gritli, S.; et al. Association of MICA-129 polymorphism with nasopharyngeal cancer risk in a tunisian population. Hum. Immunol. 2009, 70, 45–48. [CrossRef] [PubMed] 35. Li, Y.; Mariuzza, R.A. Structural basis for recognition of cellular and viral ligands by NK cell receptors. Front. Immunol. 2014, 5, 123. [CrossRef] [PubMed] 36. Schrambach, S.; Ardizzone, M.; Leymarie, V.; Sibilia, J.; Bahram, S. In vivo expression pattern of mica and micb and its relevance to auto-immunity and cancer. PLoS ONE 2007, 2, e518. [CrossRef] [PubMed] 36. Schrambach, S.; Ardizzone, M.; Leymarie, V.; Sibilia, J.; Bahram, S. In vivo expression pattern of mica and micb and its relevance to auto-immunity and cancer. PLoS ONE 2007, 2, e518. [CrossRef] [PubMed] 37. Gonzalez, S.; Lopez-Soto, A.; Suarez-Alvarez, B.; Lopez-Vazquez, A.; Lopez-Larrea, C. NKG2D ligands: Key 37. Gonzalez, S.; Lopez-Soto, A.; Suarez-Alvarez, B.; Lopez-Vazquez, A.; Lopez-Larrea, C. NKG2D ligands: Key targets of the immune response. Trends Immunol. 2008, 29, 397–403. [CrossRef] [PubMed] 8. Raulet, D.H.; Gasser, S.; Gowen, B.G.; Deng, W.; Jung, H. Regulation of ligands for the NKG2D activa receptor. Annu. Rev. Immunol. 2013, 31, 413–441. [CrossRef] [PubMed] 39. Gonzalez, S.; Groh, V.; Spies, T. Immunobiology of human NKG2D and its ligands. Curr. Top. Microbiol. Immunol. 2006, 298, 121–138. [PubMed] 40. Zou, Z.; Nomura, M.; Takihara, Y.; Yasunaga, T.; Shimada, K. Isolation and characterization of retinoic acid-inducible cDNA clones in F9 cells: A novel cDNA family encodes cell surface proteins sharing partial homology with MHC class I molecules. J. Biochem. 1996, 119, 319–328. [CrossRef] [PubMed] 41. Ogasawara, K.; Benjamin, J.; Takaki, R.; Phillips, J.H.; Lanier, L.L. Function of NKG2D in natural killer cell-mediated rejection of mouse bone marrow grafts. Nat. Immunol. 2005, 6, 938–945. [CrossRef] [PubMed] 42. Jung, H.; Hsiung, B.; Pestal, K.; Procyk, E.; Raulet, D.H. RAE-1 ligands for the NKG2D receptor are regulated b E2F i i f hi h l ll l J E M d 1 209 2409 2422 [C R f] 41. Ogasawara, K.; Benjamin, J.; Takaki, R.; Phillips, J.H.; Lanier, L.L. Function of NKG2D in natural killer cell-mediated rejection of mouse bone marrow grafts. Nat. Immunol. 2005, 6, 938–945. [CrossRef] [PubMed] 42. Jung, H.; Hsiung, B.; Pestal, K.; Procyk, E.; Raulet, D.H. References Guerra, N.; Pestal, K.; Juarez, T.; Beck, J.; Tkach, K.; Wang, L.; Raulet, D.H. A selective role of NKG2D in inflammatory and autoimmune diseases. Clin. Immunol. 2013, 149, 432–439. [CrossRef] [PubMed] 12. La Scaleia, R.; Stoppacciaro, A.; Oliva, S.; Morrone, S.; Di Nardo, G.; Santoni, A.; Cucchiara, S.; Palmieri, G. NKG2D/ligand dysregulation and functional alteration of innate immunity cell populations in pediatric IBD. Inflamm. Bowel Dis. 2012, 18, 1910–1922. [CrossRef] [PubMed] 13. Allez, M.; Skolnick, B.E.; Wisniewska-Jarosinska, M.; Petryka, R.; Overgaard, R.V. Anti-NKG2D monoclonal antibody (NNC0142-0002) in active crohn’s disease: A randomised controlled trial. Gut 2016. [CrossRef] [PubMed] 14. Neurath, M.F. Current and emerging therapeutic targets for IBD. Nat. Rev. Gastroenterol. Hepatol. 2017, 14, 269–278. [CrossRef] [PubMed] 15. Carapito, R.; Bahram, S. Genetics, genomics, and evolutionary biology of NKG2D ligands. Immunol. Rev. 2015, 267, 88–116. [CrossRef] [PubMed] 16. Champsaur, M.; Lanier, L.L. Effect of NKG2D ligand expression on host immune responses. Immunol. Rev. 2010, 235, 267–285. [CrossRef] [PubMed] 17. Upshaw, J.L.; Leibson, P.J. NKG2D-mediated activation of cytotoxic lymphocytes: Unique signaling pathways and distinct functional outcomes. Semin. Immunol. 2006, 18, 167–175. [CrossRef] [PubMed] 18. Raulet, D.H. Roles of the NKG2D immunoreceptor and its ligands. Nat. Rev. Immunol. 2003, 3, 781–790. [CrossRef] [PubMed] 19. Glienke, J.; Sobanov, Y.; Brostjan, C.; Steffens, C.; Nguyen, C.; Lehrach, H.; Hofer, E.; Francis, F. The genomic organization of NKG2C, E, F, and D receptor genes in the human natural killer gene complex. Immunogenetics 1998, 48, 163–173. [CrossRef] [PubMed] 20. Wada, H.; Matsumoto, N.; Maenaka, K.; Suzuki, K.; Yamamoto, K. The inhibitory nk cell receptor CD94/NKG2A and the activating receptor CD94/NKG2C bind the top of HLA-E through mostly shared but partly distinct sets of HLA-E residues. Eur. J. Immunol. 2004, 34, 81–90. [CrossRef] [PubMed] 21. Kondo, M.; Maruoka, T.; Otsuka, N.; Kasamatsu, J.; Fugo, K.; Hanzawa, N.; Kasahara, M. Comparative genomic analysis of mammalian NKG2D ligand family genes provides insights into their origin and evolution. Immunogenetics 2010, 62, 441–450. [CrossRef] [PubMed] Int. J. Mol. Sci. 2017, 18, 1997 13 of 18 13 of 18 22. Wu, J.; Song, Y.; Bakker, A.B.; Bauer, S.; Spies, T.; Lanier, L.L.; Phillips, J.H. An activating immunoreceptor complex formed by NKG2D and DAP10. Science 1999, 285, 730–732. [CrossRef] [PubMed] 23. Zafirova, B.; Wensveen, F.M.; Gulin, M.; Polic, B. Regulation of immune cell function and differentiation by the NKG2D receptor. Cell. Mol. Life Sci. 2011, 68, 3519–3529. [CrossRef] [PubMed] 24. References RAE-1 ligands for the NKG2D receptor are regulated by E2F transcription factors, which control cell cycle entry. J. Exp. Med. 2012, 209, 2409–2422. [CrossRef] [PubMed] 14 of 18 14 of 18 Int. J. Mol. Sci. 2017, 18, 1997 43. Caillat-Zucman, S. How NKG2D ligands trigger autoimmunity? Hum. Immunol. 2006, 67, 204–207. [CrossRef] [PubMed] 44. Groh, V.; Bruhl, A.; El-Gabalawy, H.; Nelson, J.L.; Spies, T. Stimulation of t cell autoreactivity by anomalous expression of NKG2D and its MIC ligands in rheumatoid arthritis. Proc. Natl. Acad. Sci. USA 2003, 100, 9452–9457. [CrossRef] [PubMed] 45. Lin, D.; Lavender, H.; Soilleux, E.J.; O’Callaghan, C.A. NF-kappaB regulates mica gene transcription in endothelial cell through a genetically inhibitable control site. J. Boil. Chem. 2012, 287, 4299–4310. [CrossRef] [PubMed] 46. Allegretti, Y.L.; Bondar, C.; Guzman, L.; Cueto Rua, E.; Chopita, N.; Fuertes, M.; Zwirner, N.W.; Chirdo, F.G. Broad MICA/B expression in the small bowel mucosa: A link between cellular stress and celiac disease. PLoS ONE 2013, 8, e73658. [CrossRef] [PubMed] 47. Hue, S.; Mention, J.J.; Monteiro, R.C.; Zhang, S.; Cellier, C.; Schmitz, J.; Verkarre, V.; Fodil, N.; Bahram, S.; Cerf-Bensussan, N.; et al. A direct role for NKG2D/mica interaction in villous atrophy during celiac disease. Immunity 2004, 21, 367–377. [CrossRef] [PubMed] 48. Bauer, S.; Groh, V.; Wu, J.; Steinle, A.; Phillips, J.H.; Lanier, L.L.; Spies, T. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible mica. Science 1999, 285, 727–729. [CrossRef] [PubMed] 9. Bryceson, Y.T.; March, M.E.; Ljunggren, H.G.; Long, E.O. Synergy among receptors on resting NK cell the activation of natural cytotoxicity and cytokine secretion. Blood 2006, 107, 159–166. [CrossRef] [PubM 50. Ehrlich, L.I.; Ogasawara, K.; Hamerman, J.A.; Takaki, R.; Zingoni, A.; Allison, J.P.; Lanier, L.L. Engagement of NKG2D by cognate ligand or antibody alone is insufficient to mediate costimulation of human and mouse CD8+ T cells. J. Immunol. 2005, 174, 1922–1931. [CrossRef] [PubMed] 51. Boukouaci, W.; Al-Daccak, R.; Dulphy, N.; Lauden, L.; Amokrane, K.; Fortier, C.; Marzais, F.; Bennabi, M.; Peffault de Latour, R.; Socie, G.; et al. Soluble mica-NKG2D interaction upregulates IFN-γ production by activated Cd3-Cd56+ NK cells: Potential impact on chronic graft versus host disease. Hum. Immunol. 2013, 74, 1536–1541. [CrossRef] [PubMed] 52. Poggi, A.; Zocchi, M.R. Antigen presenting cells and stromal cells trigger human natural killer lymphocytes to autoreactivity: Evidence for the involvement of natural cytotoxicity receptors (NCR) and NKG2D. Clin. Dev. Immunol. 2006, 13, 325–336. References [CrossRef] [PubMed] 53. Whitman, E.; Barber, A. NKG2D receptor activation of Nf-κB enhances inflammatory cytokine production in murine effector CD8(+) T cells. Mol. Immunol. 2015, 63, 268–278. [CrossRef] [PubMed] 54. Lanier, L.L. Dap10- and dap12-associated receptors in innate immunity. Immunol. Rev. 2009, 227, 150–160. [CrossRef] [PubMed] 55. Upshaw, J.L.; Arneson, L.N.; Schoon, R.A.; Dick, C.J.; Billadeau, D.D.; Leibson, P.J. NKG2D-mediated signaling requires a DAP10-bound GRB2-VAV1 intermediate and phosphatidylinositol-3-kinase in human natural killer cells. Nat. Immunol. 2006, 7, 524–532. [CrossRef] [PubMed] 56. Jelencic, V.; Lenartic, M.; Wensveen, F.M.; Polic, B. NKG2D: A versatile player in the immune system. Immunol. Lett. 2017. 57. Cosman, D.; Mullberg, J.; Sutherland, C.L.; Chin, W.; Armitage, R.; Fanslow, W.; Kubin, M.; Chalupny, N.J. ULBPs, novel MHC class I-related molecules, bind to cmv glycoprotein ul16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity 2001, 14, 123–133. [CrossRef] Campbell, J.A.; Trossman, D.S.; Yokoyama, W.M.; Carayannopoulos, L.N. Zoonotic orthopoxviruses encode a high-affinity antagonist of NKG2D. J. Exp. Med. 2007, 204, 1311–1317. [CrossRef] [PubMed] 59. Spear, P.; Wu, M.R.; Sentman, M.L.; Sentman, C.L. NKG2D ligands as therapeutic targets. Cancer Immun. 2013, 13, 8. [PubMed] 60. Gasser, S.; Orsulic, S.; Brown, E.J.; Raulet, D.H. The DNA damage pathway regulates innate immune system ligands of the NKG2D receptor. Nature 2005, 436, 1186–1190. [CrossRef] [PubMed] 61. Groh, V.; Wu, J.; Yee, C.; Spies, T. Tumour-derived soluble MIC ligands impair expression of NKG2D and T-cell activation. Nature 2002, 419, 734–738. [CrossRef] [PubMed] 62. Kaiser, B.K.; Yim, D.; Chow, I.T.; Gonzalez, S.; Dai, Z.; Mann, H.H.; Strong, R.K.; Groh, V.; Spies, T. Disulphide-isomerase-enabled shedding of tumour-associated NKG2D ligands. Nature 2007, 447, 482–486. [CrossRef] [PubMed] Int. J. Mol. Sci. 2017, 18, 1997 15 of 18 15 of 18 63. Baragano Raneros, A.; Suarez-Alvarez, B.; Lopez-Larrea, C. Secretory pathways generating immunosuppressive NKG2D ligands: New targets for therapeutic intervention. Oncoimmunology 2014, 3, e28497. [CrossRef] [PubMed] 4. Ullrich, E.; Koch, J.; Cerwenka, A.; Steinle, A. New prospects on the NKG2D/NKG2Dl system for onco Oncoimmunology 2013, 2, e26097. [CrossRef] [PubMed] 65. Castriconi, R.; Cantoni, C.; Della Chiesa, M.; Vitale, M.; Marcenaro, E.; Conte, R.; Biassoni, R.; Bottino, C.; Moretta, L.; Moretta, A. Transforming growth factor beta 1 inhibits expression of NKP30 and NKG2D receptors: Consequences for the NK-mediated killing of dendritic cells. Proc. Natl. Acad. Sci. USA 2003, 100, 4120–4125. [CrossRef] [PubMed] 66. Crane, C.A.; Han, S.J.; Barry, J.J.; Ahn, B.J.; Lanier, L.L.; Parsa, A.T. References TGF-β downregulates the activating receptor NKG2D on NK cells and CD8+ T cells in glioma patients. Neuro-Oncology 2010, 12, 7–13. [CrossRef] [PubMed] 67. Lee, J.C.; Lee, K.M.; Kim, D.W.; Heo, D.S. Elevated TGF-beta1 secretion and down-modulation of NKG2D underlies impaired NK cytotoxicity in cancer patients. J. Immunol. 2004, 172, 7335–7340. [CrossRef] [PubMed] 67. Lee, J.C.; Lee, K.M.; Kim, D.W.; Heo, D.S. Elevated TGF-beta1 secretion and down-modulation of NKG2D underlies impaired NK cytotoxicity in cancer patients. J. Immunol. 2004, 172, 7335–7340. [CrossRef] [PubMed] 68. Orchard, T.R.; Dhar, A.; Simmons, J.D.; Vaughan, R.; Welsh, K.I.; Jewell, D.P. MHC class I chain-like gene a (MICA) d it i ti ith i fl t b l di d i h l th th Cli E I l 68. Orchard, T.R.; Dhar, A.; Simmons, J.D.; Vaughan, R.; Welsh, K.I.; Jewell, D.P. MHC class I chain-like gene a (MICA) and its associations with inflammatory bowel disease and peripheral arthropathy. Clin. Exp. Immunol. 2001, 126, 437–440. [CrossRef] [PubMed] 69. Martinez-Chamorro, A.; Moreno, A.; Gomez-Garcia, M.; Cabello, M.J.; Martin, J.; Lopez-Nevot, M.A. Mica*A4 protects against ulcerative colitis, whereas MICA*A5.1 is associated with abscess formation and age of onset. Clin. Exp. Immunol. 2016, 184, 323–331. [CrossRef] [PubMed] 70. Lopez-Hernandez, R.; Valdes, M.; Lucas, D.; Campillo, J.A.; Martinez-Garcia, P.; Salama, H.; Lopez, M.; Salgado, G.; Botella, C.; Minguela, A.; et al. Association analysis of MICA gene polymorphism and MICA-129 dimorphism with inflammatory bowel disease susceptibility in a Spanish population. Hum. Immunol. 2010, 71, 512–514. [CrossRef] [PubMed] 71. Yun, J.; Xu, C.T.; Pan, B.R. Epidemiology and gene markers of ulcerative colitis in the Chinese. World J. Gastroenterol. WJG 2009, 15, 788–803. [CrossRef] [PubMed] 72. Perera, L.; Shao, L.; Patel, A.; Evans, K.; Meresse, B.; Blumberg, R.; Geraghty, D.; Groh, V.; Spies, T.; Jabri, B.; et al. Expression of nonclassical class I molecules by intestinal epithelial cells. Inflamm. Bowel Dis. 2007, 13, 298–307. [CrossRef] [PubMed] 73. Vadstrup, K.; Galsgaard, E.D.; Jensen, H.; Lanier, L.L.; Ryan, J.C.; Chen, S.Y.; Nolan, G.P.; Vester-Andersen, M.K.; Pedersen, J.S.; Gerwien, J.; et al. NKG2D ligand expression in crohn’s disease and NKG2D-dependent stimulation of CD8+ T cell migration. Exp. Mol. Pathol. 2017, 103, 56–70. [CrossRef] [PubMed] 74. Maasho, K.; Opoku-Anane, J.; Marusina, A.I.; Coligan, J.E.; Borrego, F. NKG2D is a costimulatory receptor for human naive CD8+ T cells. J. Immunol. 2005, 174, 4480–4484. [CrossRef] [PubMed] 75. References Moretta, L.; Montaldo, E.; Vacca, P.; del Zotto, G.; Moretta, F.; Merli, P.; Locatelli, F.; Mingari, M.C. Human natural killer cells: Origin, receptors, function, and clinical applications. Int. Arch. Allergy Immunol. 2014, 164, 253–264. [CrossRef] [PubMed] 87. Jimenez-Perez, M.I.; Jave-Suarez, L.F.; Ortiz-Lazareno, P.C.; Bravo-Cuellar, A.; Gonzalez-Ramella, O.; Aguilar-Lemarroy, A.; Hernandez-Flores, G.; Pereira-Suarez, A.L.; Daneri-Navarro, A.; del Toro-Arreola, S. Cervical cancer cell lines expressing NKG2D-ligands are able to down-modulate the NKG2D receptor on NKL cells with functional implications. BMC Immunol. 2012, 13, 7. [CrossRef] [PubMed] 88. Lundholm, M.; Schroder, M.; Nagaeva, O.; Baranov, V.; Widmark, A.; Mincheva-Nilsson, L.; Wikstrom, P. Prostate tumor-derived exosomes down-regulate NKG2D expression on natural killer cells and CD8+ T cells: Mechanism of immune evasion. PLoS ONE 2014, 9, e108925. [CrossRef] [PubMed] 89. Mincheva-Nilsson, L.; Baranov, V. Cancer exosomes and NKG2D receptor-ligand interactions: Impairing NKG2D-mediated cytotoxicity and anti-tumour immune surveillance. Semin. Cancer Biol. 2014, 28, 24–30. [CrossRef] [PubMed] 90. Eagle, R.A.; Jafferji, I.; Barrow, A.D. Beyond stressed self: Evidence for NKG2D ligand expression on healthy cells. Curr. Immunol. Rev. 2009, 5, 22–34. [CrossRef] [PubMed] 91. Groh, V.; Bahram, S.; Bauer, S.; Herman, A.; Beauchamp, M.; Spies, T. Cell stress-regulated human major histocompatibility complex class I gene expressed in gastrointestinal epithelium. Proc. Natl. Acad. Sci. USA 1996, 93, 12445–12450. [CrossRef] [PubMed] 92. Bilbao, J.R.; Martin-Pagola, A.; Perez De Nanclares, G.; Calvo, B.; Vitoria, J.C.; Vazquez, F.; Castano, L. HLA-DRB1 and mica in autoimmunity: Common associated alleles in autoimmune disorders. Ann. N. Y. Acad. Sci. 2003, 1005, 314–318. [CrossRef] [PubMed] 93. Cho, W.K.; Jung, M.H.; Park, S.H.; Baek, I.C.; Choi, H.B.; Kim, T.G.; Suh, B.K. Association of mica alleles with autoimmune thyroid disease in korean children. Int. J. Endocrinol. 2012, 2012, 235680. [CrossRef] [PubMed] 93. Cho, W.K.; Jung, M.H.; Park, S.H.; Baek, I.C.; Choi, H.B.; Kim, T.G.; Suh, B.K. Association of mica alleles with autoimmune thyroid disease in korean children. Int. J. Endocrinol. 2012, 2012, 235680. [CrossRef] [PubMed] 94. Glas, J.; Martin, K.; Brunnler, G.; Kopp, R.; Folwaczny, C.; Weiss, E.H.; Albert, E.D. MICA, MICB and 94. Glas, J.; Martin, K.; Brunnler, G.; Kopp, R.; Folwaczny, C.; Weiss, E.H.; Albert, E.D. MICA, MICB and C1_4_1 polymorphism in Crohn’s disease and ulcerative colitis. Tissue Antigens 2001, 58, 243–249. [CrossRef] [PubMed] 95. Cox, S.T.; Madrigal, J.A.; Saudemont, A. Diversity and characterization of polymorphic 5′ promoter haplotypes of mica and MICB genes. Tissue Antigens 2014, 84, 293–303. [CrossRef] [PubMed] 96. References Park, Y.P.; Choi, S.C.; Kiesler, P.; Gil-Krzewska, A.; Borrego, F.; Weck, J.; Krzewski, K.; Coligan, J.E. Complex regulation of human NKG2D-DAP10 cell surface expression: Opposing roles of the gammac cytokines and TGF-β1. Blood 2011, 118, 3019–3027. [CrossRef] [PubMed] 76. Roberts, A.I.; Lee, L.; Schwarz, E.; Groh, V.; Spies, T.; Ebert, E.C.; Jabri, B. NKG2D receptors induced by il-15 costimulate CD28-negative effector CTL in the tissue microenvironment. J. Immunol. 2001, 167, 5527–5530. [CrossRef] [PubMed] 77. Zhang, C.; Zhang, J.; Niu, J.; Zhou, Z.; Zhang, J.; Tian, Z. Interleukin-12 improves cytotoxicity of natural killer cells via upregulated expression of NKG2D. Hum. Immunol. 2008, 69, 490–500. [CrossRef] [PubMed] 78. Burgess, S.J.; Marusina, A.I.; Pathmanathan, I.; Borrego, F.; Coligan, J.E. IL-21 down-regulates NKG2D/DAP10 expression on human NK and CD8+ T cells. J. Immunol. 2006, 176, 1490–1497. [CrossRef] [PubMed] 79. Muntasell, A.; Magri, G.; Pende, D.; Angulo, A.; Lopez-Botet, M. Inhibition of NKG2D expression in nk cells by cytokines secreted in response to human cytomegalovirus infection. Blood 2010, 115, 5170–5179. [CrossRef] [PubMed] 80. Jamieson, A.M.; Diefenbach, A.; McMahon, C.W.; Xiong, N.; Carlyle, J.R.; Raulet, D.H. The role of the NKG2D immunoreceptor in immune cell activation and natural killing. Immunity 2002, 17, 19–29. [CrossRef] 81. Liu, W.; Huber, S.A. Cross-talk between CD1D-restricted NKT cells and γ∆cells in T regulatory cell response. Virol. J. 2011, 8, 32. [CrossRef] [PubMed] 80. Jamieson, A.M.; Diefenbach, A.; McMahon, C.W.; Xiong, N.; Carlyle, J.R.; Raulet, D.H. The role of the NKG2D immunoreceptor in immune cell activation and natural killing. Immunity 2002, 17, 19–29. [CrossRef] 81. Liu, W.; Huber, S.A. Cross-talk between CD1D-restricted NKT cells and γ∆cells in T regulatory cell response. Virol. J. 2011, 8, 32. [CrossRef] [PubMed] Int. J. Mol. Sci. 2017, 18, 1997 16 of 18 16 of 18 82. Kagnoff, M.F. Current concepts in mucosal immunity. III. Ontogeny and function of gamma delta t cells in the intestine. Am. J. Physiol. 1998, 274, G455–G458. [PubMed] 83. Lundqvist, C.; Hammarstrom, M.L.; Athlin, L.; Hammarstrom, S. Isolation of functionally active intraepithelial lymphocytes and enterocytes from human small and large intestine. J. Immunol. Methods 1992, 152, 253–263. [CrossRef] 84. Kuhl, A.A.; Loddenkemper, C.; Westermann, J.; Hoffmann, J.C. Role of gamma delta T cells in inflammatory bowel disease. Pathobiology 2002, 70, 150–155. [CrossRef] [PubMed] 85. Cheroutre, H.; Lambolez, F.; Mucida, D. The light and dark sides of intestinal intraepithelial lymphocytes. Nat. Rev. Immunol. 2011, 11, 445–456. [CrossRef] [PubMed] 86. References Ge, L.Q.; Jiang, T.; Zhao, J.; Chen, Z.T.; Zhou, F.; Xia, B. Upregulated mRNA expression of major histocompatibility complex class I chain-related gene A in colon and activated natural killer cells of Chinese patients with ulcerative colitis. J. Dig. Dis. 2011, 12, 82–89. [CrossRef] [PubMed] 7. Lu, M.; Xia, B.; Li, J.; Ye, M.; Zhang, X.; Tan, Q. MICB microsatellite polymorphism is associated w ulcerative colitis in Chinese population. Clin. Immunol. 2006, 120, 199–204. [CrossRef] [PubMed] 98. Li, Y.; Xia, B.; Lu, M.; Ge, L.; Zhang, X. Micb0106 gene polymorphism is associated with ulcerative colitis in central china. Int. J. Colorectal Dis. 2010, 25, 153–159. [CrossRef] [PubMed] 99. Sartor, R.B. Mechanisms of disease: Pathogenesis of Crohn’s disease and ulcerative colitis. Nat. Clin. Pract. Gastroenterol. Hepatol. 2006, 3, 390–407. [CrossRef] [PubMed] 100. Bisping, G.; Lugering, N.; Lutke-Brintrup, S.; Pauels, H.G.; Schurmann, G.; Domschke, W.; Kucharzik, T. Patients with inflammatory bowel disease (IBD) reveal increased induction capacity of intracellular interferon-gamma (IFN-gamma) in peripheral CD8+ lymphocytes co-cultured with intestinal epithelial cells. Clin. Exp. Immunol. 2001, 123, 15–22. [CrossRef] [PubMed] 101. Muller, S.; Lory, J.; Corazza, N.; Griffiths, G.M.; Z’Graggen, K.; Mazzucchelli, L.; Kappeler, A.; Mueller, C. Activated CD4+ and CD8+ cytotoxic cells are present in increased numbers in the intestinal mucosa from patients with active inflammatory bowel disease. Am. J. Pathol. 1998, 152, 261–268. [PubMed] Int. J. Mol. Sci. 2017, 18, 1997 17 of 18 17 of 18 102. Camus, M.; Esses, S.; Pariente, B.; Le Bourhis, L.; Douay, C.; Chardiny, V.; Mocan, I.; Benlagha, K.; Clave, E.; Toubert, A.; et al. Oligoclonal expansions of mucosal t cells in crohn’s disease predominate in NKG2D-expressing CD4 T cells. Mucosal Immunol. 2014, 7, 325–334. [CrossRef] [PubMed] 103. Pariente, B.; Mocan, I.; Camus, M.; Dutertre, C.A.; Ettersperger, J.; Cattan, P.; Gornet, J.M.; Dulphy, N.; Charron, D.; Lemann, M.; et al. Activation of the receptor NKG2D leads to production of Th17 cytokines in CD4+ T cells of patients with crohn’s disease. Gastroenterology 2011, 141, 217–226. [CrossRef] [PubMed] p g p y Charron, D.; Lemann, M.; et al. Activation of the receptor NKG2D leads to production of Th17 cytokines in CD4+ T cells of patients with crohn’s disease. Gastroenterology 2011, 141, 217–226. [CrossRef] [PubMed] p gy 104. Feagan, B.G.; Sandborn, W.J.; D’Haens, G.; Panes, J.; Kaser, A.; Ferrante, M.; Louis, E.; Franchimont, D.; Dewit, O.; Seidler, U.; et al. References Induction therapy with the selective interleukin-23 inhibitor risankizumab in patients with moderate-to-severe crohn’s disease: A randomised, double-blind, placebo-controlled phase 2 study. Lancet 2017, 389, 1699–1709. [CrossRef] 105. Feagan, B.G.; Sandborn, W.J.; Gasink, C.; Jacobstein, D.; Lang, Y.; Friedman, J.R.; Blank, M.A.; Johanns, J.; Gao, L.L.; Miao, Y.; et al. Ustekinumab as induction and maintenance therapy for crohn’s disease. N. Engl. J. Med. 2016, 375, 1946–1960. [CrossRef] [PubMed] 106. Meresse, B.; Chen, Z.; Ciszewski, C.; Tretiakova, M.; Bhagat, G.; Krausz, T.N.; Raulet, D.H.; Lanier, L.L.; Groh, V.; Spies, T.; et al. Coordinated induction by IL15 of a TCR-independent NKG2D signaling pathway converts CTL into lymphokine-activated killer cells in celiac disease. Immunity 2004, 21, 357–366. [CrossRef] [PubMed] 107. Allez, M.; Petryka, R.; Skolnick, B.E.; Wisniewska-Jarosinska, M.A. Mo1213 efficacy and safety of NNC0142-0002, a novel human monoclonal antibody targeting NKG2D: A randomized, double-blind, single-dose phase 2 trial in patients with crohn’s disease. Gastroenterology 2014, 146, S-587. [CrossRef] 108. Vadstrup, K.; Galsgaard, E.D.; Gerwien, J.; Vester-Andersen, M.K.; Pedersen, J.S.; Rasmussen, J.; Neermark, S.; Kiszka-Kanowitz, M.; Jensen, T.; Bendtsen, F. Validation and optimization of an ex vivo assay of intestinal mucosal biopsies in Crohn’s disease: Reflects inflammation and drug effects. PLoS ONE 2016, 11, e0155335. [CrossRef] [PubMed] 109. ClinicalTrials.gov. Safety and efficacy study of jnj-64304500 in participants with moderately to severely active crohn's disease (trident). Available online: https://clinicaltrials.gov/ct2/show/study/NCT02877134?term= JNJ-64304500&rank=2&show_locs=Y#locn(06--06) (accessed on 14 September 2017). 110. ClinicalTrials.gov. A study to investigate the safety, tolerability, pharmacokinetics and pharmacodynamics following subcutaneous injection of jnj-64304500 in healthy japanese and caucasian male participants. Available online: https://clinicaltrials.gov/ct2/show/study/NCT03002025?term=NKG2D&rank=12& view=record(06--06) (accessed on 14 September 2017). 111. Van Dullemen, H.M.; van Deventer, S.J.; Hommes, D.W.; Bijl, H.A.; Jansen, J.; Tytgat, G.N.; Woody, J. Treatment of crohn’s disease with anti-tumor necrosis factor chimeric monoclonal antibody (cA2). Gastroenterology 1995, 109, 129–135. [CrossRef] 112. Ben-Horin, S.; Chowers, Y. Review article: Loss of response to anti-TNF treatments in Crohn’s disease. Aliment. Pharmacol. Ther. 2011, 33, 987–995. [CrossRef] [PubMed] 113. Markiewicz, M.A.; Wise, E.L.; Buchwald, Z.S.; Pinto, A.K.; Zafirova, B.; Polic, B.; Shaw, A.S. Rae1epsilon ligand expressed on pancreatic islets recruits NKG2D receptor-expressing cytotoxic T cells independent of T cell receptor recognition. Immunity 2012, 36, 132–141. [CrossRef] [PubMed] p g y 114. Hart, A.L.; Ng, S.C.; Mann, E.; Al-Hassi, H.O.; Bernardo, D.; Knight, S.C. Homing of immune cells: Role in homeostasis and intestinal inflammation. Inflamm. Bowel Dis. 2010, 16, 1969–1977. [CrossRef] [PubMed] 115. References Hokari, R.; Miura, S.; Fujimori, H.; Koseki, S.; Tsuzuki, Y.; Kimura, H.; Higuchi, H.; Serizawa, H.; Granger, D.N.; Ishii, H. Altered migration of gut-derived T lymphocytes after activation with concanavalin a. Am. J. Physiol. 1999, 277, G763–G772. [PubMed] 116. Ruck, T.; Bittner, S.; Gross, C.C.; Breuer, J.; Albrecht, S.; Korr, S.; Gobel, K.; Pankratz, S.; Henschel, C.M.; Schwab, N.; et al. CD4+NKG2D+ T cells exhibit enhanced migratory and encephalitogenic properties in neuroinflammation. PLoS ONE 2013, 8, e81455. [CrossRef] 117. Mayer, L.; Sandborn, W.J.; Stepanov, Y.; Geboes, K.; Hardi, R.; Yellin, M.; Tao, X.; Xu, L.A.; Salter-Cid, L.; Gujrathi, S.; et al. Anti-IP-10 antibody (bms-936557) for ulcerative colitis: A phase II randomised study. Gut 2014, 63, 442–450. [CrossRef] [PubMed] 118. Villablanca, E.J.; Cassani, B.; von Andrian, U.H.; Mora, J.R. Blocking lymphocyte localization to the gastrointestinal mucosa as a therapeutic strategy for inflammatory bowel diseases. Gastroenterology 2011, 140, 1776–1784. [CrossRef] [PubMed] Int. J. Mol. Sci. 2017, 18, 1997 18 of 18 18 of 18 119. Isernhagen, A.; Schilling, D.; Monecke, S.; Shah, P.; Elsner, L.; Walter, L.; Multhoff, G.; Dressel, R. The mica-129Met/Val dimorphism affects plasma membrane expression and shedding of the NKG2D ligand mica. Immunogenetics 2016, 68, 109–123. [CrossRef] [PubMed] 120. Wittenbrink, M.; Spreu, J.; Steinle, A. Differential NKG2D binding to highly related human NKG2D ligands ULBP2 and RAET1G is determined by a single amino acid in the ALPHA2 domain. Eur. J. Immunol. 2009, 39, 1642–1651. [CrossRef] [PubMed] 121. Espinoza, J.L.; Nguyen, V.H.; Ichimura, H.; Pham, T.T.; Nguyen, C.H.; Pham, T.V.; Elbadry, M.I.; Yoshioka, K.; Tanaka, J.; Trung, L.Q.; et al. A functional polymorphism in the NKG2D gene modulates NK-cell cytotoxicity and is associated with susceptibility to human papilloma virus-related cancers. Sci. Rep. 2016, 6, 39231. [CrossRef] [PubMed] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

https://openalex.org/W1758854654

https://www.intechopen.com/citation-pdf-url/47785

English

High Critical Current Density MgB2

InTech eBooks

cc-by

© 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. High Critical Current Density MgB2 Wenxian Li and Shi-Xue Dou Wenxian Li and Shi-Xue Dou Additional information is available at the end of the chapter http://dx.doi.org/10.5772/59492 1. Introduction The two vibrations along the c-axis are singly degenerate modes of A2u and B1g. For the A2u mode, both Mg and B move in opposite directions along c. For B1g, the B atoms move in opposite directions, while the Mg is stationary. The other two modes along the x or y directions involving only in-plane motions are doubly degenerate. The vibration of Mg and B planes in opposite directions along the x or y directions is E1u mode. Mg atoms are stationary in the E2g mode, while B atoms exhibit a breathing vibration in the x or y directions. This mode is highly anharmonic. The theoretical vibration energy of the E2g mode is around 75 meV [10, 11], which is in agreement with the results from Raman measurements [12, 13]. MgB2 is easy to make into bulk, wire, tape, and thin film forms. However, the critical current density (Jc) of pristine MgB2 drops rapidly in high magnetic field due to the weak pinning forces and low upper critical field (Hc2). Many techniques have been employed to improve the application potential of MgB2, such as chemical doping, irradiation, thermo-mechanical processing, and magnetic shielding. Although the critical current density, upper critical field, and irreversibility field (Hirr) have been greatly increased [14], many difficulties need to be overcome for further application. The origin of the flux pinning force and relevant fabrication techniques will be discussed in this work. Lattice distortion is found to be one of the most effective sources of flux pinning centers in pure MgB2. Then, the combined effects of connec‐ tivity and lattice disorder on the flux pinning force are investigated based on nanosize SiC doped MgB2. The depairing current density, Jd, can be estimated from the Ginzburg-Landau (GL) formula: The depairing current density, Jd, can be estimated from the Ginzburg-Landau (GL) formula: ( ) ( ) ( ) 2 0 0 / 3 / 3 , é ù = F ë û d J T T pm l x (1) (1) (1) where Φ0 is the flux quantum, μ0 the permeability of vacuum, λ the penetration depth, and ξ the coherence length [15]. However, it is not the theoretical maximum [16]. With the help of optimized pinning, about 15% of Jd can be obtained at low magnetic fields in superconductors [17, 18]. 1. Introduction The highest critical transition temperature (Tc) among all the intermetallic superconductors in that was discovered MgB2 has changed the previous approaches to the theory of supercon‐ ductivity because the Tc limit in metallic superconductors had been believed to be ~ 30 K, which is predicted by the Bardeen-Cooper-Schrieffer (BCS) theory [1]. In the BCS theory of super‐ conductivity [2, 3], the expression for Tc is derived as Tc =θe(−1/λeff), where θ is always equal to the MgB2 Debye temperature, θD. λeff is defined as the electron and phonon coupling constant λ = N(EF)× V. N(EF) is the normal state electron number density the Fermi surface, and V is the average electron interaction matrix element corresponding to the attraction. A weak coupling, λ << 1, is assumed to exist between the electrons and phonons in the initial BCS theory. In this case, the value of Tc is limited to Tc ≈ 30 K. According to the BCS theory, an element or compound with larger N(EF), V, and θD has high Tc value. However, θD of MgB2 is comparable to those of other diborides and other light materials. Furthermore, the N(EF) is relatively low because of the absence of d-electrons. Thus, the unusually high Tc in MgB2 has confused researchers with respect to the origin of its superconductivity. Considerable theoretical and experimental work has been conducted to explore the superconducting mechanism in MgB2. MgB2 is the first superconductor to be proved to have two distinct superconducting gaps in its superconducting state [4]. Initially, an unconventional exotic superconducting mechanism was suggested for the material [5, 6]. Then, other researchers proposed hole superconductivity, which is similar to what occurs in high temperature superconductors (HTS), based on the fact that holes are the dominant charge carriers in the normal state [7, 8]. MgB2 has now been accepted as a phonon-mediated BCS type superconductor. The superconductivity is attributed to selective coupling between specific electronic states and specific phonons, such as the E2g mode. The unusually high Tc value arises from the strong phonon anharmonicity. Superconductors – New Developments 96 Choi et al. have calculated the phonon frequencies and electron-phonon interactions from frozen phonon calculations at all the symmetry points of the Brillouin zone [9]. Six non-acoustic modes at the Brillouin zone centre Γ are divided into four distinct phonon modes based on the point symmetry of the lattice. 1. Introduction The concept of connectivity, Acon, has been introduced to quantify the effective cross-section (σeff) for supercurrents [21, 27]: Acon = σeff / σ0, where σ0 is the geometrical cross-section. The connectivity is estimated from the phonon contribution to the normal state resistivity through Acon ≈ Δρtheo/Δρexp, with Δρtheo ≈ 9 μΩ cm. This estimate is based on the assumption that σeff is reduced equivalently in the normal and superconducting state [18]. The supercurrents are limited by the smallest effective cross-section along the conductor. A single large transverse crack can reduce strongly Jc, while only slightly increases the resistivity of a long sample. Unreacted magnesium decreases Δρexp [28] and the effective paths for supercurrents. The thin insulating layers located on grain boundaries can decrease the effective connectivity Acon, inducing high Δρexp, although this kind of defect is transparent to supercurrents. The Δρtheo values within the grains depend on the defects and strain in the grains. Sharma observed negative Δρexp in highly resistive samples [29]. Despite these objections, Acon is very useful for estimating the connectivity of samples if the resistivity is not too high. A clear correlation between the resistivity and the critical current was found in thin films [27]. It should be noted that this procedure is not always reliable, although it offers a possibility for understanding the influences of the connectivity on the supercurrent [18]. The in-situ route is a practical technique to improve the Hc2 and Jc performance of MgB2 using magnesium or MgH2 as the precursor material [30-39], which reacts with boron after mixing and compacting. Low annealing temperature generates MgB2 samples with small grains [25, 40-52] due to the poor crystallinity, and the great amount of grain boundaries result in strong pinning and high Hc2. Magnesium deficient samples can be fabricated by adjusting the stoichiometry of the precursor materials. This method can generate strong lattice strain in MgB2, which decreases Tc, and increases Hc2 [44, 53-56]. On the other hand, an excess magne‐ sium ratio in the raw materials can compensate for the evaporation loss of Mg due to its low melting temperature and the reaction loss due to the reaction with oxygen or the sheath material. The morphology and particle size of the precursor magnesium powders are crucial for the superconductivity and superconducting performance of the final MgB2 [57]. 1. Introduction The high field values for pure MgB2 are λ = 80 nm and ξ = 12 nm, respectively. Jd at zero K is estimated as ~1.3 × 108 A.cm−2. The contribution from π-band charge carriers to the depairing current density is quite low, only about 10% [16], and the interaction energy induces the difference from the high field value. The depairing current is reduced in samples with defects because of the increased λ values [18]. The grain boundaries in MgB2 do not show the weak link effect, and clean grain boundaries are not obstacle to supercurrents [19, 20]. On the other hand, dirty grain boundaries do potentially reduce the critical current [21]. Insulating phases on the grain boundaries, such as MgO, boron oxides [22] or boron carbide [23], normal conducting phases [24], porosity, and cracks [25], can further reduce the cross-section effective of supercurrents. The high porosity in in-situ prepared MgB2 is responsible for its low density, only about half (or less) of its theoretical value [26]. High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 97 The concept of connectivity, Acon, has been introduced to quantify the effective cross-section (σeff) for supercurrents [21, 27]: Acon = σeff / σ0, where σ0 is the geometrical cross-section. The connectivity is estimated from the phonon contribution to the normal state resistivity through Acon ≈ Δρtheo/Δρexp, with Δρtheo ≈ 9 μΩ cm. This estimate is based on the assumption that σeff is reduced equivalently in the normal and superconducting state [18]. The supercurrents are limited by the smallest effective cross-section along the conductor. A single large transverse crack can reduce strongly Jc, while only slightly increases the resistivity of a long sample. Unreacted magnesium decreases Δρexp [28] and the effective paths for supercurrents. The thin insulating layers located on grain boundaries can decrease the effective connectivity Acon, inducing high Δρexp, although this kind of defect is transparent to supercurrents. The Δρtheo values within the grains depend on the defects and strain in the grains. Sharma observed negative Δρexp in highly resistive samples [29]. Despite these objections, Acon is very useful for estimating the connectivity of samples if the resistivity is not too high. A clear correlation between the resistivity and the critical current was found in thin films [27]. It should be noted that this procedure is not always reliable, although it offers a possibility for understanding the influences of the connectivity on the supercurrent [18]. 1. Introduction The grain size of the initial boron powder also has a significant influence on the MgB2 samples. Some researchers have employed ball milling and mechanical alloying of the magnesium and boron mixture, reducing the grain size and enhancing the critical supercurrent [55, 58-65]. The reaction kinetics between magnesium and boron can be modified by chemical or com‐ pound dopants [66], which influence the grain shape and size [67, 68], the secondary phases [69], MgB2 density [70], and the element stoichiometry [71]. Carbon doping is one of the most promising methods to improve the superconducting performance of MgB2. The carbon sources include B4C [72, 73], carbon [52, 66, 67, 74], carbon nanotubes [75-78], nanodiamonds [78, 79], NbC [80], SiC [41, 51, 57, 66, 72, 81-89], and organic compounds [39, 47, 90]. SiC is one of the most promising dopants because it can react with magnesium and boron to form C doped MgB2 at quite low temperatures (600 oC), based on the dual reaction model [66]. Higher processing temperatures are necessary for most of the other carbon sources, leading to grain growth and worse pinning. Comparable results to those with SiC have also obtained, however, with nanoscale carbon powder [91], stearic acid [92], and carbon nanotubes [76]. Superconductors – New Developments 98 The in-situ technique also suffers from its disadvantage of low mass density, which is originates from the annealing process, because the precursor powder has a lower density than MgB2. High pressure synthesis increases the density [93, 94]. On the other hand, the density of ex- situ materials is usually close to the real density of MgB2 [94] and can be further improved by hot isostatic pressing (HIP) [91, 95, 96]. However, the grain size of ex-situ produced materials is comparatively large and inhomogeneous due to the post annealing [97, 98]. Small grains (~ 100–200 nm) have also been reported [99]. A high temperature heat treatment is necessary for the ex-situ process to improve the connectivity [99-102]. This heat treatment leads to recrys‐ tallization and Hc2 reduction [25]. It is concluded that disorder induced by the low temperature processing is insufficient for high superconducting performance of the ex-situ MgB2. Further‐ more, thermally stable defects, such as dopants, are necessary for enhancing the high field performance [18]. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk The connectivity is considered to be a critical issue for improving the Jc of MgB2 based superconductors. Furthermore, an efficient flux pinning force is crucial for high magnetic field application. Here, we employed an in-situ diffusion process to make high density MgB2 bulks and investigate the efficiency of the high flux pinning force induced by lattice thermal strain [103]. Microstructural analysis and Raman scattering measurements were employed to investigate the origins of the huge flux pinning force. Doping nano-SiC particles into MgB2 has been proven to be particularly effective in significantly enhancing Jc, Hirr, and Hc2 [14, 104, 105]. In contrast to chemical doping effects, tensile stress is believed to act as a source of strong flux pinning centers when there is no reaction between SiC and MgB2. Both the Jc and Tc are improved by thermal strain on the interface between SiC and MgB2 during the diffusion process [106] and hybrid physical-chemical vapor deposition (HPCVD) [107]. The influences of the stress field on the flux pinning force and the electron-phonon coupling constant are discussed to clarify the superconducting performance of high density SiC-MgB2 composite fabricated through the diffusion method. Crystalline B with 99.999% purity was pressed into pellets or mixed with 10wt% SiC particles and then pressed into pellets. The pellets were sealed in iron tubes and padded with 99.8% Mg powder. The Mg to B atomic ratio was 1.15:2.0. The diffusion process is time dependente. The sintering condition were 1123 K for 10 h under a flow of high purity argon gas to achieve fully reacted MgB2 bulks. Then the samples were cooled down to room temperature. X-ray diffrac‐ tion (XRD) was employed to characterize the phases, and the results were refined to deter‐ mine the a-axis and c-axis lattice parameters and the lattice distortion. Microstructure observations were performed with a transmission electron microscope (TEM). The magnetic Jc was derived from the height of the magnetization loop ΔM using the Bean model: Jc = 15ΔM/[π a3h)], where a and h are the radius and height of a cylindrical sample. The Tc value could be deduced from the temperature dependence curve of the magnetic susceptibility M(T). To observe the temperature effects of lattice strain, Raman spectra were collected using a confo‐ cal laser Raman spectrometer (Renishaw inVia plus) under a 100× microscope. The excitation laser is Ar+ 514.5 nm. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 99 The density of the pure MgB2 sample is about 1.86 g/cm3, which is about 80% of the theoretical density. This value is much higher than those of the samples made by the in- situ process, which were less than 50%. The SiC-MgB2 composite shows an even higher density of 1.91 g/cm3 due to the SiC addition. Figure 1 shows the Rietveld refinement XRD patterns of the pure MgB2 and the 10wt% SiC doped MgB2 samples. In this case, SiC did not react with magnesium and boron to form C doped MgB2 and Mg2Si. The product is a SiC-MgB2 composite. The Rietveld refinement analysis indicates that the unreacted SiC content was about 9.3wt%, similar to the SiC content in the precursor. This is consistent with the absence of Mg2Si in the XRD pattern as shown in Figure 1. The result is differ‐ ent when the SiC-doped MgB2 prepared by the in-situ technique [14, 41, 84], in which only a very small amount of SiC remaines, while Mg2Si is always present due to the reaction of Mg with SiC. The a- and c-axis lattice parameters were 3.0850 Å and 3.5230 Å for pure MgB2, and 3.0840 Å and 3.5282 Å for SiC doped MgB2, respectively. The a-axis parame‐ ters are virtually equivalent for the two samples, whereas the c-axis one is slightly stretched in the SiC-MgB2 composite. The phenomenon is different from the shortening of the a- axis parameter in in-situ processed SiC doped MgB2, while the c-axis parameter remains unchanged [14, 89]. Figure 1. Rietveld refined XRD patterns of pure MgB2 and 10wt% SiC doped MgB2 samples made by the diffusion process at 850 oC for 10 h. The a-axis parameters are virtually the same for the two samples, whereas the c-axis parame‐ tar is stretched in the SiC-MgB2 composite. Figure 1. Rietveld refined XRD patterns of pure MgB2 and 10wt% SiC doped MgB2 samples made by the diffusion process at 850 oC for 10 h. The a-axis parameters are virtually the same for the two samples, whereas the c-axis parame‐ tar is stretched in the SiC-MgB2 composite. To explain the abnormal c-axis enlargement of the SiC-MgB2 composite, the thermal expansion coefficients, α, of MgB2 and SiC are considered. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk The αSiC decreases slightly from 5×10-6/K at 1123 K to 2.5×10-6/ K at 0 K, whereas the αMgB2 drops quickly from 1.7×10-5/K at 1123 K to zero at 0 K. Based on α(T), the normalized lattice change and lattice strain in the MgB2 matrix of SiC-MgB2 composite during cooling from 1123 K to 0 K can be derived, as shown in Figure 2(b). An assumption of Figure 2(b) is that the two phases are strongly connected and the volume shrinkage of MgB2 is confined by the relatively stable SiC. The normalized lattice strain is estimated to be -0.55% in SiC-MgB2 along the c-axis at room temperature. The negative value corresponds to tensile strain in the MgB2. The large c-axis strain in the doped MgB2 results in an enlargement in the c-axis by 0.15 % in comparison with pure MgB2. As estimated from the Williamson-Hall model [112], the lattice strain is 0.208 and 0.306 along the a-axis, and 0.292 and 1.13 along the c-axis for pure and SiC-MgB2, respectively. The lattice strain along the c-axis in the SiC-MgB2 has increased from that of pure MgB2 by a factor of 4, which is attributed to the high anisotropy in the thermal expansion coefficient of MgB2. α(T), the normalized lattice change and lattice strain in the MgB2 matrix of SiC-MgB2 composite during cooling from 1123 K to 0 K can be derived, as shown in Figure 2(b). An assumption of Figure 2(b) is that the two phases are strongly connected and the volume shrinkage of MgB2 is confined by the relatively stable SiC. The normalized lattice strain is estimated to be -0.55% in SiC-MgB2 along the c-axis at room temperature. The negative value corresponds to tensile strain in the MgB2. The large c-axis strain in the doped MgB2 results in an enlargement in the c-axis by 0.15 % in comparison with pure MgB2. As estimated from the Williamson-Hall model [112], the lattice strain is 0.208 and 0.306 along the a-axis, and 0.292 and 1.13 along the c-axis for pure and SiC-MgB2, respectively. The lattice strain along the c-axis in the SiC-MgB2 has increased from that of pure MgB2 by a factor of 4, which is attributed to the high anisotropy in the thermal expansion coefficient of MgB2. Figure 3. (a) Bright field TEM image of the pure MgB2 with indexed SAD pattern (inset). 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk It is reasonable to assume that both the MgB2 and the SiC are in a stress-free state at the sintering temperature of 1123 K due to the relatively high sintering temperature over a long period of time. However, the lattice parameters are determined by the thermal strain during the cooling process. The temperature dependences of the α values for MgB2 and SiC are especially different. Figure 2(a) plots the α(T) for MgB2 and SiC along the a- and c-axes, based on the data of References [108-111]. It clearly shows the weak temperature dependence ofα(T) for SiC in both directions, whereas the changes are great for MgB2 and are characterized by high anisotropy. The averaged α(T) is also huge in MgB2 as To explain the abnormal c-axis enlargement of the SiC-MgB2 composite, the thermal expansion coefficients, α, of MgB2 and SiC are considered. It is reasonable to assume that both the MgB2 and the SiC are in a stress-free state at the sintering temperature of 1123 K due to the relatively high sintering temperature over a long period of time. However, the lattice parameters are determined by the thermal strain during the cooling process. The temperature dependences of the α values for MgB2 and SiC are especially different. Figure 2(a) plots the α(T) for MgB2 and SiC along the a- and c-axes, based on the data of References [108-111]. It clearly shows the weak temperature dependence ofα(T) for SiC in both directions, whereas the changes are great for MgB2 and are characterized by high anisotropy. The averaged α(T) is also huge in MgB2 as 100 Superconductors – New Developments 100 Figure 2. (a) Thermal expansion coefficients (α) along the a-axis and c-axis for MgB2 and SiC as a function of tempera‐ ture. The averaged α(T) values for MgB2 and SiC are plotted in the inset. (b) Plots of the normalized lattice changes for MgB2 and SiC, and the thermal strain in the matrix during cooling from 1123 K to 0 K. Figure 2. (a) Thermal expansion coefficients (α) along the a-axis and c-axis for MgB2 and SiC as a function of tempera‐ ture. The averaged α(T) values for MgB2 and SiC are plotted in the inset. (b) Plots of the normalized lattice changes for MgB2 and SiC, and the thermal strain in the matrix during cooling from 1123 K to 0 K. shown in the inset of Figure 2(a). 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk (b) High resolution electron mi‐ croscope (HREM) image of interface in pure MgB2 with FFT pattern along the [100] axis (inset). (c) Bright field TEM im‐ age of SiC-MgB2. (d) HREM image of interface in SiC-MgB2 and FFT patterns of SiC and MgB2 from each side of the interface. The dashed line shows the interface of SiC and MgB2. Figure 3. (a) Bright field TEM image of the pure MgB2 with indexed SAD pattern (inset). (b) High resolution electron mi‐ croscope (HREM) image of interface in pure MgB2 with FFT pattern along the [100] axis (inset). (c) Bright field TEM im‐ age of SiC-MgB2. (d) HREM image of interface in SiC-MgB2 and FFT patterns of SiC and MgB2 from each side of the interface. The dashed line shows the interface of SiC and MgB2. High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 101 The unreacted SiC buried in the MgB2 matrix is believed to be one of the most effective sources of strain, and the strongly connected interfaces of SiC and MgB2 are the most effective flux pinning centers. The micro morphologies can be detected using TEM to explore the defects and grain boundaries both in the pure MgB2 and in the SiC-MgB2. Figures 3(a) shows a bright field image of pure MgB2. A high density of defects, such as dislocations and lattice distortion, is observed in the MgB2 phase, and the grain size is about 100 nm,as estimated from the grain boundaries. In contrast to with the highly porous structure in the MgB2 samples [91], the samples made by the diffusion process are well connected with high density. The indexed selected area diffraction (SAD) image shows very pure polycrystalline MgB2. A high resolution grain boundary image is shown in Figure 3(b). The interface is very clean and well connected. The indexed fast Fourier transform (FFT) pattern indicates that the right part parallels the (1 1 0) plane. The micro structure of SiC-MgB2 is similar with that of pure MgB2 with high density of defects. Furthermore, nanosize SiC particle are detected in the MgB2 matrix as indicated in Figure 3(c). Figure 3(d) shows the interface of SiC and MgB2. Based on the FFT analysis, the interface is marked with a dashed line on the image. The left side is a SiC grain paralleling the (1 0 1) plane and the right side is a MgB2 grain paralleling the (0 0 1) plane. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk This kind of interface will impose tensile stress along the c-axis in MgB2 because the thermal expansion coefficient for MgB2 is highly anisotropic in the [001] direction, while that for SiC is nearly isotropic, which is responsible for the enlarged c-axis of MgB2. Figure 4. The magnetic Jc versus field at 5 K, 20 K, and 30 K for pure and nano-SiC doped samples. The inset shows the superconducting transitions of the two samples [103]. Figure 4. The magnetic Jc versus field at 5 K, 20 K, and 30 K for pure and nano-SiC doped samples. The inset shows the superconducting transitions of the two samples [103]. Based on the collective pinning model, [113], Jc is independent of the applied field in the single- vortex pinning regime (low magnetic field region: H < Hsb), where Hsb is the crossover field from single-vortex to small-bundle pinning. The Jc decreases exponentially in the small-bundle regime (high magnetic field: Hsb < H < Hirr). According to the dual model [14], the significant Based on the collective pinning model, [113], Jc is independent of the applied field in the single- vortex pinning regime (low magnetic field region: H < Hsb), where Hsb is the crossover field from single-vortex to small-bundle pinning. The Jc decreases exponentially in the small-bundle regime (high magnetic field: Hsb < H < Hirr). According to the dual model [14], the significant Superconductors – New Developments 102 effect of SiC doping on Jc comes from the high level of C substitution on the B planes, which is responsible for the reduction of the self-field Jc[104, 105]. However, the SiC-MgB2 composite sample shows not only an improved in-field Jc, but also no degradation in self-field Jc, as indicated in Figure 4. The approximate Hsb values are also indicated on the Jc curves for 20 K and 30 K, although Hsb has not been detected at 5 K due to the relatively high supercurrents. The in-situ processed SiC doped MgB2 normally shows a decrease in Tc of 1.5 to 2 K, [14, 104, 105], but this present SiC-MgB2 composite sample features a small drop of 0.6 K, as shown in the inset of Figure 4. This phenomenon is attributed to the absence of any reaction between Mg and SiC, as well as the stretched MgB2 lattice, as indicated by the XRD pattern [107]. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk To investigate whether the lattice strain is significant in SiC-MgB2 during low temperature measurements to obtain M(H) and M(T) curves, Raman spectra were collected before and after the measurements. The Raman spectra for the pure MgB2 are shown in Figure 5(a) and (b) to compare the cooling effects on the matrix. Both the spectra have been fitted with three peaks: ω1, ω2, and ω3 [114-116]. Based on the previous results, ω2 is the reflection of the E2g mode at the Г point of the Brillouin zone in the simple hexagonal MgB2 structure (space group: P6/ mmm), while ω1 and ω3 come from the lattice distortion. The effects of ω1 are not discussed in the following analysis because of its small influence on the spectra. As indicated by the fitting parameters that are shown in Figure 5, both the peak centers and the full width at half maximum (FWHM) values show negligible differences before and after the low temperature measurements because of synchronic volume fluctuation. The weak temperature dependence of the Raman spectra for pure MgB2 is in agreement with the results of Shi et al [117]. The ω2 peak of the Raman spectrum of SiC-MgB2 before the low temperature measurement has shifted to the low frequency of 585 cm-1, as shown in Figure 6(a). The FWHM of the ω2 peak increases from ~200 cm-1 to 210 cm-1. Furthermore, the FWHM of the ω3 peak increases from ~93 cm-1 to 125 cm-1. The variations in both the Raman shift and the FWHM indicate strong lattice strain in the SiC-MgB2 composite. Figure 6(b) shows the cooling effect on the Raman spectrum of SiC-MgB2. The FWHM of the ω2 peak further increases to 228 cm-1, and the frequency of the ω3 peak shifts to 770 cm-1. These results suggest that the stress field is very strong during the low temperature measurements in the SiC-MgB2 composite. Considering the stable defect struc‐ tures in the sample at room temperature and the measurement temperatures, the high Jc performance is attributed to the thermal strain. Although the interface or grain boundaries themselves are effective flux pinning centers, the thermal strain provides more efficient flux pinning force, based on the comparison of the Jc values in pure the MgB2 and the SiC-MgB2 composite. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk It should be noted that the broadened ω2 peak in SiC-MgB2 is a signal of strong electron-E2g coupling, which is responsible for the high Tc in MgB2. The electron-E2g coupling constant is estimated from the Allen equation [118]: Γ2=2πλE2gN (0)ωE2g 2 , where Γ2 is the ω2 linewidth, λE2g is the strength of the electron-E2g coupling, and N(0) is the density of states on the Fermi surface. The experimental phonon frequency and linewidth are simply and directly related to the electron-phonon coupling constant (EPC), λE2g. The total density of states (DOS) at the Fermi energy, EF, of pure MgB2 is taken as 0.354 states/eV/cell/spin. The σ band contribution is 0.15 and the π band is 0.204 [119]. N(0) is assumed to be constant for the small changes of electrons It should be noted that the broadened ω2 peak in SiC-MgB2 is a signal of strong electron-E2g coupling, which is responsible for the high Tc in MgB2. The electron-E2g coupling constant is estimated from the Allen equation [118]: Γ2=2πλE2gN (0)ωE2g 2 , where Γ2 is the ω2 linewidth, λE2g is the strength of the electron-E2g coupling, and N(0) is the density of states on the Fermi surface. The experimental phonon frequency and linewidth are simply and directly related to the High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 103 Figure 5. Fitting and experimental results for the normalized ambient Raman spectrum of MgB2 sintered at 850 oC for 10 h (a), and the cooling effect on the Raman spectrum (b) [103]. Figure 5. Fitting and experimental results for the normalized ambient Raman spectrum of MgB2 sintered at 850 oC for 10 h (a), and the cooling effect on the Raman spectrum (b) [103]. Figure 6. Fitting and experimental results for the normalized ambient Raman spectrum of SiC-MgB2 sintered at 850 oC for 10 h (a), and the cooling effect on the Raman spectrum (b) [103]. Figure 6. Fitting and experimental results for the normalized ambient Raman spectrum of SiC-MgB2 sintered at 850 oC for 10 h (a), and the cooling effect on the Raman spectrum (b) [103]. and holes in MgB2 and SiC-MgB2. Taking the fitting values of the ω2 peaks with cooling effects, the λE2g values for the pure MgB2 and SiC-MgB2 are 2.327 and 2.706, respectively. The λE2g of SiC-MgB2 is just slightly higher than that of the pure MgB2. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk However, the Tc of SiC-MgB2 is slightly decreased compared to that of the pure MgB2. The total EPC constants are degraded by the scattering effects of SiC impurities in the MgB2 matrix, which can be estimated with the and holes in MgB2 and SiC-MgB2. Taking the fitting values of the ω2 peaks with cooling effects, the λE2g values for the pure MgB2 and SiC-MgB2 are 2.327 and 2.706, respectively. The λE2g of SiC-MgB2 is just slightly higher than that of the pure MgB2. However, the Tc of SiC-MgB2 is slightly decreased compared to that of the pure MgB2. The total EPC constants are degraded by the scattering effects of SiC impurities in the MgB2 matrix, which can be estimated with the McMillan formula [120], as modified by Allen and Dynes:[121] Tc = ωlog 1.2 exp( −1.04(1 + λ) λ −μ *(1 + 0.62λ)), where ωlog =(390×ωE2g 2 ×690) is the averaged phonon frequency [122], with 390 and 690 cm-1 McMillan formula [120], as modified by Allen and Dynes:[121] Tc = ωlog 1.2 exp( −1.04(1 + λ) λ −μ *(1 + 0.62λ)), where ωlog =(390×ωE2g 2 ×690) is the averaged phonon frequency [122], with 390 and 690 cm-1 Superconductors – New Developments 104 being the phonon frequencies of the other modes in the MgB2 system [123], μ* is the Coulomb pseudopotential, taken as equal to 0.13 [124], and λ is the total EPC constant. Taking these values, λ is calculated as 0.888 in pure MgB2 and 0.886 in SiC-MgB2, respectively. Although the values are very similar, the λ of MgB2 is a little higher because of its low impurity scattering effects. It should be noted that the λ values were overestimated using the McMillan formula for the two-gap nature of MgB2. However, the overestimations do not change the dependence of λ on the strain effect due to the main contribution of the σ-band. The residual resistivity of SiC-MgB2 is 16 μΩ cm, but it is just 12 μΩ cm for pure MgB2, due to the weak impurity scattering effects. In summary, the thermal strain originating from the interface of SiC and MgB2 is one of the most effective sources of flux pinning centers to improve the supercurrent critical density. The weak temperature dependence of the thermal expansion coefficient of SiC stretches the MgB2 lattice as the temperature decreases. 2. Thermal-strain-induced high Jc in high density SiC-MgB2 bulk The thermal strain supplies much more effective flux pinning force than the interfaces and grain boundaries themselves. The low temperature effects on Raman spectra include very strong lattice stretching at the application temperature of MgB2, which has benefits from both the Jc and the Tc behaviors. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process The self-field critical current density, Jc(0), of MgB2 is much higher than for pure or doped samples processed with in-situ annealing [75, 81, 125-128]. Jc(0) values have been reported as high as 3.5×107 A/cm2 at 4.2 K and 1.6×108 A/cm2 at 2 K in highly connected thin films made by HPCVD [129, 130]. The connectivity is much lower than the theoretical value in in-situ MgB2 because the technique involves a liquid-solid phase reaction process with considerable shrinkage due to the high density of MgB2 compared to the powder mixture of Mg and B [104, 115]. Although the diffusion process could increase the Jc of bulk samples, the Jc(0) and disorder of wires and tapes cannot be improved using similar methods. Several research groups have reported possible techniques to increase the connectivity of in-situ filamentary tapes. High pressure sintering and cold high pressure densification (CHPD) can make it possible to fabricate high density wire and tape samples [96, 131, 132]. Nevertheless, neither HIP nor CHPD are suitable for long MgB2 wires and tapes, and the Jc values are still lower than needed for practical application. Kováč et al. reported a mixed in-situ/ex-situ process to make MgB2 wires, and it was found that the Jc was increased when the ex-situ powder ratio were 23% and 50% [133]. They used commercial MgB2 powder from Alfa Aesar in this process with a very wide grain size distribution ranging from submicrometer size up to 100 μm. In this work, high quality home-made ex-situ powder was used to repeat the mixed in-situ/ex-situ process to develop MgB2/Fe wires with high connectivity and strong disorder to increase both the low and high field Jc properties [134]. The home-made powder was fabricated through low temperature annealing, and the particle size was homogeneous and as small as ~200 nm. The Jc dependence on microstruc‐ High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 105 ture, connectivity, and disorder in MgB2 wires was analyzed based on the microstructure observations. MgB2 wires were fabricated by the powder-in-tube (PIT) process using a ball-milled mixture of Mg (99%) and amorphous B (99%). The in-situ wires were sintered at 750, 850, 950, and 1050 °C for 30 min in high purity Ar and marked as 750in, 850in, 950in, and 1050in, respectively. The in-situ MgB2 powder was sintered at 650 °C for 30 min in high purity Ar flow and then ball-milled. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process Then the PIT method was employed to fabricate in-situ/ex-situ combined MgB2/Fe wires using 1:3 ratio of precursor powders to form a mixture of Mg and B powders. All the green wires were annealed at 750, 850, 950, and 1050 °C for 30 min in high purity Ar and marked as 750inex, 850inex, 950inex, and 1050inex, respectively. The phases and microstructure were characterized by XRD (D/max-2200) and field emission gun scanning electron microscopy (FEG-SEM: JSM-6700F) at room temperature. The super‐ conducting properties were detected from 5 K to 305 K using a Physical Properties Measure‐ ment System (PPMS: Quantum Design). The critical superconducting transition temperature, Tc, is defined as the onset point on the magnetic moment vs. temperature curve, M(T), measured in a magnetic field of 25 Oe. The magnetic Jc was derived from the Bean model. The resistivity dependence on applied magnetic field and temperature, ρ(H, T), was measured using the four- probe method with H from 0 T to 13 T applied perpendicularly to the current direction, from 5 K to 305 K. Hc2 and Hirr were defined as the magnetic field values at 90% and 10% on the superconducting transition on the ρ(H, T) curve, respectively. The active connectivity factor, AF, was calculated based on the ρ(H, T) behavior. Figure 7. Tc dependence on sintering temperatures. The insets display the normalized magnetic moment dependence on measurement temperature for in-situ samples (upper) and for combined in-situ/ex-situ samples (lower) [134]. According to the indexed XRD patterns the samples contain a small amount of MgO The MgO Figure 7. Tc dependence on sintering temperatures. The insets display the normalized magnetic moment dependence on measurement temperature for in-situ samples (upper) and for combined in-situ/ex-situ samples (lower) [134]. According to the indexed XRD patterns, the samples contain a small amount of MgO. The MgO contents are high in 950in, 1050in, and 1050exin compared with the other samples. The broad Superconductors – New Developments 106 transition widths from the normal state to the superconducting state of these samples confirm the high impurity contents, as shown in the insets of Figure 7. The 1050in transition width is ~7 K compared with the width of ~2 K for the other samples, which is attributed to the degraded connectivity of the magnetic flux due to the high MgO content. The in-situ sintered samples show higher Tc values by 0.5 – 1 K compared with the in-situ/ex-situ ones. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process The Tc of 1050in is similar to that of 1050inex because of the high amount of impurity phase. Ball milling of the raw materials induces a great amount of defects in the in-situ/ex-situ samples, which display lower Tc compared with the in-situ samples [135, 136]. The connectivity of the in-situ/ex-situ samples is worse than those of the in-situ samples judging from their big transition widths. The low amount of Mg evaporation may be responsible for the narrow transition width of 1050inex compared with that of 1050in because the stable precursor MgB2 and low quantity of Mg in the raw materials reduce the magnesium loss during the sintering at 1050 oC. Typical SEM images of the in-situ samples are shown in Figure 8. The grain size of 750in is about 300 nm, and the grains show an isolated distribution due to independent growth. 850in, 950in, and 1050in have big clusters of grains because the increasing sintering temperatures have extended the crystal growth time. Some clusters are as big as 1 μm in 1050in. The crystallinity is enhanced at higher sintering temperatures. A high sintering temperature induces raw magnesium evaporation and MgO formation, which can be observed as small white particles without contrast under SEM in 1050in. The crystal shapes for the in-situ/ex-situ samples are irregular with a dispersed distribution of grain sizea, as shown in Figure 9. The grain size of 750inex is small because of the low sintering temperature. The recrystallization effect is triggered for 850inex, 950inex, and 1050inex, and big grains as large as 1 μm are observed. 1050in and 1050inex show similar microstructures because the high sintering temperature provides enough energy and a long time for the crystal growth. Figure 8. SEM images for the in-situ samples sintered at (a) 750 oC, (b) 850 oC, (c) 950 oC, and (d) 1050 oC. The crystal growth was improved gradually with increasing sintering temperature. Figure 8. SEM images for the in-situ samples sintered at (a) 750 oC, (b) 850 oC, (c) 950 oC, and (d) 1050 oC. The crystal growth was improved gradually with increasing sintering temperature. Figure 10 compares the Jc dependence on applied magnetic field measured at 5 K and 20 K. The Jc value of 1050in is the worst because of the high MgO content. The Jc performances of High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 107 Figure 9. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process The inset compares the self-field Jc behavior at 20 K Superconductors – New Developments Superconductors – New Developments 108 A practical quantity to evaluate the connectivity is the active area fraction, AF [21], A practical quantity to evaluate the connectivity is the active area fraction, AF [21], ( ) ideal , 300K D = D F A r r (2) where, ( ) ( ) ideal ideal ideal 300K – 40K 9 cm, D = » W r r r m (3) ( ) ideal , 300K D = D F A r r (2) where, ( ) ( ) ideal ideal ideal 300K – 40K 9 cm, D = » W r r r m (3) ( ) ideal , 300K D = D F A r r (2) (2) (3) is the resistivity of fully connected MgB2 without any disorder [18], and ( ) ( ) ( ) 300 300 – . D = c K T r r r (4) ( ) ( ) ( ) 300 300 – . D = c K T r r r (4) Figure 11. AF dependence on sintering temperature. The short dashed line indicates AF = 0.175. The upper left inset compares the resistivity of the in-situ samples. The upper right inset compares the resistivity of the combined in- situ/ex-situ samples [134]. Figure 11. AF dependence on sintering temperature. The short dashed line indicates AF = 0.175. The upper left inset compares the resistivity of the in-situ samples. The upper right inset compares the resistivity of the combined in- situ/ex-situ samples [134]. Figure 11 compares the AF of all the samples. All the samples show lower connectivity compared with those of ideal crystals, as indicated by the low AF values. The AF value of 750in is just 0.169, and high sintering temperature enhances the connectivity. It is ~ 0.26 for 850in. Higher sintering temperatures than 850 oC can improve the AF slightly, which indicates that the connectivity is easy to improve for in-situ samples. Although the MgO content is high in 1050in, its high AF value is attributed to sufficient crystallization. However, its low magnetic field Jc was degraded by MgO. The combined in-situ/ex-situ samples show rather low AF values compared with the in-situ samples. The phenomenon is consistent with the high resistivity of the in-situ/ex-situ samples, as shown in the insets of Figure 11. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process SEM images for the combined in-situ/ex-situ samples sintered at (a) 750 oC, (b) 850 oC, (c) 950 oC, and (d) 1050 oC. The grains are inhomogeneous due to the ball milled MgB2 precursor and the recrystallization effects. Figure 9. SEM images for the combined in-situ/ex-situ samples sintered at (a) 750 oC, (b) 850 oC, (c) 950 oC, and (d) 1050 oC. The grains are inhomogeneous due to the ball milled MgB2 precursor and the recrystallization effects. 750in, 850in, 750inex, and 1050inex are quite similar over the whole field range. While the Jc deteriorated in 950in and 1050in under higher magnetic fields. 850inex and 950inex show amazingly high Jc properties. The Jc values at 5 K under 8 T magnetic field are around five times higher than those of the best in-situ samples. The inset of Figure 10 compares the low field performances measured at 20 K. The in-situ samples, except for 1050in, show competitive low field Jc. The high crystal quality of 950in is of benefit for its high self-field Jc performance. 750inex and 850inex show low self-field Jc values. While 950inex and 1050inex display improved Jc, the values are lower than the self-field Jc of 950in due to the high MgO contents. The Jc performance depends on the flux pinning mechanism under different magnetic field intensities. The collective pinning model classifies the disorder-induced spatial fluctuations in the vortex lattice into three regimes based on the strength the magnetic field: e.g. single-vortex, small-bundle, large-bundle, and charge-density-wave type relaxation of the vortex lattice [113]. The practicable Jc of MgB2 falls into the single-vortex pinning region and the small-bundle pinning region in the phase diagram. The connectivity determines the Jc performance in the single-vortex pinning regime because the effective charge carrier density is responsible for the self-field Jc, while Hc2 and Hirr are responsible for the Jc performance in the small-bundle regime based on the disorder or defects. Figure 10. Jc dependence on applied magnetic field at 5 K and 20 K. The inset compares the self-field Jc behavior at 20 K [134]. Figure 10. Jc dependence on applied magnetic field at 5 K and 20 K. The inset compares the self-field Jc behavior at 20 K [134]. Jc dependence on applied magnetic field at 5 K and 20 K. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process The gradually improved Figure 11 compares the AF of all the samples. All the samples show lower connectivity compared with those of ideal crystals, as indicated by the low AF values. The AF value of 750in is just 0.169, and high sintering temperature enhances the connectivity. It is ~ 0.26 for 850in. Higher sintering temperatures than 850 oC can improve the AF slightly, which indicates that the connectivity is easy to improve for in-situ samples. Although the MgO content is high in 1050in, its high AF value is attributed to sufficient crystallization. However, its low magnetic field Jc was degraded by MgO. The combined in-situ/ex-situ samples show rather low AF values compared with the in-situ samples. The phenomenon is consistent with the high resistivity of the in-situ/ex-situ samples, as shown in the insets of Figure 11. The gradually improved High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 109 connectivity improves the AF values for the in-situ/ex-situ samples due to the improved solid state reaction. The self-field Jc performances are also improved for 950inex and 1050inex. The Jc performance in the small-bundle region depends on a strong flux pinning force. The flux pinning centers could be lattice distortion, most types of defects, and grain boundaries [110, 137]. The temperature dependences of Hc2 and Hirr determine the strength of the pinning force, as shown in Figure 12. The high field Jc performance is in agreement with the Hc2 and Hirr behavior. The 1050in sample shows the lowest Hc2 and Hirr among all the samples. 850inex and 950inex show the highest Hc2 and Hirr, which are consistent with their high Jc values in the small- bundle region. The ex-situ powder induces strong disorder and proper crystallization, which are responsible for the high Hc2 and Hirr values. Figure 12. Hc2 and Hirr of (a) the in-situ samples and (b) the combined in-situ/ex-situ samples [134]. Figure 12. Hc2 and Hirr of (a) the in-situ samples and (b) the combined in-situ/ex-situ samples [134]. In summary, both connectivity and disorder show strong influences on the Jc properties of MgB2. The connectivity is responsible for the high effective charge carrier density, which determines the self-field Jc performance. The strong flux pinning force induced by defects and disorder is responsible for the promising Jc in high magnetic field. The enhanced Jc perform‐ ances of 850inex and 950inex are attributed to the optimized connectivity and disorder. 3. High connectivity MgB2 wires fabricated by combined in-situ/ex-situ process The Jc values obtained in this work are still far below the Jd value, ~8.7×108 A/cm2 for pure MgB2. The Jc improvement in MgB2 should be explored based on the chemical doping effects and combined in-situ/ex-situ process. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process The combined in-situ/ex-situ process has proved to be a promising technique for the fabrication of practical MgB2 wires. The Jc of MgB2 superconductors has been enhanced through many different kinds of dopants or additives [125], especially different carbon sources [75, 81, Superconductors – New Developments 110 126-128, 138, 139]. In this work, a mixed in-situ/ex-situ technique was employed to develop nano-SiC doped MgB2 wires with high connectivity and strong flux pinning force to increase both the low and high field Jc properties [140]. The SiC particle size is another critical issue for introducing strong flux pinning forces into MgB2. The size of SiC used in this work is smaller than the sizes used in previous research, and the Jc dependence on sintering temperature also shows a very different trend [81, 104, 141, 142]. The powder-in-tube (PIT) process was employed to make practical MgB2 wires from a ball- milled mixture of Mg (99%), B (99%, amorphous), and SiC (< 15 nm).]The sample fabrication and characterization are similar to the techniques mentioned for the pure samples in the last section. Figure 13. XRD patterns of nano-SiC doped MgB2/Fe wires fabricated by (a) the in-situ technique and (b) the in-situ/ex- situ technique, with sintering at 750, 850, 950, and 1050 oC, respectively. All patterns were indexed with MgB2, MgO, Mg2Si, SiC, and Mg [140]. Figure 13. XRD patterns of nano-SiC doped MgB2/Fe wires fabricated by (a) the in-situ technique and (b) the in-situ/ex- situ technique, with sintering at 750, 850, 950, and 1050 oC, respectively. All patterns were indexed with MgB2, MgO, Mg2Si, SiC, and Mg [140]. Figure13 shows the XRD patterns of the two batches of samples. According to the indexed XRD patterns, all samples show quite high purity of MgB2, with only small amounts of MgO and un-reacted Mg and SiC. The un-reacted Mg can be detected because of the high content of SiC in the raw materials [104, 115, 143]. The most interesting phase change relates to the change in the Mg2Si content with sintering temperature. 750in shows very high Mg2Si content, which decreases with increasing sintering temperature and becomes a trace peak in 1050in. However, more than a trace of Mg2Si can only be found in samples sintered at lower temper‐ ature using the combined in-situ/ex-situ method, 750inex and 850inex. The variation of Mg2Si content is an important signal of the C and Si distributions in the MgB2 matrix. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process Figure 14 shows SEM images of 850in and 850inex. The 850in sample contains large slits between MgB2 clusters due to the volume contraction during the in-situ sintering of Mg and B powders. The 850inex sample shows hard-packed MgB2 clusters because the ex-situ precursor is a course of nucleating centers and releases the strain of the in-situ MgB2. The critical transition temperatures (Tc) of the two batches of samples are compared in Figure 15. It is found that the Tc values of the in-situ sintered samples are always slightly lower than those for the samples from the other batch, except for 1050in, and the Tc dependence on sintering temperature of the two batches of samples is exactly the same, which means that the High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 111 Tc depends greatly on the sintering temperature, but not on the different techniques. However, the transition widths from the normal state to the superconducting state are quite different for the two batches of samples, as shown in the inset of Figure 15. The transition widths of the in- situ sintered samples are quite broad compared with those fabricated by the combined technique. The transition width is about 4 K for 750in and becomes 3 K for the 850in and 950in samples sintered at higher temperature due to the high crystallinity. It should be noted that the transition of 1050in shows a two-step behaviour, which may be attributed to the inhomo‐ geneous carbon substitution effect or the inhomogeneous SiC distribution in the raw materials. The transition widths of all the samples made by the in-situ/ex-situ combined technique are 2.5 K, showing behaviour that is independent of the sintering temperature. This means that the crystallinity is increased through the in-situ/ex-situ combined technique because the precursor MgB2 powder is a source of high quality nucleating centres for the newly formed MgB2 during the solid-liquid reaction between the magnesium and the boron. Figure 14. SEM photos of (a) 850in and (b) 850inex [140]. Figure 14. SEM photos of (a) 850in and (b) 850inex [140]. Figure 15. Tc dependence on sintering temperature. The insets show the normalized magnetic moment dependence on the measuring temperature for the in-situ samples (upper) and for the in-situ/ex-situ samples (lower) [140]. Figure 15. Tc dependence on sintering temperature. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process 750in and 750inex show quite low Jc values in lower magnetic field. 1050in and 850inex show competitive self-field Jc. The ball-milling process used to produce the ex-situ MgB2 powder destroyes the porous structure and enhances the density of MgB2 fabricated by the in-situ/ex-situ combined process. This is because of the small particle size of SiC used in this work, which induces different reaction dynamics during the in-situ or in-situ/ex-situ processing [81, 104, 141, 142], so that the present Jc dependence on sintering temperature is quite different from what has been previously reported. It is proposed that the liquid Mg reacts with SiC first to form Mg2Si and releases free C at low sintering temperature. Then the Mg2Si reacts with B to form MgB2 and releases free Si at high sintering temperatures. Both C and Si have very small sizes and cannot be detected by XRD. The coherence length, ξ, of MgB2 is anisotropic. ξab(0) = 3.7 – 12 nm, and ξc(0) = 1.6 - 3.6 nm [110], which is shorter than the particle size of Mg2Si. The Mg2Si particles cannot be effective flux pinning centers, but are rather useless impurities in the MgB2 matrix, which decrease the density of current carriers. However, the free C and Si can be very strong flux pinning centers because of their small sizes, which are responsible for the high Jc performance in high magnetic fields. According to the collective pinning model [113], the Jc performance in the low magnetic field region depends on the density of current carriers due to its weak field dependence, while the high magnetic field Jc performance depends on the flux pinning force due to the increased high Hc2 and Hirr. The approximate Hsb values are also indicated on the Jc curves estimated at 20 K, as shown in the inset of Figure 16, where Hsb is the crossover field from single-vortex to small-bundle pinning based on the collective pinning model. However, Hsb has not been detected at 5 K due to the relatively high supercurrents [103]. The strength of the pinning force can be reflected by the dependence of Hc2 and Hirr on the normalized temperature, as shown in Figure 17. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process ξab(0) = 3.7 – 12 nm, and ξc(0) = 1.6 - 3.6 nm [110], which is shorter than the particle size of Mg2Si. The Mg2Si particles cannot be effective flux pinning centers, but are rather useless impurities in the MgB2 matrix, which decrease the density of current carriers. However, the free C and Si can be very strong flux pinning centers because of their small sizes, which are responsible for the high Jc performance in high magnetic fields. According to the collective pinning model [113], the Jc performance in the low magnetic field region depends on the density of current carriers due to its weak field dependence, while the high magnetic field Jc performance depends on the flux pinning force due to the increased high Hc2 and Hirr. The approximate Hsb values are also indicated on the Jc curves estimated at 20 K, as shown in the inset of Figure 16, where Hsb is the crossover field from single-vortex to small-bundle pinning based on the collective pinning model. However, Hsb has not been detected at 5 K due to the relatively high supercurrents [103]. The Jc dependence on the applied field is shown in Figure 16 for typical samples, which were measured at 5 K and 20 K, respectively. It should be noted that the Jc dependence on sintering temperature in this work is totally different from previously reported results, because the solid- liquid reaction dynamics is different due to the small SiC particle size, less than 15 nm, which is much smaller than the particle sizes used before. The Jc benefits from the high sintering temperature. 750in, 850in, and 750inex display non-competitive performance over the whole field range. 1050inex has very high low field Jc values but its Jc deteriorates with increasing magnetic field. The Jc properties of 950in, 850inex, and 950inex show outstandingly high Jc performances among all the samples. It is concluded that the in-situ/ex-situ combined technique only requires a lower sintering temperature to achieve high quality MgB2 wires, which is very important for industrial application in terms of energy saving and equipment simplification. The Jc values of 750inex are double those of 750in at 5 K and 20 K over the measured magnetic field range. The inset of Figure 16 displays the low field performances at 20 K to avoid the influence of the flux jumping effect. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process The insets show the normalized magnetic moment dependence on the measuring temperature for the in-situ samples (upper) and for the in-situ/ex-situ samples (lower) [140]. 112 Superconductors – New Developments The Jc dependence on the applied field is shown in Figure 16 for typical samples, which were measured at 5 K and 20 K, respectively. It should be noted that the Jc dependence on sintering temperature in this work is totally different from previously reported results, because the solid- liquid reaction dynamics is different due to the small SiC particle size, less than 15 nm, which is much smaller than the particle sizes used before. The Jc benefits from the high sintering temperature. 750in, 850in, and 750inex display non-competitive performance over the whole field range. 1050inex has very high low field Jc values but its Jc deteriorates with increasing magnetic field. The Jc properties of 950in, 850inex, and 950inex show outstandingly high Jc performances among all the samples. It is concluded that the in-situ/ex-situ combined technique only requires a lower sintering temperature to achieve high quality MgB2 wires, which is very important for industrial application in terms of energy saving and equipment simplification. The Jc values of 750inex are double those of 750in at 5 K and 20 K over the measured magnetic field range. The inset of Figure 16 displays the low field performances at 20 K to avoid the influence of the flux jumping effect. 750in and 750inex show quite low Jc values in lower magnetic field. 1050in and 850inex show competitive self-field Jc. The ball-milling process used to produce the ex-situ MgB2 powder destroyes the porous structure and enhances the density of MgB2 fabricated by the in-situ/ex-situ combined process. This is because of the small particle size of SiC used in this work, which induces different reaction dynamics during the in-situ or in-situ/ex-situ processing [81, 104, 141, 142], so that the present Jc dependence on sintering temperature is quite different from what has been previously reported. It is proposed that the liquid Mg reacts with SiC first to form Mg2Si and releases free C at low sintering temperature. Then the Mg2Si reacts with B to form MgB2 and releases free Si at high sintering temperatures. Both C and Si have very small sizes and cannot be detected by XRD. The coherence length, ξ, of MgB2 is anisotropic. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process Carbon substitution is one of the most effective methods to improve the Hc2 and Hirr because of the increased scattering by C doping, and the increased scattering can also contribute to decreased Tc and merging of the two gaps [144]. It should be noted that the Hc2 and Hirr for 750in, 850in, and 750inex have their highest values at low temperature, which means a strong flux pinning force. The poor Jc is of these samples attributed to the lower density of current carriers. Both 1050in and 1050inex show the lowest Hc2 and the lowest Hirr among all the samples. 850in, 950in, 850inex, and 950inex show competitive Hc2 and Hirr performances, which are responsible for their high Jc values under high magnetic field. High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 113 Figure 16. Jc at 5 K and 20 K. The inset indicates the Jc behavior in low magnetic field at 20 K. Hsb, the crossover field from single-vortex to small-bundle pinning, is indicated by the dashed-dotted line at its probable position on the Jc curves [140]. Figure 16. Jc at 5 K and 20 K. The inset indicates the Jc behavior in low magnetic field at 20 K. Hsb, the crossover field from single-vortex to small-bundle pinning, is indicated by the dashed-dotted line at its probable position on the Jc curves [140]. Figure 17. Comparison of Hc2 (solid symbols) and Hirr (open symbols) for (a) MgB2/Fe wires doped with nano-SiC and (b) MgB2/Fe wires doped with nano-SiC, with sintering at 750, 850, 950, and 1050 oC, respectively [140]. Figure 17. Comparison of Hc2 (solid symbols) and Hirr (open symbols) for (a) MgB2/Fe wires doped with nano-SiC and (b) MgB2/Fe wires doped with nano-SiC, with sintering at 750, 850, 950, and 1050 oC, respectively [140]. Superconductors – New Developments 114 In conclusion, high sintering temperature can improve the critical current density of small- particle-size SiC doped MgB2. The two-step reactions between Mg, SiC, and B release free C and Si to form strong flux pinning centers. The current carrier density and flux pinning force are important factors in the improvement of the Jc performance of nano-SiC doped MgB2. The current carrier density determines the Jc behavior in the single-vortex regime, where Jc is independent of applied magnetic field. 4. Nano-SiC doped MgB2 wires made by combined In-Situ/Ex-Situ process The flux pinning force determines the Jc performance in the small-bundle pinning regime, where the doping induced defects are believed to act as flux pinning centers. Nano-SiC doped MgB2/Fe wires fabricated by the combined process are worth ongoing research to develop optimized processing parameters for practical purposes. In conclusion, high sintering temperature can improve the critical current density of small- particle-size SiC doped MgB2. The two-step reactions between Mg, SiC, and B release free C and Si to form strong flux pinning centers. The current carrier density and flux pinning force are important factors in the improvement of the Jc performance of nano-SiC doped MgB2. The current carrier density determines the Jc behavior in the single-vortex regime, where Jc is independent of applied magnetic field. The flux pinning force determines the Jc performance in the small-bundle pinning regime, where the doping induced defects are believed to act as flux pinning centers. Nano-SiC doped MgB2/Fe wires fabricated by the combined process are worth ongoing research to develop optimized processing parameters for practical purposes. 5. Conclusions The diffusion method can greatly improve the critical current density compared with the normal technique, which indicates that the critical current density greatly depends on the connectivity of MgB2 grains. The combined process improves the connectivity of MgB2 grains and the compactness of the superconducting core in wires, which induces high critical current density in zero field. The flux pinning force can also be improved by dopants for magnetic field application. Further research could focus on parameter optimization of the combined process to fabricate high quality MgB2 wires. Acknowledgements The authors thank Dr. T. Silver for her useful discussions. This work is supported by the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning, the Australian Research Council (project ID: DP0770205), and Hyper Tech Research Inc. Author details Wenxian Li1,2,3* and Shi-Xue Dou2 Author details Wenxian Li1,2,3* and Shi-Xue Dou2 *Address all correspondence to: [email protected] 1 School of Materials Science & Engineering, Shanghai University, Shanghai, China 2 Institute for Superconducting & Electronic Materials, University of Wollongong, Wollongong, Australia 1 School of Materials Science & Engineering, Shanghai University, Shanghai, China 2 Institute for Superconducting & Electronic Materials, University of Wollongong, Wollongong, Australia 2 Institute for Superconducting & Electronic Materials, University of Wollongong, Wollongong, Australia 3 School of Computing, Engineering & Mathematics, University of Western Sydney, Sydney, Australia High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 115 References References [1] Nagamatsu J, Nakagawa N, Muranaka T, Zenitani Y, and Akimitsu J, Superconduc‐ tivity at 39 K in magnesium diboride, Nature 2001; 410(6824) 63-64. [2] Ginzburg VL, Nobel Lecture: On superconductivity and superfluidity (what I have and have not managed to do) as well as on the "physical minimum" at the beginning of the XXI century, Rev. Mod. Phys. 2004; 76(3) 981-998. [3] Bardeen J, Cooper LN, and Schrieffer JR, Theory of Superconductivity, Phys. Rev. 1957; 108(5) 1175-1204. [4] Canfield PC and Crabtree G, Magnesium diboride: Better late than never, Phys. To‐ day 2003; 56(3) 34-40. [5] Ivanov VA, van den Broek M, and Peeters FM, Strongly interacting sigma-electrons and MgB2 superconductivity, Solid State Commun. 2001; 120(2-3) 53-57. [6] Baskaran G, Resonating-valence-bond contribution to superconductivity in MgB2, Phys. Rev. B 2002; 65(21) 212505. [7] Hirsch JE, Hole superconductivity in MgB2: a high Tc cuprate without Cu, Phys. Lett. A 2001; 282(6) 392-398. [8] Hirsch JE and Marsiglio F, Electron-phonon or hole superconductivity in (MgB2), Phys. Rev. B 2001; 64(14) 144532. [9] Choi HJ, Cohen ML, and Louie SG, Anisotropic Eliashberg theory of MgB2: Tc, iso‐ tope effects, superconducting energy gaps, quasiparticles, and specific heat, Physica C 2003; 385(1-2) 66-74. [10] Yildirim T, Gulseren O, Lynn JW, Brown CM, Udovic TJ, Huang Q, Rogado N, Re‐ gan KA, Hayward MA, Slusky JS, He T, Haas MK, Khalifah P, Inumaru K, and Cava RJ, Giant anharmonicity and nonlinear electron-phonon coupling in MgB2: A com‐ bined first-principles calculation and neutron scattering study, Phys. Rev. Lett. 2001; 87(3) 037001. [11] Liu AY, Mazin II, and Kortus J, Beyond Eliashberg superconductivity in MgB2: An‐ harmonicity, two-phonon scattering, and multiple gaps, Phys. Rev. Lett. 2001; 87(8) 087005. [12] Hlinka J, Gregora I, Pokorny J, Plecenik A, Kus P, Satrapinsky L, and Benacka S, Pho‐ nons in MgB2 by polarized Raman scattering on single crystals, Phys. Rev. B 2001; 64(14) 140503. [13] Goncharov AF, Struzhkin VV, Gregoryanz E, Hu JZ, Hemley RJ, Mao HK, Lapertot G, Bud'ko SL, and Canfield PC, Raman spectrum and lattice parameters of MgB2 as a function of pressure, Phys. Rev. B 2001; 64(10) 100509. 116 Superconductors – New Developments [14] Dou SX, Shcherbakova O, Yeoh WK, Kim JH, Soltanian S, Wang XL, Senatore C, Flu‐ kiger R, Dhalle M, Husnjak O, and Babic E, Mechanism of enhancement in electro‐ magnetic properties of MgB2 by nano SiC doping, Phys. Rev. Lett. References 2007; 98(9) 097002. [15] Tinkham M, Introduction to Superconductivity, 2nd ed., New York: McGraw-Hill,, 1996; 123. [16] Nicol EJ and Carbotte JP, Theory of the critical current in two-band superconductors with application to MgB2, Phys. Rev. B 2005; 72(1) 014520. [17] Arcos DH and Kunchur MN, Suppressed flux motion in magnesium diboride films, Phys. Rev. B 2005; 71(18) 184516. [18] Eisterer M, Magnetic properties and critical currents of MgB2, Supercond. Sci. Tech‐ nol. 2007; 20(12) R47-R73. [19] Finnemore DK, Ostenson JE, Bud'ko SL, Lapertot G, and Canfield PC, Thermody‐ namic and transport properties of superconducting Mg10B2, Phys. Rev. Lett. 2001; 86(11) 2420-2422. [20] Kawano K, Abell JS, Kambara M, Babu NH, and Cardwell DA, Evidence for high in‐ tergranular current flow in a single-phase polycrystalline MgB2 superconductor, Appl. Phys. Lett. 2001; 79(14) 2216-2218. [21] Rowell JM, The widely variable resistivity of MgB2 samples, Supercond. Sci. Technol. 2003; 16(6) R17-R27. [22] Klie RF, Idrobo JC, Browning ND, Regan KA, Rogado NS, and Cava RJ, Direct obser‐ vation of nanometer-scale Mg- and B-oxide phases at grain boundaries in MgB2, Appl. Phys. Lett. 2001; 79(12) 1837-1839. [23] Pogrebnyakov AV, Xi XX, Redwing JM, Vaithyanathan V, Schlom DG, Soukiassian A, Mi SB, Jia CL, Giencke JE, Eom CB, Chen J, Hu YF, Cui Y, and Li Q, Properties of MgB2 thin films with carbon doping, Appl. Phys. Lett. 2004; 85(11) 2017-2019. [24] Hassler W, Birajdar B, Gruner W, Herrmann M, Perner O, Rodig C, Schubert M, Hol‐ zapfel B, Eibl O, and Schultz L, MgB2 bulk and tapes prepared by mechanical alloy‐ ing: influence of the boron precursor powder, Supercond. Sci. Technol. 2006; 19(6) 512-520. [25] Fischer C, Hassler W, Rodig C, Perner O, Behr G, Schubert M, Nenkov K, Eckert J, Holzapfel B, and Schultz L, Critical current densities of superconducting MgB2 tapes prepared on the base of mechanically alloyed precursors, Physica C 2004; 406(1-2) 121-130. [26] Yeoh WK, Kim JH, Horvat J, Dou SX, and Munroe P, Improving flux pinning of MgB2 by carbon nanotube doping and ultrasonication, Supercond. Sci. Technol. 2006; 19(2) L5-L8. High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 117 [27] Rowell JM, Xu SY, Zeng H, Pogrebnyakov AV, Li Q, Xi XX, Redwing JM, Tian W, and Pan XQ, Critical current density and resistivity of MgB2 films, Appl. Phys. Lett. 2003; 83(1) 102-104. References [28] Kim KH, Betts JB, Jaime M, Lacerda AH, Boebinger GS, Jung CU, Kim HJ, Park MS, and Lee SI, Mg as a main source for the diverse magnetotransport properties of MgB2, Phys. Rev. B 2002; 66(2) 020506. [29] Sharma PA, Hur N, Horibe Y, Chen CH, Kim BG, Guha S, Cieplak MZ, and Cheong SW, Percolative superconductivity in Mg1-xB2, Phys. Rev. Lett. 2002; 89(16) 167003. [30] Kumakura H, Kitaguchi H, Matsumoto A, and Yamada H, Upper critical field, irre‐ versibility field, and critical current density of powder-in-tube-processed MgB2/Fe tapes, Supercond. Sci. Technol. 2005; 18(8) 1042-1046. [31] Nakane T, Jiang CH, Mochiku T, Fujii H, Kuroda T, and Kumakura H, Effect of SiC nanoparticle addition on the critical current density of MgB2 tapes fabricated from MgH2, B and MgB2 powder mixtures, Supercond. Sci. Technol. 2005; 18(10) 1337-1341. [32] Hata S, Yoshidome T, Sosiati H, Tomokiyo Y, Kuwano N, Matsumoto A, Kitaguchi H, and Kumakura H, Microstructures of MgB2/Fe tapes fabricated by an powder-in- tube method using MgH2 as a precursor powder, Supercond. Sci. Technol. 2006; 19(2) 161-168. [33] Jiang CH, Hatakeyama H, and Kumakura H, Preparation of MgB2/Fe tapes with im‐ proved Jc property using MgH2 powder and a short pre-annealing and intermediate rolling process, Supercond. Sci. Technol. 2005; 18(5) L17-L22. [34] Fujii H, Togano K, and Kumakura H, Enhancement of critical current densities of powder-in-tube processed MgB2 tapes by using MgH2 as a precursor powder, Super‐ cond. Sci. Technol. 2002; 15(11) 1571-1576. [35] Jiang CH, Nakane T, Hatakeyama H, and Kumakura H, Enhanced Jc property in nano-SiC doped thin MgB2/Fe wires by a modified PIT process, Physica C 2005; 422(3-4) 127-131. [36] Jiang CH, Hatakeyama H, and Kumakura H, Effect of nanometer MgO addition on the PIT processed MgB2/Fe tapes, Physica C 2005; 423(1-2) 45-50. [37] Matsumoto A, Kumakura H, Kitaguchi H, and Hatakeyama H, Effect of SiO2 and SiC doping on the powder-in-tube processed MgB2 tapes, Supercond. Sci. Technol. 2003; 16(8) 926-930. [38] Pachla W, Morawski A, Kovac P, Husek I, Mazur A, Lada T, Diduszko R, Melisek T, Strbik V, and Kulczyk M, Properties of hydrostatically extruded in situ MgB2 wires doped with SiC, Supercond. Sci. Technol. 2006; 19(1) 1-8. 118 Superconductors – New Developments [39] Yamada H, Hirakawa M, Kumakura H, and Kitaguchi H, Effect of aromatic hydro‐ carbon addition on in situ powder-in-tube processed MgB2 tapes, Supercond. Sci. Technol. 2006; 19(2) 175-177. References [40] Goldacker W, Schlachter SI, Obst B, Liu B, Reiner J, and Zimmer S, Development and performance of thin steel reinforced MgB2 wires and low-temperature processing for further improvements, Supercond. Sci. Technol. 2004; 17(5) S363-S368. [41] Matsumoto A, Kumakura H, Kitaguchi H, Senkowicz BJ, Jewell MC, Hellstrom EE, Zhu Y, Voyles PM, and Larbalestier DC, Evaluation of connectivity, flux pinning, and upper critical field contributions to the critical current density of bulk pure and SiC-alloyed MgB2, Appl. Phys. Lett. 2006; 89(13) 132508. [42] Yamamoto A, Shimoyama J, Ueda S, Iwayama I, Horii S, and Kishio K, Effects of B4C doping on critical current properties of MgB2 superconductor, Supercond. Sci. Tech‐ nol. 2005; 18(10) 1323-1328. [43] Perner O, Habler W, Eckert R, Fischer C, Mickel C, Fuchs G, Holzapfel B, and Schultz L, Effects of oxide particle addition on superconductivity in nanocrystalline MgB2 bulk samples, Physica C 2005; 432(1-2) 15-24. [44] Jiang CH, Nakane T, and Kumakura H, Superior high-field current density in slight‐ ly Mg-deficient MgB2 tapes, Appl. Phys. Lett. 2005; 87(25) 252505. [45] Wu YF, Lu YF, Yan G, Li JS, Feng Y, Tang HP, Chen SK, Xu HL, Li CS, and Zhang PX, Improved superconducting properties in bulk MgB2 prepared by high-energy milling of Mg and B powders, Supercond. Sci. Technol. 2006; 19(11) 1215-1218. [46] Kim JH, Yeoh WK, Qin MJ, Xu X, and Dou SX, The doping effect of multiwall carbon nanotube on MgB2/Fe superconductor wire, J. Appl. Phys. 2006; 100(1) 013908. [47] Kim JH, Zhou S, Hossain MSA, Pan AV, and Dou SX, Carbohydrate doping to en‐ hance electromagnetic properties of MgB2 superconductors, Appl. Phys. Lett. 2006; 89(14) 142505. [48] Chen SK, Lockman Z, Wei M, Glowacki BA, and MacManus-Driscoll JL, Improved current densities in MgB2 by liquid-assisted sintering, Appl. Phys. Lett. 2005; 86(24) 242501. [49] Ueda S, Shimoyama J, Iwayama I, Yamamoto A, Katsura Y, Horii S, and Kishio K, High critical current properties of MgB2 bulks prepared by a diffusion method, Appl. Phys. Lett. 2005; 86(22) 222502. [50] Zhang XP, Gao ZS, Wang DL, Yu ZG, Ma YW, Awaji S, and Watanabe K, Improved critical current densities in MgB2 tapes with ZrB2 doping, Appl. Phys. Lett. 2006; 89(13) 132510. References High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 119 [51] Shcherbakova O, Dou SX, Soltanian S, Wexler D, Bhatia M, Sumption M, and Col‐ lings EW, The effect of doping level and sintering temperature on Jc(H) performance in nano-SiC doped and pure MgB2 wires, J. Appl. Phys. 2006; 99(8) 08M510. [52] Ma YW, Zhang XP, Nishijima G, Watanabe K, Awaji S, and Bai XD, Significantly en‐ hanced critical current densities in MgB2 tapes made by a scaleable nanocarbon addi‐ tion route, Appl. Phys. Lett. 2006; 88(7) 072502. [53] Ribeiro RA, Bud'ko SL, Petrovic C, and Canfield PC, Effects of boron purity, Mg stoi‐ chiometry and carbon substitution on properties of polycrystalline MgB2, Physica C 2003; 385(1-2) 16-23. [54] Liao XZ, Serquis A, Zhu YT, Peterson DE, Mueller FM, and Xu HF, Strain effect on the critical superconducting temperature of MgB2, Supercond. Sci. Technol. 2004; 17(8) 1026-1030. [55] Perner O, Eckert J, Hassler W, Fischer C, Acker J, Gemming T, Fuchs G, Holzapfel B, and Schultz L, Stoichiometry dependence of superconductivity and microstructure in mechanically alloyed MgB2, J. Appl. Phys. 2005; 97(5) 056105. [56] Serquis A, Zhu YT, Peterson EJ, Coulter JY, Peterson DE, and Mueller FM, Effect of lattice strain and defects on the superconductivity of MgB2, Appl. Phys. Lett. 2001; 79(26) 4399-4401. [57] Yamada H, Hirakawa M, Kumakura H, Matsumoto A, and Kitaguchi H, Critical cur‐ rent densities of powder-in-tube MgB2 tapes fabricated with nanometer-size Mg powder, Appl. Phys. Lett. 2004; 84(10) 1728-1730. [58] Fang H, Padmanabhan S, Zhou YX, and Salama K, High critical current density in iron-clad MgB2 tapes, Appl. Phys. Lett. 2003; 82(23) 4113-4115. [59] Fischer C, Rodig C, Hassler W, Perner O, Eckert J, Nenkov K, Fuchs G, Wendrock H, Holzapfel B, and Schultz L, Preparation of MgB2 tapes using a nanocrystalline parti‐ ally reacted precursor, Appl. Phys. Lett. 2003; 83(9) 1803-1805. [60] Strickland NM, Buckley RG, and Otto A, High critical current densities in Cu-sheath‐ ed MgB2 formed from a mechanically-alloyed precursor, Appl. Phys. Lett. 2003; 83(2) 326-328. [61] Flukiger R, Suo HL, Musolino N, Beneduce C, Toulemonde P, and Lezza P, Super‐ conducting properties of MgB2 tapes and wires, Physica C 2003; 385(1-2) 286-305. [62] Goldacker W, Schlachter SI, Liu B, Obst B, and Klimenko E, Considerations on criti‐ cal currents and stability of MgB2 wires made by different preparation routes, Physi‐ ca C 2004; 401(1-4) 80-86. References [63] Flukiger R, Lezza P, Beneduce C, Musolino N, and Suo HL, Improved transport criti‐ cal current and irreversibility fields in mono- and multifilamentary Fe/MgB2 tapes and wires using fine powders, Supercond. Sci. Technol. 2003; 16(2) 264-270. 120 Superconductors – New Developments [64] Grovenor CRM, Goodsir L, Salter CJ, Kovac P, and Husek I, Interfacial reactions and oxygen distribution in MgB2 wires in Fe, stainless steel and Nb sheaths, Supercond. Sci. Technol. 2004; 17(3) 479-484. [65] Senkowicz BJ, Giencke JE, Patnaik S, Eom CB, Hellstrom EE, and Larbalestier DC, Improved upper critical field in bulk-form magnesium diboride by mechanical alloy‐ ing with carbon, Appl. Phys. Lett. 2005; 86(20) 202502. [66] Dou SX, Shcherbakova O, Yoeh WK, Kim JH, Soltanian S, Wang XL, Senatore C, Flu‐ kiger R, Dhalle M, Husnjak O, and Babic E, Mechanism of enhancement in electro‐ magnetic properties of MgB2 by nano SiC doping, Phys. Rev. Lett. 2007; 98(9) 097002. [67] Zhao Y, Feng Y, Shen TM, Li G, Yang Y, and Cheng CH, Cooperative doping effects of Ti and C on critical current density and irreversibility field of MgB2, J. Appl. Phys. 2006; 100(12) 123902. [68] Xu X, Kim JH, Yeoh WK, Zhang Y, and Dou SX, Improved Jc of MgB2 superconductor by ball milling using different media, Supercond. Sci. Technol. 2006; 19(11) L47-L50. [69] Haigh S, Kovac P, Prikhna TA, Savchuk YM, Kilburn MR, Salter C, Hutchison J, and Grovenor C, Chemical interactions in Ti doped MgB2 superconducting bulk samples and wires, Supercond. Sci. Technol. 2005; 18(9) 1190-1196. [70] Ma YW, Kumakura H, Matsumoto A, Hatakeyama H, and Togano K, Improvement of critical current density in Fe-sheathed MgB2 tapes by ZrSi2, ZrB2 and WSi2 doping, Supercond. Sci. Technol. 2003; 16(8) 852-856. [71] Kumar D, Pennycook SJ, Narayan J, Wang H, and Tiwari A, Role of silver addition in the synthesis of high critical current density MgB2 bulk superconductors, Supercond. Sci. Technol. 2003; 16(4) 455-458. [72] Yamamoto A, Shimoyama J, Ueda S, Katsura Y, Iwayama I, Horii S, and Kishio K, Universal relationship between crystallinity and irreversibility field of MgB2, Appl. Phys. Lett. 2005; 86(21) 212502. [73] Lezza P, Senatore C, and Flukiger R, Improved critical current densities in B4C doped MgB2 based wires, Supercond. Sci. Technol. 2006; 19(10) 1030-1033. References [74] Yeoh WK, Kim JH, Horvat J, Xu X, Qin MJ, Dou SX, Jiang CH, Nakane T, Kumakura H, and Munroe P, Control of nano carbon substitution for enhancing the critical cur‐ rent density in MgB2, Supercond. Sci. Technol. 2006; 19(6) 596-599. [75] Dou SX, Yeoh WK, Horvat J, and Ionescu M, Effect of carbon nanotube doping on critical current density of MgB2 superconductor, Appl. Phys. Lett. 2003; 83(24) 4996-4998. [76] Kim JH, Yeoh WK, Qin MJ, Xu X, Dou SX, Munroe P, Kumakura H, Nakane T, and Jiang CH, Enhancement of in-field Jc in MgB2/Fe wire using single- and multiwalled carbon nanotubes, Appl. Phys. Lett. 2006; 89(12) 122510. 121 High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 121 [77] Kovac P, Husek I, Skakalova V, Meyer J, Dobrocka E, Hirscher M, and Roth S, Trans‐ port current improvements of in situ MgB2 tapes by the addition of carbon nano‐ tubes, silicon carbide or graphite, Supercond. Sci. Technol. 2007; 20(1) 105-111. [78] Cheng CH, Yang Y, Munroe P, and Zhao Y, Comparison between nano-diamond and carbon nanotube doping effects on critical current density and flux pinning in MgB2, Supercond. Sci. Technol. 2007; 20(3) 296-301. [79] Cheng CH, Zhang H, Zhao Y, Feng Y, Rui XF, Munroe P, Zeng HM, Koshizuka N, and Murakami M, Doping effect of nano-diamond on superconductivity and flux pinning in MgB2, Supercond. Sci. Technol. 2003; 16(10) 1182-1186. [80] Gao ZS, Zhang XP, Wang DL, Liu X, Li XH, Ma YW, and Mossang E, Effects of NbC addition on the critical current density of MgB2 tapes, Supercond. Sci. Technol. 2007; 20(1) 57-61. [81] Dou SX, Soltanian S, Horvat J, Wang XL, Zhou SH, Ionescu M, Liu HK, Munroe P, and Tomsic M, Enhancement of the critical current density and flux pinning of MgB2 superconductor by nanoparticle SiC doping, Appl. Phys. Lett. 2002; 81(18) 3419-3421. [82] Soltanian S, Wang XL, Horvat J, Dou SX, Sumption MD, Bhatia M, Collings EW, Munroe P, and Tomsic M, High transport critical current density and large Hc2 and Hirr in nanoscale SiC doped MgB2 wires sintered at low temperature, Supercond. Sci. Technol. 2005; 18(5) 658-666. [83] Chen SK, Tan KS, Glowacki BA, Yeoh WK, Soltanian S, Horvat J, and Dou SX, Effect of heating rates on superconducting properties of pure MgB2, carbon nanotube- and nano-SiC-doped MgB2/Fe wires, Appl. Phys. Lett. 2005; 87(18) 182504. References [84] Dou SX, Braccini V, Soltanian S, Klie R, Zhu Y, Li S, Wang XL, and Larbalestier D, Nanoscale-SiC doping for enhancing Jc and Hc2 in superconducting MgB2, J. Appl. Phys. 2004; 96(12) 7549-7555. [85] Ma YW, Zhang XP, Xu AX, Li XH, Xiao LY, Nishijima G, Awaji S, Watanabe K, Jiao YL, Xiao L, Bai XD, Wu KH, and Wen HH, The effect of ZrSi2 and SiC doping on the microstructure and Jc-B properties of PIT processed MgB2 tapes, Supercond. Sci. Technol. 2006; 19(1) 133-137. [86] Sumption MD, Bhatia M, Rindfleisch M, Tomsic M, and Collings EW, Transport and magnetic Jc of MgB2 strands and small helical coils, Appl. Phys. Lett. 2005; 86(10) 102501. [87] Kumakura H, Kitaguchi H, Matsumoto A, and Hatakeyama H, Upper critical fields of powder-in-tube-processed MgB2/Fe tape conductors, Appl. Phys. Lett. 2004; 84(18) 3669-3671. [88] Li S, White T, Laursen K, Tan TT, Sun CQ, Dong ZL, Li Y, Zho SH, Horvat J, and Dou SX, Intense vortex pinning enhanced by semicrystalline defect traps in self-aligned nanostructured MgB2, Appl. Phys. Lett. 2003; 83(2) 314-316. 122 Superconductors – New Developments [89] Sumption MD, Bhatia M, Rindfleisch M, Tomsic M, Soltanian S, Dou SX, and Col‐ lings EW, Large upper critical field and irreversibility field in MgB2 wires with SiC additions, Appl. Phys. Lett. 2005; 86(9) 092507. [90] Hossain MSA, Kim JH, Wang XL, Xu X, Peleckis G, and Dou SX, Enhancement of flux pinning in a MgB2 superconductor doped with tartaric acid, Supercond. Sci. Technol. 2007; 20(1) 112-116. [91] Liao XZ, Serquis A, Zhu YT, Civale L, Hammon DL, Peterson DE, Mueller FM, Nes‐ terenko VF, and Gu Y, Defect structures in MgB2 wires introduced by hot isostatic pressing, Supercond. Sci. Technol. 2003; 16(7) 799-803. [92] Gao ZS, Ma YW, Zhang XP, Wang DL, Yu ZG, Watanabe K, Yang HA, and Wen HH, Strongly enhanced critical current density in MgB2/Fe tapes by stearic acid and stea‐ rate doping, Supercond. Sci. Technol. 2007; 20(5) 485-489. [93] Wen HH, Li SL, Zhao ZW, Jin H, Ni YM, Ren ZA, Che GC, and Zhao ZX, Magnetic relaxation and critical current density of the new superconductor MgB2, Supercond. Sci. Technol. 2002; 15(3) 315-319. References [94] Prikhna TA, Gawalek W, Savchuk YM, Moshchil VE, Sergienko NV, Habisreuther T, Wendt M, Hergt R, Schmidt C, Dellith J, Melnikov VS, Assmann A, Litzkendorf D, and Nagorny PA, High-pressure synthesis of MgB2 with addition of Ti, Physica C 2004; 402(3) 223-233. [95] Serquis A, Liao XZ, Zhu YT, Coulter JY, Huang JY, Willis JO, Peterson DE, Mueller FM, Moreno NO, Thompson JD, Nesterenko VF, and Indrakanti SS, Influence of mi‐ crostructures and crystalline defects on the superconductivity of MgB2, J. Appl. Phys. 2002; 92(1) 351-356. [96] Serquis A, Civale L, Hammon DL, Liao XZ, Coulter JY, Zhu YT, Jaime M, Peterson DE, Mueller FM, Nesterenko VF, and Gu Y, Hot isostatic pressing of powder in tube MgB2 wires, Appl. Phys. Lett. 2003; 82(17) 2847-2849. [97] Eyidi D, Eibl O, Wenzel T, Nickel KG, Schlachter SI, and Goldacker W, Supercon‐ ducting properties, microstructure and chemical composition of MgB2 sheathed ma‐ terials, Supercond. Sci. Technol. 2003; 16(7) 778-788. [98] Pan AV, Zhou SH, Liu HK, and Don SX, Properties of superconducting MgB2 wires: versus reaction technique, Supercond. Sci. Technol. 2003; 16(5) 639-644. [99] Serquis A, Civale L, Hammon DL, Coulter JY, Liao XZ, Zhu YT, Peterson DE, and Mueller FM, Microstructure and high critical current of powder-in-tube MgB2, Appl. Phys. Lett. 2003; 82(11) 1754-1756. [100] Serquis A, Civale L, Hammon DL, Liao XZ, Coulter JY, Zhu YT, Peterson DE, and Mueller FM, Role of excess Mg and heat treatments on microstructure and critical current of MgB2 wires, J. Appl. Phys. 2003; 94(6) 4024-4031. High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 123 [101] Suo HL, Beneduce C, Dhalle M, Musolino N, Genoud JY, and Flukiger R, Large transport critical currents in dense Fe- and Ni-clad MgB2 superconducting tapes, Appl. Phys. Lett. 2001; 79(19) 3116-3118. [102] Grasso G, Malagoli A, Ferdeghini C, Roncallo S, Braccini V, Siri AS, and Cimberle MR, Large transport critical currents in unsintered MgB2 superconducting tapes, Appl. Phys. Lett. 2001; 79(2) 230-232. [103] Li WX, Zeng R, Lu L, and Dou SX, Effect of thermal strain on Jc and Tc in high density nano-SiC doped MgB2, J. Appl. Phys. 2011; 109(7) 07E108. [104] Li WX, Zeng R, Lu L, Li Y, and Dou SX, The combined influence of connectivity and disorder on Jc and Tc performances in MgxB2+10 wt % SiC, J. Appl. Phys. 2009; 106(9) 093906. References [105] Li WX, Zeng R, Lu L, Zhang Y, Dou SX, Li Y, Chen RH, and Zhu MY, Improved su‐ perconducting properties of powder-in-tube processed Mg1.15B2/Fe wires with nano- size SiC addition, Physica C 2009; 469(15-20) 1519-1522. [106] Zeng R, Dou SX, Lu L, Li WX, Kim JH, Munroe P, Zheng RK, and Ringer SP, Ther‐ mal-strain-induced enhancement of electromagnetic properties of SiC-MgB2 compo‐ sites, Appl. Phys. Lett. 2009; 94(4) 042510. [107] Pogrebnyakov AV, Redwing JM, Raghavan S, Vaithyanathan V, Schlom DG, Xu SY, Li Q, Tenne DA, Soukiassian A, Xi XX, Johannes MD, Kasinathan D, Pickett WE, Wu JS, and Spence JCH, Enhancement of the superconducting transition temperature of MgB2 by a strain-induced bond-stretching mode softening, Phys. Rev. Lett. 2004; 93(14) 147006. [108] Neumeier JJ, Tomita T, Debessai M, Schilling JS, Barnes PW, Hinks DG, and Jorgen‐ sen JD, Negative thermal expansion of MgB2 in the superconducting state and anom‐ alous behavior of the bulk Gruneisen function, Phys. Rev. B 2005; 72(22) 220505. [109] Li Z and Bradt RC, Thermal expansion of the hexagonal (4H) polytype of SiC, J. Appl. Phys. 1986; 60(2) 612-614. [110] Buzea C and Yamashita T, Review of the superconducting properties of MgB2, Super‐ cond. Sci. Technol. 2001; 14(11) R115-R146. [111] Jorgensen JD, Hinks DG, and Short S, Lattice properties of MgB2 versus temperature and pressure, Phys. Rev. B 2001; 6322(22) 224522. [112] Williamson GK and Hall WH, X-ray line broadening from filed aluminium and wolf‐ ram, Acta Metall. Mater. 1953; 1)22-31. [113] Blatter G, Feigelman MV, Geshkenbein VB, Larkin AI, and Vinokur VM, Vortices in high-temperature superconductors, Rev. Mod. Phys. 1994; 66(4) 1125-1388. 124 Superconductors – New Developments [114] Li WX, Li Y, Chen RH, Zeng R, Zhu MY, Jin HM, and Dou SX, Electron-phonon cou‐ pling properties in MgB2 observed by Raman scattering, J. Phys.-Condens. Matter 2008; 20(25) 255235. [115] Li WX, Li Y, Chen RH, Zeng R, Dou SX, Zhu MY, and Jin HM, Raman study of ele‐ ment doping effects on the superconductivity of MgB2, Phys. Rev. B 2008; 77(9) 094517. [116] Li WX, Zeng R, Poh CK, Li Y, and Dou SX, Magnetic scattering effects in two-band superconductor: the ferromagnetic dopants in MgB2, J. Phys.-Condens. Matter 2010; 22(13) 135701. [117] Shi L, Zhang HR, Chen L, and Feng Y, The Raman spectrum and lattice parameters of MgB2 as a function of temperature, J. Phys.-Condens. Matter 2004; 16(36) 6541-6550. References [118] Allen PB, Neutron spectroscopy of superconductors, Phys. Rev. B 1972; 6(7) 2577-2579. [119] Kortus J, Dolgov OV, Kremer RK, and Golubov AA, Band filling and interband scat‐ tering effects in MgB2: Carbon versus aluminum doping, Phys. Rev. Lett. 2005; 94(2) 027002. [120] McMillan WL, Transition temperature of strong-coupled superconductors, Phys. Rev. 1968; 167(2) 331-334. [121] Allen PB and Dynes RC, Transition-temperature of strong-coupled superconductors reanalyzed, Phys. Rev. B 1975; 12(3) 905-922. [122] Kortus J, Mazin II, Belashchenko KD, Antropov VP, and Boyer LL, Superconductivi‐ ty of metallic boron in MgB2, Phys. Rev. Lett. 2001; 86(20) 4656-4659. [123] Osborn R, Goremychkin EA, Kolesnikov AI, and Hinks DG, Phonon density of states in MgB2, Phys. Rev. Lett. 2001; 87(1) 017005. [124] Brinkman A, Golubov AA, Rogalla H, Dolgov OV, Kortus J, Kong Y, Jepsen O, and Andersen OK, Multiband model for tunneling in MgB2 junctions, Phys. Rev. B 2002; 65(18) 180517. [125] Collings EW, Sumption MD, Bhatia M, Susner MA, and Bohnenstiehl SD, Prospects for improving the intrinsic and extrinsic properties of magnesium diboride supercon‐ ducting strands, Supercond. Sci. Technol. 2008; 21(10) 103001. [126] Li WX, Li Y, Chen RH, Yeoh WK, and Dou SX, Effect of magnetic field processing on the microstructure of carbon nanotubes doped MgB2, Physica C 2007; 4601)570-571. [127] Yeoh WK, Horvat J, Dou SX, and Munroe P, Effect of carbon nanotube size on super‐ conductivity properties of MgB2, IEEE Trans. Appl. Supercond. 2005; 15(2) 3284-3287. High Critical Current Density MgB2 http://dx.doi.org/10.5772/59492 125 [128] Li WX, Li Y, Zhu MY, Chen RH, Xu X, Yeoh WK, Kim JH, and Dou SX, Benzoic acid doping to enhance electromagnetic properties of MgB2 superconductors, IEEE Trans. Appl. Supercond. 2007; 17(2) 2778-2781. [129] Zhuang CG, Meng S, Zhang CY, Feng QR, Gan ZZ, Yang H, Jia Y, Wen HH, and Xi XX, Ultrahigh current-carrying capability in clean MgB2 films, J. Appl. Phys. 2008; 104(1) 013924. [130] Zeng XH, Pogrebnyakov AV, Zhu MH, Jones JE, Xi XX, Xu SY, Wertz E, Li Q, Redw‐ ing JM, Lettieri J, Vaithyanathan V, Schlom DG, Liu ZK, Trithaveesak O, and Schu‐ bert J, Superconducting MgB2 thin films on silicon carbide substrates by hybrid physical-chemical vapor deposition, Appl. Phys. Lett. 2003; 82(13) 2097-2099. [131] Flukiger R, Hossain MSA, and Senatore C, Strong enhancement of Jc and Birr in bina‐ ry MgB2 wires after cold high pressure densification, Supercond. Sci. Technol. 2009; 22(8) 085002. References [132] Hossain MSA, Senatore C, Flukiger R, Rindfleisch MA, Tomsic MJ, Kim JH, and Dou SX, The enhanced Jc and Birr of MgB2 wires and tapes alloyed with C4H6O5 (malic acid) after cold high pressure densification, Supercond. Sci. Technol. 2009; 22(9) 095004. [133] Kovac P, Reissner M, Melisek T, Husek I, and Mohammad S, Current densities of MgB2 wires by combined ex process, J. Appl. Phys. 2009; 106(1) 013910. [134] Li WX, Zeng R, Zhang Y, Xu X, Li Y, and Dou SX, Evolution of electromagnetic prop‐ erties and microstructure with sintering temperature for MgB2/Fe wires made by combined In-Situ/Ex-Situ process, IEEE Trans. Appl. Supercond. 2011; 21(3) 2635-2638. [135] Xu X, Kim JH, Dou SX, Choi S, Lee JH, Park HW, Rindfleish M, and Tomsic M, A correlation between transport current density and grain connectivity in MgB2/Fe wire made from ball-milled boron, J. Appl. Phys. 2009; 105(10) 103913. [136] Romano G, Vignolo M, Braccini V, Malagoli A, Bernini C, Tropeano M, Fanciulli C, Putti M, and Ferdeghini C, High-energy ball milling and synthesis temperature study to improve superconducting properties of MgB2 ex-situ tapes and wires, IEEE Trans. Appl. Supercond. 2009; 19(3) 2706-2709. [137] Li WX, Chen RH, Li Y, Zhu MY, Jin HM, Zeng R, Dou SX, and Lu B, Raman study on the effects of sintering temperature on the Jc(H) performance of MgB2 superconduc‐ tor, J. Appl. Phys. 2008; 103(1) 013511. [138] Chen RH, Zhu MY, Li Y, Li WX, Jin HM, and Dou SX, Effect of pulsed magnetic field on critical current in carbon-nanotube-doped MgB2 wires, Acta Physica Sinica 2006; 55(9) 4878-4882. 126 Superconductors – New Developments [139] Zhang XP, Ma YW, Gao ZS, Yu ZG, Watanabe K, and Wen HH, Effect of nanoscale C and SiC doping on the superconducting properties of MgB2 tapes, Acta Physica Sini‐ ca 2006; 55(9) 4873-4877. [140] Li WX, Zeng R, Wang JL, Li Y, and Dou SX, Dependence of magnetoelectric proper‐ ties on sintering temperature for nano-SiC-doped MgB2/Fe wires made by combined in situ/ex situ process, J. Appl. Phys. 2012; 111(7) E7135-E7135. [141] Dou SX, Pan AV, Zhou S, Ionescu M, Wang XL, Horvat J, Liu HK, and Munroe PR, Superconductivity, critical current density, and flux pinning in MgB2-x(SiC)x/2 super‐ conductor after SiC nanoparticle doping, J. Appl. Phys. 2003; 94(3) 1850-1856. References [142] Dou SX, Pan AV, Zhou S, Ionescu M, Liu HK, and Munroe PR, Substitution-induced pinning in MgB2 superconductor doped with SiC nano-particles, Supercond. Sci. Technol. 2002; 15(11) 1587-1591. [143] Zhang Y, Dou SX, Lu C, Zhou H, and Li WX, Effect of Mg/B ratio on the supercon‐ ductivity of MgB2 bulk with SiC addition, Phys. Rev. B 2010; 81(9) 094501. [144] Angst M, Bud'ko SL, Wilke RHT, and Canfield PC, Difference between Al and C doping in anisotropic upper critical field development in MgB2, Phys. Rev. B 2005; 71(14) 144512.

https://openalex.org/W2784828076

https://biblio.ugent.be/publication/8549229/file/8549231

English

Editorial Seminars in MSK Radiology

Journal of the Belgian Society of Radiology

cc-by

Dear participants and readers of the JBSR, devoted to young radiologists. During this session, promising talents in MSK radiology have the opportunity to share their work, which were carefully selected by the scientific board after submission of an abstract for an orig­ inal work or a pictorial review. This initiative will certainly boost their enthusiasm and interest in MSK radiology. Indeed, young talents form the backbone of the future of our subspecialty discipline. This special volume of the Journal of Belgian Society of Radiology is dedicated to the scientific program of the fourth edition of the “Seminar in Musculoskeletal Radi­ ology” to be held in Brussels on December 15–16, 2017 (www.mskradiology.be). We wish to express on behalf of the whole faculty a warm welcome to all delegates of this annual meeting. We would like to express our deepest gratitude to all speakers who provided a well-written manuscript on the contents of their lectures, resulting in an issue of JBSR that may serve as very useful reference text in our daily clinical practice. A special word of thanks goes to the edi­ tor-in-chief of the JBSR, Dr. A. Nchimi, for giving us the opportunity to distribute this special printed edition of the journal for the first time since the start of this annual symposium. The congress is held under the auspices of the Belgian Society of Radiology (BSR) and the European Society of MSK Radiology (ESSR) and has been granted with 0.9 points (CAT 3) for the E.S.S.R. Diploma. Compared to the three previous editions, this year’s symposium will take place in a new location, the “Bowling Stone”, in Wemmel, Brussels, a modern meeting center equipped with cutting-edge audio and video facilities. The venue is easily accessible by car and from Brussels National Airport. The meeting center is very close to the Expo area of Brussels and the Atomium, a seminal totem in the skyline of Brussels and one of the prominent symbols of the city. Apart from the scientific program, the city of Brussels offers so many attractions, including the marvelous his­ torical Grand Place, the Peeing Boy (Manneken Pis), the Atomium, its many museums such as Magritte Museum etc., and promenades around the “Art Nouveau” quarters. Moreover, Brussels during the wintertime has a very spe­ cial Christmas atmosphere, which make strolling through the city and shopping an unforgettable experience. * HUDERF, BE † AZ Sint-Maarten Duffel-Mechelen and University (Hospital) Antwerp/Ghent, BE Corresponding author: Paolo Simoni ([email protected]) GUEST EDITORIAL Keywords: Education; Diagnotic Imaging; Musculoskeletal bone; Musculoskeletal joint Simoni, P and Vanhoenacker, F. Editorial Seminars in MSK Radiology. Journal of the Belgian Society of Radiology. 2017; 101(S2): 4, pp. 1–2. DOI: https://doi.org/10.5334/jbr-btr.1468 Simoni, P and Vanhoenacker, F. Editorial Seminars in MSK Radiology. Journal of the Belgian Society of Radiology. 2017; 101(S2): 4, pp. 1–2. DOI: https://doi.org/10.5334/jbr-btr.1468 Dear participants and readers of the JBSR, In keeping with the tradition of “Seminars in MSK radi­ ology”, an outstanding national and international faculty will deliver a variety of state-of-the-art and clinically ori­ ented lectures for radiologists and other specialists with special interest in MSK imaging. We are particularly happy to welcome leading musculoskletal radiologists from dif­ ferent European countries, emphasizing the European dimension of this annual meeting. We wish you a very pleasant stay in Brussels, a fruitful educational experience, and a nice reading of this JBSR issue. Prof. Paolo Simoni, President of the Congress Prof. Filip Vanhoenacker, President of European Society of Skeletal Radiology (ESSR)(2018) Competing Interests The authors have no competing interests to declare. Prof. Paolo Simoni, President of the Congress Prof. Paolo Simoni, The fourth edition of this symposium will focus on imag­ ing of the upper and lower limbs respectively, with special attention to ultrasound and interventional ­procedures. Other sessions will be dedicated to arthritis and trauma. Prof. Filip Vanhoenacker, President of European Society of Skeletal Radiology (ESSR)(2018) The category “Ethics and economy” comprises an impor­ tant part of this meeting and covers subjects which are relevant for MSK radiologists. Competing Interests The authors have no competing interests to declare. Simoni and Vanhoenacker: Editorial Seminars in MSK Radiology Competing Interests Probably, the most exciting novelty of this year’s edition is the introduction of a separate “Young Stars Session” Competing Interests The authors have no competing interests to declare. The authors have no competing interests to declare. Art. 4, pp. 2 of 2 Art. 4, pp. 2 of 2 Simoni and Vanhoenacker: Editorial Seminars in MSK Radiology Simoni and Vanhoenacker: Editorial Seminars in MSK Radiology How to cite this article: Simoni, P and Vanhoenacker, F. Editorial Seminars in MSK Radiology. Journal of the Belgian Society of Radiology. 2017; 101(S2): 4, pp. 1–2. DOI: https://doi.org/10.5334/jbr-btr.1468 Published: 16 December 2017 Published: 16 December 2017 Copyright: © 2017 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/. OPEN ACCESS Journal of the Belgian Society of Radiology is a peer-reviewed open access journal published by Ubiquity Press. OPEN ACCESS

https://openalex.org/W4225513537

https://researchonline.lshtm.ac.uk/id/eprint/4666080/7/Rees_etal_2021_Estimating-the-duration-of-seropositivity.pdf

English

Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies

Wellcome open research

cc-by

RESEARCH ARTICLE    Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Eleanor M. Rees 1,2, Naomi R. Waterlow 1, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group, Rachel Lowe 1,2, Adam J. Kucharski1 1Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK 2Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Second version: 22 Nov 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.2 Latest published: 21 Dec 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 v3 Abstract Background: The duration of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still uncertain, but it is of key clinical and epidemiological importance. Seasonal human coronaviruses (HCoV) have been circulating for longer and, therefore, may offer insights into the long-term dynamics of reinfection for such viruses. Methods: Combining historical seroprevalence data from five studies covering the four circulating HCoVs with an age-structured reverse catalytic model, we estimated the likely duration of seropositivity following seroconversion. Results: We estimated that antibody persistence lasted between 0.9 (95% Credible interval: 0.6 - 1.6) and 3.8 (95% CrI: 2.0 - 7.4) years. Furthermore, we found the force of infection in older children and adults (those over 8.5 [95% CrI: 7.5 - 9.9] years) to be higher compared with young children in the majority of studies. Conclusions: These estimates of endemic HCoV dynamics could provide an indication of the future long-term infection and reinfection patterns of SARS-CoV-2. Keywords Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Barnaby Flower , Imperial College London, London, UK Oxford University Clinical Research Unit, 2. Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Barnaby Flower , Imperial College London, London, UK Oxford University Clinical Research Unit, Vietnam, Ho Chi Minh City, Vietnam 2. Sereina A. Herzog , Medical University of Graz, Graz, Austria 3. First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Second version: 22 Nov 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.2 Latest published: 21 Dec 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 v3 RESEARCH ARTICLE    Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Eleanor M. Rees 1,2, Naomi R. Waterlow 1, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group, Rachel Lowe 1,2, Adam J. Kucharski1 1Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK 2Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Second version: 22 Nov 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.2 Latest published: 21 Dec 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 v3 Abstract Background: The duration of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still uncertain, but it is of key clinical and epidemiological importance. Seasonal human coronaviruses (HCoV) have been circulating for longer and, therefore, may offer insights into the long-term dynamics of reinfection for such viruses. Methods: Combining historical seroprevalence data from five studies covering the four circulating HCoVs with an age-structured reverse catalytic model, we estimated the likely duration of seropositivity following seroconversion. Results: We estimated that antibody persistence lasted between 0.9 (95% Credible interval: 0.6 - 1.6) and 3.8 (95% CrI: 2.0 - 7.4) years. Furthermore, we found the force of infection in older children and adults (those over 8.5 [95% CrI: 7.5 - 9.9] years) to be higher compared with young children in the majority of studies. Conclusions: These estimates of endemic HCoV dynamics could provide an indication of the future long-term infection and reinfection patterns of SARS-CoV-2. Keywords Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Barnaby Flower , Imperial College London, London, UK Oxford University Clinical Research Unit, 2. Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 RESEARCH ARTICLE    Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Eleanor M. Rees 1,2, Naomi R. Waterlow 1, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group, Rachel Lowe 1,2, Adam J. Kucharski1 1Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK 2Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Second version: 22 Nov 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.2 Latest published: 21 Dec 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 v3 Abstract Background: The duration of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still uncertain, but it is of key clinical and epidemiological importance. Seasonal human coronaviruses (HCoV) have been circulating for longer and, therefore, may offer insights into the long-term dynamics of reinfection for such viruses. Methods: Combining historical seroprevalence data from five studies covering the four circulating HCoVs with an age-structured reverse catalytic model, we estimated the likely duration of seropositivity following seroconversion. Results: We estimated that antibody persistence lasted between 0.9 (95% Credible interval: 0.6 - 1.6) and 3.8 (95% CrI: 2.0 - 7.4) years. Furthermore, we found the force of infection in older children and adults (those over 8.5 [95% CrI: 7.5 - 9.9] years) to be higher compared with young children in the majority of studies. Conclusions: These estimates of endemic HCoV dynamics could provide an indication of the future long-term infection and reinfection patterns of SARS-CoV-2. Keywords Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Barnaby Flower , Imperial College London, London, UK Oxford University Clinical Research Unit, 2. Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 RESEARCH ARTICLE    Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Eleanor M. Rees 1,2, Naomi R. Waterlow 1, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group, Rachel Lowe 1,2, Adam J. Abstract Background: The duration of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still uncertain, but it is of key clinical and epidemiological importance. Seasonal human coronaviruses (HCoV) have been circulating for longer and, therefore, may offer insights into the long-term dynamics of reinfection for such viruses. Methods: Combining historical seroprevalence data from five studies covering the four circulating HCoVs with an age-structured reverse catalytic model, we estimated the likely duration of seropositivity following seroconversion. Angkana T. Huang , University of Florida, Gainesville, USA 1. Angkana T. Huang , University of Florida, Gainesville, USA 1. Results: We estimated that antibody persistence lasted between 0.9 (95% Credible interval: 0.6 - 1.6) and 3.8 (95% CrI: 2.0 - 7.4) years. Furthermore, we found the force of infection in older children and adults (those over 8.5 [95% CrI: 7.5 - 9.9] years) to be higher compared with young children in the majority of studies. Conclusions: These estimates of endemic HCoV dynamics could provide an indication of the future long-term infection and reinfection patterns of SARS-CoV-2. RESEARCH ARTICLE    Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Eleanor M. Rees 1,2, Naomi R. Waterlow 1, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group, Rachel Lowe 1,2, Adam J. Kucharski1 1Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK 2Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Second version: 22 Nov 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.2 Latest published: 21 Dec 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 v3 Abstract Background: The duration of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still uncertain, but it is of key clinical and epidemiological importance. Seasonal human coronaviruses (HCoV) have been circulating for longer and, therefore, may offer insights into the long-term dynamics of reinfection for such viruses. Methods: Combining historical seroprevalence data from five studies covering the four circulating HCoVs with an age-structured reverse catalytic model, we estimated the likely duration of seropositivity following seroconversion. Results: We estimated that antibody persistence lasted between 0.9 (95% Credible interval: 0.6 - 1.6) and 3.8 (95% CrI: 2.0 - 7.4) years. Furthermore, we found the force of infection in older children and adults (those over 8.5 [95% CrI: 7.5 - 9.9] years) to be higher compared with young children in the majority of studies. Conclusions: These estimates of endemic HCoV dynamics could provide an indication of the future long-term infection and reinfection patterns of SARS-CoV-2. Keywords Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Barnaby Flower , Imperial College London, London, UK Oxford University Clinical Research Unit, 2. Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Kucharski1 1Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene & Tropical Medicine, London, UK 2Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, London, UK First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Second version: 22 Nov 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.2 Latest published: 21 Dec 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 v3 Abstract Background: The duration of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still uncertain, but it is of key clinical and epidemiological importance. Seasonal human coronaviruses (HCoV) have been circulating for longer and, therefore, may offer insights into the long-term dynamics of reinfection for such viruses. Methods: Combining historical seroprevalence data from five studies covering the four circulating HCoVs with an age-structured reverse catalytic model, we estimated the likely duration of seropositivity following seroconversion. Results: We estimated that antibody persistence lasted between 0.9 (95% Credible interval: 0.6 - 1.6) and 3.8 (95% CrI: 2.0 - 7.4) years. Furthermore, we found the force of infection in older children and adults (those over 8.5 [95% CrI: 7.5 - 9.9] years) to be higher compared with young children in the majority of studies. Conclusions: These estimates of endemic HCoV dynamics could provide an indication of the future long-term infection and reinfection f Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Barnaby Flower , Imperial College 2. Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Barnaby Flower , Imperial College London, London, UK Oxford University Clinical Research Unit, Vietnam, Ho Chi Minh City, Vietnam 2. Sereina A. Herzog , Medical University of Graz, Graz, Austria 3. Open Peer Review Approval Status 1 2 3 version 3 (revision) 21 Dec 2021 view version 2 (revision) 22 Nov 2021 view view view version 1 03 Jun 2021 view view view Angkana T. Huang , University of Florida, Gainesville, USA Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand 1. Introduction Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a novel beta coronavirus, was first detected in December 2019 and has since spread globally causing high morbidity and mortality. There is evidence of some short-term sterilising immunity (protection against reinfection and symptoms) fol- lowing infection with SARS-CoV-21, but also some reports of reinfection2. However, there is currently limited evidence on the duration of immunity conferred by SARS-CoV-2 infection. Given the limited duration of SARS-CoV-2 circulation to date, the dynamics of antibody responses of seasonal human coronaviruses (HCoV) could provide insights into the possible long-term potential for reinfections3. The duration of immunity following infection is of both clinical and epidemiological importance, as it provides information as to how long previously infected individuals may no longer be at risk of infection and disease, as well as influencing the long-term dynamics of epidemics4 and enabling the interpretation of population-wide serological data5. If infections are fully immunising – as is the case for patho- gens like measles and varicella zoster – then seroprevalence would be expected to accumulate over time14, and hence with age, with little waning of responses. The dynamics can therefore be captured with catalytic models of seroconversion15, which enables estimation of the force of infection (FOI, the rate at which susceptible individuals acquire infection and seroconvert). In contrast, when individuals serorevert, i.e. their immunity wanes by the progressive loss of protective antibodies against a disease over time, ‘reverse catalytic models’ can jointly estimate FOI and waning of immunity16. Variation in FOI with age may further complicate the dynamics, particularly if a high infec- tion rate in children is followed by a lower rate in adults as well as waning of seroprevalence. To understand how serocon- version, waning and age-variation in infection risk could shape population-level seroprevalence, we combine age-stratified data with age-structured reverse catalytic models, and estimate the likely duration of seropositivity following seroconversion for the four seasonal coronaviruses. There are four circulating HCoVs: HCoV-NL63 and HCoV- 229E (alpha coronaviruses), HCoV-OC43 and HCoV-HKU1 (beta coronaviruses). HCoV-OC43 and HCoV-229E were first identified in the 1960s, but HCoV-NL63 and HCov-HKU1 were not identified until 2004 and 2005 respectively6,7. Like SARS-CoV-2, these typically cause respiratory tract infections. A small number of human challenge studies have looked at the duration of immunity to these viruses. Keywords Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Wellcome Open Research 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Page 2 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 a seasonal HCoV (of 216 with confirmed first infection), and the time between reinfection ranged from 7 to 56 weeks. None of these reinfections were with the same strain, providing some evidence of lasting immunity11. However, a community surveil- lance study of 483 participants conducted in Kenya in 2010 over six months found evidence of high numbers of repeat infections of HCoV-NL63 (20.9%), HCoV-OC43 (5.7%), and HCoV-229E (4.0%). The majority of these reinfections showed reduced virus replication in the second infection, and a lower proportion of individuals had symptoms following the second infection12. Furthermore, another study conducted in New York City which included 191 participants found that rein- fections with the same strain can occur within one year13. Care should be taken with the interpretation of these studies since we do not know the background exposure rates, and this will influence the estimates of duration of immunity. Amendments from Version 2 This article has been updated in response to reviewer comments. We have added standard error estimates to WAIC and LOO. Any further responses from the reviewers can be found at the end of the article REVISED Any further responses from the reviewers can be found at the end of the article Introduction Callow et al.8 found that six out of nine participants were reinfected when chal- lenged with HCoV-229E again one year later, as measured by a rise in IgG antibodies and viral shedding. However, the period of viral shedding was shorter following the second inocula- tion, and none of the participants developed symptoms. Reed9 found that reinfection did not occur when participants were re-inoculated with a homologous strain approximately one year following infection, but participants had partial immunity against reinfection with a heterologous strain. Taken together these results suggest that immunity against infection with a homologous strain could last at least one year8,9. Keywords Keywords Seasonal coronavirus, Seroprevalence, Catalytic model, waning immunity Seasonal coronavirus, Seroprevalence, Catalytic model, waning immunity Page 1 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Any reports and responses or comments on the article can be found at the end of the article. Corresponding author: Eleanor M. Rees ([email protected]) Author roles: Rees EM: Conceptualization, Data Curation, Formal Analysis, Investigation, Methodology, Visualization, Writing – Original Draft Preparation; Waterlow NR: Formal Analysis, Methodology, Writing – Review & Editing; Lowe R: Formal Analysis, Methodology, Supervision, Visualization, Writing – Review & Editing; Kucharski AJ: Conceptualization, Formal Analysis, Methodology, Supervision, Visualization, Writing – Review & Editing Competing interests: No competing interests were disclosed. Grant information: AK was supported by Wellcome Trust and the Royal Society (grant Number 206250/Z/17/Z). EMR receives funding from the UK Medical Research Council (MR/N013638/1). NRW receives funding from the UK Medical Research Council (MR/N013638/1). RL was supported by a Royal Society Dorothy Hodgkin Fellowship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Copyright: © 2021 Rees EM et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. How to cite this article: Rees EM, Waterlow NR, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group et al. Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies [version 3; peer review: 3 approved] Wellcome Open Research 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.3 First published: 03 Jun 2021, 6:138 https://doi.org/10.12688/wellcomeopenres.16701.1 Corresponding author: Eleanor M. Rees ([email protected]) Author roles: Rees EM: Conceptualization, Data Curation, Formal Analysis, Investigation, Methodology, Visualization, Writing – Original Draft Preparation; Waterlow NR: Formal Analysis, Methodology, Writing – Review & Editing; Lowe R: Formal Analysis, Methodology, Supervision, Visualization, Writing – Review & Editing; Kucharski AJ: Conceptualization, Formal Analysis, Methodology, Supervision, Visualization, Writing – Review & Editing ompeting interests: No competing interests were disclo How to cite this article: Rees EM, Waterlow NR, Centre for the Mathematical Modelling of Infectious Diseases COVID-19 Working Group et al. Methods Human seroprevalence from four different human coronavirus strains (229E, HKU1, NL63, and OC43) were identified in a recent systematic review7. Studies which did not include esti- mates for individuals under 10 years old17 were excluded, as well as studies with which only reported two age groups18. A total of six different studies were included, covering the four seasonal HCoVs, with some studies reporting on multi- ple strains19–24. Two studies were reported separately for two different strains, but the overall study population was the same21,22. A summary of these studies is presented in Table 1. The different assays used in each study for the different strains is shown, and where the antibody detected was speci- fied this is included in the table. To account for maternal immu- nity individuals aged ≤1 year were excluded. The full dataset used for this analysis can be found as underlying data25. There are also a small number of cohort and community-based surveillance studies which have looked at reinfection of seasonal HCoV. One study looked at HCoV reinfection in a small cohort of ten individuals over 35 years and found the median reinfection times to be 30 months, but with reinfection often occurring at 12 months10. A larger study looking at data from Flu Watch, a community cohort study which measures the incidence and transmission of respiratory viruses, found that between 2006 and 2011, eight subjects were reinfected with To explore the duration of antibody persistence for different seasonal coronaviruses, where detectable antibodies is defined as seropositivity, we developed age-structured reverse catalytic models. The basic reverse catalytic model follows individuals from birth and assumes that there is a constant FOI (λ), Page 3 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Table 1. Characteristics of studies used to fit the model. Table 1. Characteristics of studies used to fit the model. Methods Strain Author (year published) Pubmed ID Sample size Country/ region Years sampled Assay Antigen Assay cut-off HCoV-HKU1 Chan (2009) 19342289 709 Hong Kong Not specified ELISA (IgG) S protein Mean + 3SD (OD>0.495) Zhou (2013)a 24040960 789 China 1999 – 2011 IFA (IgG) S protein >1:20 HCoV-OC43 Zhou (2013)a 24040960 789 China 1999 – 2011 IFA (IgG) S protein >1:20 Monto (1974)c 4816305 910 USA 1965 – 1969 CF or HI Whole virus <1:8 to >1:8 or 4-fold rise Sarateanu (1980) 6248465 3,016 Germany 1974 – 1976 HI Whole virus >1:8 HCoV-NL63 Shao (2007)b 17889596 243 USA 2003 – 2004 ELISA (IgG) N protein OD>0.2 at dilution of 1:80 or greater Zhou (2013)a 24040960 789 China 1999 – 2011 IFA (IgG) S protein >1:20 HCoV-229E Shao (2007)b 17889596 243 USA 2003 – 2004 ELISA (IgG) N protein OD>0.2 at dilution of 1:80 or greater Zhou (2013)a 24040960 789 China 1999 – 2011 IFA (IgG) S protein >1:20 Cavallaro (1970)c 5504709 307 USA 1966 Neutralization Whole virus >1:4 Human coronavirus (HCoV), Enzyme-linked immunosorbent assays (ELISA), immunofluorescence assays (IFA), complement fixation (CF), hemagglutination inhibition assays (HI), Immunoglobulin G (IgG), standard deviation (SD), optical density (OD). Studies which occurred in the same setting are denoted by the superscripts, a, b and c. Human coronavirus (HCoV), Enzyme-linked immunosorbent assays (ELISA), immunofluorescence assays (IFA), complement fixation (CF), hemagglutination inhibition assays (HI), Immunoglobulin G (IgG), standard deviation (SD), optical density (OD). Studies which occurred in the same setting are denoted by the superscripts, a, b and c. which is independent of age (a) and calendar year, and that immunity (as measured by serological status) wanes over time, at a rate ω. This model also assumes that the mortal- ity rate for susceptible and infectious individuals is the same. The expression for the proportion of individuals age a who are seropositive, z(a), in the reverse catalytic model is as follows: vary by study and strain, to account for local differences in population-level transmission dynamics, while the average rate of waning within a given individual was assumed to be universal and was jointly estimated across all studies and strains. This means that one overall estimate of waning was obtained. Some of the studies occurred in the same setting, and so the under- lying contact patterns were presumed to be the same (in total we identified five settings). Methods Therefore, the relative change in FOI (α) and the age at cut-off (a0) were jointly estimated across settings. This model assumes no cross-protection between strains. Annual attack rates were calculated after estimating the FOI using the following expression, ( ) ( ) (1 ) a z a e λ ω λ λ ω − + = − + ( ) ( ) (1 ) a z a e λ ω λ λ ω − + = − + where λ is the FOI, ω is seropositivity waning rate and a is age. The duration of antibody persistence was estimated as follows: where λ is the FOI, ω is seropositivity waning rate and a is age. The duration of antibody persistence was estimated as follows: Duration of antibody persistence = 1/ω Duration of antibody persistence = 1/ω In our analysis, we allowed λ1 to Page 4 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 For the age at cut-off (a0), uniform priors from 0 to 20 years were chosen as we were interested in the difference in FOI in children and young adults. For the relative change in FOI (α) we did not have any prior information. Therefore, we selected a prior with median 1, which presumes no difference between FOI in young compared with FOI in old and allowed for a range of plausible values using a gamma distribution (shape = 5, scale = 0.2). included individuals ≤1 year. The impact of the assay used in the study on the estimate of waning was also explored, where FOI was allowed to vary by study, alpha and the age at cut-off varied by setting and waning varied by assay (ELISA, IFA, HI and neutralisation). Finally, the primary model (age-varying FOI model) was fitted using only half the data (seropreva- lence studies from two strains), to explore whether the results from one study was heavily influencing the results. For this model, waning, the relative change in FOI (α) and age at cut-off (a0) were held across all studies. A description of these models is presented in Table 2. included individuals ≤1 year. The impact of the assay used in the study on the estimate of waning was also explored, where FOI was allowed to vary by study, alpha and the age at cut-off varied by setting and waning varied by assay (ELISA, IFA, HI and neutralisation). Finally, the primary model (age-varying FOI model) was fitted using only half the data (seropreva- lence studies from two strains), to explore whether the results from one study was heavily influencing the results. For this model, waning, the relative change in FOI (α) and age at cut-off (a0) were held across all studies. A description of these models is presented in Table 2. Several sensitivity analyses were conducted to assess the robust- ness of these results. First, the choice of priors for the FOI was explored, and a less informed prior was tested (FOI ~ Normal (mean 0.3, standard deviation 0.5)). Second, waning was estimated by strain, instead of being jointly fitted across all studies. Duration of antibody persistence = 1/ω 1 . Attack rate e λ − = − We then extended the reverse catalytic model to allow for a dif- ferent FOI by age. The expressions for seroprevalence in the reverse catalytic model with age-varying FOI are as follows: To reflect uncertainty in current knowledge about the transmis- sion dynamics of HCoVs, weakly informative distributions were chosen as priors for ω, the rate of waning over time. Spe- cifically uniform priors from 0 to 5 years. For the FOI, there is little information on the attack rate of HCoVs. However, there have been several systematic reviews and meta-analyses looking at influenza in unvaccinated individuals which have reported the attack rates to range between 15.2% – 22.5% in children and 3.5% – 10.7% in adults26–28. Modelling stud- ies using serological influenza data predicted estimates from 20 – 60%29,30. Based on the epidemiology of these viruses in children31, we expect the attack rate for HCoV may be lower. Therefore, we selected a Gamma distribution, with a mean of 0.3 (shape = 1.2 and scale = 0.25) and this corresponds to an attack rate of 26% and covers a range of plausible values. 1 1 2 0 0 0 0 1 1 1 2 2 2 2 1 ( ) ( ) ( ) ( ) ( ) (1 ) ( ) ( (1 ) ) ( ) a a a a z a e when a a z a e e when a a λ ω λ ω λ ω λ λ ω λ λ λ λ λ λ ω ω ω − − + − − + + = − < + = − − + ≥ + + + 2 1 λ λ α = Where z(a) is those who are seropositive at age a, λ1 is the FOI in young age groups, λ2 is FOI in the old age group, ω is waning, a is age, a0 is the age cut-off used to define the young and old group, and the relative change in FOI, α, is the change in FOI in the older age group. ~ ( , ), ija ija ija y Binomial P N Duration of antibody persistence = 1/ω Page 5 of 33 Page 5 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 where Nija is total number of individuals by age group, strain and study, and Pija is the proportion of individuals who are seropositive. The inference was implemented in RJags (version 4–10)32. The Gelman-Rubin statistic was used to evaluate MCMC convergence, and a threshold of <1.1 was chosen. The effective sample size (ESS), which is the esti- mated number of independent samples accounting for autocor- relations generated by the MCMC run, was checked, and an ESS >200 was used. All analysis and calculations were per- formed using R version 3.6.1. Model selection was based on the lowest value of the widely applicable information criterion (WAIC) and the leave-one-out cross validation (LOO) using Pareto-smoothed importance sampling33,34. WAIC and LOO were estimated using the R package Loo (version 2.4.1)34. All code is available here at GitHub25. having only four parameters by study, our model could capture the overall trends in most studies (Figure 1). Waning was jointly fitted across all studies and strains to obtain one overall esti- mate, and the duration of antibody persistence was estimated to be 3.75 (95% credible interval [CrI]: 1.96 – 7.38) years (Table 3). The FOI across all studies and strains in the young age group ranged from 0.02 (95% CrI: 0.01 – 0.05) to 1.06 (95% CrI: 0.57 – 1.68). The cut-off (age at which the FOI changes) ranged between 2.35 (95% CrI: 0.31 – 17.51) to 16.58 (95% CrI: 7.71 – 19.81) years. The relative change in FOI (Alpha) which measures the relative value of FOI in the young age group compared with the older age group ranged from 0.72 (95% CrI: 0.3 – 1.17) to 2.48 (95% CrI: 1.96 – 2.99). For three of the study settings, the FOI in the older age group was higher (Figure 2). A sensitivity analysis was conducted using less informative priors for the FOI parameters, where a normal distribution was used (extended data Figure 1, Table 135). This model estimated a shorter duration of anti- body persistence [0.93 (95% CrI: 0.60 – 1.64) years]. The FOI across all studies and strains were higher, ranging from 0.09 Duration of antibody persistence = 1/ω The relative change in FOI (α) and age at cut-off (a0) were then held across all studies (instead of allowing them to vary by setting) to explore the impact on the estimate for waning. The impact of excluding the youngest age groups (≤1 year) was also explored, and a model was run which Bayesian inference was used to fit the sero-catalytic models to the seroprevalence data, using Markov chain Monte Carlo (MCMC) with the Gibbs sampling algorithm to estimate model parameters. To do so, we used the following binomial likelihood representing seropositivity by age (a), study (i) and strain (j) ~ ( , ), ija ija ija y Binomial P N Table 2. Description of models explored. Table 2. Description of models explored. Model Priors Number of parameters Main model: Reverse catalytic model with age-varying FOI (alpha and cut-off varying across settings) - More informed priors FOI ~ gamma(shape = 1.2, scale = 0.25) Waning ~ uniform(0,5) Alpha ~ gamma(shape = 5, scale = 0.2) Cut-off ~ uniform(0,20) 21 Reverse catalytic model with age-varying FOI (alpha and cut-off varying across settings) - less informed priors FOI ~ normal(0.3,0.5) Waning ~ uniform(0,5) Alpha ~ gamma(shape = 5, scale = 0.2) Cut-off ~ uniform(0,20) 21 Reverse catalytic model with age-varying FOI (alpha and cut-off varying across settings, waning varying by strain) FOI ~ gamma(shape = 1.2, scale = 0.25) Waning ~ uniform(0,5) Alpha ~ gamma(shape = 5, scale = 0.2) Cut-off ~ uniform(0,20) 24 Reverse catalytic model with age-varying FOI (alpha and cut-off held across settings) FOI ~ gamma(shape = 1.2, scale = 0.25) Waning ~ uniform(0,5) Alpha ~ gamma(shape = 5, scale = 0.2) Cut-off ~ uniform(0,20) 13 Reverse catalytic model with age-varying FOI (alpha and cut-off varying across settings) including data <1 year FOI ~ gamma(shape = 1.2, scale = 0.25) Waning ~ uniform(0,5) Alpha ~ gamma(shape = 5, scale = 0.2) Cut-off ~ uniform(0,20) 21 Reverse catalytic model with age-varying FOI (alpha and cut-off varying across settings, waning varying by assay) FOI ~ gamma(shape = 1.2, scale = 0.25) Waning ~ uniform(0,5) Alpha ~ gamma(shape = 5, scale = 0.2) Cut-off ~ uniform(0,20) 24 Reverse catalytic model FOI ~ gamma(shape = 1.2, scale = 0.25) Waning ~ uniform(0,5) 11 Force of infection (FOI), relative change in FOI (Alpha,α), age at which the FOI changes (Cut-off). Results Strain First Author FOI (youngest age group) Relative change in FOI (Alpha) Age at which the FOI changes (cut-off) Waning HCoV-229E Shao 0.40 (0.26 – 0.64) 0.78 (0.35 – 1.68) 9.5 (0.59 – 19.47) 0.27 (0.14 - 0.51) Zhou 1.06 (0.57 – 1.68) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) Cavallaro 0.11 (0.06 – 0.3) 0.72 (0.3 - 1.17) 9.14 (0.57 - 19.28) HCoV-HKU1 Chan 0.02 (0.01 - 0.05) 2.27 (1.44 – 3.45) 16.58 (7.71 – 19.81) Zhou 0.59 (0.32 – 0.89) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) HCoV-OC43 Zhou 0.64 (0.35 – 0.96) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) Monto 0.07 (0.04 – 0.19) 0.72 (0.3 – 1.17) 9.14 (0.57 – 19.28) Sarateanu 0.19 (0.11 – 0.35) 2.48 (1.96 – 2.99) 9.93 (7.34 – 14.84) HCoV-NL63 Zhou 0.50 (0.27 – 0.74) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) Shao 0.41 (0.26 – 0.67) 0.78 (0.35 – 1.68) 9.5 (0.59 – 19.47) Human coronavirus (HCoV), force of infection (FOI). 2. Posterior estimates for the relative change in FOI (alpha) from the age-varying reverse catalytic model for each etting. The alpha estimate from the model where alpha and cut-off were simultaneously estimated across studies is shown s “combined”. The prior is shown as a dashed line. Table 3. Parameter estimates from the age-varying FOI reverse catalytic model (median [95% CrI]). FOI was allowed to vary across study, while waning was simultaneously estimated across all studies. The relative change in FOI (Alpha) and the cut-off were allowed to vary across study settings. Results Using a reverse catalytic model, which allowed the FOI to change in individuals by age, we estimated the duration of antibody persistence for the four seasonal HCoVs. Despite ure 1. Reverse catalytic model with age-varying FOI. The points are the observed proportion of seropositive individuals from each dy (with confidence intervals), i.e. the data that was fit to. The lines are the seroprevalence curves, sampled from the fitted model, where shaded region represents the 95% credible interval of the predictive posterior distribution. FOI was allowed to vary by study, whilst the ative change in FOI (Alpha) and cut-off were allowed to vary by setting. Waning was jointly fit across all studies and strains. Figure 1. Reverse catalytic model with age-varying FOI. The points are the observed proportion of seropositive individuals from each study (with confidence intervals), i.e. the data that was fit to. The lines are the seroprevalence curves, sampled from the fitted model, where the shaded region represents the 95% credible interval of the predictive posterior distribution. FOI was allowed to vary by study, whilst the l ti h i FOI (Al h ) d t ff ll d t b tti W i j i tl fit ll t di d t i Figure 1. Reverse catalytic model with age-varying FOI. The points are the observed proportion of seropositive individuals from each study (with confidence intervals), i.e. the data that was fit to. The lines are the seroprevalence curves, sampled from the fitted model, where the shaded region represents the 95% credible interval of the predictive posterior distribution. FOI was allowed to vary by study, whilst the relative change in FOI (Alpha) and cut-off were allowed to vary by setting. Waning was jointly fit across all studies and strains. Page 6 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Table 3. Parameter estimates from the age-varying FOI reverse catalytic model (median [95% CrI]). FOI was allowed to vary across study, while waning was simultaneously estimated across all studies. The relative change in FOI (Alpha) and the cut-off were allowed to vary across study settings. Results Strain First Author FOI (youngest age group) Relative change in FOI (Alpha) Age at which the FOI changes (cut-off) Waning HCoV-229E Shao 0.40 (0.26 – 0.64) 0.78 (0.35 – 1.68) 9.5 (0.59 – 19.47) 0.27 (0.14 - 0.51) Zhou 1.06 (0.57 – 1.68) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) Cavallaro 0.11 (0.06 – 0.3) 0.72 (0.3 - 1.17) 9.14 (0.57 - 19.28) HCoV-HKU1 Chan 0.02 (0.01 - 0.05) 2.27 (1.44 – 3.45) 16.58 (7.71 – 19.81) Zhou 0.59 (0.32 – 0.89) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) HCoV-OC43 Zhou 0.64 (0.35 – 0.96) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) Monto 0.07 (0.04 – 0.19) 0.72 (0.3 – 1.17) 9.14 (0.57 – 19.28) Sarateanu 0.19 (0.11 – 0.35) 2.48 (1.96 – 2.99) 9.93 (7.34 – 14.84) HCoV-NL63 Zhou 0.50 (0.27 – 0.74) 1.57 (0.8 – 2.65) 2.35 (0.31 – 17.51) Shao 0.41 (0.26 – 0.67) 0.78 (0.35 – 1.68) 9.5 (0.59 – 19.47) Human coronavirus (HCoV), force of infection (FOI). Table 3. Parameter estimates from the age-varying FOI reverse catalytic model (median [95% CrI]). FOI was allowed to vary across study, while waning was simultaneously estimated across all studies. The relative change in FOI (Alpha) and the cut-off were allowed to vary across study settings. Table 3. Parameter estimates from the age-varying FOI reverse catalytic model (median [95% CrI]). FOI was allowed to vary across study, while waning was simultaneously estimated across all studies. The relative change in FOI (Alpha) and the cut-off were allowed to vary across study settings Human coronavirus (HCoV), force of infection (FOI). Figure 2. Posterior estimates for the relative change in FOI (alpha) from the age-varying reverse catalytic model for each study setting. The alpha estimate from the model where alpha and cut-off were simultaneously estimated across studies is shown in grey as “combined”. The prior is shown as a dashed line. Figure 2. Posterior estimates for the relative change in FOI (alpha) from the age-varying reverse catalytic mod study setting. The alpha estimate from the model where alpha and cut-off were simultaneously estimated across studi Figure 2. Posterior estimates for the relative change in FOI (alpha) from the age-varying reverse catalytic model for each study setting. The alpha estimate from the model where alpha and cut-off were simultaneously estimated across studies is shown in grey as “combined”. The prior is shown as a dashed line. Results Page 7 of 33 Page 7 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 all studies, ranging from 0.04 (95% CrI: 0.02 - 0.07) to 2.92 (95% CrI: 2.08 - 4.01); extended data Table 5, Figure 435). The estimates for the relative change in FOI were found to be very similar to the model which excluded this age group. (95% CrI: 0.04 - 0.16) to 3.22 (95% CrI: 1.95 - 4.85), with six studies reporting FOI estimates > one, which is equivalent to an attack rate of >63%. The relative change in FOI and cut-off were similar for both models. This model had a lower WAIC (536.4 compared with 545.9) and LOO (546.0 compared with 557.8), which suggests that this model may have an improved fit compared with the model with more informed priors, how- ever, large standard errors (SE) were reported for both WAIC and LOO. Furthermore, the high FOI estimates indicate that this model may be less plausible (Table 4). As an additional sensitivity we allowed the waning estimate to vary by strain (extended data Table 235). This model estimated the duration of antibody persistence to be similar for all strains, ranging from 2.26 (1.06 – 5.07) years for HCoV-OC43 to 4.09 (1.91 – 9.60) years for HCoV-229E. As an additional sensitivity analysis, we refit the models using data for only two strains at a time, and estimated the FOI, waning and the relative change in FOI (extended data Table 635). We found that although the results varied, the overall trends were the same, indicating that the model did not rely heavily on one dataset. The duration of antibody persistence varied from 1.80 years (95% CrI: 1.17 - 2.67) to 5.26 years (95% CrI: 2.53 - 13.56). Finally, we explored the impact of the different assays used in the studies on the waning estimates. We allowed the waning estimate to vary by assay (extended data Table 7, Figure 535), whilst allowing FOI to vary by study, and alpha and cut-off to vary by setting. This model estimated the duration of anti- body persistence to be similar for ELISA (2.63 [95% CrI: 0.94-9.09] years), HI (1.08 [95% CrI: 0.44-3.33] years) and neutralisation (1.28 [95% CrI: 0.25-50.0] years) assays, but longer for IFA (7.69 [95% CrI: 3.03-14.29] years). Results The credible intervals were wide, likely due to the small number of studies by assay. When the relative change in FOI and cut-off parameters were simultaneously estimated by setting (extended data Figure 2, Table 335) the duration of antibody persistence was estimated to be shorter, 2.20 (95% CrI: 1.57 - 3.08) years, although the confidence intervals overlap with the main model. The FOI ranged from 0.04 (95% CrI: 0.03 - 0.06) to 0.88 (95% CrI: 0.67 - 1.19). The overall model WAIC (622.1 compared with 545.9) and LOO (632.5 compared with 557.8) were higher, indi- cating that this model did not have as much support, although the SEs reported were large for WAIC and LOO (Table 4). To demonstrate the relationship between FOI and seropositivity at age 30, we created simulated scenarios under different sero- catalytic models. Using the parameters for the relative change in FOI and waning estimated from the age-varying reverse catalytic model (where the relative change in FOI and the age at cut-off were simultaneously estimated across settings), we simulated the proportion of individuals aged 30 years that would be seropositive using a range of FOI estimates to show how the proportion changes using the different models. The catalytic model, which does not allow for seroreversion, results in the highest estimates of seropositivity at age 30 with increasing FOI. The age-varying FOI model results in higher estimates of seropositivity at age 30 compared with the reverse catalytic model. This is due to the FOI which was estimated to be almost twice as high in the older age We also tested a basic reverse catalytic model, where the FOI was not allowed to vary by age, and this model estimated a longer duration of antibody persistence (7.69 [95% CrI: 6.25 - 9.09] years; extended data Table 4, Figure 335). The WAIC (717.2) and LOO (718.5) values for the basic reverse catalytic model were higher compared with the other models, indicating that this basic model did not have strong support among the models considered (Table 4). To explore the effect of excluding the youngest ages (≤1 year), a sensitivity analysis was done where these individuals were included within the analysis. The duration of antibody per- sistence was found to be slightly shorter (2.04 [95% CrI: 0.1.28 -1.4.76] years) and the FOI was found to be higher for Table 4. Results Comparison of duration of antibody persistence estimates from the different models explored. Model Reverse catalytic model with  age-varying FOI (alpha and cut-off varying across settings) - More informed priors Reverse catalytic model with  age-varying FOI (alpha and cut-off varying across settings) - less informed priors Reverse catalytic model with  age-varying FOI (alpha and cut-off held across settings) Reverse catalytic model Duration of antibody persistence (years) 3.75 (95% CrI: 1.96 – 7.38) 0.93 (95% CrI: 0.60 – 1.64) 2.20 (95% CrI: 1.57 – 3.08) 7.69 (95% CrI: 6.25 – 9.09) WAIC 545.9 (SE: 100.2) 536.4 (SE: 99.6) 622.1 (SE: 103.3) 717.2 (SE: 156.8) LOO 557.8 (SE: 102.3) 546.0 (SE: 100.6) 632.5 (SE: 105.6) 718.5 (SE: 151.8) Force of infection (FOI), relative change in FOI (Alpha), age at which the FOI changes (Cut-off), widely applicable information criterion (WAIC), leave-one-out cross validation (LOO), Standard Error (SE). e 4. Comparison of duration of antibody persistence estimates from the different models explored. Force of infection (FOI), relative change in FOI (Alpha), age at which the FOI changes (Cut-off), widely applicable information criterion (WAIC), leave-one-out cross validation (LOO), Standard Error (SE). Page 8 of 33 Page 8 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 group (with age at cut-off 8.49 [7.52 – 9.94] years) in the age-varying FOI model (Figure 3A). We further explored the relationship between FOI, attack rates and the estimated number of infections by age. We used the pooled estimate across all studies of FOI to estimate the proportion exposed at a given age to provide an indication of how many infections we might expect to see by age under our modelling assumptions (Figure 3B). We estimate that by two years, over 50% of the population will have at least one infection, and by age ten over 75% will have had more than four infections. month period. The reverse catalytic model assumes that waning occurs at a constant rate, however, individuals may become reinfected within a shorter time period than average, and conversely some will take longer. Some evidence also exists for the duration of immunity to SARS-CoV-2. A recent survey of health care workers in Oxford, UK, found that pro- tection against reinfection with SARS-CoV-2 lasts at least six months36, whilst another study of health care workers from across the UK conducted by Public Health England found that immunity lasts for at least five months2. Discussion To date, there has been limited evidence about the duration of immunity to SARS-CoV-2. Given the inevitable right censoring of data during an emerging infectious disease pandemic, understanding the duration of protection following infection with HCoV could help provide insights which will be relevant to SARS-CoV-2. Using an age-varying reverse catalytic model, we estimated the overall duration of immunity, as measured by seropositivity, to be between 0.9 (95% CrI: 0.6 - 1.6) years and 3.8 (95% CrI: 2.0 - 7.4) years for HCoV’s. When waning was estimated by strain, we found comparable estimates of the duration of seropositivity, indicating that the assumption that waning is similar across strains holds true. Previous studies have produced varied estimates for the duration of immunity for HCoVs. One study estimated the median duration of immunity to be 2.5 years10, and Reed found immunity lasts at least one year9. However, several studies have reported reinfection occurring in less than one year8,11–13. Aldridge et al.11 found that reinfection with HCoV did not occur with the same strain, but Kiyuka et al.12 found reinfec- tion frequently occurred with the same strain within a six More informed priors for the FOI based on attack rates for influenza, resulted in higher estimates for the duration of seropositivity. When we used less informed priors for the FOI, a lower estimate of duration of seropositivity was obtained. However, this model produced higher estimates of FOI, with six studies reported FOI estimates in the young age group greater than one (attack rate >63%). There is limited infor- mation on the attack rate of seasonal HCoV, however there have been numerous studies looking at influenza. Previous systematic reviews have estimated the attack rate of influenza to be between 3.5% and 22.5%26–28, whilst modelling studies have estimated this to be higher, 20 – 60%29,30. Based on reporting rates of seasonal HCoV we would expect the attack rate to be lower than influenza. Therefore, this suggests that Figure 3. (A) Proportion of individuals age 30 who are seropositive for different estimates of force of infection (FOI). The catalytic model is shown in red, the reverse catalytic model in green, and the reverse catalytic model with age-varying FOI is shown in blue. Model estimates were used for the parameter values (relative change in FOI (alpha),1.93 [1.69 – 2.19]; waning, 0.45 [0.32 – 0.64]; cut-off, 8.49 [7.52 – 9.94]). Results This seems to align with what is known about reinfection in seasonal HCoVs. However, these studies only followed up individuals for six months and five months respectively, and longer follow-up times are needed. Future studies could also work to untangle the relationship between seroreversion as a result of waning homo- typic antibody responses and antigenic evolution leading to a mismatch between prior immunity and circulating viruses37. Discussion All the studies (except for Chan et al.23 who did not report this information), included within this analysis collected data over at least a six-month period. For this reason, the timing of data collection is unlikely to have biased our results. We also assume an overall FOI by age, and we do not account for differences in population susceptibility, for example health care workers or immunocompromised individuals. Despite these limitations, the duration of immunity estimated in this study is in line with literature estimates, suggesting the age-varying reverse catalytic model was able to capture overall dynamics. the results from the model with less informative priors are less plausible. Maternally derived immunity may also have a role, protecting young infants from infection38. We tested this with a model which included individuals ≤1 year. This resulted in a shorter estimate of the duration of antibody persistence, and a higher FOI, suggesting that maternal immunity may be important. A wide range of different assays were used in the studies we considered in our analysis, including enzyme-linked immu- nosorbent assays (ELISA), immunofluorescence assays (IFA), western blots, and complement fixation (CF), hemagglutination inhibition assays (HAI) and neutralisation assays. Neutrali- sation assays are considered to be the gold standard as they measure the ability of the sera to inhibit viral processes7,39. Only Cavallaro and Monto22 used a neutralisation assay. Other assays, such as ELISA and IFA, do not assess the functionality of the antigen, but instead detect the presence of antibodies in a sample. Zhou et al.19 used IFA to detect levels of IgG anti- bodies. When we allowed the waning estimate to vary by assay, we found a similar estimates of antibody persistence for ELISA, HI and neutralisation assays, ranging from 1.1 years to 2.6 years, and these are comparable to the estimates from the main model. However, for IFA, we observed a longer estimate of 7.7 years (CrI: 3.0-14.3). Due to the small number of studies, the credible intervals were large, particularly for the IFA and neutralisation assay, which only had one study setting for each assay. This highlights the need for more studies, and better standardisation of assays. A recent study provided evidence that IgG antibodies in SARS-CoV-2 are correlated with neutralising antibodies, and may therefore act as a correlate of sterilising immunity40, whilst another study suggested that neutralizing antibodies may be correlated with protection against reinfection1. Discussion (B) Estimated proportion of individuals experiencing infections by age estimated from the age-varying reverse catalytic model (more informed priors) using the pooled median estimate across studies for FOI (0.46), and median estimates for waning (0.45), alpha (1.93) and cut-off (8.49). Page 9 of 33 Figure 3. (A) Proportion of individuals age 30 who are seropositive for different estimates of force of infection (FOI). The catalytic model is shown in red, the reverse catalytic model in green, and the reverse catalytic model with age-varying FOI is shown in blue. Model estimates were used for the parameter values (relative change in FOI (alpha),1.93 [1.69 – 2.19]; waning, 0.45 [0.32 – 0.64]; cut-off, 8.49 [7.52 – 9.94]). (B) Estimated proportion of individuals experiencing infections by age estimated from the age-varying reverse catalytic model (more informed priors) using the pooled median estimate across studies for FOI (0.46), and median estimates for waning (0.45), alpha (1.93) and cut-off (8.49). Page 9 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 and beta (HCoV-OC43 and HCoV-HKU1) coronaviruses, but minimal reactivity between alpha and beta coronaviruses7. However, it is not clear whether cross-reactivity equates to cross-protection. False positives due to cross-reactivity would lead to an over-estimation of seroprevalence in a setting. This would lead to a higher plateau in older ages, and therefore generally lead to an over-estimation of both the FOI and the duration of antibody persistence. We also did not account for seasonality within this model, which may have under-estimated our FOI. Ferrari et al.44 found that ignoring seasonality may overemphasize the role of adults in the transmission, however, this was observed in measles in Niger, with outbreak peaks ranging over severalorders of magnitude, and long periods between epidemics. Theepidemic profile is different for seasonal coronaviruses, and therefore, this is unlikely to apply in this context. Whitaker & Farrington45 found that accounting for seasonality resultingfrom past epidemics only had a marginal effect on the estimates, and that regular epidemic dynamics do not strongly bias the catalytic model. The time of year data collection occurred may influence seropositivity estimates, particularly given that the duration of antibody persistence is estimated to range between 0.9 (95% CrI: 0.6 - 1.6) years and 3.8 (95% CrI: 2.0 - 7.4) years. Data collection during high transmission periods would lead to an overestimate of both the FOI and the duration of antibody persistence. Discussion Therefore, although antibody prevalence does not equate to immunity for seasonal HCoVs, prevalence of IgG antibody may be a good correlate of immu- nity. However, all of these assays only assess humoural immunity, and it is thought that cellular immunity also has a role SARS-CoV-2, and so it is likely to be also important in seasonal HCoVs41–43. Numerous studies have looked at the age pattern of HCoV patients presenting to hospital and healthcare settings, and predominantly found that the burden of disease is higher in younger children and the elderly46–48. However, it is likely that these age groups may have more severe symptoms and are therefore more likely to be reported. In contrast, seroprevalence data makes it possible to examine the whole population for evidence of past exposure, and hence can provide a clearer understanding of the underlying transmission dynamics of disease, rather than just the resulting burden. The seroprevalence surveys included in this study were conducted in different countries and settings (USA, China, Germany and Hong Kong), as well as in different time- periods (ranging from 1965–2011). It is likely that there are differences in social structure and contact patterns between these settings. Furthermore, individual level data was not available for these studies, and instead aggregated data was used. Finer resolution, particularly for the younger age groups, would have helped to provide more certainty with these estimates. In addition, we did not take into consideration cross-protection between seasonal coronavirus strains. There is some evidence of cross protective immunity between seasonal coronavirus strains, and in settings where there is co-circulating HCoV strains, this may lead to a higher prevalence. There is also evidence that there is cross-reactivity between different coronaviruses, which may lead to false positive results. A recent systematic review found that there was some cross-reactivity that occurred within alpha (HCoV-229E and HCoV-NL63) In this study, when the relative change in FOI and the age of cut-off were simultaneously estimated across studies, we found that the FOI was estimated to be twice as high in the older age group (in this case, those over 8.49 [CrI: 7.52 - 9.94] years), compared with the younger age group. A similar pattern was observed for three of five settings when the relative change in FOI and the cut-off age were allowed to vary by setting. Extended data Zenodo: Extended data: Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies. https://doi.org/10.5281/zenodo.578401835. This project contains the following extended data • SupplementaryMaterial.pdf • SupplementaryMaterial.pdf • Sensitivity analysis: Less informed priors for FOI (sup- plementary Table 1 and Figure 1) • Sensitivity analysis: Waning estimated by strain (sup- plementary Table 2) • Sensitivity analysis: Alpha and cut-off jointly simul- taneously by study (supplementary Table 3 and Figure 2) The results from this study are in accordance with what studies have observed in children during the coronavirus disease 2019 (COVID-19) pandemic, with low numbers of cases reported in young age groups, and several large seroprevalence studies have reported lower seroprevalence in children compared with adults53,54. As well as differences in contact structure, this could be explained in part by reduced susceptibility to acqui- sition of infection; a meta-analysis of contact tracing studies found that children had 56% (31% – 71%) lower odds of becoming an infected contact compared with adults55. • Reverse catalytic model (supplementary Figure 3 and Table 4) • Sensitivity analysis: Including the youngest age groups (<1 year) (supplementary Table 5 and Figure 4) • Sensitivity analysis: Refitting the model using data from only two strains (supplementary Table 6) • Sensitivity analysis: Waning estimated by assay (supplementary Table 7 and Figure 5) • Sensitivity analysis: Waning estimated by assay (supplementary Table 7 and Figure 5) The duration of immunity to SARS-CoV-2 is still largely unknown and is of significance for the interpretation of population wide serological data, the understanding of the long-term dynamics of the epidemic, as well as of clinical importance. Given the long-term circulation of seasonal HCoVs, data on these related coronaviruses could provide indications of the possible future dynamics of SARS-CoV-2. With infection likely to become endemic in parts of the world, the duration of antibody-mediated immune responses will be particularly impor- tant in shaping transmission patterns in years to come. Using seroprevalence data, in this study we estimated the duration of seropositivity to seasonal HCoVs following seroconver- sion to be between 0.9 (95% CrI: 0.6 - 1.6) years and 3.8 (95% CrI: 2.0 - 7.4) years. We allowed the FOI to vary by age group and found it to be lower in young children (≤8.5 years) compared with older children and adults, which is corroborated with what has been observed in the COVID-19 pandemic. Extended data This suggests individuals in settings with endemic HCoVs accumulate multiple infections over the course of their lifetime, punctuated by periods of waning seropositivity against circulating viruses. Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0). Discussion This suggests that older children and adults may be important for the transmission of seasonal HCoVs in some settings. A previous study looking at social mixing patterns in Europe49 found that children are expected to have the highest inci- dence during the initial stages of an epidemic as a result of Page 10 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 GNU General Public License v3.0. their social mixing patterns, and this is what is found for some diseases, such as seasonal influenza, where there is evidence young children drive transmission50,51. However, a more recent study looking at a large scale dataset of movement and contact patterns in the United Kingdom data found contact inten- sity was highest in the 18–30 year age group when looking at all types of contacts (conversational, which was defined as face-to-face conversation of three or more words, and physi- cal), although for physical alone, those aged 5–9 years had the highest contact52. Therefore, any association between contact intensity and transmission will depend on the contacts consid- ered, particularly if a pathogen is more commonly spread via conversational contacts or via prolonged physical contacts. One possible explanation for the higher FOI we estimate in older age groups is that conversational contacts – which are typically higher in volume but lower in duration and intensity – could be more important for the transmission of seasonal HCoVs. Software availability Source code available from: https://github.com/erees/ seasonalHCoV Archived source code at time of publication: https://doi. org/10.5281/zenodo.570776425 Archived source code at time of publication: https://doi. org/10.5281/zenodo.570776425 License: GNU General Public License v3.0. Acknowledgements Members of the Centre for Mathematical Modelling of Infectious Diseases (CMMID) COVID-19 Working Group (random order): Hamish P Gibbs, Alicia Rosello, Nicholas G. Davies, Kevin van Zandvoort, Timothy W Russell, Graham Medley, Damien C Tully, Sebastian Funk, Emilie Finch, Kaja Abbas, Fiona Yueqian Sun, Rosanna C Barnard, Matthew Quaife, Oliver Brady, Joel Hellewell, David Hodgson, Carl A B Pearson, W John Edmunds, Simon R Procter, William Waites, Billy J Quilty, Jiayao Lei, James D Munday, Kathleen O’Reilly, Rosalind M Eggo, Christopher I Jarvis, Fabienne Krauer, Mihaly Koltai, Sophie R Meakin, Mark Jit, Anna M Foss, Yalda Jafari, Nikos I Bosse, Emily S Nightingale, Katherine E. Atkins. Thibaut Jombart. Frank G Sandmann, Samuel Clifford, Petra Klepac, Stefan Flasche, C Julian Villabona-Arenas, Yung-Wai Desmond Chan, Yang Acknowledgements Members of the Centre for Mathematical Modelling of Infectious Diseases (CMMID) COVID-19 Working Group (random order): Infectious Diseases (CMMID) COVID-19 Working Group (random order): Hamish P Gibbs, Alicia Rosello, Nicholas G. Davies, Kevin van Zandvoort, Timothy W Russell, Graham Medley, Damien C Tully, Sebastian Funk, Emilie Finch, Kaja Abbas, Fiona Yueqian Sun, Rosanna C Barnard, Matthew Quaife, Oliver Brady, Joel Hellewell, David Hodgson, Carl A B Pearson, W John Edmunds, Simon R Procter, William Waites, Billy J Quilty, Jiayao Lei, James D Munday, Kathleen O’Reilly, Rosalind M Eggo, Christopher I Jarvis, Fabienne Krauer, Mihaly Koltai, Sophie R Meakin, Mark Jit, Anna M Foss, Yalda Jafari, Nikos I Bosse, Emily S Nightingale, Katherine E. Atkins. Thibaut Jombart. Frank G Sandmann, Samuel Clifford, Petra Klepac, Stefan Flasche, C Julian Villabona-Arenas, Yung-Wai Desmond Chan, Yang This project contains the following underlying data: 1. Addetia A, Crawford KHD, Dingens A, et al.: Neutralizing Antibodies Correlate with Protection from SARS-CoV-2 in Humans during a Fishery Vessel Outbreak with a High Attack Rate. J Clin Microbiol. 2020; 58(11): e02107–20. PubMed Abstract | Publisher Full Text | Free Full Text Data availability statement Underlying data Publisher Full Text 6. Vabret A, Dina J, Gouarin S, et al.: Detection of the New Human Coronavirus HKU1: A Report of 6 Cases. Clin Infect Dis. 2006; 42(5): 634–9. PubMed Abstract | Publisher Full Text | Free Full Text 7. Huang AT, Garcia-Carreras B, Hitchings MDT, et al.: A systematic review of antibody mediated immunity to coronaviruses: kinetics, correlates of protection, and association with severity. Nat Commun. 2020; 11(1): 4704. PubMed Abstract | Publisher Full Text | Free Full Text 8. Callow KA, Parry HF, Sergeant M, et al.: The time course of the immune response to experimental coronavirus infection of man. Epidemiol Infect. 1990; 105(2): 435–46. PubMed Abstract | Publisher Full Text | Free Full Text 9. Reed SE: The behaviour of recent isolates of human respiratory coronavirus in vitro and in volunteers: Evidence of heterogeneity among 229E-related strains. J Med Virol. 1984; 13(2): 179–92. PubMed Abstract | Publisher Full Text | Free Full Text 10. Edridge AWD, Kaczorowska J, Hoste ACR, et al.: Seasonal coronavirus protective immunity is short-lasting. Nat Med. 2020; 26(11): 1691–3. PubMed Abstract | Publisher Full Text 2. Hall V, Foulkes S, Charlett A, et al.: Do antibody positive healthcare workers have lower SARS-CoV-2 infection rates than antibody negative healthcare workers? Large multi-centre prospective cohort study (the SIREN study), England: June to November 2020. medRxiv. 2021; 2021.01.13.21249642. Publisher Full Text 11. Aldridge RW, Lewer D, Beale S, et al.: Seasonality and immunity to laboratory-confirmed seasonal coronaviruses (HCoV-NL63, HCoV-OC43, and HCoV-229E): results from the Flu Watch cohort study [version 2; peer review: 2 approved]. Wellcome Open Res. 2020; 5: 52. PubMed Abstract | Publisher Full Text | Free Full Text 3. Guthmiller JJ, Wilson PC: Remembering seasonal coronaviruses. Science. 2020; 370(6522): 1272–3. PubMed Abstract | Publisher Full Text 3. Guthmiller JJ, Wilson PC: Remembering seasonal coronaviruses. Science. 2020; 370(6522): 1272–3. PubMed Abstract | Publisher Full Text 12. Kiyuka PK, Agoti CN, Munywoki PK, et al.: Human Coronavirus NL63 Molecular Epidemiology and Evolutionary Patterns in Rural Coastal Kenya. J Infect Dis. 2018; 217(11): 1728–39. PubMed Abstract | Publisher Full Text | Free Full Text 4. Saad-Roy CM, Wagner CE, Baker RE, et al.: Immune life history, vaccination, and the dynamics of SARS-CoV-2 over the next 5 years. Science. 2020; 370(6518): 811–8. 4. Saad-Roy CM, Wagner CE, Baker RE, et al.: Immune life history, vaccination, and the dynamics of SARS-CoV-2 over the next 5 years. Science. Data availability statement Underlying data Davies (UKRI Research England, NIHR: NIHR200929, UK MRC: MC_PC_ 19065), Kevin van Zandvoort (Elrha R2HC/UK FCDO/Wellcome Trust/NIHR, FCDO/Wellcome Trust: Epidemic Preparedness Coronavirus research programme 221303/Z/20/Z), Timothy W Russell (Wellcome Trust: 206250/Z/17/Z), Graham Medley (B&MGF: NTD Modelling Consortium OPP1184344), Sebastian Funk, (Wellcome Trust: 210758/Z/18/Z), Emilie Finch (MRC: MR/N013638/1), Kaja Abbas (BMGF: INV-016832; OPP1157270), Fiona Yueqian Sun (NIHR: 16/137/109), Rosanna C Barnard (European Commission: 101003688), Matthew Quaife (ERC Starting Grant: #757699, B&MGF: INV-001754), Oliver Brady (Wellcome Trust: 206471/Z/17/Z), Joel Hellewell (Wellcome Trust: 210758/Z/18/Z), David Hodgson (NIHR: 1R01AI141534-01A1), Carl A B Pearson (B&MGF: NTD Modelling Consortium OPP1184344, FCDO/Wellcome Trust: Epidemic Preparedness Coronavirus research programme 221303/ Z/20/Z), W John Edmunds (European Commission: 101003688, UK MRC: MC_PC_19065, NIHR: PR-OD-1017-20002), Simon R Procter (B&MGF: INV-016832), William Waites (MRC: MR/V027956/1), Billy J Quilty (NIHR: 16/137/109, NIHR: 16/136/46, B&MGF: OPP1139859), Jiayao Lei (B&MGF: Thibaut Jombart (Global Challenges Research Fund: ES/ P010873/1, UK Public Health Rapid Support Team, NIHR: Health Protection Research Unit for Modelling Methodology HPRU- 2012-10096, UK MRC: MC_PC_19065), Frank G Sandmann (NIHR: NIHR200929), Samuel Clifford (Wellcome Trust: 208812/Z/17/Z, UK MRC: MC_PC_19065), Petra Klepac (Royal Society: RP\EA\180004, European Commission: 101003688), Stefan Flasche (Wellcome Trust: 208812/Z/17/Z), C Julian Villabona-Arenas (ERC: SG 757688), Yang Liu (B&MGF: INV- 003174, NIHR: 16/137/109, European Commission: 101003688, UK MRC: MC_PC_19065), Kiesha Prem (B&MGF: INV-003174, European Commission: 101003688), Sam Abbott (Wellcome Trust: 210758/Z/18/Z), Akira Endo (Nakajima Foundation), Amy Gimma (European Commission: 101003688), Gwenan M Knight (UK MRC: MR/P014658/1) References 1. Addetia A, Crawford KHD, Dingens A, et al.: Neutralizing Antibodies Correlate with Protection from SARS-CoV-2 in Humans during a Fishery Vessel Outbreak with a High Attack Rate. J Clin Microbiol. 2020; 58(11): e02107–20. PubMed Abstract | Publisher Full Text | Free Full Text 2. Hall V, Foulkes S, Charlett A, et al.: Do antibody positive healthcare workers have lower SARS-CoV-2 infection rates than antibody negative healthcare workers? Large multi-centre prospective cohort study (the SIREN study), England: June to November 2020. medRxiv. 2021; 2021.01.13.21249642. Publisher Full Text 3. Guthmiller JJ, Wilson PC: Remembering seasonal coronaviruses. Science. 2020; 370(6522): 1272–3. PubMed Abstract | Publisher Full Text 4. Saad-Roy CM, Wagner CE, Baker RE, et al.: Immune life history, vaccination, and the dynamics of SARS-CoV-2 over the next 5 years. Science. 2020; 370(6518): 811–8. PubMed Abstract | Publisher Full Text | Free Full Text 5. Bobrovitz N, Arora RK, Cao C, et al.: Global seroprevalence of SARS-CoV-2   antibodies: a systematic review and meta-analysis. medRxiv. 2020; 2020.11.17.20233460. Data availability statement Underlying data Zenodo: erees/seasonalHCoV: First release. https://doi.org/ 10.5281/zenodo.570776425 Zenodo: erees/seasonalHCoV: First release. https://doi.org/ 10.5281/zenodo.570776425 This project contains the following underlying data: Data extracted from Huang et al.7 (“41467_2020_18450_ MOESM7_ESM-1.csv”) Page 11 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Liu, Kiesha Prem, Sam Abbott, Akira Endo, Amy Gimma, Gwenan M Knight Liu, Kiesha Prem, Sam Abbott, Akira Endo, Amy Gimma, Gwenan M Knight INV-003174), James D Munday (Wellcome Trust: 210758/Z/18/Z), Kathleen O’Reilly (B&MGF: OPP1191821), Rosalind M Eggo (HDR UK: MR/S003975/1, UK MRC: MC_PC_19065, NIHR: NIHR200908), Christopher I Jarvis (Global Challenges Research Fund: ES/P010873/1), Fabienne Krauer (Innovation Fund: 01VSF18015, Wellcome Trust: UNS110424), Mihaly Koltai (Wellcome Trust: 221303/Z/20/Z), Sophie R Meakin (Wellcome Trust: 210758/Z/18/Z), Mark Jit (B&MGF: INV-003174, NIHR: 16/137/109, NIHR: NIHR200929, European Commission: 101003688), Yalda Jafari (UKRI: MR/V028456/1), Nikos I Bosse (HPRU: NIHR200908), Emily S Nightingale (B&MGF: OPP1183986), Katherine E. Atkins (ERC: SG 757688), CMMID COVID-19 working group funding statements: Hamish P Gibbs (EDCTP2: RIA2020EF-2983-CSIGN, UK DHSC/UK Aid/NIHR: PR-OD-1017-20001), Alicia Rosello (NIHR: PR-OD-1017-20002), Nicholas G. Data availability statement Underlying data 2020; 370(6518): 811–8. ( ) PubMed Abstract | Publisher Full Text | Free Full Text 13. Galanti M, Shaman J: Direct Observation of Repeated Infections With Endemic Coronaviruses. J Infect Dis. 2021; 223(3): 409–415. PubMed Abstract | Publisher Full Text | Free Full Text 5. Bobrovitz N, Arora RK, Cao C, et al.: Global seroprevalence of SARS-CoV-2   antibodies: a systematic review and meta-analysis. medRxiv. 2020; 2020.11.17.20233460. Publisher Full Text 5. Bobrovitz N, Arora RK, Cao C, et al.: Global seroprevalence of SARS-CoV-2   antibodies: a systematic review and meta-analysis. medRxiv. 2020; 2020.11.17.20233460. Publisher Full Text 14. Antia A, Ahmed H, Handel A, et al.: Heterogeneity and longevity of antibody memory to viruses and vaccines. PLoS Biol. 2018; 16(8): e2006601. PubMed Abstract | Publisher Full Text | Free Full Text 6. Vabret A, Dina J, Gouarin S, et al.: Detection of the New Human Coronavirus HKU1: A Report of 6 Cases. Clin Infect Dis. 2006; 42(5): 634–9. PubMed Abstract | Publisher Full Text | Free Full Text 6. Vabret A, Dina J, Gouarin S, et al.: Detection of the New Human Coronavirus HKU1: A Report of 6 Cases. Clin Infect Dis. 2006; 42(5): 634–9. PubMed Abstract | Publisher Full Text | Free Full Text 15. Melegaro A, Jit M, Gay N, et al.: What types of contacts are important for the spread of infections?: using contact survey data to explore European mixing patterns. Epidemics. 2011; 3(3–4): 143–51. PubMed Abstract | Publisher Full Text 7. Huang AT, Garcia-Carreras B, Hitchings MDT, et al.: A systematic review of antibody mediated immunity to coronaviruses: kinetics, correlates of protection, and association with severity. Nat Commun. 2020; 11(1): 4704. PubMed Abstract | Publisher Full Text | Free Full Text 16. Hens N, Aerts M, Faes C, et al.: Seventy-five years of estimating the force of infection from current status data. Epidemiol Infect. 2010; 138(6): 802–12. PubMed Abstract | Publisher Full Text 8. Callow KA, Parry HF, Sergeant M, et al.: The time course of the immune response to experimental coronavirus infection of man. Epidemiol Infect. 1990; 105(2): 435–46. PubMed Abstract | Publisher Full Text | Free Full Text antigenically to escape antibody immunity. Microbiology. 2020. Publisher Full Text Immunol. 2008; 15(12): 1805–10. PubMed Abstract | Publisher Full Text | Free Full Text 38. Dijkman R, Jebbink MF, Bahia el Idrissi N, et al.: Human Coronavirus NL63 and 229E Seroconversion in Children. J Clin Microbiol. 2008; 46(7): 2368–73. PubMed Abstract | Publisher Full Text | Free Full Text 18. Liang FY, Lin LC, Ying TH, et al.: Immunoreactivity characterisation of the three structural regions of the human coronavirus OC43 nucleocapsid protein by Western blot: implications for the diagnosis of coronavirus infection. J Virol Methods. 2013; 187(2): 413–20. PubMed Abstract | Publisher Full Text | Free Full Text 39. Kellam P, Barclay W: The dynamics of humoral immune responses following SARS-CoV-2 infection and the potential for reinfection. J Gen Virol. 2020; 101(8): 791–7. PubMed Abstract | Publisher Full Text | Free Full Text 19. Zhou W, Wang W, Wang H, et al.: First infection by all four non-severe acute respiratory syndrome human coronaviruses takes place during childhood. BMC Infect Dis. 2013; 13: 433. PubMed Abstract | Publisher Full Text | Free Full Text 40. Iyer AS, Jones FK, Nodoushani A, et al.: Dynamics and significance of the antibody response to SARS-CoV-2 infection. medRxiv. 2020. Publisher Full Text 20. Shao X, Guo X, Esper F, et al.: Seroepidemiology of group I human coronaviruses in children. J Clin Virol. 2007; 40(3): 207–13. PubMed Abstract | Publisher Full Text | Free Full Text 41. Dan JM, Mateus J, Kato Y, et al.: Immunological memory to SARS-CoV-2 assessed for up to 8 months after infection. Science. 2021; 371(6529): eabf4063. PubMed Abstract | Publisher Full Text | Free Full Text 21. Monto AS, Lim SK: The Tecumseh Study of Respiratory Illness. VI. Frequency of and Relationship between Outbreaks of Coronavirus Infection. J Infect Dis. 1974; 129(3): 271–6. PubMed Abstract | Publisher Full Text | Free Full Text 42. Cox RJ, Brokstad KA: Not just antibodies: B cells and T cells mediate immunity to COVID-19. Nat Rev Immunol. 2020; 20(10): 581–2. PubMed Abstract | Publisher Full Text | Free Full Text 22. Cavallaro JJ, Monto AS: Community-wide outbreak of infection with a 229E- like coronavirus in Tecumseh, Michigan. J Infect Dis. 1970; 122(4): 272–9. PubMed Abstract | Publisher Full Text | Free Full Text 43. Ni L, Ye F, Cheng ML, et al.: Detection of SARS-CoV-2-Specific Humoral and Cellular Immunity in COVID-19 Convalescent Individuals. Immunity. 2020; 52(6): 971–977.e3. ( ) PubMed Abstract | Publisher Full Text | Free Full Text ( ) PubMed Abstract | Publisher Full Text | Free Full Text 49. Mossong J, Hens N, Jit M, et al.: Social Contacts and Mixing Patterns Relevant to the Spread of Infectious Diseases. PLoS Med. 2008; 5(3): e74. PubMed Abstract | Publisher Full Text | Free Full Text 30. Minter A, Hoschler K, Jagne YJ, et al.: Estimation of Seasonal Influenza Attack Rates and Antibody Dynamics in Children Using Cross-Sectional Serological Data. J Infect Dis. 2020; jiaa338. PubMed Abstract | Publisher Full Text 50. Baguelin M, Flasche S, Camacho A, et al.: Assessing Optimal Target Populations for Influenza Vaccination Programmes: An Evidence Synthesis and Modelling Study. PLoS Med. 2013; 10(10): e1001527. PubMed Abstract | Publisher Full Text | Free Full Text 31. Liu WK, Liu Q, Chen DH, et al.: Epidemiology of Acute Respiratory Infections in Children in Guangzhou: A Three-Year Study. PLoS One. 2014; 9(5): e96674. PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text 51. Kissler SM, Viboud C, Grenfell BT, et al.: Symbolic transfer entropy reveals the age structure of pandemic influenza transmission from high-volume influenza-like illness data. J R Soc Interface. 2020; 17(164): 20190628. PubMed Abstract | Publisher Full Text | Free Full Text 32. Plummer M, Stukalov A, Denwood M: rjags: Bayesian Graphical Models using MCMC. 2019; (accessed 16 Dec 2020). Reference Source 33. Vehtari A, Gelman A, Gabry J: Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. 2016. Reference Source 52. Klepac P, Kucharski AJ, Conlan AJK, et al.: Contacts in context: large-scale setting-specific social mixing matrices from the BBC Pandemic project. medRxiv. Epidemiology. 2020. Publisher Full Text 34. Vehtari A, Gabry J, Magnusson M, et al.: loo: Efficient Leave-One-Out Cross- Validation and WAIC for Bayesian Models. 2020; (accessed 15 Nov 2021). Reference Source 53. Munro APS, Faust SN: COVID-19 in children: current evidence and key questions. Curr Opin Infect Dis. 2020; 33(6): 540–7. PubMed Abstract | Publisher Full Text 35. Rees EM, Waterlow NR, Lowe R, et al.: Extended Data: Estimating the duration of seropositivity of human seasonal coronaviruses using seroprevalence studies. 2021. http://www.doi.org/10.5281/zenodo.5784018 54. Marks M, Millat-Martinez P, Ouchi D, et al.: Transmission of COVID-19 in 282 clusters in Catalonia, Spain: a cohort study. Lancet Infect Dis. 2021; 21(5): 629–636. PubMed Abstract | Publisher Full Text 17. Severance EG, Bossis I, Dickerson FB, et al.: Development of a Nucleocapsid- Based Human Coronavirus Immunoassay and Estimates of Individuals Exposed to Coronavirus in a U.S. Metropolitan Population. Clin Vaccine 9. Reed SE: The behaviour of recent isolates of human respirator Page 12 of 33 Page 12 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 antigenically to escape antibody immunity. Microbiology. 2020. Publisher Full Text antigenically to escape antibody immunity. Microbiology. 2020. Publisher Full Text PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text 23. Chan CM, Tse H, Wong SSY, et al.: Examination of seroprevalence of coronavirus HKU1 infection with S protein-based ELISA and neutralization assay against viral spike pseudotyped virus. J Clin Virol. 2009; 45(1): 54–60. PubMed Abstract | Publisher Full Text | Free Full Text 44. Ferrari MJ, Djibo A, Grais RF, et al.: Episodic outbreaks bias estimates of age-specific force of infection: a corrected method using measles as an example. Epidemiol Infect. 2010; 138(1): 108–16. PubMed Abstract | Publisher Full Text | Free Full Text 24. Sarateanu DE, Ehrengut W: A two year serological surveillance of coronavirus infections in Hamburg. Infection. 1980; 8(2): 70–2. PubMed Abstract | Publisher Full Text | Free Full Text p p f PubMed Abstract | Publisher Full Text | Free Full Text g f PubMed Abstract | Publisher Full Text | Free Full Text 45. Whitaker HJ, Farrington CP: Estimation of infectious disease parameters from serological survey data: the impact of regular epidemics. Stat Med. 2004; 23(15): 2429–43. PubMed Abstract | Publisher Full Text 25. erees: erees/seasonalHCoV: Second release (v2.0.0). Zenodo. 2021. http://www.doi.org/10.5281/zenodo.5707764 26. Somes MP, Turner RM, Dwyer LJ, et al.: Estimating the annual attack rate of seasonal influenza among unvaccinated individuals: A systematic review and meta-analysis. Vaccine. 2018; 36(23): 3199–207. PubMed Abstract | Publisher Full Text 46. Fischer N, Dauby N, Bossuyt N, et al.: Monitoring of human coronaviruses in Belgian primary care and hospitals, 2015–20: a surveillance study. Lancet Microbe. 2021. Publisher Full Text 27. Jayasundara K, Soobiah C, Thommes E, et al.: Natural attack rate of influenza in unvaccinated children and adults: a meta-regression analysis. BMC Infect Dis. 2014; 14: 670. PubMed Abstract | Publisher Full Text | Free Full Text 47. Zhang SF, Tuo JL, Huang XB, et al.: Epidemiology characteristics of human coronaviruses in patients with respiratory infection symptoms and phylogenetic analysis of HCoV-OC43 during 2010-2015 in Guangzhou. PLoS One. 2018; 13(1): e0191789. b d b | bli h ll | ll ; PubMed Abstract | Publisher Full Text | Free Full Text 28. Turner D, Wailoo A, Nicholson K, et al.: Systematic review and economic decision modelling for the prevention and treatment of influenza A and B. Health Technol Assess. 2003; 7(35): iii–iv, xi–xiii, 1–170. PubMed Abstract | Publisher Full Text l 48. Fowlkes A, Giorgi A, Erdman D, et al.: Viruses Associated With Acute Respiratory Infections and Influenza-like Illness Among Outpatients From the Influenza Incidence Surveillance Project, 2010-2011. J Infect Dis. 2014; 209(11): 1715–25. PubMed Abstract | Publisher Full Text | Free Full Text 29. Kucharski AJ, Lessler J, Cummings DAT, et al.: Timescales of influenza A/H3N2 antibody dynamics. PLoS Biol. 2018; 16(8): e2004974. PubMed Abstract | Publisher Full Text | Free Full Text l ( ) PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text 36. Lumley SF, O’Donnell D, Stoesser NE, et al.: Antibodies to SARS-CoV-2 are associated with protection against reinfection. medRxiv. 2020; 2020.11.18.20234369. Publisher Full Text 55. Viner RM, Mytton OT, Bonell C, et al.: Susceptibility to SARS-CoV-2 infection amongst children and adolescents compared with adults: a systematic review and meta-analysis. medRxiv. Public and Global Health. 2020. Publisher Full Text 37. Eguia R, Crawford KHD, Stevens-Ayers T, et al.: A human coronavirus evolves Page 13 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Sereina A. Herzog No comments on the current version 3. Version 2 Reviewer Report 26 November 2021 © 2021 Huang A. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Angkana T. Huang 1 Department of Biology and Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA 2 Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand I thank the authors for thoroughly considering and addressing my comments. I only have one Open Peer Review Current Peer Review Status: © 2022 Herzog S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. https://doi.org/10.21956/wellcomeopenres.19266.r47157 © 2021 Flower B. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Reviewer Report 23 November 2021 Reviewer Report 23 November 2021 Angkana T. Huang 1 Department of Biology and Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA 2 Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand I thank the authors for thoroughly considering and addressing my comments. I only have one Page 14 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 minor point to make regarding my suggestion of ELPD but is not meant to hold off the state of the approval. Specifically, it would be great if the authors could also include the 95% CI (in addition to the point estimates of the performance metric) to demonstrate that the superiority was well supported (i.e., whether the intervals were non-overlapping). Competing Interests: No competing interests were disclosed. Reviewer Expertise: infectious disease modeling Barnaby Flower 1 Department of Infectious Disease, Faculty of Medicine, Imperial College London 1 Department of Infectious Disease, Faculty of Medicine, Imperial College London, London, UK 2 Oxford University Clinical Research Unit, Vietnam, Ho Chi Minh City, Vietnam No further comments. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Author Response 15 Dec 2021 Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK hank you for taking the time to review the manuscript again. Thank you for the suggestion, we have now included the standard error estimates as a measure of the variance for both the WAIC and LOO. The standard errors for these point estimates are large, and we have highlighted this in the text. Competing Interests: No competing interests were disclosed. Competing Interests: No competing interests were disclosed. Reviewer Expertise: Infectious disease modeling, biostatistics I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Version 1 Reviewer Report 14 October 2021 https://doi.org/10.21956/wellcomeopenres.18416.r46084 © 2021 Herzog S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Reviewer Report 22 November 2021 https://doi.org/10.21956/wellcomeopenres.19266.r4715 Reviewer Report 22 November 2021 © 2021 Herzog S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Sereina A. Herzog Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria The authors have addressed the comments I had, and I have no further comments on the revised version. Competing Interests: No competing interests were disclosed. Reviewer Expertise: Infectious disease modeling, biostatistics I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Version 1 Reviewer Report 14 October 2021 https://doi.org/10.21956/wellcomeopenres.18416.r46084 © 2021 Herzog S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Sereina A. Herzog Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria Rees et al. provide estimations for the force of infection (FOI) and seropositivity waning rate regarding four different human coronaviruses (HCoV) strains using reverse catalytic model with age-varying FOI. The limitations of the modelling approach as well as the limitations about the datasets are well discussed Sereina A. Herzog The authors have addressed the comments I had, and I have no further comments on the revised version. Competing Interests: No competing interests were disclosed. Competing Interests: No competing interests were disclosed. Reviewer Expertise: SARSCoV2 antibody seroprevalence, SARSCoV2 diagnostics, Hepatitis C Clinical Trialist. No modeling expertise. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Page 15 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Sereina A. Herzog Sereina A. Herzog Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria Rees et al. provide estimations for the force of infection (FOI) and seropositivity waning rate regarding four different human coronaviruses (HCoV) strains using reverse catalytic model with age-varying FOI. The limitations of the modelling approach as well as the limitations about the datasets are well discussed. Rees et al. provide estimations for the force of infection (FOI) and seropositivity waning rate regarding four different human coronaviruses (HCoV) strains using reverse catalytic model with age-varying FOI. The limitations of the modelling approach as well as the limitations about the datasets are well discussed. Some minor points that should be addressed: Some minor points that should be addressed: p The authors give in the introduction examples about pathogens which are ‘fully immunising’ like pertussis. To my knowledge pertussis natural infection protects longer than vaccination 1. The authors give in the introduction examples about pathogens which are ‘fully immunising’ like pertussis. To my knowledge pertussis natural infection protects longer than vaccination 1. Page 16 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 but not lifelong (see Wendelboe et al., 20051). Please rephrase the sentence accordingly. Please clarify in the method section what is meant with “[…] were excluded, as well as studies with small number of data points.” Is ‘small number’ referring to number of different age (groups) reported in the study or number of total observations in a study, and what is ‘small’? 2. The full dataset used is not available on Zenodo (reference 25) but on GitHub (reference 33) [Method section, 1st paragraph, last sentence] – please check throughout the text those two references (e.g. last sentence Method section GitHub is mentioned with reference number 25). 3. In the method section it is stated that for the age at the cut-off time (a0) also uniform priors from 0 to 5 years are also investigated, however, I don’t see this prior appearing in the Table 2 with the different models and priors or in the Result section. 4. Can the authors clarify in the discussion why they did not investigate a sensitivity analysis about the different assays used in the studies, e.g. Sereina A. Herzog allowing to vary waning according to assays used? 5. It is not clear which model was used to produce Figure 3B and which FOI estimates were used (from the main model, all models?) and how they were pooled. The description and interpretation of Figure 3B should be deepened. 6. Some minor points which could improve readability: d In the 3rd paragraph of the introduction, the authors provide the number of observations for the studies mentioned in this paragraph except for the studies conducted in Kenya and in New York. It would help to put the results in perspective to also have the numbers for those two studies. 1. Assay description: Table 1 indicates a very rough description of the assay used in the study. For interested readers it would be helpful to have more information in the supplementary material about the assays like exact assay name, was it qualitative/quantitative, sensitivity/specificity, what cut-off values were used, what is the lower limit of detection (LLOD), what happened with equivocal results (count towards being positive or negative), etc. 2. Referring to tables and figures in the supplementary material: change ‘extended data Figure 1, Table 1’ to ‘Supplementary Figure 1, Table 1’ etc. 3. Table 1-4: add the abbreviations used within the table in the table legend (e.g. Table 1 HCoV). 4. Table 2 - 5th model: maybe rephrase ‘using all data’ to ‘using also data on <1 year’. 5. Table 3: I would suggest to extend the column titles a bit: “FOI”: clarifying that it is the one for the younger age group (i.e. lambda1); “Cut-off”: clarifying that his is about splitting observations into two age groups and adding the unit years. 6. Table 3: I would suggest to extend the column titles a bit: “FOI”: clarifying that it is the one for the younger age group (i.e. lambda1); “Cut-off”: clarifying that his is about splitting observations into two age groups and adding the unit years. 6. Page 17 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Table 4: Why are not all models from Table 2 reappearing in this table? 7. References 1. Wendelboe AM, Van Rie A, Salmaso S, Englund JA: Duration of immunity against pertussis after natural infection or vaccination.Pediatr Infect Dis J. Sereina A. Herzog 2005; 24 (5 Suppl): S58-61 PubMed Abstract | Publisher Full Text Is the work clearly and accurately presented and does it cite the current literature? Partly Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: Infectious disease modeling, biostatistics I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Author Response 17 Nov 2021 Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK Thank you for taking the time to read our article. We have added an additional sensitivity analysis by assay and clarified and amended the technical points you raised. Please see our detailed responses to your comments below. Some minor points that should be addressed: The authors give in the introduction examples about pathogens which are ‘fully immunising’ like pertussis. To my knowledge pertussis natural infection protects 1. Thank you for bringing this to our attention, we have now corrected this. 2. Please clarify in the method section what is meant with “[…] were excluded, as well as studies with small number of data points.” Is ‘small number’ referring to number of different age (groups) reported in the study or number of total observations in a study, and what is ‘small’? We excluded Liang et al. 2013 since they only reported two age groups (ages 0 and 18-25). This has now been clarified in the text: “Studies which did not include estimates for individuals under 10 years old were excluded, as well as studies with which only reported two age groups.” 3. The full dataset used is not available on Zenodo (reference 25) but on GitHub (reference 33) [Method section, 1st paragraph, last sentence] – please check throughout the text those two references (e.g. last sentence Method section GitHub is mentioned with reference number 25). Thank you – these references have now been corrected throughout the text. Thank you – these references have now been corrected throughout the text. 4. In the method section it is stated that for the age at the cut-off time (a0) also uniform priors from 0 to 5 years are also investigated, however, I don’t see this prior appearing in the Table 2 with the different models and priors or in the Result section. This is referred to as “cut-off” in all the tables and text. Uniform priors between 0 and 20 were chosen for the cut-off (a0), and these are stated in Table 2. I have now clarified the description or the priors in the methods and added further information below the table to avoid confusion. Methods: “For the age at cut-off (a 0), uniform priors from 0 to 20 years were chosen as we were interested in the difference in FOI in children and young adults.” 5. Can the authors clarify in the discussion why they did not investigate a sensitivity analysis about the different assays used in the studies, e.g. allowing to vary waning according to assays used? Following yours and the other reviewers suggestion we have now included an additional sensitivity in the supplementary material, where we estimated the duration of antibody persistence by assay. We found that when we allowed waning to vary by assay (ELISA, IFA, HI and Neutralisation) we estimated the duration of antibody persistence to be between 1.1 years to 2.6 years for ELISA, HI and neutralisation. Are all the source data underlying the results available to ensure full reproduc Yes We found that when we allowed waning to vary by assay (ELISA, IFA, HI and longer than vaccination but not lifelong (see Wendelboe et al., 20051). Please rephrase the sentence accordingly. Are all the source data underlying the results available to ensure full reproduc Yes Page 18 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 longer than vaccination but not lifelong (see Wendelboe et al., 20051). Please rephrase the sentence accordingly. Thank you for bringing this to our attention, we have now corrected this. 2. Please clarify in the method section what is meant with “[…] were excluded, as well as studies with small number of data points.” Is ‘small number’ referring to number of different age (groups) reported in the study or number of total observations in a study, and what is ‘small’? We excluded Liang et al. 2013 since they only reported two age groups (ages 0 and 18-25). This has now been clarified in the text: “Studies which did not include estimates for individuals under 10 years old were excluded, as well as studies with which only reported two age groups.”   3. The full dataset used is not available on Zenodo (reference 25) but on GitHub (reference 33) [Method section, 1st paragraph, last sentence] – please check throughout the text those two references (e.g. last sentence Method section GitHub is mentioned with reference number 25). Thank you – these references have now been corrected throughout the text. 4. In the method section it is stated that for the age at the cut-off time (a0) also uniform priors from 0 to 5 years are also investigated, however, I don’t see this prior appearing in the Table 2 with the different models and priors or in the Result section. This is referred to as “cut-off” in all the tables and text. Uniform priors between 0 and 20 were chosen for the cut-off (a0), and these are stated in Table 2. I have now clarified the description or the priors in the methods and added further information below the table to avoid confusion. Methods: “For the age at cut-off (a 0), uniform priors from 0 to 20 years were chosen as we were interested in the difference in FOI in children and young adults.” 5. Can the authors clarify in the discussion why they did not investigate a sensitivity analysis about the different assays used in the studies, e.g. allowing to vary waning according to assays used? Following yours and the other reviewers suggestion we have now included an additional sensitivity in the supplementary material, where we estimated the duration of antibody persistence by assay. Thank you for bringing this to our attention, we have now corrected this. For IFA we observed a longer estimate of 7.7 (CrI: 3.0-14.3) years. However, we found wide credible intervals, particularly for the neutralisation and IFA which only had one study setting per assay. This highlights the need for more studies looking at the seroprevalence of HCoVs, and for better standardisation of assays. Page 19 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 "Finally, we explored the impact of the different assays used in the studies on the waning estimates. We allowed the waning estimate to vary by assay (extended data Table 7, Figure 525), whilst allowing FOI to vary by study, and alpha and cut-off to vary by setting. This model estimated the duration of antibody persistence to be similar for ELISA (2.63 [95% CrI: 0.94-9.09] years), HI (1.08 [95% CrI: 0.44-3.33] years) and neutralisation (1.28 [95% CrI: 0.25-50.0] years) assays, but longer for IFA (7.69 [95% CrI: 3.03-14.29] years). The credible intervals were wide, likely due to the small number of studies by assay." "When we allowed the waning estimate to vary by assay, we found a similar estimates of antibody persistence for ELISA, HI and neutralisation assays, ranging from 1.1 years to 2.6 years, and these are comparable to the estimates from the main model. However, for IFA, we observed a longer estimate of 7.7 years (CrI: 3.0-14.3). Due to the small number of studies, the credible intervals were large, particularly for the IFA and neutralisation assay, which only had one study setting for each assay. This highlights the need for more studies, and better standardisation of assays." 6. It is not clear which model was used to produce Figure 3B and which FOI estimates were used (from the main model, all models?) and how they were pooled. The description and interpretation of Figure 3B should be deepened. Thank you for this comment. We have clarified within the figure legend which model was used for Fig 3B. We have also expanded the description and interpretation of the figure within the results section. “To demonstrate the relationship between FOI and seropositivity at age 30, we created simulated scenarios under different sero-catalytic models. This has now been added to the introduction. Thank you for bringing this to our attention, we have now corrected this. Using the parameters for the relative change in FOI and waning estimated from the age-varying reverse catalytic model (where the relative change in FOI and the age at cut-off were simultaneously estimated across settings), we simulated the proportion of individuals aged 30 years that would be seropositive using a range of FOI estimates to show how the proportion changes using the different models. The catalytic model, which does not allow for seroreversion, results in the highest estimates of seropositivity at age 30 with increasing FOI. The age-varying FOI model results in higher estimates of seropositivity at age 30 compared with the reverse catalytic model. This is due to the FOI which was estimated to be almost twice as high in the older age group (with age at cut-off 8.49 [7.52 – 9.94] years) in the age-varying FOI model ( Figure 3A). We further explored the relationship between FOI, attack rates and the estimated number of infections by age. We used the pooled estimate across all studies of FOI to estimate the proportion exposed at a given age to provide an indication of how many infections we might expect to see by age under our modelling assumptions ( Figure 3B). We estimate that by two years, over 50% of the population will have at least one infection, and by age ten over 75% will have had more than four infections.” Some minor points which could improve readability: d Some minor points which could improve readability: d In the 3rd paragraph of the introduction, the authors provide the number of observations for the studies mentioned in this paragraph except for the studies conducted in Kenya and in New York. It would help to put the results in perspective to also have the numbers for those two studies. 1. In the 3rd paragraph of the introduction, the authors provide the number of observations for the studies mentioned in this paragraph except for the studies conducted in Kenya and in New York. It would help to put the results in perspective to also have the numbers for those two studies. 1. This has now been added to the introduction. Page 20 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Assay description: Table 1 indicates a very rough description of the assay used in the study. Thank you for bringing this to our attention, we have now corrected this. Table 3: I would suggest to extend the column titles a bit: “FOI”: clarifying that it is the one for the younger age group (i.e. lambda1); “Cut-off”: clarifying that his is about splitting observations into two age groups and adding the unit years. Thank you for your suggestion, the table headings have been changed. 7. Table 4: Why are not all models from Table 2 reappearing in this table? Since we explored many different models, we chose to focus this table only on the models which we felt were the most relevant, particular since this table is a comparison summary table, and the detailed information for all models is provided elsewhere in the manuscript. Competing Interests: No competing interests were disclosed. Referring to tables and figures in the supplementary material: change ‘extended da Figure 1, Table 1’ to ‘Supplementary Figure 1, Table 1’ etc. 3. Referring to tables and figures in the supplementary material: change ‘extended data Figure 1, Table 1’ to ‘Supplementary Figure 1, Table 1’ etc. 3. Thank you for your suggestion, however, this is the naming convention chosen by Wellcome Open research. Referring to tables and figures in the supplementary material: change ‘extended data Figure 1, Table 1’ to ‘Supplementary Figure 1, Table 1’ etc. 3. Thank you for your suggestion, however, this is the naming convention chosen by Wellcome Open research. 4. Table 1-4: add the abbreviations used within the table in the table legend (e.g. Table 1 HCoV). Thank you for bringing this to our attention, we have now corrected this. For interested readers it would be helpful to have more information in the supplementary material about the assays like exact assay name, was it qualitative/quantitative, sensitivity/specificity, what cut-off values were used, what is the lower limit of detection (LLOD), what happened with equivocal results (count towards being positive or negative), etc. 2. Assay description: Table 1 indicates a very rough description of the assay used in the study. For interested readers it would be helpful to have more information in the supplementary material about the assays like exact assay name, was it qualitative/quantitative, sensitivity/specificity, what cut-off values were used, what is the lower limit of detection (LLOD), what happened with equivocal results (count towards being positive or negative), etc. 2. Thank you, I have added some further information to Table 1 regarding the assay characteristics (antigen used and assay cut-off used in the study). I have also added an assay-specific sensitivity analysis and added further details regarding different assays in the discussion section. There is limited information in the literature on sensitivity and specificity of these assays, likely because these pathogens aren’t tested routinely in the community. Furthermore, given seroprevalence studies include asymptomatic and mildly symptomatic individuals, it is uncertain how these assays perform as validation is usually conducted in acute cases. Thank you, I have added some further information to Table 1 regarding the assay characteristics (antigen used and assay cut-off used in the study). I have also added an assay-specific sensitivity analysis and added further details regarding different assays in the discussion section. There is limited information in the literature on sensitivity and specificity of these assays, likely because these pathogens aren’t tested routinely in the community. Furthermore, given seroprevalence studies include asymptomatic and mildly symptomatic individuals, it is uncertain how these assays perform as validation is usually conducted in acute cases. Referring to tables and figures in the supplementary material: change ‘extended data Figure 1, Table 1’ to ‘Supplementary Figure 1, Table 1’ etc. 3. Thank you for your suggestion, however, this is the naming convention chosen by Wellcome Open research. 4. Table 1-4: add the abbreviations used within the table in the table legend (e.g. Table 1 HCoV). These have been added. 5. Table 2 - 5th model: maybe rephrase ‘using all data’ to ‘using also data on <1 year’. This has been amended. 6. These have been added. Since we explored many different models, we chose to focus this table only on the models which we felt were the most relevant, particular since this table is a comparison summary table, and the detailed information for all models is provided elsewhere in the manuscript. Competing Interests: No competing interests were disclosed. Reviewer Report 12 October 2021 Page 21 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Barnaby Flower y 1 Department of Infectious Disease, Faculty of Medicine, Imperial College London, London, UK 2 Oxford University Clinical Research Unit, Vietnam, Ho Chi Minh City, Vietnam Department of Infectious Disease, Faculty of Medicine, Imperial College London, London, UK 2 Oxford University Clinical Research Unit, Vietnam, Ho Chi Minh City, Vietnam Rees et al. provide a timely analysis of duration of natural immunity against seasonal human coronaviruses. They use an age-structured catalytic model which inputs historical seroprevalence data from five studies to estimate likely duration of seropositivity following seroconversion. They estimate HCoV antibodies last between 0.9 and 3.8 years, and find that the rate at which susceptible individuals acquire infection and seroconvert is higher in older children (>8.49 years) and adults than it is in younger children. The duration of seropositivity is consistent with previous estimates by Edridge et al. and tallies with emerging data concerning SARSCoV2. I am not a modeler and defer to my modeling colleague's review regarding aspects of the methodology. However to the non-specialist the methods appear robust and reproducible and they are transparently described in this well-written paper. As with all models, the outputs are only as good as the inputs, and in this case, data has been extracted from just 5 studies performed between 1965 and 2011 in four countries (Hong Kong, China, USA and Germany). The studies employ different assays, namely: enzyme-linked immunosorbent assays (ELISA), immunofluorescence assays (IFA), western blots, and complement fixation (CF), hemagglutination inhibition assays (HAI) and neutralisation assays which have different sensitivities and specificities, and I suspect this will lead to heterogeneous under- estimation of actual antibody positivity rates, particularly in comparison to current estimates of SARSCoV2 using highly sensitive ELISAs. The study would be strengthened by assay specific sensitivity analysis. The second limitation regards not accounting for seasonality. The authors point out that other studies have found that ignoring seasonality may overemphasize the role of adults in the transmission. Given that HCoVs are detected at much higher rates in the winter and spring seasons this makes it difficult to interpret one of the study's major findings of a higher force of infection in adults. This is mentioned in the 4th paragraph of the discussion but not in the context of the finding that the FOI was estimated to be twice as high in the older age group (6th paragraph) and perhaps this should be better tied together. https://doi.org/10.21956/wellcomeopenres.18416.r46089 © 2021 Flower B. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Are the conclusions drawn adequately supported by the results? Partly Competing Interests: No competing interests were disclosed. Reviewer Expertise: SARSCoV2 antibody seroprevalence, SARSCoV2 diagnostics, Hepatitis C Clinica Trialist. No modeling expertise. Reviewer Expertise: SARSCoV2 antibody seroprevalence, SARSCoV2 diagnostics, Hepatitis C Clinical Trialist. No modeling expertise. Barnaby Flower While these considerations might improve the paper, the limitations are well discussed and the manuscript is well worthy of indexing and dissemination. One minor point: In paragraph 4 of the methods it says: p g p y "There is little information on the attack rate of HCoVs, but there have been several systematic reviews and meta-analyses looking at unvaccinated individuals which have reported the attack rate to range between 15.2% – 22.5% in children and 3.5% – 10.7% in adults26–28." Page 22 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 This sentence is missing 'influenza'. Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Partly Competing Interests: No competing interests were disclosed. Reviewer Expertise: SARSCoV2 antibody seroprevalence, SARSCoV2 diagnostics, Hepatitis C Cli Trialist. No modeling expertise. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Author Response 17 Nov 2021 Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK Thank you for your comments on this manuscript. As you suggest, we have added an additional sensitivity analysis by assay and expanded our discussion around seasonality. Please see our detailed responses to your comments below. As with all models, the outputs are only as good as the inputs, and in this case, data has been extracted from just 5 studies performed between 1965 and 2011 in four countries (Hong Kong, China, USA and Germany). The studies employ different assays, This sentence is missing 'influenza'. Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Partly Results: "Finally, we explored the impact of the different assays used in the studies on the waning estimates. We allowed the waning estimate to vary by assay (extended data Table 7, Figure 525), whilst allowing FOI to vary by study, and alpha and cut-off to vary by setting. This model estimated the duration of antibody persistence to be similar for ELISA (2.63 [95% CrI: 0.94-9.09] years), HI (1.08 [95% CrI: 0.44-3.33] years) and neutralisation (1.28 [95% CrI: 0.25-50.0] years) assays, but longer for IFA (7.69 [95% CrI: 3.03-14.29] years). The credible intervals were wide, likely due to the small number of studies by assay." Discussion: "When we allowed the waning estimate to vary by assay, we found a similar estimates of antibody persistence for ELISA, HI and neutralisation assays, ranging from 1.1 years to 2.6 years, and these are comparable to the estimates from the main model. However, for IFA, we observed a longer estimate of 7.7 years (CrI: 3.0-14.3). Due to the small number of studies, the credible intervals were large, particularly for the IFA and neutralisation assay, which only had one study setting for each assay. This highlights the need for more studies, and better standardisation of assays." The second limitation regards not accounting for seasonality. The authors point out that other studies have found that ignoring seasonality may overemphasize the role of adults in the transmission. Given that HCoVs are detected at much higher rates in the winter and spring seasons this makes it difficult to interpret one of the study's major findings of a higher force of infection in adults. This is mentioned in the 4th paragraph of the discussion but not in the context of the finding that the FOI was estimated to be twice as high in the older age group (6th paragraph) and perhaps this should be better tied together. Thank you for your comment. The seasonality of HCoVs is an important consideration, and we have now expanded this in the discussion: “The time of year data collection occurred may influence our results, particularly given that the duration of antibody persistence is estimated to range between 0.9 (95% CrI: 0.6 - 1.6) years and 3.8 (95% CrI: 2.0 - 7.4) years. Data collection during high transmission periods would lead to an over estimate of both the FOI and the duration of antibody persistence. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Author Response 17 Nov 2021 Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK Thank you for your comments on this manuscript. As you suggest, we have added an additional sensitivity analysis by assay and expanded our discussion around seasonality. Please see our detailed responses to your comments below. As with all models, the outputs are only as good as the inputs, and in this case, data has been extracted from just 5 studies performed between 1965 and 2011 in four countries (Hong Kong, China, USA and Germany). The studies employ different assays, namely: enzyme-linked immunosorbent assays (ELISA), immunofluorescence assays (IFA), western blots, and complement fixation (CF), hemagglutination inhibition assays (HAI) and neutralisation assays which have different sensitivities and specificities, and I suspect this will lead to heterogeneous under-estimation of actual antibody positivity rates, particularly in comparison to current estimates of SARSCoV2 using highly sensitive ELISAs. The study would be strengthened by assay specific sensitivity analysis. Page 23 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Thank you for this suggestion, we have added an additional sensitivity looking at how the estimates of waning antibody duration change by assay. We found that when we allowed waning to vary by assay (ELISA, IFA, HI and Neutralisation) we estimated the duration of antibody persistence to be between 1.1 years to 2.6 years for ELISA, HI and neutralisation. For IFA we observed a longer estimate of 7.7 (CrI: 3.0-14.3) years. However, we found wide credible intervals, particularly for the neutralisation and IFA which only had one study setting per assay. This highlights the need for more studies looking at the seroprevalence of HCoVs, and for better standardisation of assays. Results: All the studies (except for Chan et al.23 who did not report this information), included within this analysis collected data over at least a six-month period. For this reason, the timing of data collection and the seasonality is unlikely to have biased our results.” Page 24 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 We have also further considered the Ferrari et al. paper in the context of our results. Their paper looked at measles outbreaks in Niger, which have a very different epidemic profile, with erratic outbreaks that vary by orders of magnitude, and large periods between epidemics. The epidemic profile is different for seasonal coronaviruses and we think that regular cyclic outbreaks do not strongly bias the model, as found by Whitaker & Farrington. We have expanded this within the discussion: “We also did not account for seasonality within this model, which may have under-estimated our FOI. Ferrari et al.42 found that ignoring seasonality may overemphasize the role of adults in the transmission, however, this was observed in measles in Niger, with outbreak peaks ranging over several orders of magnitude, and long periods between epidemics. The epidemic profile is different for seasonal coronaviruses, and therefore, this is unlikely to apply in this context.” While these considerations might improve the paper, the limitations are well discussed and the manuscript is well worthy of indexing and dissemination. One minor point: In paragraph 4 of the methods it says: "There is little information on the attack rate of HCoVs, but there have been several systematic reviews and meta-analyses looking at unvaccinated individuals which have reported the attack rate to range between 15.2% – 22.5% in children and 3.5% – 10.7% in adults26–28." In paragraph 4 of the methods it says: "There is little information on the attack rate of HCoVs, but there have been several systematic reviews and meta-analyses looking at unvaccinated individuals which have reported the attack rate to range between 15.2% – 22.5% in children and 3.5% – 10.7% in adults26–28." This sentence is missing 'influenza'. Thank you, this has now been corrected. Thank you, this has now been corrected. Competing Interests: No competing interests were disclosed. Reviewer Report 28 June 2021 Reviewer Report 28 June 2021 https://doi.org/10.21956/wellcomeopenres.18416.r44311 © 2021 Huang A. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Angkana T. Huang It is optional whether the a path. 4) The authors may consider additional metrics of model performance log predictive density (ELPD) which is based on approximate leave-one https://doi.org/10.1007/s11222-016-9696-4. The measure provides unc performance estimates which may help determine the superiority amo especially for ones where DIC were very similar. 5) It would be helpful to provide an equation to explicitly show the link waning rates and the duration of antibody persistence which were rep paper, and is the main finding of the study. 6) In contrast to the previous point, the authors provided an equation rates but barely touched upon this topic in the paper except for one oc May there be ideas that the authors wished to discuss but were not ful The authors provided ample background to understand the motivation and the approach of their work. Below are some minor points that I view should be addressed. 1) Please consider using the term ‘seropositivity waning rate’ instead of ‘immunity waning rate’ for ω as many of the assays in the dataset are binding assays which may not directly translate to protection, though correlated. 2) I appreciate that the authors included multiple sensitivity analyses in their study. I would like to suggest one more where waning rates are assay specific. The assays measure different functional aspects of the antibodies which may involve different subsets of the induced repertoires. Some hints exist in the authors’ results. Shao (2007) and Chan (2009) used IgG ELISA while Monto (1974) and Sarateanu (1980) were HI based. Looking at plots of the observed data, these tend to exhibit more prominent wanes. In Table S6, Model 1, 5, and 6 where Monto (1974) and Sarateanu (1980) were included, the estimated waning rates were much higher than the others. It could be that these were specific to HCoV-OC43, but looking at Table S2, the estimates for -NL63 and -HKU1 were similar to -OC43 (considering both the point estimates and credible intervals). Posterior density plots from code provided by the authors showed that combined cutoff of the Cavallaro (1970) and the Monto (1974) study in the main model, though converged, is bimodal. I wonder if this resulted from the model trying to accommodate the enforced single waning rate between studies which used different assays. This assessment will likely help the field reconcile discrepancies in serosurveys that were measured using different assays. Angkana T. Huang 1 Department of Biology and Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA 2 Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand Leveraging data on other human coronaviruses (HCoV) that have been circulating for longer, the authors estimated a range of antibody waning rates that might apply to SARS-CoV-2. They performed multiple sensitivity analyses to assess the robustness of their estimates. Their results also pointed towards transmission being higher in older individuals. Page 25 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 The authors provided ample background to understand the motivation work. Below are some minor points that I view should be addressed. 1) Please consider using the term ‘seropositivity waning rate’ instead o ω as many of the assays in the dataset are binding assays which may n protection, though correlated. 2) I appreciate that the authors included multiple sensitivity analyses in suggest one more where waning rates are assay specific. The assays m aspects of the antibodies which may involve different subsets of the in hints exist in the authors’ results. Shao (2007) and Chan (2009) used Ig and Sarateanu (1980) were HI based. Looking at plots of the observed more prominent wanes. In Table S6, Model 1, 5, and 6 where Monto (1 were included, the estimated waning rates were much higher than the these were specific to HCoV-OC43, but looking at Table S2, the estimat were similar to -OC43 (considering both the point estimates and credib density plots from code provided by the authors showed that combine (1970) and the Monto (1974) study in the main model, though converg this resulted from the model trying to accommodate the enforced sing studies which used different assays. This assessment will likely help the discrepancies in serosurveys that were measured using different assay 3) As the authors pointed out, data for some strains were from the sam for co-circulation. The cross-reactivity between strains, specifically with would affect the seropositivity. I acknowledge that literature is slim on reactivity is and the limited data may make it hard to infer from the da adding some discussion around how this would change the estimated readers adjust their expectations on what the true waning rates could intuition may involve extending the models to include cross-reactivity i circulation, imposing different sets of informative priors to the cross-re how those affect the inferred waning rates. Angkana T. Huang For example: p In the Methods, “The different assays used in each study for the different strains is shown, and where the antibody detected was specified this is included in the table.” ○ In the Results, “When the relative change in FOI and cut-off parameters were simultaneously estimated by setting (extended data Figure 2, Table 3) the duration of antibody persistence was estimated to be shorter, …”. ○ According to the captions, it occurs to me that the pairings are as follows: Table S1 with Figure S1, Table S3 with Figure S2, Table S5 with Figure S4. The pair of Table S2 is not obvious. It is unclear for Figure S3 which analysis this belongs to. Please also double check that the Extended data section in the main text matches the materials provided in the supplement. ○ “The inference was impleented in RJags…” The figure legend of Figure 3A says “reverse catalytic & age varying FOI” while the caption says “reverse catalytic model with time-varying FOI”. The colors also do not seem to match the legend (blue vs green vs grey). May be helpful to choose colors that are more different, especially that the lines in the legend keys are quite thin. ○ It is unclear which model was used to generate Figure 3B. I am guessing the “reverse catalytic model” from its matching color but explicit statements would be helpful to the readers. ○ It is unclear which model was used to generate Figure 3B. I am guessing the “reverse catalytic model” from its matching color but explicit statements would be helpful to the readers. ○ Please consider breaking this portion in the Discussion into a new sentence, “... and a meta- analysis of contact tracing studies found that children had 56% (31% – 71%) lower odds of becoming an infected contact compared with adults52.” ○ Please consider breaking this portion in the Discussion into a new sentence, “... and a meta- analysis of contact tracing studies found that children had 56% (31% – 71%) lower odds of becoming an infected contact compared with adults52.” ○ Angkana T. Huang 3) As the authors pointed out, data for some strains were from the same settings. This is evidence for co-circulation. The cross-reactivity between strains, specifically within alpha- and beta-CoVs, would affect the seropositivity. I acknowledge that literature is slim on what the degree of cross- reactivity is and the limited data may make it hard to infer from the data. Anyhow, please consider adding some discussion around how this would change the estimated waning rate as it will help readers adjust their expectations on what the true waning rates could be. One way to gain such intuition may involve extending the models to include cross-reactivity in studies with known co- circulation, imposing different sets of informative priors to the cross-reactivity rate, and assess how those affect the inferred waning rates. It is optional whether the authors would take on this path. 4) The authors may consider additional metrics of model performance, for instance, the expected log predictive density (ELPD) which is based on approximate leave-one-out cross-validation, https://doi.org/10.1007/s11222-016-9696-4. The measure provides uncertainties around the performance estimates which may help determine the superiority among the suite of models, especially for ones where DIC were very similar. 5) It would be helpful to provide an equation to explicitly show the link between the estimated waning rates and the duration of antibody persistence which were reported throughout the paper, and is the main finding of the study. 6) In contrast to the previous point, the authors provided an equation linking the FOI to the attack rates but barely touched upon this topic in the paper except for one occurrence in the Discussion. May there be ideas that the authors wished to discuss but were not fully expressed in the text? Page 26 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 7) It is unclear what the motivation for the analysis which led to Figure 3 was. Providing some context, especially how that would better our projections on SARS-CoV-2 would be helpful. 8) Lastly, there are some truncated sentences or missing/inconsistent information in the paper that needs to be fixed. Is the work clearly and accurately presented and does it cite the current literature? Partly Is the work clearly and accurately presented and does it cite the current literature? Partly Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? If applicable, is the statistical analysis and its interpretation appropriate? Yes Thank you for this suggestion, we have added an additional sensitivity looking at how the estimates of waning antibody duration change by assay. We found that when we allowed waning to vary by assay (ELISA, IFA, HI and Neutralisation) we estimated the duration of antibody persistence to be between 1.1 years to 2.6 years for ELISA, HI and neutralisation. For IFA we observed a longer estimate of 7.7 (CrI: 3.0-14.3) years (See Extended data Figure 5 and Table 7). However, we found wide credible intervals, particularly for the neutralisation and IFA which only had one study setting per assay. This highlights the need for more studies looking at the I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Author Response 17 Nov 2021 Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK Thank you for taking the time to review our article. We have added an additional sensitivity analysis by assay and technical details to the manuscript, as you suggest. Please see our detailed responses to your comments below. 1) Please consider using the term ‘seropositivity waning rate’ instead of ‘immunity waning rate’ for ω as many of the assays in the dataset are binding assays which may not directly translate to protection, though correlated. Thank you for this comment. We agree with the point you raise, and have amended “immunity waning rate” to “seropositivity waning rate” as suggested. 2) I appreciate that the authors included multiple sensitivity analyses in their study. I would like to suggest one more where waning rates are assay specific. The assays measure different functional aspects of the antibodies which may involve different subsets of the induced repertoires. Some hints exist in the authors’ results. Shao (2007) and Chan (2009) used IgG ELISA while Monto (1974) and Sarateanu (1980) were HI based. Looking at plots of the observed data, these tend to exhibit more prominent wanes. In Table S6, Model 1, 5, and 6 where Monto (1974) and Sarateanu (1980) were included, the estimated waning rates were much higher than the others. If applicable, is the statistical analysis and its interpretation appropriate? Yes Are the conclusions drawn adequately supported by the results? Page 27 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: infectious disease modeling I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Author Response 17 Nov 2021 Eleanor Rees, London School of Hygiene & Tropical Medicine, London, UK Thank you for taking the time to review our article. We have added an additional sensitivity analysis by assay and technical details to the manuscript, as you suggest. Please see our detailed responses to your comments below. 1) Please consider using the term ‘seropositivity waning rate’ instead of ‘immunity waning rate’ for ω as many of the assays in the dataset are binding assays which may not directly translate to protection, though correlated. Thank you for this comment. We agree with the point you raise, and have amended “immunity waning rate” to “seropositivity waning rate” as suggested. 2) I appreciate that the authors included multiple sensitivity analyses in their study. I would like to suggest one more where waning rates are assay specific. The assays measure different functional aspects of the antibodies which may involve different subsets of the induced repertoires. Some hints exist in the authors’ results. Shao (2007) and Chan (2009) used IgG ELISA while Monto (1974) and Sarateanu (1980) were HI based. Looking at plots o the observed data, these tend to exhibit more prominent wanes. In Table S6, Model 1, 5, and 6 where Monto (1974) and Sarateanu (1980) were included, the estimated waning rates were much higher than the others. It could be that these were specific to HCoV-OC43, but looking at Table S2, the estimates for -NL63 and -HKU1 were similar to -OC43 (considering both the point estimates and credible intervals). Posterior density plots from code provided by the authors showed that combined cutoff of the Cavallaro (1970) and the Monto (1974) study in the main model, though converged, is bimodal. I wonder if this resulted from the model trying to accommodate the enforced single waning rate between studies which used different assays. This assessment will likely help the field reconcile discrepancies in serosurveys that were measured using different assays. seroprevalence of HCoVs, and for better standardisation of assays. "Finally, we explored the impact of the different assays used in the studies on the waning estimates. We allowed the waning estimate to vary by assay (extended data Table 7, Figure 525), whilst allowing FOI to vary by study, and alpha and cut-off to vary by setting. This model estimated the duration of antibody persistence to be similar for ELISA (2.63 [95% CrI: 0.94-9.09] years), HI (1.08 [95% CrI: 0.44-3.33] years) and neutralisation (1.28 [95% CrI: 0.25-50.0] years) assays, but longer for IFA (7.69 [95% CrI: 3.03-14.29] years). The credible intervals were wide, likely due to the small number of studies by assay." "When we allowed the waning estimate to vary by assay, we found a similar estimates of antibody persistence for ELISA, HI and neutralisation assays, ranging from 1.1 years to 2.6 years, and these are comparable to the estimates from the main model. However, for IFA, we observed a longer estimate of 7.7 years (CrI: 3.0-14.3). Due to the small number of studies, the credible intervals were large, particularly for the IFA and neutralisation assay, which only had one study setting for each assay. This highlights the need for more studies, and better standardisation of assays." 3) As the authors pointed out, data for some strains were from the same settings. This is evidence for co-circulation. The cross-reactivity between strains, specifically within alpha- and beta-CoVs, would affect the seropositivity. I acknowledge that literature is slim on what the degree of cross-reactivity is and the limited data may make it hard to infer from the data. Anyhow, please consider adding some discussion around how this would change the estimated waning rate as it will help readers adjust their expectations on what the true waning rates could be. One way to gain such intuition may involve extending the models to include cross-reactivity in studies with known co-circulation, imposing different sets of informative priors to the cross-reactivity rate, and assess how those affect the inferred waning rates. It is optional whether the authors would take on this path. Thank you for this comment. We have added some additional discussion surrounding cross- reactivity and cross-protection to the discussion and explored how this might have impacted our results. Since we only have a limited number of datasets and data points, we don’t feel that we have sufficient data to include cross-reactivity within this analysis. If applicable, is the statistical analysis and its interpretation appropriate? Yes Due to the small number of studies, the credible intervals were large, particularly for the IFA and neutralisation assay, which only had one study setting for each assay. This highlights the need for more studies, and better standardisation of assays."                         3) As the authors pointed out, data for some strains were from the same settings. This is evidence for co-circulation. The cross-reactivity between strains, specifically within alpha- and beta-CoVs, would affect the seropositivity. I acknowledge that literature is slim on what the degree of cross-reactivity is and the limited data may make it hard to infer from the data. Anyhow, please consider adding some discussion around how this would change the estimated waning rate as it will help readers adjust their expectations on what the true waning rates could be. One way to gain such intuition may involve extending the models to include cross-reactivity in studies with known co-circulation, imposing different sets of informative priors to the cross-reactivity rate, and assess how those affect the inferred waning rates. It is optional whether the authors would take on this path. Thank you for this comment. We have added some additional discussion surrounding cross- reactivity and cross-protection to the discussion and explored how this might have impacted our results. Since we only have a limited number of datasets and data points, we don’t feel that we have sufficient data to include cross-reactivity within this analysis. Discussion: "In addition, we did not take into consideration cross-protection between seasonal coronavirus strains. There is some evidence of cross protective immunity between seasonal coronavirus strains, and in settings where there is co-circulating HCoV strains, this may lead to a higher prevalence. There is also evidence that there is cross-reactivity between different coronaviruses, which may lead to false positive results. A recent systematic review found that there was some cross-reactivity that occurred within alpha (HCoV-229E and HCoV-NL63) and beta (HCoV-OC43 and HCoV-HKU1) coronaviruses, but minimal reactivity between alpha and beta coronaviruses7 . However, it is not clear whether cross-reactivity equates to cross-protection. False positives due to cross-reactivity would lead to an over-estimation of seroprevalence in a setting This would lead If applicable, is the statistical analysis and its interpretation appropriate? Yes It could be that these were specific to HCoV-OC43, but looking at Table S2, the estimates for -NL63 and -HKU1 were similar to -OC43 (considering both the point estimates and credible intervals). Posterior density plots from code provided by the authors showed that combined cutoff of the Cavallaro (1970) and the Monto (1974) study in the main model, though converged, is bimodal. I wonder if this resulted from the model trying to accommodate the enforced single waning rate between studies which used different assays. This assessment will likely help the field reconcile discrepancies in serosurveys that were measured using different assays. Thank you for this suggestion, we have added an additional sensitivity looking at how the estimates of waning antibody duration change by assay. We found that when we allowed waning to vary by assay (ELISA, IFA, HI and Neutralisation) we estimated the duration of antibody persistence to be between 1.1 years to 2.6 years for ELISA, HI and neutralisation. For IFA we observed a longer estimate of 7.7 (CrI: 3.0-14.3) years (See Extended data Figure 5 and Table 7). However, we found wide credible intervals, particularly for the neutralisation and IFA which only had one study setting per assay. This highlights the need for more studies looking at the Page 28 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 seroprevalence of HCoVs, and for better standardisation of assays. Results: "Finally, we explored the impact of the different assays used in the studies on the waning estimates. We allowed the waning estimate to vary by assay (extended data Table 7, Figure 525), whilst allowing FOI to vary by study, and alpha and cut-off to vary by setting. This model estimated the duration of antibody persistence to be similar for ELISA (2.63 [95% CrI: 0.94-9.09] years), HI (1.08 [95% CrI: 0.44-3.33] years) and neutralisation (1.28 [95% CrI: 0.25-50.0] years) assays, but longer for IFA (7.69 [95% CrI: 3.03-14.29] years). The credible intervals were wide, likely due to the small number of studies by assay." Discussion: "When we allowed the waning estimate to vary by assay, we found a similar estimates of antibody persistence for ELISA, HI and neutralisation assays, ranging from 1.1 years to 2.6 years, and these are comparable to the estimates from the main model. However, for IFA, we observed a longer estimate of 7.7 years (CrI: 3.0-14.3). seroprevalence of HCoVs, and for better standardisation of assays. The relative change in FOI and cut-off were similar for both models. This model had a lower DIC (476.8 compared with 480.8; with a DIC difference of 4), WAIC (536.2 compared with 546.4), and LOO (546.9 compared with 560.9), which suggests that this model has an improved fit compared with the model with more informed priors. However, the high FOI estimates indicate that this model may be less plausible ( Table 4).” “When the relative change in FOI and cut-off parameters were simultaneously estimated by setting (extended data Figure 2, Table 3 25 ) the duration of antibody persistence was estimated to be shorter, 2.20 (95% CrI: 1.57 - 3.08) years, although the confidence intervals overlap with the main model. The FOI ranged from 0.04 (95% CrI: 0.03 - 0.06) to 0.88 (95% CrI: 0.67 - 1.19). The overall model DIC (548.2 compared with 480.8), WAIC (622.0 compared with 546.5) and LOO (633.2 compared with 560.9) were higher, indicating that this model did not have as much support (Table 4).” 5) It would be helpful to provide an equation to explicitly show the link between the estimated waning rates and the duration of antibody persistence which were reported throughout the paper, and is the main finding of the study. The duration of antibody persistence was estimated a 1/seropositivity waning rate. I have now included this as an equation in the text. “The duration of antibody persistence was estimated as follows: Duration of antibody persistence = 1/ ω” 4) The authors may consider additional metrics of model performance, for instance, the     (ELPD) which is based on approximate leave-one-out cross-validation, https://doi.org/10.1007/s11222-016-9696-4. The measure provides uncertainties around the performance estimates which may help determine the superiority among the suite of models, especially for ones where DIC were very similar. Thank you for the suggestion. We agree that DIC as a metric has limitations for model selection, particularly given our DIC estimates were so similar, which is why in the main results and abstract we chose to report a range of estimates from the different models, rather than focussing simply on just one model. We have now added leave-one-out cross validation (LOO) to estimate the out of sample prediction ability. Additionally, we have added WAIC, which offers advantages over the DIC for Bayesian models, since it uses the entire posterior distribution. seroprevalence of HCoVs, and for better standardisation of assays. "In addition, we did not take into consideration cross-protection between seasonal coronavirus strains. There is some evidence of cross protective immunity between seasonal coronavirus strains, and in settings where there is co-circulating HCoV strains, this may lead to a higher prevalence. There is also evidence that there is cross-reactivity between different coronaviruses, which may lead to false positive results. A recent systematic review found that there was some cross-reactivity that occurred within alpha (HCoV-229E and HCoV-NL63) and beta (HCoV-OC43 and HCoV-HKU1) coronaviruses, but minimal reactivity between alpha and beta coronaviruses7 . However, it is not clear whether cross-reactivity equates to cross-protection. False positives due to cross-reactivity would lead to an over-estimation of seroprevalence in a setting. This would lead to a higher plateau in older ages, and therefore would generally lead to an over-estimation of both the FOI and the duration of antibody persistence." Page 29 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 4) The authors may consider additional metrics of model performance, for instance, the     (ELPD) which is based on approximate leave-one-out cross-validation, https://doi.org/10.1007/s11222-016-9696-4. The measure provides uncertainties around the performance estimates which may help determine the superiority among the suite of models, especially for ones where DIC were very similar. Thank you for the suggestion. We agree that DIC as a metric has limitations for model selection, particularly given our DIC estimates were so similar, which is why in the main results and abstract we chose to report a range of estimates from the different models, rather than focussing simply on just one model. We have now added leave-one-out cross validation (LOO) to estimate the out of sample prediction ability. Additionally, we have added WAIC, which offers advantages over the DIC for Bayesian models, since it uses the entire posterior distribution. We have reported these for all models, and found the same general trends as observed using DIC. “A sensitivity analysis was conducted using less informative priors for the FOI parameters, where a normal distribution was used (extended data Figure 1, Table 1 25 ). This model estimated a shorter duration of antibody persistence [0.93 (95% CrI: 0.60 - 1.64) years]. The FOI across all studies and strains were higher, ranging from 0.09 (95% CrI: 0.04 - 0.16) to 3.22 (95% CrI: 1.95 - 4.85), with six studies reporting FOI estimates > one, which is equivalent to an attack rate of >63%. seroprevalence of HCoVs, and for better standardisation of assays. Our primary motivation for exploring attack rates was to inform our priors for the FOI estimates using attack rates from influenza (reported on in the methods). We realise we could have been clearer, so we have clarified this in the methods. We also compare the attack rates estimated for seasonal HCoV with influenza in the results and discussion to ensure that model parameter estimates were within plausible ranges. “For the FOI, there is little information on the attack rate of HCoVs. However, there have been several systematic reviews and meta-analyses looking at influenza in unvaccinated individuals which have reported the attack rates to range between 15.2% – 22.5% in children and 3.5% – 10.7% in adults 26– 28 . Modelling studies using serological influenza data predicted estimates from 20 – 60% 29, 30 . Based on the epidemiology of these viruses in children31, we expect the attack rate for HCoV may be lower. Therefore, we selected a Gamma distribution, with a mean of 0.3 (shape = 1.2 and scale = 0.25) and this corresponds to an attack rate of 26% and covers a range of plausible values. For the age at cut-off (a 0), uniform priors from 0 to 20 years were chosen as we were interested in the difference in FOI in children and young adults.” Results: “A sensitivity analysis was conducted using less informative priors for the FOI parameters, where a normal distribution was used (extended data Figure 1, Table 1 25 ). This model estimated a shorter duration of antibody persistence [0.93 (95% CrI: 0.60 - 1.64) years]. The FOI across all studies and strains were higher, ranging from 0.09 (95% CrI: 0.04 - 0.16) to 3.22 (95% CrI: 1.95 - 4.85), with six studies reporting FOI estimates > one, which is equivalent to an attack rate of >63%. The relative change in FOI and cut-off were similar for both models. This model had a lower DIC (476.8 compared with 480.8; with a DIC difference of 4), WAIC (536.2 compared with 546.4), and LOO (546.9 compared with 560.9), which suggests that this model has an improved fit compared with the model with more informed priors. However, the high FOI estimates indicate that this model may be less plausible ( Table 4).” “When we used less informed priors for the FOI, a lower estimate of duration of seropositivity was obtained. seroprevalence of HCoVs, and for better standardisation of assays. We have reported these for all models, and found the same general trends as observed using DIC. “A sensitivity analysis was conducted using less informative priors for the FOI parameters, where a normal distribution was used (extended data Figure 1, Table 1 25 ). This model estimated a shorter duration of antibody persistence [0.93 (95% CrI: 0.60 - 1.64) years]. The FOI across all studies and strains were higher, ranging from 0.09 (95% CrI: 0.04 - 0.16) to 3.22 (95% CrI: 1.95 - 4.85), with six studies reporting FOI estimates > one, which is equivalent to an attack rate of >63%. The relative change in FOI and cut-off were similar for both models. This model had a lower DIC (476.8 compared with 480.8; with a DIC difference of 4), WAIC (536.2 compared with 546.4), and LOO (546.9 compared with 560.9), which suggests that this model has an improved fit compared with the model with more informed priors. However, the high FOI estimates indicate that this model may be less plausible ( Table 4).” “When the relative change in FOI and cut-off parameters were simultaneously estimated by setting (extended data Figure 2, Table 3 25 ) the duration of antibody persistence was estimated to be shorter, 2.20 (95% CrI: 1.57 - 3.08) years, although the confidence intervals overlap with the main model. The FOI ranged from 0.04 (95% CrI: 0.03 - 0.06) to 0.88 (95% CrI: 0.67 - 1.19). The overall model DIC (548.2 compared with 480.8), WAIC (622.0 compared with 546.5) and LOO (633.2 compared with 560.9) were higher, indicating that this model did not have as much support (Table 4).” 5) It would be helpful to provide an equation to explicitly show the link between the estimated waning rates and the duration of antibody persistence which were reported throughout the paper, and is the main finding of the study. The duration of antibody persistence was estimated a 1/seropositivity waning rate. I have now included this as an equation in the text. Page 30 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 6) In contrast to the previous point, the authors provided an equation linking the FOI to the attack rates but barely touched upon this topic in the paper except for one occurrence in the Discussion. May there be ideas that the authors wished to discuss but were not fully expressed in the text? seroprevalence of HCoVs, and for better standardisation of assays. We used estimate across all studies of FOI to estimate the proportion exposed at a given ag indication of how many infections we might expect to see by age under our modell assumptions ( Figure 3B). We estimate that by two years, over 50% of the populatio least one infection, and by age ten over 75% will have had more than four infection   8) Lastly, there are some truncated sentences or missing/inconsistent informa paper that needs to be fixed. For example: Thank you for highlighting these, these have all been fixed. In the Methods, “The different assays used in each study for the differen shown, and where the antibody detected was specified this is included i ○ This has been amended. In the Results, “When the relative change in FOI and cut-off parameters simultaneously estimated by setting (extended data Figure 2, Table 3) th antibody persistence was estimated to be shorter, …”. ○ This has been amended. According to the captions, it occurs to me that the pairings are as follow Table S1 with Figure S1, Table S3 with Figure S2, Table S5 with Figure S4 Table S2 is not obvious. It is unclear for Figure S3 which analysis this be Please also double check that the Extended data section in the main tex materials provided in the supplement. ○ The contents of the extended data can be found at the end of the manuscript unde data” section: The supplementary contains the following: seroprevalence of HCoVs, and for better standardisation of assays. However, this model produced higher estimates of FOI, with six studies reported FOI estimates in the young age group greater than one (attack rate >63%). There is limited information on the attack rate of seasonal HCoV, however there have been numerous studies looking at influenza. Previous systematic reviews have estimated the attack rate of influenza to be between 3.5% and 22.5% 26– 28 , whilst modelling studies have estimated this to be higher, 20 – 60% 29, 30 . Based on reporting rates of seasonal HCoV we would expect the attack rate to be lower than influenza. Therefore, this suggests that the results from the model with less informative priors are less plausible.” 7) It is unclear what the motivation for the analysis which led to Figure 3 was. Providing some context, especially how that would better our projections on SARS-CoV-2 would be Page 31 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 helpful. Thank you for this comment. Our motivation for this figure was to look at the relati between FOI, attack rates, and the number of infections you might expect to observ ages. As you suggest, we have expanded the interpretation of this figure in the last the results: “To demonstrate the relationship between FOI and seropositivity at age 30, we crea scenarios under different sero-catalytic models. Using the parameters for the relati FOI and waning estimated from the age-varying reverse catalytic model (where the change in FOI and the age at cut-off were simultaneously estimated across settings simulated the proportion of individuals aged 30 years that would be seropositive u FOI estimates to show how the proportion changes using the different models. The model, which does not allow for seroreversion, results in the highest estimates of se age 30 with increasing FOI. The age-varying FOI model results in higher estimates seropositivity at age 30 compared with the reverse catalytic model. This is due to th was estimated to be almost twice as high in the older age group (with age at cut-of 9.94] years) in the age-varying FOI model ( Figure 3A). We further explored the rela between FOI, attack rates and the estimated number of infections by age. helpful. Thank you for this comment. Our motivation for this figure was to look at the relationship between FOI, attack rates, and the number of infections you might expect to observe in younger ages. As you suggest, we have expanded the interpretation of this figure in the last paragraph of the results: “To demonstrate the relationship between FOI and seropositivity at age 30, we created simulated scenarios under different sero-catalytic models. Using the parameters for the relative change in FOI and waning estimated from the age-varying reverse catalytic model (where the relative change in FOI and the age at cut-off were simultaneously estimated across settings), we simulated the proportion of individuals aged 30 years that would be seropositive using a range of FOI estimates to show how the proportion changes using the different models. The catalytic model, which does not allow for seroreversion, results in the highest estimates of seropositivity at age 30 with increasing FOI. The age-varying FOI model results in higher estimates of seropositivity at age 30 compared with the reverse catalytic model. This is due to the FOI which was estimated to be almost twice as high in the older age group (with age at cut-off 8.49 [7.52 – 9.94] years) in the age-varying FOI model ( Figure 3A). We further explored the relationship between FOI, attack rates and the estimated number of infections by age. We used the pooled estimate across all studies of FOI to estimate the proportion exposed at a given age to provide an indication of how many infections we might expect to see by age under our modelling assumptions ( Figure 3B). We estimate that by two years, over 50% of the population will have at least one infection, and by age ten over 75% will have had more than four infections.” 8) Lastly, there are some truncated sentences or missing/inconsistent information in the paper that needs to be fixed. For example: 8) Lastly, there are some truncated sentences or missing/inconsistent information in the paper that needs to be fixed. For example: 8) Lastly, there are some truncated sentences or missing/inconsistent information in the paper that needs to be fixed. For example: Thank you for highlighting these, these have all been fixed. helpful. ○ The contents of the extended data can be found at the end of the manuscript under the “extended data” section: The supplementary contains the following: Page 32 of 33 Wellcome Open Research 2021, 6:138 Last updated: 23 MAR 2022 • Sensitivity analysis: Less informed priors for FOI (supplementary Table 1 and Figure 1) ○ • Sensitivity analysis: Waning estimated by strain (supplementary Table 2) ○ • Sensitivity analysis: Alpha and cut-off jointly simultaneously by study (supplementary Table 3 and Figure 2) ○ • Reverse catalytic model (supplementary Figure 3 and Table 4) ○ • Sensitivity analysis: Including the youngest age groups (<1 year) (supplementary Table 5 and Figure 4) ○ • Sensitivity analysis: Refitting the model using data from only two strains (supplementary Table 6) ○ • Sensitivity analysis: Waning estimated by assay (supplementary Table 7 and Figure 5) ○ ○ We have also included subheadings within the supplementary material to make it easier to • Sensitivity analysis: Less informed priors for FOI (supplementary Table 1 and Figure 1) ○ • Sensitivity analysis: Waning estimated by strain (supplementary Table 2) ○ • Sensitivity analysis: Alpha and cut-off jointly simultaneously by study (supplementary Table 3 and Figure 2) ○ • Reverse catalytic model (supplementary Figure 3 and Table 4) ○ • Sensitivity analysis: Including the youngest age groups (<1 year) (supplementary Table 5 and Figure 4) ○ • Sensitivity analysis: Refitting the model using data from only two strains (supplementary Table 6) ○ • Sensitivity analysis: Waning estimated by assay (supplementary Table 7 and Figure 5) ○ ○ We have also included subheadings within the supplementary material to make it easier to “The inference was impleented in RJags…” ○ This has been amended. The figure legend of Figure 3A says “reverse catalytic & age varying FOI” while the caption says “reverse catalytic model with time-varying FOI”. The colors also do not seem to match the legend (blue vs green vs grey). May be helpful to choose colors that are more different, especially that the lines in the legend keys are quite thin. ○ This has been amended. The figure legend of Figure 3A says “reverse catalytic & age varying FOI” while the caption says “reverse catalytic model with time-varying FOI”. The colors also do not seem to match the legend (blue vs green vs grey). helpful. In the Methods, “The different assays used in each study for the different strains is shown, and where the antibody detected was specified this is included in the table.” ○ This has been amended. In the Results, “When the relative change in FOI and cut-off parameters were simultaneously estimated by setting (extended data Figure 2, Table 3) the duration of antibody persistence was estimated to be shorter, …”. ○ This has been amended. According to the captions, it occurs to me that the pairings are as follows: Table S1 with Figure S1, Table S3 with Figure S2, Table S5 with Figure S4. The pair of Table S2 is not obvious. It is unclear for Figure S3 which analysis this belongs to. Please also double check that the Extended data section in the main text matches the materials provided in the supplement. ○ The contents of the extended data can be found at the end of the manuscript under the “extended data” section: The supplementary contains the following: Thank you for highlighting these, these have all been fixed. In the Methods, “The different assays used in each study for the different strains is shown, and where the antibody detected was specified this is included in the table.” ○ This has been amended. Thank you for highlighting these, these have all been fixed. In the Methods, “The different assays used in each study for the different strains is shown, and where the antibody detected was specified this is included in the table.” ○ This has been amended. In the Results, “When the relative change in FOI and cut-off parameters were simultaneously estimated by setting (extended data Figure 2, Table 3) the duration of antibody persistence was estimated to be shorter, …”. ○ This has been amended. According to the captions, it occurs to me that the pairings are as follows: Table S1 with Figure S1, Table S3 with Figure S2, Table S5 with Figure S4. The pair of Table S2 is not obvious. It is unclear for Figure S3 which analysis this belongs to. Please also double check that the Extended data section in the main text matches the materials provided in the supplement. helpful. May be helpful to choose colors that are more different, especially that the lines in the legend keys are quite thin. ○ I have updated the figure caption to clarify the models used to create Figure 3. I have changed the colours to allow for a clearer distinction between the models. It is unclear which model was used to generate Figure 3B. I am guessing the “reverse catalytic model” from its matching color but explicit statements would be helpful to the readers. ○ It is unclear which model was used to generate Figure 3B. I am guessing the “reverse catalytic model” from its matching color but explicit statements would be helpful to the readers. ○ I have updated the figure caption to clarify the model used to create Fig 3b. “Figure 3. ( A) Proportion of individuals age 30 who are seropositive for different estimates of force of infection (FOI). The catalytic model is shown in red, the reverse catalytic model in green, and the reverse catalytic model with age-varying FOI is shown in blue. Model estimates were used for the parameter values (relative change in FOI (alpha),1.93 [1.69 – 2.19]; waning, 0.45 [0.32 – 0.64]; cut-off, 8.49 [7.52 – 9.94]). ( B) Estimated proportion of individuals experiencing infections by age estimated from the age-varying reverse catalytic model (more informed priors) using the pooled median estimate across studies for FOI (0.46), and median estimates for waning (0.45), alpha (1.93) and cut-off (8.49).” Please consider breaking this portion in the Discussion into a new sentence, “... and a meta-analysis of contact tracing studies found that children had 56% (31% – 71%) lower odds of becoming an infected contact compared with adults52.” ○ This has been amended. Please consider breaking this portion in the Discussion into a new sentence, “... and a meta-analysis of contact tracing studies found that children had 56% (31% – 71%) lower odds of becoming an infected contact compared with adults52.” ○ This has been amended. Competing Interests: No competing interests were disclosed. Page 33 of 33

https://openalex.org/W4220724878

https://discovery.ucl.ac.uk/10146001/1/Makin_1-s2.0-S2211124722002595-main.pdf

English

Beyond body maps: Information content of specific body parts is distributed across the somatosensory homunculus

Cell reports

cc-by

In brief The human body is represented in a topographic manner in the primary somatosensory cortex. Muret et al. show that the information related to each body part is not limited to its highly selective region within the topographic map and is instead more broadly distributed than previously considered. Correspondence [email protected] Correspondence [email protected] Beyond body maps: Information content of specific body parts is distributed across the somatosensory homunculus Beyond body maps: Information content of specific body parts is distributed across the somatosensory homunculus Authors Dollyane Muret, Victoria Root, Paulina Kieliba, Danielle Clode, Tamar R. Makin Article Article Authors Graphical abstract Dollyane Muret, Victoria Root, Paulina Kieliba, Danielle Clode, Tamar R. Makin INTRODUCTION While the organizing principles of S1 remain so far relatively un- questioned, other intrinsic organizing principles beyond somato- topy, e.g., representation of ethologically relevant actions, were suggested to underlie the general organization of body-related motor maps in the primary motor cortex (hereafter M1) (Graziano and Aflalo, 2007). Yet we know that M1 and S1 are tightly coupled, both anatomically and functionally (Matyas et al., 2010; Catani et al., 2012; Kumar et al., 2019). For instance, recent evidence in rodents (Auffret etal., 2018; Halleyetal.,2020) and non-human pri- mates (Widener and Cheney, 1997; Baldwin et al., 2018; Mayer et al., 2019) demonstrate that S1 stimulation can evoke move- ments or affect muscle activity. Active touch and habitual motor behavior were also suggested to impact S1 organization (Demp- sey-Jones et al., 2019; Cybulska-Klosowicz et al., 2020). More- over, the M1 and S1 hand regions were found to share similar representational features (Ejaz et al., 2015; Wesselink et al., 2019), which are in both cases better explained by inter-finger co-use in daily life than by topographic organization (Ejaz et al., 2015). Thus, despite their physiological differences, S1 and M1 may share more functional organization than previously thought, especially in terms of information content. This raises the question of whether S1 could also contain additional representational pat- terns underlying its topographic organization. Contrary to its motor counterpart, the primary somatosensory cortex (hereafter S1) is considered highly topographically orga- nized, with relatively high levels of selectivity within each body part’s representation (Schieber, 2001; Cunningham et al., 2013; Huber et al., 2020). This perspective over S1 organization arises from a long-lasting mapping tradition, initiated in the 19th cen- tury (Fritsch and Hitzig, 1870; Ferrier, 1873; Penfield and Bol- drey, 1937) and continued since then in electrophysiology (Merzenich et al., 1978; Kaas et al., 1979; Baldwin et al., 2017), cortical stimulation (Roux et al., 2018; Sun et al., 2021), and neuroimaging studies (Nakamura et al., 1998; Ger- mann et al., 2020; Saadon-Grosman et al., 2020; Willoughby et al., 2021). This conventional mapping approach assigns brain function to a given cortical area by selecting the most responsive body part for a set of neurons or voxels in a winner-takes-all manner. SUMMARY The homunculus in primary somatosensory cortex (S1) is famous for its body part selectivity, but this domi- nant feature may eclipse other representational features, e.g., information content, also relevant for S1 orga- nization. Using multivariate fMRI analysis, we ask whether body part information content can be identified in S1 beyond its primary region. Throughout S1, we identify significant representational dissimilarities between body parts but also subparts in distant non-primary regions (e.g., between the hand and the lips in the foot region and between different face parts in the foot region). Two movements performed by one body part (e.g., the hand) could also be dissociated well beyond its primary region (e.g., in the foot and face regions), even within Brodmann area 3b. Our results demonstrate that information content is more distributed across S1 than selectivity maps suggest. This finding reveals underlying information contents in S1 that could be har- nessed for rehabilitation and brain-machine interfaces. Beyond body maps: Information content of specific body parts is distributed across the somatosensory homunculus Dollyane Muret,1,5,* Victoria Root,1,2 Paulina Kieliba,1 Danielle Clode,1,3 and Tamar R. Makin1,4 1Institute of Cognitive Neuroscience, University College London, 17 Queen Square, London WC1N 3AZ, UK 2Wellcome Centre of Integrative Neuroimaging, University of Oxford, Oxford OX3 9DU, UK 3Dani Clode Design, 40 Hillside Road, London SW2 3HW, UK 4Wellcome Trust Centre for Neuroimaging, University College London, London WC1N 3AR, UK 5Lead contact Dollyane Muret,1,5,* Victoria Root,1,2 Paulina Kieliba,1 Danielle Clode,1,3 and Tamar R. Makin1,4 1Institute of Cognitive Neuroscience, University College London, 17 Queen Square, London WC1N 3AZ, UK 2Wellcome Centre of Integrative Neuroimaging, University of Oxford, Oxford OX3 9DU, UK 3Dani Clode Design, 40 Hillside Road, London SW2 3HW, UK 4Wellcome Trust Centre for Neuroimaging, University College London, London WC1N 3AR, UK 5Lead contact *Correspondence: [email protected] https://doi.org/10.1016/j.celrep.2022.110523 *Correspondence: [email protected] https://doi.org/10.1016/j.celrep.2022.110523 Cell Reports 38, 110523, March 15, 2022 ª 2022 The Authors. 1 This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Highlights g g d We replicate the high univariate selectivity profile of the somatosensory homunculus d We use multivariate fMRI analysis to identify information content beyond selectivity d Significant body part and action-related content are found throughout the homunculus d Functional information is available, even in regions selective to other body parts Muret et al., 2022, Cell Reports 38, 110523 March 15, 2022 ª 2022 The Authors. https://doi.org/10.1016/j.celrep.2022.110523 ll ll ll OPEN ACCESS Article Other recent evi- dence in humans hints at the existence of distributed patterns of functional connectivity throughout S1 (Ngo et al., 2021; Thomas et al., 2021), as well as distributed processing underlying and interrelating finger representations. For example, focal anes- thesia of a finger was found to affect the representation of all fin- gers (Wesselink et al., 2020) and intraneural microstimulation of single afferent units elicited widespread activity in the S1 hand re- gion (Sanchez Panchuelo et al., 2016). Altogether, these recent reports stress the need to investigate the distribution of represen- tational information content throughout S1 homunculus. g Recent methodological advancement (e.g., multivariate pattern analysis) allows to identify representational features beyond selectivity and thus provide an opportunity to reassess the homunculus. Here, we take advantage of these methods to investigate whether information content can be identified in S1 beyond the primary region of a given body part, as defined by conventional mapping criteria. We asked healthy participants to perform a series of sensorimotor paradigms in the scanner: (1) in- dividual finger movements (hereafter finger task), (2) movements of specific facial parts (hereafter face task), or (3) two different ac- tions (i.e., squeeze or push an object) with each of three body parts (i.e., lips, hand, and feet; hereafter body task). Using con- ventional univariate analyses on an independent dataset, we first defined individual S1 regions of interest showing high selectivity to face, hand, and foot movements. We then used representa- tional similarity analysis (RSA) to index information content by quantifying multivoxel representational dissimilarities between actions and body parts. Cross-validated Mahalanobis distances provide a quantification of these dissimilarities, where distances that are greater than zero reflect significant information content (note that we deliberately avoid the term ‘‘representational con- tent,’’ since it could imply functional relevance). We found task- relevant information content was distributed across S1, demon- strating an intrinsic organization to S1 beyond somatotopy. Article the selectivity of our independent ROIs (Figure 1A), we extracted the average univariate activity level obtained for each body part (movement versus rest) in the contralateral hemisphere and found that each ROI was highly selective to its primary body part, showing significant activity for this body part only (one- sample t tests versus zero, primary body parts: all t(21) R 10.67, all p < 0.001, all d R 2.28, 95% confidence interval [CI] [1.59, 2.93]; non-primary body parts: all t(21) % 1.00, all d % 0.21, 95% CI [
0.14, 0.57]; Figure 1B). We then used RSA to quantify the dissimilarity between activity patterns evoked by each movement (see Figure S3A for similar analysis using the ab- solute difference between univariate activity levels). One-sample t tests with Bonferroni-corrected alpha levels (a = 0.017, cor- rected for the three comparisons across body parts) confirmed that the representational dissimilarities were significantly greater than zero for pairs of body parts involving the primary body part of each ROI (all t(21) R 16.21; all p < 0.001; all d R 3.45; 95% CI [2.50, N]). Interestingly, significant dissimilarities were also found for cortically remote pairs of body parts, where both body parts were non-primary to the ROI (in leg ROI: hand-lips: t(21) = 9.93, p < 0.001, d = 2.12, 95% CI [1.46, N]; in hand ROI: feet-lips: t(21) = 6.51, p < 0.001, d = 1.39, 95% CI [0.88, N]; in face ROI: feet-hand: t(21) = 10.59, p < 0.001, d = 2.26, 95% CI [1.57, N]; Figure 1C). Thus, despite being highly selective to their primary body part, each ROI contained robust information con- tent about non-primary body parts. This first evidence suggests that non-primary and cortically distant body parts may contribute information to a given region of the homunculus. (BOLD) responses in humans revealed the presence of an under- lying inverted homunculus (Tal et al., 2017), similar to M1 (Zeharia et al., 2012), suggesting a distribution of activity patterns across the homunculus. In addition, recent reports in rodents show that the information content arising from different tactile inputs pro- vided to a digit could be decoded even from a non-adjacent digit representation (Enander and Jo¨ rntell, 2019). Information from different body subparts is distributed across S1 We next assessed how information from different body subparts is distributed across S1. Two tasks were used for that purpose: a face task involving bilateral movements and a finger task per- formed with each hand (see STAR Methods). First, to verify the selectivity of the individual ROIs used for each task, we quanti- fied the univariate activity level obtained for each subpart (versus rest) in the contralateral ROIs. Alpha was adjusted to 0.013 and 0.01, corrected for four and five comparisons (respectively) across face and hand subparts. Activity levels (averaged across hemispheres; see STAR Methods) were significantly greater than zero for all face subparts in the face ROI (face ROI: all t(21) R 3.51; all p % 0.002; all d R 0.75; 95% CI [0.34, 1.14]; Figure 2A, blue) and for the five fingers in the hand ROI (all t(18) R 7.44; all p < 0.001; all d R 1.71; 95% CI [1.09, 2.29]; Figure 2A, red). Face subparts did not significantly activate the hand and leg ROIs (all t(21) %
0.86; all d %
0.18; 95% CI [
0.54, 0.17]; Figure 2A, blue). For some fingers, significant positive activity levels (or trends) were found in the face and leg ROIs, as shown in red in Figure 2A (significant fingers: all t(18) R 2.70, all p % 0.015, all d R 0.62, 95% CI [0.20, 1.02]; other fingers: all t(18) % 2.47, all p R 0.024, all d % 0.57, 95% CI [0.15, 0.97]). Thus, while the face task shows high selectivity to the face ROI, finger-related activity seems to be more distributed across ROIs, though activ- ity levels were still low on average, i.e., below the liberal 2.3 threshold usually used to threshold individual data (dotted line in Figure 2A). RESULTS Our main approach was to identify distant and highly selective regions of the S1 homunculus at the individual participant level to assess their univariate activity levels and multivariate informa- tion content. Toward this end, within an S1 anatomical landmark (black contours in Figure 1A), we defined three regions of interest (ROIs) for each participant in each hemisphere, based on the 50 most selective voxels from an independent body localizer of the foot, hand, and face (see Figures 1A, S2A, and S3B for consis- tency maps across participants). INTRODUCTION While this approach has been hugely beneficial, e.g., for understanding (Kaas et al., 1979) and restoring (Bensmaia and Miller, 2014; Flesher et al., 2016; Ar- menta Salas et al., 2018; Chandrasekaran et al., 2021) brain function, it may also eclipse the presence of other organizing principles that may also bear relevance for S1 function. In particular, the underlying (weaker) inputs that tend to be ne- glected in mapping approaches may also provide functional contributions, even if secondary to the dominant input. Recent evidence indeed points toward a more complex orga- nization of S1, beyond its topographic organization. For example, an imaging study of the negative blood-oxygen-level-dependent Article ll OPEN ACCESS across S1 Individual ROIs in the hemisphere contralateral to the dominant hand were converted to Montreal Neurological Institute (MNI) space and projected onto an inflated surface. The color code represents the number of participants with overlapping ROIs in the standard MNI space. The black contour shows the anatomical delineation of S1 used to restrict the ROI definition, based on a probabilistic atlas. (B) Univariate activity levels (versus rest) for the three body parts (green feet; red hand; blue lips) within each ROI Only the primary body part of each ROI exhibited activity levels significantly above zero. (C) Multivariate dissimilarities. The left plots are a multidimensional scaling (MDS) depiction of the representational dissimilarity between the three body parts (green, feet; red, hand; blue, lips) in each ROI. Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a.u.) observed for the three pairs of body parts in each ROI (yellow, feet-hand; cyan, feet-lips; magenta, hand-lips). Data are represented as mean ± SEM. Gray dots represent individual participants. Asterisks indicate a significant difference relative to zero; *p < 0.017 ***p < 0.001. See Figure S2 for similar analysis in the non-dominant hemisphere and Figure S5 for example individual univariate activity maps. We then investigated the multivariate pattern of dissimilarity between the four facial subparts (Figure 2B, blue) and the five fin- gers (Figure 2B, red) across S1 ROIs. A qualitative observation of the multidimensional scaling (MDS) plots (Figure 2B) suggests that the representational structure of the face and hand, whose canonical representations are seen in their respective primary ROIs, is preserved across ROIs. This was confirmed by quantita- tive assessment of dissimilarities. To reduce the number of com- parisons, cross-validated representational dissimilarities from different pairs of subparts (i.e., face parts or fingers) were group- ed according to the subpart’s cortical neighborhood (i.e., adja- cent versus non-adjacent; see inset in Figure 3). Alpha was adjusted to 0.025 to account for two comparisons for the adja- cent and non-adjacent dissimilarities. We found that, for both tasks, dissimilarities between subparts’ activity patterns were all significantly above zero, regardless of their neighborhood (Fig- ure 3B). across S1 This was true not only in their respective primary ROI (all t R 12.41; all p < 0.001; all d R 2.85; 95% CI [1.96, N]) but also in remote parts of the homunculus, such as the hand ROI or the leg ROI for the face task (all t(21) R 5.85; all p < 0.001; all d R 1.29; 95% CI [0.80, N]; Figure 3B, blue) and the face ROI or the leg ROI for the finger task (all t(18) R 3.18; all p % 0.003; all d R 0.73; 95% CI [0.29, N]; Figure 3B, red). These results replicate the previous observation that information about body parts is not restricted to their primary S1 region. across S1 The left plots are a multidimensional scaling (MDS) depiction of the representational dissimilarity between the three body parts (green feet; red hand; blue lips) in each ROI Ellipses indicate between participant standard errors The right histograms show the cross validated dissimilarity g g , y, p y p (A) Consistency maps across participants of the S1 regions of interest (ROIs) for the body task (n = 22). Individual ROIs in the hemisphere contralateral to the dominant hand were converted to Montreal Neurological Institute (MNI) space and projected onto an inflated surface. The color code represents the number of participants with overlapping ROIs in the standard MNI space. The black contour shows the anatomical delineation of S1 used to restrict the ROI definition, based on a probabilistic atlas. (B) Univariate activity levels (versus rest) for the three body parts (green, feet; red, hand; blue, lips) within each ROI. Only the primary body part of each ROI exhibited activity levels significantly above zero. (C) Multivariate dissimilarities. The left plots are a multidimensional scaling (MDS) depiction of the representational dissimilarity between the three body parts (green, feet; red, hand; blue, lips) in each ROI. Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a.u.) observed for the three pairs of body parts in each ROI (yellow, feet-hand; cyan, feet-lips; magenta, hand-lips). Data are represented as mean ± SEM. Gray dots represent individual participants. Asterisks indicate a significant difference relative to zero; *p < 0.017; , y, p y p articipants of the S1 regions of interest (ROIs) for the body task (n = 22). Individual ROIs in the hemisphere contralateral to the to Montreal Neurological Institute (MNI) space and projected onto an inflated surface. The color code represents the number of OIs in the standard MNI space. The black contour shows the anatomical delineation of S1 used to restrict the ROI definition, based (A) Consistency maps across participants of the S1 regions of interest (ROIs) for the body task (n = 22). Individual ROIs in th dominant hand were converted to Montreal Neurological Institute (MNI) space and projected onto an inflated surface. The col participants with overlapping ROIs in the standard MNI space. The black contour shows the anatomical delineation of S1 used t on a probabilistic atlas. (A) Consistency maps across participants of the S1 regions of interest (ROIs) for the body task (n = 22). across S1 We first focused on the body task to assess how information from different body parts is distributed across S1. To confirm 2 Cell Reports 38, 110523, March 15, 2022 A i l Figure 1. Regions of interest, selectivity, and multivariate information content related to specific body parts across S1 homunculus (A) Consistency maps across participants of the S1 regions of interest (ROIs) for the body task (n = 22). Individual ROIs in the hemisphere contralateral to the dominant hand were converted to Montreal Neurological Institute (MNI) space and projected onto an inflated surface. The color code represents the number of participants with overlapping ROIs in the standard MNI space. The black contour shows the anatomical delineation of S1 used to restrict the ROI definition, based on a probabilistic atlas. (B) Univariate activity levels (versus rest) for the three body parts (green, feet; red, hand; blue, lips) within each ROI. Only the primary body part of each ROI exhibited activity levels significantly above zero. (C) Multivariate dissimilarities. The left plots are a multidimensional scaling (MDS) depiction of the representational dissimilarity between the three body parts (green, feet; red, hand; blue, lips) in each ROI. Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a u ) observed for the three pairs of body parts in each ROI (yellow feet hand; cyan feet lips; magenta hand lips) Article gure 1. Regions of interest, selectivity, and multivariate information content related to specific body parts acro erest, selectivity, and multivariate information content related to specific body parts across S1 homunculus Figure 1. Regions of interest, selectivity, and multivariate information content related to specific body parts across S1 homunculus (A) Consistency maps across participants of the S1 regions of interest (ROIs) for the body task (n = 22). Individual ROIs in the hemisphere contralateral to the dominant hand were converted to Montreal Neurological Institute (MNI) space and projected onto an inflated surface. The color code represents the number of participants with overlapping ROIs in the standard MNI space. The black contour shows the anatomical delineation of S1 used to restrict the ROI definition, based on a probabilistic atlas. (B) Univariate activity levels (versus rest) for the three body parts (green, feet; red, hand; blue, lips) within each ROI. Only the primary body part of each ROI exhibited activity levels significantly above zero. (C) Multivariate dissimilarities. Topographic features from different body subparts are distributed across S1 A significant difference between adjacent and non-adjacent univariate content was found in the primary ROI of each task (both t R
7.51; both p < 0.001; both d R
1.62; 95% CI [
2.25,
0.97]; Figure 3A). In addition, a significant topo- graphic difference was found for the face task in the hand ROI (t(21) =
3.30; p = 0.003; d =
0.70; 95% CI [
1.16,
0.23]). No univariate topographic features were observed for other comparisons and ROIs (all p R 0.099; all d %
0.40; 95% CI [
0.86, 0.07]), despite the higher activity levels found for the finger than for the face task in non-primary ROIs (t(39) =
4.29; p < 0.001; d =
1.35; 95% CI [
2.02,
0.66]; Figure 2A). Alto- gether, these results suggest that the univariate information con- tent does not appear to be consistently topographically orga- nized outside of its primary ROI. multivariate differences between adjacent and non-adjacent subparts across ROIs for both the face and finger tasks. Univar- iate content was defined as the absolute difference between ac- tivity levels evoked by pairs of subparts in the different ROIs (see inset in Figure 3). A significant difference between adjacent and non-adjacent univariate content was found in the primary ROI of each task (both t R
7.51; both p < 0.001; both d R
1.62; 95% CI [
2.25,
0.97]; Figure 3A). In addition, a significant topo- graphic difference was found for the face task in the hand ROI (t(21) =
3.30; p = 0.003; d =
0.70; 95% CI [
1.16,
0.23]). No univariate topographic features were observed for other comparisons and ROIs (all p R 0.099; all d %
0.40; 95% CI [
0.86, 0.07]), despite the higher activity levels found for the finger than for the face task in non-primary ROIs (t(39) =
4.29; p < 0.001; d =
1.35; 95% CI [
2.02,
0.66]; Figure 2A). Alto- gether, these results suggest that the univariate information con- We then compared the representational dissimilarities between adjacent and non-adjacent subparts, expecting adjacent subparts to be more similar if topographic information of body subparts is preserved across the homunculus.Similar to the univariate results, a significant difference between adjacent and non-adjacent subparts was found in the primary ROIs for both tasks (both t R
13.59; both p < 0.001; both d R
2.90; 95% CI [
3.85,
1.92]). Topographic features from different body subparts are distributed across S1 Gray dots represent individual participants. *p < 0.013 and 0.010 for the face and finger tasks, respectively (alpha corrected for four and five comparisons, respectively); #p < 0.025 and 0.020 for the face and finger tasks, respectively; ***p < 0.001. The color code for the respective subparts is depicted in the inset. Figure 2. Selectivity and multivariate representational patterns of body subparts across S1 homunculus for face and finger tasks (A) Univariate activity levels (versus rest) averaged across hemispheres obtained for the different subparts of the face task (shades of blue) and of the finger task (shades of red) within each ROI. The dotted line marks the 2.3 individual threshold. (B) MDS plots illustrating the representational structure contained in the face (shades of blue) and hand (shades of red) activity across S1 ROIs (averaged across hemispheres). The canonical hand and face representational structures are observed, respectively, in the hand and face ROIs (i.e., primary ROIs). Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.013 and 0.010 for the face and finger tasks, respectively (alpha corrected for four and five comparisons, respectively); #p < 0.025 and 0.020 for the face and finger tasks, respectively; ***p < 0.001. The color code for the respective subparts is depicted in the inset. (A) Univariate activity levels (versus rest) averaged across hemispheres obtained for the different subparts of the face task (s (shades of red) within each ROI. The dotted line marks the 2.3 individual threshold. (B) MDS plots illustrating the representational structure contained in the face (shades of blue) and hand (shades of red) activ hemispheres). The canonical hand and face representational structures are observed, respectively, in the hand and face R Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.013 and 0.010 for the face and finger for four and five comparisons, respectively); #p < 0.025 and 0.020 for the face and finger tasks, respectively; ***p < 0.001 subparts is depicted in the inset. multivariate differences between adjacent and non-adjacent subparts across ROIs for both the face and finger tasks. Univar- iate content was defined as the absolute difference between ac- tivity levels evoked by pairs of subparts in the different ROIs (see inset in Figure 3). Topographic features from different body subparts are distributed across S1 To assess whether the topographical content was preserved throughout S1, we next investigated the univariate and Cell Reports 38, 110523, March 15, 2022 3 Figure 2. Selectivity and multivariate representational patterns of body subparts across S1 homunculus for face and finger tasks (A) Univariate activity levels (versus rest) averaged across hemispheres obtained for the different subparts of the face task (shades of blue) and of the finger task (shades of red) within each ROI. The dotted line marks the 2.3 individual threshold. (B) MDS plots illustrating the representational structure contained in the face (shades of blue) and hand (shades of red) activity across S1 ROIs (averaged across hemispheres). The canonical hand and face representational structures are observed, respectively, in the hand and face ROIs (i.e., primary ROIs). Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.013 and 0.010 for the face and finger tasks, respectively (alpha corrected for four and five comparisons, respectively); #p < 0.025 and 0.020 for the face and finger tasks, respectively; ***p < 0.001. The color code for the respective subparts is depicted in the inset. Article ll OPEN ACCESS Article ty and multivariate representational patterns of body subparts across S1 homunculus for face and finger tasks levels (versus rest) averaged across hemispheres obtained for the different subparts of the face task (shades of blue) and of the each ROI. The dotted line marks the 2.3 individual threshold. Figure 2. Selectivity and multivariate representational patterns of body subparts across S1 homunculus for ctivity and multivariate representational patterns of body subparts across S1 homunculus for face and finger ta vity levels (versus rest) averaged across hemispheres obtained for the different subparts of the face task (shades of blue) and of Figure 2. Selectivity and multivariate representational patterns of body subparts across S1 homunculus for face and finger tasks (A) Univariate activity levels (versus rest) averaged across hemispheres obtained for the different subparts of the face task (shades of blue) and of the finger task (shades of red) within each ROI. The dotted line marks the 2.3 individual threshold. (B) MDS plots illustrating the representational structure contained in the face (shades of blue) and hand (shades of red) activity across S1 ROIs (averaged across hemispheres). The canonical hand and face representational structures are observed, respectively, in the hand and face ROIs (i.e., primary ROIs). Data are represented as mean ± SEM. 4 Cell Reports 38, 110523, March 15, 2022 Topographic features from different body subparts are distributed across S1 Importantly, we found significant evidence for topograph- icalcontentforbothtasksinthenon-primaryROIs(alltR
3.41;all p % 0.003; all d R
0.73; 95% CI [
1.19,
0.25];Figure3B), witha trend found for the face task in the leg ROI (t(21) =
2.03; p = 0.055; d =
0.43; 95% CI [
0.87, 0.01]). These multivariate results reveal that topographical information content about body parts, and the hand in particular, can be observed throughout the homunculus. 4 Cell Reports 38, 110523, March 15, 2022 ll OPEN ACCESS Article Figure 3. Univariate and multivariate topo- graphic content related to body subparts across S1 homunculus for face and finger tasks Figure 3. Univariate and multivariate topo- graphic content related to body subparts across S1 homunculus for face and finger tasks (A) Univariate topographic content defined as the absolute difference between activity levels evoked by adjacent (dotted bars) and non-adjacent (hatched bars) subparts in the different ROIs for the face task (blue) and the finger task (red). (B) Multivariate topographic content measured by the cross-validated representational dissimilarity (a.u.) between activity patterns evoked by adjacent (dotted bars) and non-adjacent (hatched bars) subparts in the different ROIs for the face task (blue) and the finger task (red). Data are represented as mean ± SEM. Gray dots represent individual participants. The matrix in the inset illustrates how adjacent and non-adjacent content is computed using the fingers as an example (D, digit). Black asterisks indicate a significant dif- ference between adjacent and non-adjacent body subparts: *p < 0.05; #p < 0.1; ***p < 0.001. Gray as- terisks indicate a significant difference relative to zero: *p < 0.025; ***p < 0.001. d R 1.00; 95% CI [1.00, N]) but also in non-primary ROIs for hand and feet movements (all p % 0.002, all d R 0.69, 95% CI [0.29, N]; except for a trend for feet movements in face ROI: t(21) = 2.09, p = 0.024, d = 0.45, 95% CI [0.07, N]; Figure 4B). These results suggest that ac- tion-related information content from the hand and the feet seems to be distributed across the homunculus (see STAR Methods for potential explanations for the lack of lips information). Cell Reports 38, 110523, March 15, 2022 5 Two actions done with the same body part can be differentiated throughout BA3b Finally, we investigated the profile of ac- tion-related information content specif- ically in BA3b, known to show the greatest level of selectivity in S1 (Powell and Mountcastle, 1959; Martuzzi et al., 2014; Schel- lekens et al., 2021), alongside the univariate activity levels used to classically determine body maps (Figure 5). To this end, BA3b’s strip was segmented into 29 bands of equal height (Fig- ure 5A) that were then used to calculate activity levels and multi- variate dissimilarities at the individual level. Consistent with our previous ROI selectivity analysis, the highest activity levels for each body part, based on univariate analysis, lay within their independently defined primary ROI (gray shades in Figure 5B). As we found before, multivariate dissimilarities were qualitatively apparent beyond the regions where activity levels can be observed (Figure 5C). To reduce the number of comparisons, the peak(s) dissimilarity between two different movements with each primary body part (dotted black lines in Figure 5C) was used to test whether action dissimilarities obtained in the Two actions done with the same body part can be DISCUSSION corresponding BA3b band using the other non-primary body parts were significantly greater than zero. Alpha was corrected to 0.025 to account for the two comparisons for the two non-pri- mary body parts. For instance, while feet activity levels were observed solely within the first few medial bands of BA3b, dis- similarities between the two actions performed with the feet were significantly greater than zero at the two peaks observed for the hand (both z R 208.00; both p % 0.003; both d R 0.64; 95% CI [0.35, N]) but also at the peak observed for the lips (t(21) = 3.35; p = 0.002; d = 0.71; 95% CI [0.31, N]). Similarly, dis- similarities between the two actions performed with the hand were significantly greater than zero at the peaks observed for both the feet and lips (both t R 2.48; both p % 0.011; both d R 0.53; 95% CI [0.15, N]). These results emphasize the avail- ability of body part information across S1. Therefore, we find that the information content about body part actions is much more widely distributed then can be inferred by delineating the univar- iate selectivity profiles. corresponding BA3b band using the other non-primary body parts were significantly greater than zero. Alpha was corrected to 0.025 to account for the two comparisons for the two non-pri- mary body parts. For instance, while feet activity levels were observed solely within the first few medial bands of BA3b, dis- similarities between the two actions performed with the feet were significantly greater than zero at the two peaks observed for the hand (both z R 208.00; both p % 0.003; both d R 0.64; 95% CI [0.35, N]) but also at the peak observed for the lips (t(21) = 3.35; p = 0.002; d = 0.71; 95% CI [0.31, N]). Similarly, dis- similarities between the two actions performed with the hand were significantly greater than zero at the peaks observed for both the feet and lips (both t R 2.48; both p % 0.011; both d R 0.53; 95% CI [0.15, N]). These results emphasize the avail- ability of body part information across S1. Therefore, we find that the information content about body part actions is much more widely distributed then can be inferred by delineating the univar- iate selectivity profiles. Two actions done with the same body part can be The left plots are a MDS depiction of the representational dissimilarity between the two actions (gray ellipses, squeeze; black ellipses, push) performed with the non-primary body parts in each ROI (green, feet; red, hand; blue, lips). Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a.u.) observed for the two actions performed with each body part in each ROI (green, feet; red, hand; blue, lips). Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.017; #p < 0.033; ***p < 0.001. See Figure S4 for similar analysis in the non-dominant hemisphere. p ) (B) Multivariate dissimilarities. The left plots are a MDS depiction of the representational dissimilarity between the two actions (gray ellipses, squeeze; black ellipses, push) performed with the non-primary body parts in each ROI (green, feet; red, hand; blue, lips). Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a.u.) observed for the two actions performed with each body part in each ROI (green, feet; red, hand; blue, lips). # ellipses, push) performed with the non-primary body parts in each ROI (green, feet; red, hand; blue, lips). Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a.u.) observed for the two actions performed with each body part in each ROI (green, feet; red, hand; blue, lips). Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.017; #p < 0.033; ***p < 0.001. See Figure S4 for similar analysis in the non-dominant hemisphere. blue, lips). Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.017; #p < 0.033; ***p < 0.001. See Figure S4 for similar analysis in the non-dominant hemisphere. Two actions done with the same body part can be differentiated in non-primary regions of the homunculus We then assessed how information from different actions done with a given body part is distributed across S1. For that pur- pose, we compared the squeeze and push conditions per- formed with each of three body parts (i.e., feet, hand, and lips; see objects in Figure S1). Alpha was adjusted to 0.017, correcting for the three comparisons across body parts. Activ- ity levels evoked by these actions were significantly different only when performed with the primary body part of each ROI (all p % 0.008; Figure 4A). No significant differences were observed in non-primary ROIs (all p R 0.050), except for a trend for feet movements in the hand ROI (t(21) =
2.55; p = 0.019; Figure 4A). However, multivariate representational dissimilarities between the two actions were significantly greater than zero not only in primary ROIs (all p < 0.001; all Cell Reports 38, 110523, March 15, 2022 5 ll OPEN ACCESS Figure 4. Univariate and multivariate content related to different actions performed with a given body part across S1 homunculus (A) Univariate activity levels (versus rest) for the two actions (gray hatched, squeeze; black hatched, push) performed with each body part (green, feet; red, hand; blue, lips) within each ROI. (B) Multivariate dissimilarities. The left plots are a MDS depiction of the representational dissimilarity between the two actions (gray ellipses, squeeze; black ellipses, push) performed with the non-primary body parts in each ROI (green, feet; red, hand; blue, lips). Ellipses indicate between-participant standard errors. The right histograms show the cross-validated dissimilarity (a.u.) observed for the two actions performed with each body part in each ROI (green, feet; red, hand; blue, lips). Data are represented as mean ± SEM. Gray dots represent individual participants. *p < 0.017; #p < 0.033; ***p < 0.001. See Figure S4 for similar analysis in the non-dominant hemisphere Article ll OPEN ACCESS Figure 4. Univariate and multivariate content related to different actions performed with a given body part across S1 homunculus (A) Univariate activity levels (versus rest) for the two actions (gray hatched, squeeze; black hatched, push) performed with each body part (green, feet; red, h blue, lips) within each ROI. blue, lips) within each ROI. (B) Multivariate dissimilarities. 6 Cell Reports 38, 110523, March 15, 2022 DISCUSSION Due to its highly selective profile, conventional mapping proced- ures providing a ‘‘parcellated’’—all or nothing (i.e., winner takes all)—view over S1 have dominated our conceptualization of its functional organization (Roux et al., 2018; Willoughby et al., 2021). Consequently, alteration of map boundaries have been commonly interpreted as cortical reorganization, with the limita- tions previously discussed (Muret and Makin, 2021). Using con- ventional univariate analyses, together with multivariate RSA, we investigated the distribution of information content underlying S1 topographic organization. We found that S1 contained signifi- cant task-relevant information content beyond the primary re- gion of a given body part, as defined by conventional mapping criteria. Even though, as expected from somatotopy, information content was more pronounced in primary regions, cortically distant body parts but also body subparts (e.g., fingers) could be consistently discriminated throughout S1. Perhaps most 6 Cell Reports 38, 110523, March 15, 2022 6 ll OPEN ACCESS Figure 5. Regions of interest, selectivity, and multivariate information content related to the two actions across BA3b’s strip (A) Illustration of the segmentation of BA3b’s strip into 29 bands of similar height (i.e., 2.0911 mm). The black outlines represent the surrounding S1 Brodmann areas, respectively, BA3a, BA1, and BA2 (from left to right), based on a probabilistic atlas. The color code represents the band number (1–29). CS, central sulcus. (B) Univariate activity levels (versus rest) observed for the three body parts throughout BA3b strip (green, feet; red, hand; blue, lips). (C) Multivariate cross-validated dissimilarities (a.u.) observed between the two actions (i.e., squeeze and push) for the three pairs of body parts throughout BA3b strip (green, feet; red, hand; blue, lips). The peak dissimilarity for each primary body part (dotted black lines) was used to test whether dissimilarities obtained in the corresponding band of BA3b for the other non-primary body parts were significantly greater than zero; *p < 0.025; ***p < 0.001. Data are represented as mean (curves) ± SEM (shades around each curve). Gray shades in the background of each plot represent the location of our leg, hand, and face ROIs. gions of interest, selectivity, and multivariate information content related to the two actions across BA3b’s stri Figure 5. Regions of interest, selectivity, and multivariate information content related to the two actio (A) Illustration of the segmentation of BA3b’s strip into 29 bands of similar height (i.e., 2.0911 mm). DISCUSSION The black outlines represent the surrounding S1 Brodmann areas, respectively, BA3a, BA1, and BA2 (from left to right), based on a probabilistic atlas. The color code represents the band number (1–29). CS, central sulcus. (B) Univariate activity levels (versus rest) observed for the three body parts throughout BA3b strip (green, feet; red, hand; blue, lips). (C) Multivariate cross-validated dissimilarities (a.u.) observed between the two actions (i.e., squeeze and push) for the three pairs of body parts throughout BA3b strip (green, feet; red, hand; blue, lips). The peak dissimilarity for each primary body part (dotted black lines) was used to test whether dissimilarities obtained in the corresponding band of BA3b for the other non-primary body parts were significantly greater than zero; *p < 0.025; ***p < 0.001. Data are represented as mean (curves) ± SEM (shades around each curve). Gray shades in the background of each plot represent the location of our leg, hand, and face ROIs. strikingly, different actions performed with the hand or the feet could be differentiated at remote extremities of the homunculus. Overall, our results suggest a widespread distribution of informa- tion content across the S1 homunculus that goes beyond what can be expected from its selectivity profile and emphasizes the need to consider S1 more as a whole than as a patchwork of in- dependent body maps. Our results also stress the need to further investigate the functional relevance of the distributed information and its potential for rehabilitation, augmentation, or brain-ma- chine interfaces. not restricted or compartmentalized by anatomical septa (Fang et al., 2002; Kuehn et al., 2017). Finally, when present (i.e., for hand and feet), information content did not seem to decrease lin- early with cortical distance (e.g., feet dissimilarities appeared relatively constant outside the feet region; see Figure 5). This observation might indicate that the distributed information re- ported here is more likely to arise from thalamo-cortical projec- tions (Rausell et al., 1998) than from horizontal cortico-cortical connections within S1 (Ne´ gyessy et al., 2013). However, horizon- tal connections could still contribute to some extent (Johnson and Frostig, 2015). In addition, potential top-down modulation from higher order regions that are integrating signal across body parts could also contribute to the distribution of information across S1 (Cerkevich and Kaas, 2019). Further work would be required to tease apart these potential origins. Cell Reports 38, 110523, March 15, 2022 7 Article In contrast, the information we detect is likely related to patterns of negative BOLD responses previously observed in S1 (Tal et al., 2017), responses that were linked to neuronal activity rather than blood-flow-stealing effects (Shmuel et al., 2006; Schridde et al., 2008; Mullinger et al., 2014). We add to this body of previ- ous findings by demonstrating that body part information can be found across the homunculus and is not restricted by functional boundaries or affected by cortical proximity outside the primary region. Nevertheless, we do find that qualitatively greater sharing of information exists between the hand and the feet, as well as between the hand and the mouth. This could be driven by the topographic relationship of the primary regions (where the hand is roughly equidistant between the feet and the mouth). Alternatively, this might reflect differences in the functional us- age of body parts: the hands and upper limbs are our main effector to reach and interact with other body parts and for functional coordination with the mouth and the lower limbs for goal-directed behaviors (e.g., for feeding and locomotion and balancing, respectively). Microstimulation in the S1 hand region most often elicits sen- sations on the patients’ hand (Flesher et al., 2016; Armenta Salas et al., 2018; Chandrasekaran et al., 2021), and stroke in the M1 hand region results in motor impairment of the hand (Darling et al., 2016). Such selectivity in the sensations induced by focal S1 microstimulation showcases the potential lack of functional relevance of the content reported here. In line with these studies, it is important to clarify that we do not negate the notion of a pri- mary function for a given S1 region. But it is also worth noting that exogeneous stimulation of S1 is an artificial procedure (eliciting unnatural percepts) that is unlikely to tap into the full ecological functioning of S1 and that perceptual assessments in these clin- ical observations in humans are relatively crude (Richards et al., 2015). For instance, potential (albeit weaker) sensations to other body parts have not been investigated with the refined psycho- physical tests that would be required to detect the more subtle effects that we uncover. As such, these clinical reports are not incompatible with the alternative eventuality that the latent activ- ity, which comprises the information content we are reporting, might also be functionally relevant. Article Thus, an active paradigm allows us to take full advantage of sensorimotor information, relevant for motor planning (Sun et al., 2015) and encompass signal arising from the efference copy (London and Miller, 2013). Other recent studies indicate that information content in S1 can be probed even in the absence of sensorimotor inputs. For example, visual observation of roughness exploration with a finger (Kim et al., 2018) or even imagined tactile percepts (Bashford et al., 2021) could be decoded in S1. As such, while the specific contribution of the abundance of tactile, propriocep- tive, and even cognitive (Meftah et al., 2002) inputs provided by an active paradigm needs to be further dissociated, all of it is crucial for S1 function. Therefore, the use of an active paradigm is arguably more appropriate to investigate ecological represen- tational motifs. face subparts (see Figure 3). The lack of action-related informa- tion for the lips could arise from a lower extent and richness of sensory feedbacks since, contrary to the other body parts, most participants did not manipulate an object during the two lips actions (see STAR Methods). Alternatively, the lips could require less coordination with other body parts, resulting in less representational overlap. Specifically, when coordinating actions, the face is most often the recipient of targeted actions, but not the supplier, contrary to the limbs. This idea is also compatible with the observation of higher resting-state func- tional connectivity between the hand and feet regions than be- tween the face and the other regions in BA3b (Thomas et al., 2021). In our view, the observation of widespread information content for body (sub)parts and actions across S1 is not surprising since some extent of distributed tactile information in S1 was already documented by Penfield and Boldrey, based on electrical stim- ulation (1937). We also do not believe these findings could be discarded as an fMRI artifact due to contribution of blood-steal- ing effects (Woolsey et al., 1996; Harel et al., 2002; Devor et al., 2005), where local increase in blood flow also results in a decreased blood flow in the immediate surrounding areas. This is because we observe abundant information content remotely from the primary region, which likely extends beyond the spatial scale of these effects (Woolsey et al., 1996; Devor et al., 2005). Article ll OPEN ACCESS important to note that recent studies using similar multivariate analyses in the hand region showed that the representational multivariate structure, as well as the univariate topographic map, were comparable between active paradigm and passively applied tactile stimulation to individual fingers (Berlot et al., 2019; Sanders et al., 2019). Comparable discriminability of hand ges- tures or postures was also previously reported in humans using electrocorticography, with similar decoding abilities in S1 (Ches- tek et al., 2013; Branco et al., 2017; Li et al., 2017) and in M1 (Branco et al., 2017). Conversely, recent studies showed that finger movements and effector information can be decoded in S1 during motor planning, well before movement execution (Ariani et al., 2020; Gale et al., 2021). Thus, an active paradigm allows us to take full advantage of sensorimotor information, relevant for motor planning (Sun et al., 2015) and encompass signal arising from the efference copy (London and Miller, 2013). Other recent studies indicate that information content in S1 can be probed even in the absence of sensorimotor inputs. For example, visual observation of roughness exploration with a finger (Kim et al., 2018) or even imagined tactile percepts (Bashford et al., 2021) could be decoded in S1. As such, while the specific contribution of the abundance of tactile, propriocep- tive, and even cognitive (Meftah et al., 2002) inputs provided by an active paradigm needs to be further dissociated, all of it is crucial for S1 function. Therefore, the use of an active paradigm is arguably more appropriate to investigate ecological represen- tational motifs. important to note that recent studies using similar multivariate analyses in the hand region showed that the representational multivariate structure, as well as the univariate topographic map, were comparable between active paradigm and passively applied tactile stimulation to individual fingers (Berlot et al., 2019; Sanders et al., 2019). Comparable discriminability of hand ges- tures or postures was also previously reported in humans using electrocorticography, with similar decoding abilities in S1 (Ches- tek et al., 2013; Branco et al., 2017; Li et al., 2017) and in M1 (Branco et al., 2017). Conversely, recent studies showed that finger movements and effector information can be decoded in S1 during motor planning, well before movement execution (Ariani et al., 2020; Gale et al., 2021). 8 Cell Reports 38, 110523, March 15, 2022 DISCUSSION Altogether, the widespread distribution of information content even within BA3b suggests that the common spatial definition of body part representation within S1 may only reflect one (if dominant) aspect of S1 organization. The widespread availability of body part information was further confirmed by focusing the analysis on discriminating different actions done with the same body part along the most topographically organized sub-region of S1, BA3b. This analysis revealed that, while information content was the highest in pri- mary regions and substantially reduced in non-primary regions, a significant amount of information (related to body parts and ac- tions) persisted throughout BA3b. In particular, it is interesting to note that conventional functional ‘‘boundaries’’ between body maps, as defined by contrasting univariate activity (e.g., between hand and face; Kuehn et al., 2017), did not seem to abruptly disrupt the distribution of information content. Moreover, our data suggest that functional activity and information content is In contrast to the hand and feet, lips’ action-related informa- tion content appeared to be more restricted to the face region (see Figure 5). This result contrasts with the widespread topo- graphical content we found throughout the homunculus for the Cell Reports 38, 110523, March 15, 2022 7 Article A ti l INCLUSION AND DIVERSITY We worked to ensure gender balance in the recruitment of human subjects. One or more of the authors of this paper self-identifies as a member of the LGBTQ+ community. Received: September 20, 2021 Revised: December 10, 2021 Accepted: February 21, 2022 Published: March 15, 2022 AUTHOR CONTRIBUTIONS D.M. and T.R.M. conceived the study. D.C. designed the objects. D.M. collected the body and face datasets, and P.K. and D.C. collected the finger dataset. V.R. pre-processed the face dataset, D.M. pre-processed the body dataset and analyzed the body and face datasets, and P.K. pre-processed and analyzed the finger dataset. D.M. and T.R.M. wrote the manuscript with input from all co-authors. T.R.M. secured funding. SUPPLEMENTAL INFORMATION Supplemental information can be found online at https://doi.org/10.1016/j. celrep.2022.110523. ACKNOWLEDGMENTS One important consideration is that we may have found this distributed information because multivariate techniques are too sensitive. For example, the existence of distributed tactile infor- mation outside of the sensorimotor cortex was recently detected in neurons as far as in the rodent primary visual cortex (Enander et al., 2019). As such, being able to decode (or differentiate) across conditions does not necessarily mean the brain is actually using this information (i.e., functionally represented). Even if not functionally valuable, this information could be exploited for brain-machine interfaces (Flesher et al., 2021), where specific parts of the homunculus might not be as directly accessible (e.g., the medial foot region). Finally, the distribution of informa- tion across the homunculus and redundancy of information that it might entail could prove particularly useful for solving the issue of ‘‘resource reallocation’’ that augmentation techniques are currently facing (Dominijanni et al., 2021). This work was supported by an ERC Starting Grant (715022 EmbodiedTech) and a Wellcome Trust Senior Research Fellowship (215575/Z/19/Z), awarded to T.R.M. We thank Arabella Bouzigues and Maria Kromm for their substantial help in terms of recruitment and data collection; we also thank Adriana Zai- nurin, Esther Teo, Christine Tan, and Mathew Kollamkulam for help with data collection. The authors declare no competing interests. The authors declare no competing interests. Article For example, multiple reports of ‘‘referred sensations’’ across body parts have been docu- mented, either under lab-based manipulations (Badde et al., 2019; Amoruso et al., 2021) or spontaneously (e.g., Katz and Melzack, 1987; Ramachandran et al., 1992; Borsook et al., 1998; Moore et al., 2000; McCabe et al., 2003; Soler et al., 2010). Distributing the content of information throughout S1 could allow for an increased number of combinations and pat- terns throughout body parts (Hoffmann et al., 2018), which might be more ecologically relevant, considering that we rarely use body parts independently from each other. In other words, this distributed information could provide a way to support It could be argued that the use of active paradigms in the pre- sent study have resulted in more distributed information than would be obtained with a passive tactile paradigm. For example, the distributed information content reported here might result from S1’s role in supporting selective movement generation. It has been suggested that latent activity in M1 could contribute to inhibit movement in other body parts not involved in the task (Zeharia et al., 2012) or to afford better motor coordination across body parts (Graziano et al., 2002). Similarly, latent activity in S1 could serve a role for predicting and encoding whole-body sensory feedback expected and perceived during actions involving multiple body parts. Therefore, a passive stimulation of body parts may not necessarily produce widespread informa- tion content in S1. While this needs to be investigated, it is 8 Cell Reports 38, 110523, March 15, 2022 Article ll OPEN ACCESS ll OPEN ACCESS B fMRI low-level analysis B Definition of regions of interest (ROIs) B Univariate analysis B Multivariate representational similarity analysis B BA3b analysis B Statistical analysis B Group-level ROI visualisations coordinated movements between body parts and to give context to their resulting sensory inputs in a coherent manner. Even if the distributed information underlying the traditional homunculus may not serve a functional role under normal circumstances, it could represent an underlying ‘‘scaffolding’’ for plasticity to take place, such as following congenital (Hahamy et al., 2017) or acquired (Pons et al., 1991) deprivation. For example, latent activity could be harnessed to restore the deprived primary function by potentiating any residual, now latent, activity (Qi et al., 2014). This idea could open perspectives for rehabilitative strategies. Article coordinated movements between body parts and to give context to their resulting sensory inputs in a coherent manner. Even if the distributed information underlying the traditional homunculus may not serve a functional role under normal circumstances, it could represent an underlying ‘‘scaffolding’’ for plasticity to take place, such as following congenital (Hahamy et al., 2017) or acquired (Pons et al., 1991) deprivation. For example, latent activity could be harnessed to restore the deprived primary function by potentiating any residual, now latent, activity (Qi et al., 2014). This idea could open perspectives for rehabilitative strategies. REFERENCES Detailed methods are provided in the online version of this paper and include the following: Amoruso, E., Terhune, D.B., Kromm, M., Kirker, S., Muret, D., and Makin, T.R. (2021). Reassessing referred sensations following peripheral deafferentation and the role of cortical reorganisation. Preprint at medRxiv. https://doi.org/ 10.1101/2021.12.08.21267128. d KEY RESOURCES TABLE DECLARATION OF INTERESTS To conclude, our results suggest that information in S1 might be a lot more distributed than selectivity profiles and winner- takes-all mapping approaches lead us to presume. While the functional consequences of this widespread information need to be further investigated, it reveals yet unexplored underlying in- formation contents. This abundant information could be har- nessed for better functional integration across body parts but also between brain and artificial body parts for rehabilitation, restoration, and augmentation purposes. Article Fang, P.C., Jain, N., and Kaas, J.H. (2002). Few intrinsic connections cross the hand-face border of area 3b of New World monkeys. J. Comp. Neurol. 454, 310–319. Branco, M.P., Freudenburg, Z.V., Aarnoutse, E.J., Bleichner, M.G., Vansteen- sel, M.J., and Ramsey, N.F. (2017). Decoding Hand Gestures from Primary So- matosensory Cortex Using High-Density ECoG. Neuroimage 147, 130–142. https://doi.org/10.1016/j.neuroimage.2016.12.004. Ferrier, D. (1873). Experimental researches in cerebral physiology and pathol- ogy. Br. Med. J. 1, 457. Catani, M., Dell’acqua, F., Vergani, F., Malik, F., Hodge, H., Roy, P., Valab- regue, R., and Thiebaut de Schotten, M. (2012). Short frontal lobe connections of the human brain. Cortex 48, 273–291. https://doi.org/10.1016/j.cortex. 2011.12.001. Fischl, B., Liu, A., and Dale, A.M. (2001). Automated manifold surgery: con- structing geometrically accurate and topologically correct models of the hu- man cerebral cortex. IEEE Trans. Med. Imaging 20, 70–80. https://doi.org/ 10.1109/42.906426. Cerkevich, C.M., and Kaas, J.H. (2019). Corticocortical projections to area 1 in squirrel monkeys (Saimiri sciureus). Eur. J. Neurosci. 49, 1024–1040. https:// doi.org/10.1111/ejn.13884. Flesher, S.N., Downey, J.E., Weiss, J.M., Hughes, C.L., Herrera, A.J., Tyler- Kabara, E.C., Boninger, M.L., Collinger, J.L., and Gaunt, R.A. (2021). A brain-computer interface that evokes tactile sensations improves robotic arm control. Science 372, 831–836. Chandrasekaran, S., Bickel, S., Herrero, J.L., Kim, J.W., Markowitz, N., Espi- nal, E., Bhagat, N.A., Ramdeo, R., Xu, J., Glasser, M.F., et al. (2021). Evoking highly focal percepts in the fingertips through targeted stimulation of sulcal re- gions of the brain for sensory restoration. Brain Stimul. 14, 1184–1196. https:// doi.org/10.1016/j.brs.2021.07.009. Flesher, S.N., Collinger, J.L., Foldes, S.T., Weiss, J.M., Downey, J.E., Tyler- Kabara, E.C., Bensmaia, S.J., Schwartz, A.B., Boninger, M.L., and Gaunt, R.A. (2016). Intracortical microstimulation of human somatosensory cortex. Sci. Transl. Med. 8, 361ra141–ra211. https://doi.org/10.1126/scitranslmed. aaf8083. Chestek, C.A., Gilja, V., Blabe, C.H., Foster, B.L., Shenoy, K.V., Parvizi, J., and Henderson, J.M. (2013). Hand posture classification using electrocorticogra- phy signals in the gamma band over human sensorimotor brain areas. J. Neural. Eng. 10, 026002. https://doi.org/10.1088/1741-2560/10/2/026002. Friston, K.J., Fletcher, P., Josephs, O., Holmes, A., Rugg, M.D., and Turner, R. (1998). Event-related fMRI: characterizing differential responses. Neuroimage 7, 30–40. https://doi.org/10.1006/nimg.1997.0306. Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (Aca- demic), p. 54. Fritsch, G.T., and Hitzig, E. (1870). On the electrical excitability of the cere- brum. International Classics in Epilepsy and Behavior 15, P123–P130. Cunningham, D.A., Machado, A., Yue, G.H., Carey, J.R., and Plow, E.B. (2013). Article A ti l enhances touch perception and learning in other fingers. J. Exp. Psychol. Gen. 148, 713–727. https://doi.org/10.1037/xge0000514. Baldwin, M.K., Cooke, D.F., and Krubitzer, L. (2017). Intracortical microstimu- lation maps of motor, somatosensory, and posterior parietal cortex in tree shrews (Tupaia belangeri) reveal complex movement representations. Cereb. Cortex 27, 1439–1456. https://doi.org/10.1093/cercor/bhv329. Devor, A., Ulbert, I., Dunn, A.K., Narayanan, S.N., Jones, S.R., Andermann, M.L., Boas, D.A., and Dale, A.M. (2005). Coupling of the cortical hemodynamic response to cortical and thalamic neuronal activity. Proc. Natl. Acad. Sci. U S A 102, 3822–3877. https://doi.org/10.1073/pnas.0407789102. Baldwin, M.K.L., Cooke, D.F., Goldring, A.B., and Krubitzer, L. (2018). Repre- sentations of fine digit movements in posterior and anterior parietal cortex re- vealed using long-train intracortical microstimulation in macaque monkeys. Diedrichsen, J., Provost, S., and Zareamoghaddam, H. (2016). On the Distribu- tion of Cross-Validated Mahalanobis Distances. Preprint at arXiv. 1607.01371. Cereb. Cortex 28, 4244–4263. https://doi.org/10.1093/cercor/bhx279. Bashford, L., Rosenthal, I., Kellis, S., Pejsa, K., Kramer, D., Lee, B., Liu, C., and Andersen, R.A. (2021). The neurophysiological representation of imagined so- matosensory percepts in human cortex. J. Neurosci. 41, 2177–2185. https:// doi.org/10.1523/jneurosci.2460-20.2021. Dominijanni, G., Shokur, S., Salvietti, G., Buehler, S., Palmerini, E., Rossi, S., De Vignemont, F., D’Avella, A., Makin, T.R., Prattichizzo, D., and Micera, S. (2021). Enhancing human bodies with extra robotic arms and fingers: the neu- ral resource allocation problem. Preprint at arxiv. 2103.17252. Bensmaia, S.J., and Miller, L.E. (2014). Restoring sensorimotor function through intracortical interfaces: progress and looming challenges. Nat. Rev. Neurosci. 15, 313–325. https://doi.org/10.1038/nrn3724. Ejaz, N., Hamada, M., and Diedrichsen, J. (2015). Hand use predicts the struc- ture of representations in sensorimotor cortex. Nat. Neurosci. 18, 1034–1040. https://doi.org/10.1038/nn.4038. Berlot, E., Prichard, G., O’Reilly, J., Ejaz, N., and Diedrichsen, J. (2019). Ipsilat- eral finger representations in the sensorimotor cortex are driven by active movement processes, not passive sensory input. J. Neurophysiol. 121, 418–426. https://doi.org/10.1152/jn.00439.2018. Enander, J.M.D., Spanne, A., Mazzoni, A., Bengtsson, F., Oddo, C.M., and Jo¨ rntell, H. (2019). Ubiquitous neocortical decoding of tactile input patterns. Front. Cell Neurosci. 13, 1–17. https://doi.org/10.3389/fncel.2019.00140. Enander, J.M.D., and Jo¨ rntell, H. (2019). Somatosensory cortical neurons decode tactile input patterns and location from both dominant and non-domi- nant digits. Cell Rep. 26, 3551–3560.e4. https://doi.org/10.1016/j.celrep.2019. 02.099. Borsook, D., Becerra, L., Fishman, S., Edwards, A., Jennings, C.L., Stojanovic, M., Papinicolas, L., Ramachandran, V.S., Gonzalez, R.G., and Breiter, H. (1998). Acute plasticity in the human somatosensory cortex following amputa- tion. Neuroreport 9, 1013–1017. d KEY RESOURCES TABLE d KEY RESOURCES TABLE d RESOURCE AVAILABILITY B Lead contact B Materials availability B Data and code availability d EXPERIMENTAL MODEL AND SUBJECT DETAILS d METHOD DETAILS B Scanning procedures B Body and face tasks B Finger task B Functional localiser B MRI data acquisition d QUANTIFICATION AND STATISTICAL ANALYSIS B fMRI pre-processing d RESOURCE AVAILABILITY B Lead contact B Materials availability B Data and code availability d EXPERIMENTAL MODEL AND SUBJECT DETAILS d METHOD DETAILS B Scanning procedures B Body and face tasks B Finger task B Functional localiser B MRI data acquisition d QUANTIFICATION AND STATISTICAL ANALYSIS B fMRI pre-processing d RESOURCE AVAILABILITY d RESOURCE AVAILABILIT B Lead contact B Materials availability B Lead contact Ariani, G., Pruszynski, J.A., and Diedrichsen, J. (2020). Motor Planning Brings Human Primary Somatosensory Cortex into Movement-specific Preparatory States. Preprint at bioRxiv. https://doi.org/10.1101/2020.12.17.423254. Ariani, G., Pruszynski, J.A., and Diedrichsen, J. (2020). Motor Planning Brings Human Primary Somatosensory Cortex into Movement-specific Preparatory States. Preprint at bioRxiv. https://doi.org/10.1101/2020.12.17.423254. B Materials availability B Data and code availability Armenta Salas, M., Bashford, L., Kellis, S., Jafari, M., Jo, H., Kramer, D., Shan- field, K., Pejsa, K., Lee, B., Liu, C.Y., et al. (2018). Proprioceptive and cuta- neous sensations in humans elicited by intracortical microstimulation. Elife 7, 1–11. https://doi.org/10.7554/eLife.32904. d EXPERIMENTAL MODEL AND SUBJECT DETAILS Auffret, M., Ravano, V.L., Rossi, G.M.C., Hankov, N., Petersen, M.F.A., and Petersen, C.C.H. (2018). Optogenetic Stimulation of Cortex to Map Evoked Whisker Movements in Awake Head-Restrained Mice. Neuroscience 368, 199–213. https://doi.org/10.1016/j.neuroscience.2017.04.004. B Functional localiser B MRI data acquisition d QUANTIFICATION AND STATISTICA B fMRI pre-processing d QUANTIFICATION AND STATISTICAL ANALYSIS B fMRI pre-processing Badde, S., Ro¨ der, B., and Heed, T. (2019). Feeling a touch to the hand on the foot. Curr. Biol. 29, 1491–1497.e4. https://doi.org/10.1016/j.cub.2019.02.060. B fMRI pre-processing Cell Reports 38, 110523, March 15, 2022 9 Article ll OPEN ACCESS Article ll OPEN ACCESS a natural stimulus. Hum. Brain Mapp. 35, 213–226. https://doi.org/10.1002/ hbm.22172. a natural stimulus. Hum. Brain Mapp. 35, 213–226. https://doi.org/10.1002/ hbm.22172. Greve, D.N., and Fischl, B. (2009). Accurate and robust brain image alignment using boundary-based registration. Neuroimage 48, 63–72. https://doi.org/10. 1016/j.neuroimage.2009.06.060. Matyas, F., Sreenivasan, V., Marbach, F., Wacongne, C., Barsy, B., Mateo, C., Aronoff, R., and Petersen, C.C. (2010). Motor control by sensory cortex. Sci- ence 330, 1240–1243. Hahamy, A., Macdonald, S.N., van den Heiligenberg, F., Kieliba, P., Emir, U., Malach, R., Johansen-Berg, H., Brugger, P., Culham, J.C., and Makin, T.R. (2017). Representation of multiple body parts in the missing-hand territory of congenital one-handers. Curr. Biol. 27, 1350–1355. https://doi.org/10.1016/j. cub.2017.03.053. Mayer, A., Baldwin, M.K.L., Cooke, D.F., Lima, B.R., Padberg, J., Lewenfus, G., Franca, J.G., and Krubitzer, L. (2019). The multiple representations of com- plex digit movements in primary motor cortex form the building blocks for complex grip types in capuchin monkeys. J. Neurosci. 39, 6684–6695. https://doi.org/10.1523/JNEUROSCI.0556-19.2019. Halley, A.C., Baldwin, M.K.L., Cooke, D.F., Englund, M., and Krubitzer, L. (2020). Distributed motor control of limb movements in rat motor and somato- sensory cortex: the sensorimotor amalgam revisited. Cereb. Cortex 30, 6296– 6312. https://doi.org/10.1093/cercor/bhaa186. Halley, A.C., Baldwin, M.K.L., Cooke, D.F., Englund, M., and Krubitzer, L. (2020). Distributed motor control of limb movements in rat motor and somato- McCabe, C.S., Haigh, R.C., Halligan, P.W., and Blake, D.R. (2003). Referred sensations in patients with complex regional pain syndrome type 1. Rheuma- tology (Oxford) 42, 1067–1073. https://doi.org/10.1093/rheumatology/ keg298. Harel, N., Lee, S.P., Nagaoka, T., Kim, D.S., and Kim, S.G. (2002). Origin of negative blood oxygenation level-dependent fMRI signals. J. Cereb. Blood Flow Metab. 22, 908–917. https://doi.org/10.1097/00004647-200208000- 00002. Meftah, el-M, Shenasa, J., and Chapman, C.E. (2002). Effects of a cross- modal manipulation of attention on somatosensory cortical neuronal re- sponses to tactile stimuli in the monkey. J. Neurophysiol. 88, 3133–3149. https://doi.org/10.1152/jn.00121.2002. Hoffmann, M., Straka, I., Vavre
cka, F.M., and Metta, G. (2018). Robotic homunculus: learning of artificial skin representation in a humanoid robot moti- vated by primary somatosensory cortex. IEEE Trans. Cogn. Dev. Syst. 10, 163–176. https://doi.org/10.1109/TCDS.2017.2649225. Merzenich, M.M., Kaas, J.H., Sur, M., and Lin, C.S. (1978). Double represen- tation of the body surface within cytoarchitectonic areas 3b and 1 in ‘‘SI’’ in the Owl monkey (Aotus trivirgatus). J. Comp. Neurol. 181, 41–73. Huber, L., Finn, E.S., Handwerker, D.A., Bo¨ nstrup, M., Glen, D.R., Kashyap, S., Ivanov, D., Petridou, N., Marrett, S., Goense, J., et al. Article (2020). Sub-millimeter fmri reveals multiple topographical digit representations that form action maps in human motor cortex. Neuroimage 208, 116463. https://doi.org/10. 1016/j.neuroimage.2019.116463. Moore, C.I., Stern, C.E., Dunbar, C., Kostyk, S.K., Gehi, A., and Corkin, S. (2000). Referred phantom sensations and cortical reorganization after spinal cord injury in humans. Proc. Natl. Acad. Sci. U S A 97, 14703–14708. https://doi.org/10.1073/pnas.250348997. Jenkinson, M., Bannister, P., Brady, M., and Smith, S. (2002). Improved opti- mization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 17, 825–841. https://doi.org/10.1006/nimg.2002. 1132. Mullinger, K.J., Mayhew, S.D., Bagshaw, A.P., Bowtell, R., and Francis, S.T. (2014). Evidence that the Negative BOLD Response Is Neuronal in Origin: A Simultaneous EEG-BOLD-CBF Study in Humans. NeuroImage 94, 263–274. https://doi.org/10.1016/j.neuroimage.2014.02.029. Jenkinson, M., Beckmann, C.F., Behrens, T.E., Woolrich, M.W., and Smith, S.M. (2012). FSL. Neuroimage 62, 782–790. https://doi.org/10.1016/j.neuro- image.2011.09.015. Muret, D., and Makin, T.R. (2021). The homeostatic homunculus: rethinking deprivation-triggered reorganisation. Curr. Opin. Neurobiol. 67, 115–122. https://doi.org/10.1016/j.conb.2020.08.008. Johnson, B.A., and Frostig, R.D. (2015). Long, intrinsic horizontal axons radi- ating through and beyond rat barrel cortex have spatial distributions similar to horizontal spreads of activity evoked by whisker stimulation. Brain Struct. Funct. 221. https://doi.org/10.1007/s00429-015-1123-7. Nakamura, A., Yamada, T., Goto, A., Kato, T., Ito, K., Abe, Y., Kachi, T., and Kakigi, R. (1998). Somatosensory homunculus as drawn by MEG. Neuroimage 7, 377–386. Ne´ gyessy, L., Pa´ lfi, E., Ashaber, M., Palmer, C., Ja´ kli, B., Friedman, R.M., Chen, L.M., and Roe, A.W. (2013). Intrinsic horizontal connections process global tactile features in the primary somatosensory cortex: neuroanatomical evidence. J. Comp. Neurol. 521, 2798–2817. https://doi.org/10.1002/cne. 23317. Kaas, J.H., Nelson, R.J., Sur, M., Lin, C.S., and Merzenich, M.M. (1979). Mul- tiple representations of the body within the primary somatosensory cortex of primates. Science 204, 521–523. https://doi.org/10.1126/science.107591. Katz, J., and Melzack, R. (1987). Referred sensations in chronic pain patients. Pain 28, 51–59. https://doi.org/10.1016/0304-3959(87)91059-1. Ngo, G.N., Ngo, G.N., Haak, K.V., Beckmann, C.F., and Menon, R.S. (2021). Mesoscale hierarchical organization of primary somatosensory cortex captured by resting-state-fMRI in humans. Neuroimage 235, 118031. https://doi.org/10.1016/j.neuroimage.2021.118031. Kieliba, P., Clode, D., Maimon-Mor, R.O., and Makin, T.R. (2021). Robotic hand augmentation drives changes in neural body representation. Sci. Ro- botics 6, eabd7935. https://doi.org/10.1126/scirobotics.abd7935. Nili, H., Wingfield, C., Walther, A., Su, L., Marslen-Wilson, W., and Kriege- skorte, N. (2014). A toolbox for representational similarity analysis. PLoS Com- put. Biol. 10, e1003553. https://doi.org/10.1371/journal.pcbi.1003553. Kim, J., B€ulthoff, I., and B€ulthoff, H.H. (2018). Article Functional somatotopy revealed across multiple cortical regions using a model of complex motor task. Brain Res. 1531, 25–36. https://doi.org/10.1016/j. brainres.2013.07.050. Gale, D.J., Flanagan, J.R., and Gallivan, J.P. (2021). Human somatosensory cortex is modulated during motor planning. J. Neurosci. 41, 5909–5922. https://doi.org/10.1523/jneurosci.0342-21.2021. Cybulska-Klosowicz, A., Tremblay, F., Jiang, W., Bourgeon, S., Meftah, E.M., and Chapman, C.E. (2020). Differential effects of the mode of touch, active and passive, on experience-driven plasticity in the S1 cutaneous digit representa- tion of adult macaque monkeys. J. Neurophysiol. 123, 1072–1089. https://doi. org/10.1152/jn.00014.2019. Germann, J., Chakravarty, M.M., Collins, D.L., and Petrides, M. (2020). Tight coupling between morphological features of the central sulcus and somato- motor body representations: a combined anatomical and functional mri study. Cereb Cortex 30 1843–1854 https://doi org/10 1093/cercor/bhz208 Glasser, M.F., Coalson, T.S., Robinson, E.C., Hacker, C.D., Harwell, J., Ya- coub, E., Ugurbil, K., Andersson, J., Beckmann, C.F., Jenkinson, M., Smith, S.M., and Van Essen, D.C. (2016). A multi-modal parcellation of human cere- bral cortex. Nature 536, 171–178. Dale, A.M., Fischl, B., and Sereno, M.I. (1999). Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 9, 179–194. https:// doi.org/10.1006/nimg.1998.0395. Darling, W.G., Pizzimenti, M.A., Rotella, D.L., Hynes, S.M., Ge, J., Stilwell- Morecraft, K., and Morecraft, R.J. (2016). Sensorimotor cortex injury effects on recovery of contralesional dexterous movements in macaca mulatta. Exp. Neurol. 281, 37–52. https://doi.org/10.1016/j.expneurol.2016.04.004. Graziano, M.S., and Aflalo, T.N. (2007). Mapping behavioral repertoire onto the cortex. Neuron 56, 239–251. https://doi.org/10.1016/j.neuron.2007.09.013. Graziano, M.S., Taylor, C.S., and Moore, T. (2002). Complex movements evoked by microstimulation of precentral cortex. Neuron 34, 841–851. https://doi.org/10.1016/S0896-6273(02)00698-0. Dempsey-Jones, H., Themistocleous, A.C., Carone, D., Ng, T.W.C., Harrar, V., and Makin, T.R. (2019). Blocking tactile input to one finger using anaesthetic 10 Cell Reports 38, 110523, March 15, 2022 Article Article ll OPEN ACCESS A ti l Sun, H., Blakely, T.M., Darvas, F., Wander, J.D., Johnson, L.A., Su, D.K., Miller, K.J., Fetz, E.E., and Ojemann, J.G. (2015). Sequential activation of premotor, primary somatosensory and primary motor areas in humans during cued finger movements. Clin. Neurophysiol. 126, 2150–2161. https://doi.org/10.1016/j. clinph.2015.01.005. Sun, H., Blakely, T.M., Darvas, F., Wander, J.D., Johnson, L.A., Su, D.K., Miller, K.J., Fetz, E.E., and Ojemann, J.G. (2015). Sequential activation of premotor, Rausell, E., Bickford, L., Manger, P.R., Woods, T.M., and Jones, E.G. (1998). Extensive divergence and convergence in the thalamocortical projection to monkey somatosensory cortex. J. Neurosci. 18, 4216–4232. primary somatosensory and primary motor areas in humans during cued finger movements. Clin. Neurophysiol. 126, 2150–2161. https://doi.org/10.1016/j. clinph.2015.01.005. Richards, C.L., Malouin, F., and Nadeau, S. (2015). Stroke Rehabilitation: Clin- ical Picture, Assessment, and Therapeutic challenge. Prog. Brain Res. 218, 253–280. https://doi.org/10.1016/bs.pbr.2015.01.003. Tal, Z., Geva, R., and Amedi, A. (2017). Positive and negative somatotopic BOLD responses in contralateral versus Ipsilateral Penfield homunculus. Cereb. Cortex 27, 962–980. https://doi.org/10.1093/cercor/bhx024. Root, V., Muret, D., Arribas, M., Amoruso, E., Thornton, J., Tarall-Jozwiak, A., Tracey, I., and Makin, T.R. (2021). Investigating Facial Information Content in the Hand Area of Individuals with a Congenital and Acquired Missing Hand. Preprint at bioRxiv. https://doi.org/10.1101/2021.07.05.451126. Root, V., Muret, D., Arribas, M., Amoruso, E., Thornton, J., Tarall-Jozwiak, A., Tracey, I., and Makin, T.R. (2021). Investigating Facial Information Content in Thomas, J., Sharma, D., Mohanta, S., and Jain, N. (2021). Resting-State func- tional networks of different topographic representations in the somatosensory cortex of macaque monkeys and humans. Neuroimage 228, 117694. https:// doi.org/10.1016/j.neuroimage.2020.117694. the Hand Area of Individuals with a Congenital and Acquired Missing Hand. Preprint at bioRxiv. https://doi.org/10.1101/2021.07.05.451126. Roux, F.E., Djidjeli, I., and Durand, J.B. (2018). Functional architecture of the somatosensory homunculus detected by electrostimulation. J. Physiol. 596, 941–956. https://doi.org/10.1113/JP275243. Walther, A., Nili, H., Ejaz, N., Alink, A., Kriegeskorte, N., and Diedrichsen, J. (2016). Reliability of dissimilarity measures for multi-voxel pattern analysis. Neuroimage 137, 188–200. https://doi.org/10.1016/j.neuroimage.2015.12. 012. Saadon-Grosman, N., Arzy, S., and Loewenstein, Y. (2020). Hierarchical cortical gradients in somatosensory processing. Neuroimage 222, 117257. https://doi.org/10.1016/j.neuroimage.2020.117257. Wesselink, D.B., van den Heiligenberg, F.M., Ejaz, N., Dempsey-Jones, H., Cardinali, L., Tarall-Jozwiak, A., Diedrichsen, J., and Makin, T.R. (2019). Obtaining and maintaining cortical hand representation as evidenced from acquired and congenital handlessness. Elife 8, 1–19. https://doi.org/10. 7554/eLife.37227. Sanchez Panchuelo, R.M., Ackerley, R., Glover, P.M., Bowtell, R.W., Wess- berg, J., Francis, S.T., and McGlone, F. (2016). Article Mapping quantal touch using 7 tesla functional magnetic resonance imaging and single-unit intraneural mi- crostimulation. Elife 5, 1–19. https://doi.org/10.7554/eLife.12812. Sanders, Z.-B., Wesselink, D.B., Dempsey-Jones, H., and Makin, T.R. (2019). Similar Somatotopy for Active and Passive Digit Representation in Primary So- matosensory Cortex. Preprint at bioRxiv. https://doi.org/10.1101/754648. Wesselink, D.B., Sanders, Z.B., Edmondson, L.R., Dempsey-Jones, H., Kie- liba, P., Kikkert, S., Themistocleous, A.C., Emir, U., Diedrichsen, J., Saal, H.P., and Makin, T.R. (2020). Malleability of the cortical hand map following a single finger nerve block. Preprint at bioRxiv. https://doi.org/10.1101/2020. 10.16.338640. Schellekens, W., Thio, M., Badde, S., Winawer, J., Ramsey, N., and Petridou, N. (2021). A touch of hierarchy: population receptive fields reveal fingertip inte- gration in Brodmann areas in human primary somatosensory cortex. Brain Struct. Funct. 226, 2099–2112. https://doi.org/10.1007/s00429-021-02309-5. Schieber, M.H. (2001). Constraints on somatotopic organization in the primary motor cortex. J. Neurophysiol. 86, 2125–2143. https://doi.org/10.1152/jn. 2001.86.5.2125. Schellekens, W., Thio, M., Badde, S., Winawer, J., Ramsey, N., and Petridou, N. (2021). A touch of hierarchy: population receptive fields reveal fingertip inte- gration in Brodmann areas in human primary somatosensory cortex. Brain Struct. Funct. 226, 2099–2112. https://doi.org/10.1007/s00429-021-02309-5. Wesselink, D., and Maimon-Mor, R.O. (2018). RSA toolbox extension for FSL. https://github.com/ronimaimon/rsatoolbox. Schieber, M.H. (2001). Constraints on somatotopic organization in the primary motor cortex. J. Neurophysiol. 86, 2125–2143. https://doi.org/10.1152/jn. 2001.86.5.2125. Widener, G.L., and Cheney, P.D. (1997). Effects on muscle activity from micro- stimuli applied to somatosensory and motor cortex during voluntary move- ment in the monkey. J. Neurophysiol. 77, 2446–2465. https://doi.org/10. 1152/jn.1997.77.5.2446. Schridde, U., Khubchandani, M., Motelow, J.E., Sanganahalli, B.G., Hyder, F., and Blumenfeld, H. (2008). Negative BOLD with large increases in neuronal ac- tivity. Cereb. Cortex 18, 1814–1827. https://doi.org/10.1093/cercor/bhm208. Wiech, K., Jbabdi, S., Lin, C.S., Andersson, J., and Tracey, I. (2014). Differen- tial structural and resting state connectivity between insular subdivisions and other pain-related brain regions. Pain 155, 2047–2055. https://doi.org/10. 1016/j.pain.2014.07.009. Shmuel, A., Augath, M., Oeltermann, A., and Logothetis, N.K. (2006). Negative functional MRI response correlates with decreases in neuronal activity in mon- key visual area V1. Nat. Neurosci. 9, 569–577. https://doi.org/10.1038/nn1675. Wiestler, T., and Diedrichsen, J. (2013). Skill learning strengthens cortical rep- resentations of motor sequences. Elife 2, e00801–e00820. https://doi.org/10. 7554/eLife.00801. Smith, S.M. (2002). Fast robust automated brain extraction. Hum. Brain Mapp. 17, 143–155. https://doi.org/10.1002/hbm.10062. Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E., Johansen-Berg, H., Bannister, P.R., De Luca, M., Drobnjak, I., Flitney, D.E., et al. (2004). 12 Cell Reports 38, 110523, March 15, 2022 Article Decoding visual roughness perception: an fMRI study. Somatosens Mot Res 35, 212–217. https://doi. org/10.1080/08990220.2018.1527761. Kuehn, E., Dinse, J., Jakobsen, E., Long, X., Scha¨ fer, A., Bazin, P.L., Villringer, A., Sereno, M.I., and Margulies, D.S. (2017). Body topography parcellates hu- man sensory and motor cortex. Cereb. Cortex 27, 3790–3805. https://doi.org/ 10.1093/cercor/bhx026. Penfield, W., and Boldrey, E. (1937). Somatic motor and sensory representa- tion in the cerebral cortex of man as studied by electrical stimulation. Brain 60, 389–443. Pons, T.P., Garraghty, P.E., Ommaya, A.K., Kaas, J.H., Taub, E., and Mishkin, M. (1991). Massive cortical reorganization after sensory deafferentation in adult macaques. Science 252, 1857–1860. Kumar, N., Manning, T.F., and Ostry, D.J. (2019). Somatosensory cortex par- ticipates in the consolidation of human motor memory. PLoS Biol. 17, e3000469–21. https://doi.org/10.1371/journal.pbio.3000469. Powell, T.P., and Mountcastle, V.B. (1959). Some aspects of the functional or- ganization of the cortex of the postcentral gyrus of the monkey: a correlation of findings obtained in a single unit analysis with cytoarchitecture. Bull. Johns Hopkins Hosp. 105, 133–162. Li, Y., Zhang, S., Jin, Y., Cai, B., Controzzi, M., Zhu, J., Zhang, J., and Zheng, X. (2017). Gesture decoding using ECoG signals from human sensorimotor cor- tex: a pilot study. Behav. Neurol. 2017, 1–12. https://doi.org/10.1155/2017/ 3435686. Qi, H.X., Kaas, J.H., and Reed, J.L. (2014). The reactivation of somatosensory cortex and behavioral recovery after sensory loss in mature primates. Front. Syst. Neurosci. 8, 84–114. https://doi.org/10.3389/fnsys.2014.00084. London, B.M., and Miller, L.E. (2013). Responses of somatosensory area 2 neurons to actively and passively generated limb movements. J. Neurophysiol. 109, 1505–1513. https://doi.org/10.1152/jn.00372.2012. Ramachandran, V.S., Rogers-Ramachandran, D., and Stewart, M. (1992). Perceptual correlates of massive cortical reorganization. Science 258, 1159– 1160. Martuzzi, R., van der Zwaag, W., Farthouat, J., Gruetter, R., and Blanke, O. (2014). Human finger somatotopy in areas 3b, 1, and 2: a 7T fMRI study using Cell Reports 38, 110523, March 15, 2022 11 Article Data and code availability y d The data generated during this study will be available from the Open Science Framework upon publication (https://osf.io/ g3y5u/). d The data generated during this study will be available from the Open Science Framework upon publication (https://osf.io/ g3y5u/). d This paper does not report original code. d Any additional information required to reanalyse the data reported in this paper is available from the lead contact upon request. d Any additional information required to reanalyse the data reported in this paper is available from the d Any additional information required to reanalyse the data reported in this paper is available from the lead contact upon request. Article Advances in functional and structural MR image analysis and im- plementation as FSL. Neuroimage 23, S208–S219. https://doi.org/10.1016/j. neuroimage.2004.07.051. Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E., Johansen-Berg, H., Bannister, P.R., De Luca, M., Drobnjak, I., Flitney, D.E., Willoughby, W.R., Thoenes, K., and Bolding, M. (2021). Somatotopic arrange- ment of the human primary somatosensory cortex derived from functional magnetic resonance imaging. Front. Neurosci. 14, 1–10. https://doi.org/10. 3389/fnins.2020.598482. et al. (2004). Advances in functional and structural MR image analysis and im- plementation as FSL. Neuroimage 23, S208–S219. https://doi.org/10.1016/j. neuroimage.2004.07.051. Woolsey, T.A., Rovainen, C.M., Cox, S.B., Henegar, M.H., Liang, G.E., Liu, D., Moskalenko, Y.E., Sui, J., and Wei, L. (1996). Neuronal units linked to micro- vascular modules in cerebral cortex: response elements for imaging the brain. Cereb. Cortex 6, 647–660. https://doi.org/10.1093/cercor/6.5.647. Soler, M.D., Kumru, H., Vidal, J., Pelayo, R., Tormos, J.M., Fregni, F., Navarro, X., and Pascual-Leone, A. (2010). Referred sensations and neuropathic pain following spinal cord injury. Pain 150, 192–198. https://doi.org/10.1016/j. pain.2010.04.027. Sun, F., Zhang, G., Ren, L., Yu, T., Ren, Z., Gao, R., and Zhang, X. (2021). Func- tional organization of the human primary somatosensory cortex: a stereo-elec- troencephalography study. Clin. Neurophysiol. 132, 487–497. https://doi.org/ 10.1016/j.clinph.2020.11.032. Zeharia, N., Hertz, U., Flash, T., and Amedi, A. (2012). Negative blood oxygen- ation level dependent homunculus and somatotopic information in primary motor cortex and supplementary motor area. Proc. Natl. Acad. Sci. U S A 109, 18565–18570. https://doi.org/10.1073/pnas.1119125109. 12 Cell Reports 38, 110523, March 15, 2022 ll OPEN ACCESS ll OPEN ACCESS REAGENT or RESOURCE SOURCE IDENTIFIER Deposited data fMRI data This paper https://osf.io/g3y5u/ https://doi.org/10.17605/OSF.IO/G3Y5U Software and algorithms FSL v. 6.00 Smith et al., 2004; Jenkinson et al., 2012 https://fsl.fmrib.ox.ac.uk Connectome Workbench v. 1.4.2 https://www.humanconnectome.org https://www.humanconnectome.org Matlab v. R2016a Mathworks https://www.mathworks.com Freesurfer v. 7.1.1 Dale et al. (1999); Fischl et al. 2001 https://freesurfer.net JASP v. 0.14 https://github.com/jasp-stats/jasp- desktop/commit/f2cb223fef6b15089fd 6e9da54825d370c1ece57 https://jasp-stats.org/ The study did not generate new materials. The study did not generate new materials. Scanning procedures Scanning procedures Each dataset comprised three or four functional task-related block-design runs, a functional localiser, a structural scan and field maps. Body and face tasks Each dataset comprised three or four functional task-related block-design runs, a functional localiser, a structural scan and field maps. EXPERIMENTAL MODEL AND SUBJECT DETAILS Twenty-two healthy volunteers [mean age = 45.55 ± 9.47 (SD) years; 10 women; 6 left-handed] took part in the body and face tasks and a further nineteen healthy volunteers [mean age = 23.16 ± 4.34 (SD) years; 11 women; all right-handed] took part in the finger task. To account for age-related differences, age was added as a covariate in statistical analyses. Participants reported no senso- rimotor disorders and had no counterindications for magnetic resonance imaging. All participants gave written informed consent before participating. The protocols were approved by the NHS National Research Ethics Service approval (18/LO/0474) for the body and face tasks and UCL Research Ethics Committee (REC: 12921/001) for the finger task and were performed in accordance with the Declaration of Helsinki. The face and hand datasets were recently used for other purposes (Kieliba et al., 2021; Root et al., 2021). Materials availability The study did not generate new materials. RESOURCE AVAILABILITY Lead contact For further information and requests for resources should be directed to and will be fulfilled by the lead contact, Dollyane Muret ([email protected]). Lead contact fMRI pre-processing Functional data was pre-processed in FSL-FEAT (v. 6.00) and included the following steps: motion correction using MCFLIRT (Jen- kinson et al., 2002); brain extraction using BET (Smith, 2002); high-pass temporal filtering with a cut-off of 150s, 119s and 150s for the body, face and finger tasks respectively and 280s for the functional localiser; and finally spatial smoothing using a Gaussian kernel with a full width at half maximum of 3 mm for the three tasks, and 5 mm for the functional localiser. Field maps were used for distortion correction of the functional data from the body and face tasks and the functional localiser collected for these participants. For each participant, a midspace between the different functional runs of each task was calculated, i.e., the average space in which the images are minimally reorientated. Each functional run was then aligned to the midspace and registered to each individual structural T1 scan using FMRIB’s Linear Image Registration Tool (FLIRT), optimised using Boundary-Based Registration (Greve and Fischl, 2009). Where specified, functional and structural data were transformed to MNI152 space using FMRIB’s Nonlinear Registration Tool (FNIRT). Finger task Participants performed an active finger tapping task using a button box. Each finger movement was repeated at 1Hz over a period of 9s per block, with 4 blocks per finger per run in a semi-counterbalanced order and 4 runs in total. Instructions and pace were provided visually. Ten vertical bars, representing the fingers, flashed individually in green at a frequency of 1 Hz. Task performance was moni- tored online. Two additional conditions involving the feet and lips were also included but will not be further described as they were not included in the main analysis (see Kieliba et al., 2021 for analysis of these conditions). MRI data acquisition MRI images were acquired using a 3T Prisma MRI scanner (Siemens, Erlangen, Germany) with a 32-channel head coil. Functional data were obtained using a multiband T2*-weighted pulse sequence with a between-slice acceleration factor of 4 and no in-slice ac- celeration. The following acquisition parameters were used: TR = 1450 ms; TE = 35 ms; flip angle =70; voxel size = 2 mm isotropic; imaging matrix = 106 3 106; FOV = 212 mm. 72 slices were oriented in the transversal plane covering the entire brain. A total of 216, 172 and 346 volumes were collected per participant for each run of the body, face and finger tasks respectively. Field-maps were acquired for field unwarping. A T1-weighted sequence (MPRAGE, TR = 2530 ms; TE = 3.34 ms; flip angle = 7; voxel size = 1 mm isotropic) was used to obtain anatomical images. QUANTIFICATION AND STATISTICAL ANALYSIS MRI analysis was implemented using tools from FSL (v. 6.00, Smith et al., 2004; Jenkinson et al., 2012), Connectome Workbench software (v. 1.4.2, humanconnectome.org) in combination with bash and Matlab scripts (v. R2016a, mathworks.com), both devel- oped in-house and as part of the RSA Toolbox (Nili et al., 2014). Cortical surface reconstructions were produced using FreeSurfer (v. 7.1.1; Dale et al. 1999; Fischl et al. 2001, freesurfer.net). Body and face tasks Two additional conditions involving the non-domi- nant arm were also included but will not be described in the main text since we focus on the hemisphere contralateral to each body part. See supplemental information for similar analyses and results in the non-dominant hemisphere for the body task (Figures S2 and S4). For the face task, participants were instructed to perform one of four movements: raise the eyebrows (i.e., forehead), flare nostrils (i.e., nose), puckering lips (i.e., lips), and tap the tongue to the roof of the mouth (i.e., tongue). Two additional conditions involving the body parts (i.e., feet, dominant hand and lips), resulting in a total of six conditions. Two additional conditions involving the non-domi- nant arm were also included but will not be described in the main text since we focus on the hemisphere contralateral to each body part. See supplemental information for similar analyses and results in the non-dominant hemisphere for the body task (Figures S2 and part. See supplemental information for similar analyses and results in the non-dominant hemisphere for the body task (Figures S2 and S4). For the face task, participants were instructed to perform one of four movements: raise the eyebrows (i.e., forehead), flare nostrils (i.e., nose), puckering lips (i.e., lips), and tap the tongue to the roof of the mouth (i.e., tongue). Two additional conditions involving the left and right thumbs were also included but will not be further described as they were not included in the main analysis (see Root et al., 2021 for analysis of these conditions). For both tasks, instructions and pace were provided visually via a screen, resulting in 5 cycles of movement per 8 sec block. In addition, each movement block was repeated 4 times per run, which also comprised 5 blocks of rest used as baseline. Conditions were pseudo-randomly distributed, such that each condition was equally preceded by all other conditions. Three and four functional runs were acquired for the face and body tasks, respectively. To confirm that appropriate movements were made at the instructed times, task performance was visually monitored online. Functional localiser Participants were visually instructed to move one of five body parts: right or left hand (open/closing the fingers), right or left toes (wiggling the toes) or lips (puckering the lips). Movements were repeated at a constant instructed pace for a period of 12s, interleaved with 12s of rest. Each condition was repeated 4 times in a pseudo-random order. Here again, participants practiced the movements before entering the scanner and task performance was visually monitored online. Body and face tasks y The two tasks were performed within the same experimental session. Prior to entering the scanner room, participants were thor- oughly instructed, and all movements were practiced in front of the experimenter to ensure they were performed correctly. For the body task, participants were instructed to perform one of two actions (i.e., squeeze or push; Figure S1) with one of three different Cell Reports 38, 110523, March 15, 2022 e1 Cell Reports 38, 110523, March 15, 2022 e1 Article A ti l ll OPEN ACCESS body parts (i.e., feet, dominant hand and lips), resulting in a total of six conditions. Two additional conditions involving the non-domi- nant arm were also included but will not be described in the main text since we focus on the hemisphere contralateral to each body part. See supplemental information for similar analyses and results in the non-dominant hemisphere for the body task (Figures S2 and S4). For the face task, participants were instructed to perform one of four movements: raise the eyebrows (i.e., forehead), flare nostrils (i.e., nose), puckering lips (i.e., lips), and tap the tongue to the roof of the mouth (i.e., tongue). Two additional conditions involving the left and right thumbs were also included but will not be further described as they were not included in the main analysis (see Root et al., 2021 for analysis of these conditions). body parts (i.e., feet, dominant hand and lips), resulting in a total of six conditions. Two additional conditions involving the non-domi- nant arm were also included but will not be described in the main text since we focus on the hemisphere contralateral to each body part. See supplemental information for similar analyses and results in the non-dominant hemisphere for the body task (Figures S2 and S4). For the face task, participants were instructed to perform one of four movements: raise the eyebrows (i.e., forehead), flare nostrils (i.e., nose), puckering lips (i.e., lips), and tap the tongue to the roof of the mouth (i.e., tongue). Two additional conditions involving the left and right thumbs were also included but will not be further described as they were not included in the main analysis (see Root et al., 2021 for analysis of these conditions). body parts (i.e., feet, dominant hand and lips), resulting in a total of six conditions. Definition of regions of interest (ROIs) Since we were interested in investigating the information content of highly selective regions across the S1 homunculus, we used the functional localiser to select highly selective voxels to toe, hand and lip movements within anatomical S1 masks. The functional ROI was restricted by anatomical criteria, as detailed below. Although M1 is expected to be largely activated during each movement, M1 topography tends to be less well-defined and thus information content more widespread (Schieber, 2001; Graziano and Aflalo, 2007). We therefore primarily focus on the more topographically selective S1, though we wish to note that marginal contribution from M1 may have affected our S1 activity profiles due to their spatial proximity, the probabilistic nature of our anatomical masks and spatial smoothing of the data. First, S1 was defined on a template surface using probabilistic cytoarchitectonic maps, by selecting only nodes that belonged to the grey matter of Brodmann areas (BAs) 3a, 3b, 1 and 2 with maximal probability (Wiestler and Diedrichsen, 2013). This S1 anatom- ical mask was then split into three anatomical sub-regions. A node approximately 1 cm below and above the hand knob was defined as an anatomical hand sub-region. Note that this criterion defined a more conservative hand region than was done in previous work (Wiestler and Diedrichsen, 2013; Wesselink et al., 2019; Kieliba et al., 2021). A gap of 1 cm was then defined above and below this anatomical hand sub-region, and the remaining medial and lateral parts of S1 were used as the other two anatomical sub-regions. Structural T1-weighted images were used to reconstruct the pial and white-grey matter surfaces using Freesurfer. Surface co- registration across hemispheres was done using spherical alignment. The three anatomical S1 sub-regions were then projected into the individual brains via the reconstructed individual anatomical surfaces. To exclude any possible contribution from neighbour- ing more integrative regions that contain information from multiple body parts, we further trimmed in participant’s structural space: i) the medial sub-region by removing any overlap with BA5L and BA5M, and ii) the lateral sub-region by removing any overlap with S2. BA5L, BA5M and S2 were defined in MNI152 space using the Juelich Histological Atlas thresholded at 25% maximum probability (Wiech et al., 2014). BA5L, BA5M and S2 were then registered to participants’ structural space using an inversion of the nonlinear registration carried about by FNIRT and used to trim our anatomical sub-regions. fMRI low-level analysis For the functional localiser, each condition (i.e., right/left hand, right/left toes, lips) was contrasted against all other conditions to identify the most selective voxels. The activity patterns associated with these five contrasts were then registered to the structural space of each individual and to the functional space of each task using FLIRT to define regions of interest. Definition of regions of interest (ROIs) These trimmed anatomical sub-regions were then registered to functional space of each task using FLIRT, excluding voxels partially overlapping with the central sulcus (thresholding at 0.5 for body and face tasks, and 0.2 for the finger task) to minimise M1 contribution, and used to mask the functional localiser contrasts. The medial S1 sub-region was used to mask the toe contrasts, the central hand sub-region to mask the hand contrasts and the lateral S1 sub-region to mask the lip contrast. Within each of these anatomical sub-regions, we then selected the 50 most activated voxels for the corresponding contrasts (all contrasts vs all other body parts, see section ‘fMRI low-level analysis’). This provided us with the most selective Leg, Hand and Face ROIs for each indi- vidual, while ensuring the same ROI size across participants and regions. fMRI low-level analysis Voxel-wise General Linear Model (GLM) was applied to the data using FEAT to obtain statistical parametric maps for each movement. For each task, the design comprised a regressor of interest for each movement convolved with a double-gamma hemodynamic response function (Friston et al., 1998) and its temporal derivative. The six motion parameters were included as regressors of no e2 Cell Reports 38, 110523, March 15, 2022 Article ll OPEN ACCESS interest. Large head movements between volumes (>0.9 mm for body and face tasks, > 1 mm for finger task) were defined as motion outliers and regressed out, by adding additional regressors of no interest to the GLM [body task: mean proportion of volumes excluded = 0.45 ± 0.76 (SD) %; face task: mean proportion of volumes excluded = 0.36 ± 0.67 (SD) %; finger task: mean proportion of volumes excluded = 0.32 ± 0.77 (SD) %]. interest. Large head movements between volumes (>0.9 mm for body and face tasks, > 1 mm for finger task) were defined as motion outliers and regressed out, by adding additional regressors of no interest to the GLM [body task: mean proportion of volumes excluded = 0.45 ± 0.76 (SD) %; face task: mean proportion of volumes excluded = 0.36 ± 0.67 (SD) %; finger task: mean proportion of volumes excluded = 0.32 ± 0.77 (SD) %]. For each task, a contrast relative to rest was set up for each movement, resulting in 8 contrasts for the body task (i.e., lip, dominant hand, non-dominant arm, right/left foot x Squeeze or Push, each vs rest), in 6 contrasts for the face task (i.e., forehead, nose, lips, tongue; and left/right thumb, not used here), and in 12 contrasts for the finger task (i.e., each digit of each hand; and feet and lips, not used here). The estimates from the total number of functional runs for each task (3 for face task, 4 for body and finger tasks) were then averaged voxel-wise at the individual level using fixed effects model. For the face task, each estimates’ average was masked prior to cluster formation with a sensorimotor mask, defined as the precentral and postcentral gyrus from the Harvard Cortical Atlas. The sensorimotor mask was registered to the individuals structural scan using an inversion of the nonlinear registration by FNIRT. Multivariate representational similarity analysis Representational Similarity Analysis (RSA; Nili et al., 2014) was used to assess the multivariate relationship between the contralateral activity patterns generated by each movement. The dissimilarity between activity patterns within each S1 ROI (i.e., Leg, Hand and Face) was computed at the individual level for each pair of movements using cross-validated squared Mahalanobis distance (Walther et al., 2016). Multidimensional noise normalisation was used to increase reliability of distance estimates (noisier voxels are down- weighted), based on the voxel’s covariance matrix calculated from the GLM residuals. Due to cross-validation, the expected value of the distance is zero (or negative) if two patterns are not statistically different from each other, and significantly greater than zero if the two representational patterns are different (Diedrichsen et al., 2016). Larger distances for movement pairs therefore suggest greater information content. The resulting representational pairwise distances (i.e., 8 for the body task, 6 for the face task and 10 for the finger task) were extracted. For the body task, the dissimilarities obtained between pairs of body parts when performing similar actions (e.g., dissimilarity between lip squeeze and feet squeeze) were averaged across actions (e.g., previous example averaged with dissimilarity between lip push and feet push) to define overall dissimilarity between body parts. For the face and finger tasks, since face movements evoked bilateral activity and finger movements were performed with each hand, and since no major significant differences were observed between hemispheres (see above), dissimilarities from the two hemispheres were averaged within indi- vidual participants. To further reduce the number of comparisons while still assessing the topographical content, dissimilarities from different pairs of subparts (i.e., face parts or fingers) were grouped according to the subpart’s cortical neighborhood (i.e., adjacent vs non-adjacent). Multidimensional scaling (MDS) was used to project the higher-dimensional representational dissimilarity matrices into lower-dimensional space, whilst preserving pairwise dissimilarities, for visualisation purposes only. To illustrate what activity the RSA analysis is relying upon, we computed the unthresholded S1 univariate maps of participants showing dissimilarities close to the group median for the different contents of interest (e.g., median lips-hand dissimilarity in the Leg ROI for the body task, Fig- ure S5A). Analysis was conducted on an adapted version of the RSA Toolbox in MATLAB (Nili et al., 2014), customised for FSL (Wes- selink and Maimon-Mor, 2018). Univariate analysis The z statistic time series from the 50 voxels of each ROI obtained for each movement were extracted and averaged. These averaged values were used to assess the selectivity of our ROIs. Univariate information content was defined as the absolute difference between the averaged univariate activity evoked by two movements in a given ROI. For the body task, the two absolute differences obtained between pairs of body parts when performing the same action (i.e., squeeze or push) were averaged to define an overall difference between body parts. For the face and finger tasks, since face movements evoked bilateral activity and finger movements were performed with each hand, and since no major differences were observed across hemispheres [three-way ANOVA Face task: all F % 1.30, all p R 0.267, except for a triple interaction Hemi*ROIs*Face subparts (F(3.12,65.49) = 3.06, p = 0.033) revealing a significant difference between hemispheres for the lips (z(21) = 45.00, p = 0.007) and for the nose in the Leg ROI only (t(21) =
2.78, p = 0.011); three-way ANOVA Finger task: all F % 1.20, all p R 0.293], absolute differences from the two hemispheres were averaged within Cell Reports 38, 110523, March 15, 2022 e3 Article A ti l ll OPEN ACCESS ll OPEN ACCESS participants. To further reduce the number of comparisons while still assessing the topographical content, absolute difference from different pairs of subparts (i.e., face parts or fingers) were grouped according to the subpart’s cortical neighborhood (i.e., adjacent vs non-adjacent). participants. To further reduce the number of comparisons while still assessing the topographical content, absolute difference from different pairs of subparts (i.e., face parts or fingers) were grouped according to the subpart’s cortical neighborhood (i.e., adjacent vs non-adjacent). BA3b analysis The same analyses as the ones performed in the S1 ROIs described above were performed throughout BA3b’s strip. BA3b was defined on the same template surface as S1 using probabilistic cytoarchitectonic maps, by selecting the nodes showing at least 50% maximum probability for the grey matter of BA3b (Wiestler and Diedrichsen, 2013). Due to the use of surface-based ROIs where BA3b does not directly neighbour M1 (BA4p), this analysis is unlikely to be contaminated by M1 contributions. BA3b’s strip was then segmented into 30 bands, each 2.09 mm high in the medio-lateral direction (see Figure 5A). Similar to the S1 ROIs, these bands were then projected into the individual brains via the reconstructed individual anatomical surfaces and registered to participants’ functional space of each task using FLIRT. On average, each band contained 46.12 voxels ±16.22 (SD). Univariate and multivariate analyses were then repeated in each of these bands. Note that the most medial band contained very few and discontinuous voxels that pre- vented from getting reliable RSA dissimilarities, and was thus excluded from further analysis. e4 Cell Reports 38, 110523, March 15, 2022 Cell Reports 38, 110523, March 15, 2022 e5 Statistical analysis Resulting consistency maps were then projected to a group cortical surface (Glasser et al., 2016) using Connectome Workbench (v1.4.2) (see Figures 1 and S2 for body task; see Figure S3B for face and finger tasks). Article ll OPEN ACCESS Article A ti l ll OPEN ACCESS ll OPEN ACCESS Statistical analysis All statistical analyses were carried out using JASP (v. 0.14). Two-tailed one-sample t-tests versus zero were used to assess signif- icant activity levels in each ROIs. Alpha levels were Bonferroni corrected for the number of tests performed across conditions within each ROI (i.e., alpha = 0.017 corrected for three comparisons for the body task, alpha = 0.013 corrected for four comparisons for the face task and alpha = 0.010 corrected for five comparisons for the finger task). Since negative dissimilarity measures represent noise levels, one-tailed one-sample t-tests versus zero were used to test the significance of representational dissimilarities as well as ab- solute differences in activity levels in each ROIs. Here again, alpha levels were Bonferroni corrected for the number of tests performed within each ROI (i.e., corrected for three comparisons for the body and action dissimilarities, and for two comparisons for the adjacent and non-adjacent dissimilarities for the face and finger tasks). Paired t-tests were used to compare adjacent and non-adjacent con- ditions for the face and finger tasks. In each case, a trend was defined when p values were inferior to twice the corrected alpha level. Data normality was assessed using Shapiro-Wilk test. Effect sizes were computed using Cohen’s d (Cohen, 1988), and based on benchmarks suggested by Cohen, a large effect size was defined as greater than 0.8. Wilcoxon signed-rank t-tests were used when data violated normality assumptions. Two three-way rmANOVAs with the factors Hemisphere, ROI and Subpart were applied to univariate activity levels from the face and finger tasks to assess Hemisphere effect or interaction. Since no major differences were observed across hemispheres, univariate and multivariate data from each hemisphere were averaged. Greenhouse-Geisser correc- tion was applied when data did not follow sphericity assumption. All group data are expressed as means ± SEM, except mentioned otherwise. e4 Cell Reports 38, 110523, March 15, 2022 Group-level ROI visualisations S1 ROIs of each participant were projected to MNI152 space using the nonlinear registration carried about by FNIRT. Participant information regarding hand dominance were used to sagittal-flip data, such that the ROIs contralateral to the dominant hand were always represented in the left hemisphere. ROIs of all participants were then concatenated into a single volume to produce a consistency map (i.e., how many participants have their ROIs overlapping in the MNI space). Group-level ROI visualisations Group-level ROI visualisations p S1 ROIs of each participant were projected to MNI152 space using the nonlinear registration carried about by FNIRT. Participant information regarding hand dominance were used to sagittal-flip data, such that the ROIs contralateral to the dominant hand were always represented in the left hemisphere. ROIs of all participants were then concatenated into a single volume to produce a consistency map (i.e., how many participants have their ROIs overlapping in the MNI space). Resulting consistency maps were then projected to a group cortical surface (Glasser et al., 2016) using Connectome Workbench (v1.4.2) (see Figures 1 and S2 for body task; see Figure S3B for face and finger tasks). S1 ROIs of each participant were projected to MNI152 space using the nonlinear registration carried about by FNIRT. Participant information regarding hand dominance were used to sagittal-flip data, such that the ROIs contralateral to the dominant hand were always represented in the left hemisphere. ROIs of all participants were then concatenated into a single volume to produce a consistency map (i.e., how many participants have their ROIs overlapping in the MNI space). Resulting consistency maps were then projected to a group cortical surface (Glasser et al., 2016) using Connectome Workbench (v1.4.2) (see Figures 1 and S2 for body task; see Figure S3B for face and finger tasks).

https://openalex.org/W2145347299

https://zenodo.org/record/2023184/files/article.pdf

English

LIV. <i>Some observations in reply to the paper of Prof. C. G. Knott “On Swan's Prism Photometer, etc.”</i>

˜The œLondon, Edinburgh and Dublin philosophical magazine and journal of science

public-domain

Philosophical Magazine Series 5 SSN: 1941-5982 (Print) 1941-5990 (Online) Journal homepage: http://www.tandfonline.com/loi/tphm16 Philosophical Magazine Series 5 O. Lummer & E. Brodhun O. Lummer & E. Brodhun To cite this article: O. Lummer & E. Brodhun (1900) LIV. Some observations in reply to the paper of Prof. C. G. Knott “On Swan's Prism Photometer, etc.” , Philosophical Magazine Series 5, 49:301, 541-543, DOI: 10.1080/14786440009463879 Date: 20 June 2016, At: 04:36 Communleated by ~he Authors. i 23 y t Zeitschriftf. Instrkde, ix. pp. 23-25 (1889). b d i 41 50 (1889) f f p Ibid. ix. pp. 41-50 (1889). f f , pp ( ) Ibid. ix. pp. 41-50 (1889). To link to this article: http://dx.doi.org/10.1080/14786440009463879 To link to this article: http://dx.doi.org/10.1080/14786440009463879 Published online: 21 Apr 2009. Submit your article to this journal Article views: 4 View related articles Download by: [The University Of Melbourne Libraries] Download by: [The University Of Melbourne Libraries] [ 5tl ] LIV. Some Observations in reply to tire Paper of Prof. C. G. Krwtt " On Swan's Prism Photometer~ etc." By O. LUMMER and E. BROI)t~U~ ~'. U U NDER the title " On Swan's Prism-Photometer, com- monly called Lummer and Brodhun's Photometers" Prof. C. G. Knott calls attention, in the January number of this periodical, to the fact that Swan had already, in 1859, " invented, described, constructed, and used ' the photometer" afterwards " re-invented" byus. The description given by Swan of his photometer-cube at that time is also reprinted. Downloaded by [The University Of Melbourne Libraries] at 04:36 20 June 2016 p p ~av we be permitted to offer some observations correcting certain points in the above-named paper which seem to us to be misleading ? g It is, first of all, certainly much to be regretted that this publication has been delayed until now, although Prof. Knott says the Swan photometer has been " familiar to all officially connected with the Edinburgh University Physical Labor- atory tbr some years pas~." y y p Now, however, that Mr. Swan's precedence in the field is brought to onr attention after this long interval, we desire to be among the first to acknowledge it. But the statement that the photometer which we introduced into the technical world is nothing moro than a re-inventlon of Swan's photo- meter seems equally unfair to us. Fig. 1. On the basis of theoretical eonsideratlons we formulated the conditions necessary for a "Gleiehheits-Photometer" of the greatest possible sensitiveness. Following theso in our preliminary paper ~, only thag photometer-cube was described which thlfilled the theoretical requirement.s of an ideal "grease- s ~ot" fi~. 11. This preliminary paper was followed, how- eP, er b ( our " Photomet,isehe Untersuehungen" ~/, in which , Y the advantages and disadvantages of the various possible f f , p Ibid. ix. pp. 41-50 (1889). 542 On the Pvlsm Photometer. prism combinations were discussed. Among these is to be found the one which, as we now know, was invented by Swan fort,y-one years ago. g "~ Photom. Untersuchungen, it. : " Lichtmessung durch Schittzung gleicher IIelligkeitsuntelschiedc (Ko trastphotometer)" Zeitschri/'g f I,.~/rl,,le, ix pp 461 -1~5 (1889). - " " "- "' To link to this article: http://dx.doi.org/10.1080/14786440009463879 This form was in our opinion, however, imperfect, and unsuited to technical requirements for the following reasons :--Apart from the fact ~hat the balsam spot is not perfectly transparent, its edges do not afford ~hat exireme sharpness which can be aitMned by ground surfaces, and they deteriorate by continually drying up. In consequence of this the balsam-spot cube, as well as some of the other prism-combinations perhaps s,lperior to it, has never been recommended by ns, and we doubt whether any conslder.~ble number have found their way into practical use in technical indusiry. Should photometers with balsanl cubes, nevertheless, be now in use, they should of course be called Swan's photometers. But let us not forget that only with the most imperfect of all our described forms does the cube of Swan coincide, and the coincidence only extends to our " Gteichheit's " principle * Downloaded by [The University Of Melbourne Libraries] at 04:36 20 June 2016 p p But we did not stop with the " Gleichheits-Wiirfel;" we have als'o constructed a prism-combination which not only satisfies all the conditions of the " Gleichheits" principle, but "dso allows the more sensitive "contrast-principle" t, first Downloaded by [The University Of Melbourne Lib Fig. 2. A I P 1T I 7- worked out by us, to be realized. The construction of this prism combination with the help of a sand-blast is described in our paper, " Photom. Untersuehungen I.," under cube No. 5. Its fiehl has commonly the appearance indicated in fig. 2. The different field-sections (I.-IV.) are separated * It may be remarked here that the cube form had been already used earlier by Fuchs for photometrical purposes, as mentioned in our paper. As criterion lie used the disappearance of the interference-bands, which arise in the thin air-layer between the two prisms near the limit of total reflexion, and which are complementary to each other in reflected and refracted light. "~ Photom. Untersuchungen, it. : " Lichtmessung durch Schittzung gleicher IIelligkeitsuntelschiedc (Ko trastphotometer)" Zeitschri/'g f I,.~/rl,,le, ix pp 461 -1~5 (1889). - " " "- "' Fig. 2. A I P 1T I 7- worked out by us, to be realized. The construction of this prism combination with the help of a sand-blast is described in our paper, " Photom. Untersuehungen I.," under cube No. 5. Its fiehl has commonly the appearance indicated in fig. 2. * It may be remarked here that the cube form had been already used earlier by Fuchs for photometrical purposes, as mentioned in our paper. As criterion lie used the disappearance of the interference-bands, which arise in the thin air-layer between the two prisms near the limit of total reflexion, and which are complementary to each other in reflected and refracted light. * Photom. Untersuchungen, iv. : Zeitschrififi Instrkde, xli. pp. 41-50 (1892). t Communicated by Lord Kelvin ; having been read before the Royal Society, Feb. 1, 1900. :~ T'he experime~tal results of the paper were obtained two years ago. Various circumstances have prevented earlier publication ; and it was originally intended to carry the investigation further before publishing. Want of opportunity, however, makes this difficult for the present ; and we therefore deem it advisable to put our results on record just now, as they stand. The present investigation, as well as the former work referred to ~n the text above, has been assisted by grants from the Government Grant Fund. w " On Thermal Radiatiml iu Absolute Measure," J. T. l~ottonfley, w " On Thermal Radiatiml iu Absolute Measure," J. T. l~ottonfley, lh)v. 5oc. Prec. a~ld t'hil. Trans. 1884 -lg93. To link to this article: http://dx.doi.org/10.1080/14786440009463879 The different field-sections (I.-IV.) are separated 543 On T~ermal Radiation in Absolute Measure. from each other by hair sharp edges, which disappear com- pletely, so that the whole fi_eld ABCD looks like a uniform, diffusely illmninated surface. But as soon as one introduces thin glass plates at proper points in front of the two illumi- nated sides of the cube the " prittciple of contrast" is added to the "principle of equality." The fields L and II. are of uniform intensity, in consequence of which their separating- edge pq disappears; the fields III and IV. have another intensity, but the difference between I. and IV. being thB same as the difference between II. and III., they show equal contrast. from each other by hair sharp edges, which disappear com- pletely, so that the whole fi_eld ABCD looks like a uniform, diffusely illmninated surface. But as soon as one introduces thin glass plates at proper points in front of the two illumi- nated sides of the cube the " prittciple of contrast" is added to the "principle of equality." The fields L and II. are of uniform intensity, in consequence of which their separating- edge pq disappears; the fields III and IV. have another intensity, but the difference between I. and IV. being thB same as the difference between II. and III., they show equal contrast. Downloaded by [The University Of Melbourne Libraries] at 04:36 20 June 2016 In the treatise ~ entitled " Die photometrischen Apparato der Reichsansialt fiir den technischen Gebrauch," our photo- meter supplied with this ~' Gleichheits and Contrast-Wiirfel" is described and illustrated in the form in which it is con- structed for technical use according to our designs by the Berlin firm, Franz Sehmidt and tlaensch. Whoever com- pares this photometer with Swan's prism-photometer will with an impartial judgment hardly agree with Prof. Knott when he exclaims, with refbrence to our photometer, "Let us now, who know its value, not forget that it is 'Swan's' photometer." LV. T/,ermal Radiation in Absolute Measttre. By J. T. BOTTO~LEY, .~I.A., D.Sc., F.R.8., and J. C. BEATTIE, D.Sc., F.R.S.E.r r|,H II r|,HE experiments:[: described in the fbllowing paper form a II continuation of researches on tbermal radiation by one of the present authors, the results of which have been com- municated to the Royal Society fi'om time to time since 1884 ~. The main object of the present experiments was to push ibrward the inquiry as to the amount, and the relative quality,

https://openalex.org/W2531198037

https://journals.plos.org/plosgenetics/article/file?id=10.1371/journal.pgen.1006193&type=printable

English

Epigenetics in Cancer: A Hematological Perspective

PLOS genetics

cc-by

OPEN ACCESS OPEN ACCESS Citation: Stahl M, Kohrman N, Gore SD, Kim TK, Zeidan AM, Prebet T (2016) Epigenetics in Cancer: A Hematological Perspective. PLoS Genet 12(10): e1006193. doi:10.1371/journal.pgen.1006193 Editor: David J. Kwiatkowski, Brigham and Women’s Hospital, UNITED STATES Published: October 10, 2016 Copyright: © 2016 Stahl et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Citation: Stahl M, Kohrman N, Gore SD, Kim TK, Zeidan AM, Prebet T (2016) Epigenetics in Cancer: A Hematological Perspective. PLoS Genet 12(10): e1006193. doi:10.1371/journal.pgen.1006193 Epigeneticsin Cancer: A Hematological Perspective Maximilian Stahl1, Nathan Kohrman1, Steven D. Gore1, Tae Kon Kim1, Amer M. Zeidan1, Thomas Prebet2* 1 Department of Internal Medicine, Section of Hematology, Yale School of Medicine, New Haven, Connecticut, United States of America, 2 Department of Internal Medicine, Section of Hematology, Yale Cancer Center at Yale University, New Haven, Connecticut, United States of America Abstract For several decades, we have known that epigenetic regulation is disrupted in cancer. Recently, an increasing body of data suggests epigenetics might be an intersection of cur- rent cancer research trends: next generation sequencing, immunology, metabolomics, and cell aging. The new emphasis on epigenetics is also related to the increasing produc- tion of drugs capable of interfering with epigenetic mechanisms and able to trigger clinical responses in even advanced phase patients. In this review, we will use myeloid malignan- cies as proof of concept examples of how epigenetic mechanisms can trigger or promote oncogenesis. We will also show how epigenetic mechanisms are related to genetic aberra- tions, and how they affect other systems, like immune response. Finally, we will show how we can try to influence the fate of cancer cells with epigenetic therapy. a1111 Mechanism of Epigenetics Epigenetics is defined as heritable changes in gene expression that are not due to any alteration in the DNA sequence [3,4]. Epigenetic modifications are placed by epigenetic writers and removed by erasers in a dynamic but highly regulated manner [5]. Many different DNA and histone modifications have been identified to determine the epigenetic landscape (Fig 1) [6–8]. The DNA (5-cytosine)-methytransferases(DNMTs) add methyl groups to cytosine in CpG dinucleotides in DNA and the TET family of proteins catalyze 5-methylcytosine to 5-hydroxy- methylcytosine [9–11]. Histone acetylation is associated with elevated transcription, while deacetylated histones are associated with gene repression. Acetylation removes the positive charge on the histones, Histone acetylation is associated with elevated transcription, while deacetylated histones are associated with gene repression. Acetylation removes the positive charge on the histones, which leads to a decrease in the interaction of the N termini of histones with the negatively charged phosphate groups of DNA. Subsequently condensed chromatin (heterochromatin) is transformed into a more relaxed structure (euchromatin), which leads to increased levels of gene transcription [6]. Histone acetylases and histone deacetylases (HDACs) add and remove acetyl groups from histones and are critical regulators of gene expression [12]. Methyltransferase Enhancer of Zeste Homologue 2 (EZH2) is an integral part of the poly- comb repressive complex 2 (PRC2), which maintains transcriptional silencing through post- translational histone modifications [11]. Transcriptional silencing is initiated by recruitment of PRC2, which, through EZH2, induces mono-,di-, and trimethylation of lysine 27 of histone H3 (H3K27). PRC1 recognizes H3K27me3 and mediates ubiquitylation (Ub) of lysine 119 of his- tone H2A (H2AK119), which is thought to lead to the recruitment of DNMTs to target loci and silencing of gene expression (Fig 1) [13]. Lastly, metabolic input is mediated by the isocitrate dehydrogenase enzyme (IDH), which catalyzes the conversion of isocitrate to alpha-ketogluatarate (α-KG) [14]. Dioxygenase enzymes, which include the TET family of enzymes and the Jumonji –C domain-containing (JMJC) family of histone lysine demethylases, are α-KG dependent enzymes (Fig 1) [15]. The application of new molecular techniques, namely, next generation sequencing (NGS) coupled with DNA methylation profiling as well as chromatin immunoprecipitation (ChIP-- Seq) [16–18] and epigenome editing technology based on CRISPR-Cas9 approaches [19,20], allowed researchers to characterize the impact of epigenetic modification not only on promo- tors but on the entire genome [21]. Typical patterns of histone modifications exhibited at pro- moters and regulatory domains (insulators, enhancers, repressors) have been identified [22]. In this review, we will present the current trends in epigenetic research encompassing the biology of epigenetics, interactions with other cancer mechanisms, and drug development. Research on myeloid malignancies will be used to illustrate these different topics. PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Introduction Over the past two decades, the connection between cancer and epigenetic regulation has been a promising venue for research. From the first evidence of the epigenetic silencing of tumor sup- pressor genes’ promotors, we now have a more complex and multidimensional picture, inte- grating several layers of (dys)regulated DNA methylation, histone modification,and micro RNA modulation. Maybe most important, epigenetic regulation has emerged as an intersection of several key hallmarks of cancer such as immunology, metabolism, or aging [1,2]. Editor: David J. Kwiatkowski, Brigham and Women’s Hospital, UNITED STATES Published: October 10, 2016 Copyright: © 2016 Stahl et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Editor: David J. Kwiatkowski, Brigham and Women’s Hospital, UNITED STATES Copyright: © 2016 Stahl et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Many of these discoveries were initially describedin context of hematological malignancies, and, acknowledging significant exceptions, their counterparts in solid tumors have not been so easy to demonstrate. Similarly, the benefit of epigenetic targeting has been identified in myelo- dysplastic syndromes and acute myeloid leukemias with the use of DNA hypomethylating agents and, to a lesser extent, histone deacetylase inhibitors. As our tools to study epigenetics progressed, our arsenal of epigenetic-targeted drugs started to expand. Hematology is at the cutting edge of research on the development of drugs targeting epigenetic regulators, including DOT1L, BET proteins, LSD1, and IDH1/2 inhibitors. Funding: The authors received no specific funding for this work. Competing Interests: The authors have declared that no competing interests exist. 1 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Fig 1. Epigenetics in hematological malignancies. Epigenetic regulation, dysregulation and therapeutic targets. DNA (red) forms a Fig 1. Epigenetics in hematological malignancies. Epigenetic regulation, dysregulation and therapeutic targets. DNA (red) forms a complex with histone proteins (light blue) to form nucleosomes. Each nucleosome consists of DNA wrapped around a unit of eight histone proteins. Epigenetic regulation: Epigenetic marks are placed both on the DNA and histones by epigenetic writers (in green). DNMT3A, TET2, EZH1/2, and the histone acetylase are examples of epigenetic writers. EZH2 is part of the Polycomb Repressive Complex 2 (PRC2), which also contains ASXL1, EED, SUZ12, and RBAP48. Epigenetic marks are removed by epigenetic erasers (in red), e.g., Lysine specific demythylases (LSD) and histone deacetylases (HDAC). IDH1/2 provide metabolic input by providing a-KG, which is an important substrate for the catalytic domain of other epigenetic regulators like TET2. Finally, epigenetic marks are recognized by epigenetic regulators though special reader domains (in blue), which lead to the recruitment of epigenetic regulators to DNA and histones. Examples of reader domains are the plant homeodomain (PHD) finger proteins and the bromodomain and extraterminal (BET) family of proteins. The BET family has four members, including bromodomain-containing proteins 2, 3, and 4 (BRD2, BRD3, and BRD4) and BRDT. The wild-type mixed-lineage leukemia (MLL) gene is post-translationally cleaved into N-terminal and C-terminal fragments that re-associate to form the MLL complex. The C-terminal fragment contains a SET domain, which methylates H3K4, and the N-terminal fragment contains PHD fingers and a bromodomain, which serve as reader domains. Epigenetic dysregulation: The complex epigenetic regulatory program is disturbed in hematologic malignancies by mutations in epigenetic regulators (indicated by red thunderbolt) or by the recruitment of large multi-protein complexes like the MLL fusion complex (purple circle). Translocations involving the MLL gene account for the vast majority of infantile and approximately 10% of adult leukemias. Following translocation, a fragment of the N-terminal portion of MLL is fused in frame with a translocation partner, leading to the formation of novel MLL- fusion protein complexes, including the super elongation complex (SEC) and the DOT1-Like Histone H3K79 Methyltransferase (DOT1L) complex. The DOT1L complex leads to misdirected H3K79 methylation, which has been shown to sustain the expression of key pro-leukemic genes such as the HOXA genes and MEIS1. The SEC complex phosphorylates RNA polymerase II (POL II) facilitating its recruitment to the promoters of crucial oncogenes such as MYC, BCL2, and CDK6. Metabolic dysregulation is caused by IDH1/2 mutations, which leads to the production of an abnormal metabolite in the cell, 2-hydroxyglutarate (2HG), which can inhibit the hydroxylation of 5-mC by TET2. Therapeutic targeting: Food and Drug Administration (FDA)-approved are DNMT3A inhibitors for AML and MDS and HDAC inhibitors for T cell lymphoma and multiple myeloma, respectively (in orange). Several investigational drugs (in yellow) are in different stages of preclinical and clinical development. Adopted from Semin Hematol. 2015 Jul;52(3):172–83 [8] Fig 1. Epigenetics in hematological malignancies. Epigenetic regulation, dysregulation and therapeutic targets. DNA (red) forms a complex with histone proteins (light blue) to form nucleosomes. Each nucleosome consists of DNA wrapped around a unit of eight histone proteins. Epigenetic regulation: Epigenetic marks are placed both on the DNA and histones by epigenetic writers (in green). DNMT3A, TET2, EZH1/2, and the histone acetylase are examples of epigenetic writers. EZH2 is part of the Polycomb Repressive Complex 2 (PRC2), which also contains ASXL1, EED, SUZ12, and RBAP48. Epigenetic marks are removed by epigenetic erasers (in red), e.g., Lysine specific demythylases (LSD) and histone deacetylases (HDAC). IDH1/2 provide metabolic input by providing a-KG, which is an important substrate for the catalytic domain of other epigenetic regulators like TET2. Finally, epigenetic marks are recognized by epigenetic regulators though special reader domains (in blue), which lead to the recruitment of epigenetic regulators to DNA and histones. Examples of reader domains are the plant homeodomain (PHD) finger proteins and the bromodomain and extraterminal (BET) family of proteins. The BET family has four members, including bromodomain-containing proteins 2, 3, and 4 (BRD2, BRD3, and BRD4) and BRDT. The wild-type mixed-lineage leukemia (MLL) gene is post-translationally cleaved into N-terminal and C-terminal fragments that re-associate to form the MLL complex. The C-terminal fragment contains a SET domain, which methylates H3K4, and the N-terminal fragment contains PHD fingers and a bromodomain, which serve as reader domains. Epigenetic dysregulation: The complex epigenetic regulatory program is disturbed in hematologic malignancies by mutations in epigenetic regulators (indicated by red thunderbolt) or by the recruitment of large multi-protein complexes like the MLL fusion complex (purple circle). Translocations involving the MLL gene account for the vast majority of infantile and approximately 10% of adult leukemias. Following translocation, a fragment of the N-terminal portion of MLL is fused in frame with a translocation partner, leading to the formation of novel MLL- fusion protein complexes, including the super elongation complex (SEC) and the DOT1-Like Histone H3K79 Methyltransferase (DOT1L) complex. The DOT1L complex leads to misdirected H3K79 methylation, which has been shown to sustain the expression of key pro-leukemic genes such as the HOXA genes and MEIS1. The SEC complex phosphorylates RNA polymerase II (POL II) facilitating its recruitment to the promoters of crucial oncogenes such as MYC, BCL2, and CDK6. Metabolic dysregulation is caused by IDH1/2 mutations, which leads to the production of an abnormal metabolite in the cell, 2-hydroxyglutarate (2HG), which can inhibit the hydroxylation of 5-mC by TET2. Therapeutic targeting: Food and Drug Administration (FDA)-approved are DNMT3A inhibitors for AML and MDS and HDAC inhibitors for T cell lymphoma and multiple myeloma, respectively (in orange). Several investigational drugs (in yellow) are in different stages of preclinical and clinical development. Adopted from Semin Hematol. 2015 Jul;52(3):172–83 [8] doi:10.1371/journal.pgen.1006193.g001 Mechanism of Epigenetics Many chromatin regulators also survey the epigenetic landscape using specializeddomains to dock at specific domains within the genome, leading to recruitment of functional complexes regulating DNA transcription [5,23]. Many “writers” and “erasers” possess this chromatin “reader” ability in addition to their catalytic activity and respond to information conveyed by upstream signaling cascades. These regulators use complex three-dimensional binding pockets (e.g., bromodomain, PHD finger), which allow readers with similar binding domains to dock at different modifiedresidues or at the same amino acid displaying a different modification state (Fig 1) [24]. The multifaceted mechanism that chromatin readers use to decipher the epi- genetic landscape is exemplified by the fact that many readers have more than one reader domain, and binding to chromatin is influenced by neighboring histone modifications [5,25]. Single-cell epigenetic profiling will further promote our understanding of epigenetic regulation by addressing the issue of epigenetic heterogeneity of cancer [26]. 2 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 doi:10.1371/journal.pgen.1006193.g001 of genomic DNA increases as the tumor progresses from a benign proliferation of cells to an invasive cancer [32,33]. Global hypomethylation promotes tumorigenesis by the generation of chromosomal insta- bility (promoting chromosomal deletions and rearrangements), reactivation of transposable elements (further disrupting the genome), and loss of imprinting [34–39]. Furthermore, gene inactivation through hypermethylation of the CpG islands in the promoter region has been identified for many tumor suppressor genes, including the retinoblastoma tumor-suppressor gene (Rb), the von Hippel-Lindau tumor-suppressor gene (VHL), p16INK4a, the breast-cancer susceptibility gene 1 (BRCA1), and the MutL homolog 1 gene (hMLH1) [4,28,40–43]. Profiles of hypermethylation of the CpG islands in tumor-suppressor genes are specific to the cancer type so that each tumor can be assigned a specific,defining DNA “hypermethylome.” [44–46]. In acute myelogenous leukemia (AML), large-scale, genome-wide DNA methylation profiling reveals the existence of distinct DNA methylation patterns and identifies novel, biologically, and clinically relevant defined AML subgroups [47]. For example, the function of the basic leu- cine zipper transcription factor CCAAT/enhancer binding protein-α (C/EBPα), one of the cru- cial transcription factors for myeloid cell development, is frequently abrogated in AML by mutations but also through epigenetic modification through hypermethylation of the CEBPA promoter [48–52]. Furthermore, hypermethylation of CCCTC-binding factor (CTCF) sites has been shown to disrupt the function of insulators, which separate different genomic loops from each other [53–55]. In IDH mutated gliomas, this mechanism leads to the close interaction of FIP1L1 gene and Platelet-Derived Growth Factor Receptor, Alpha Polypeptide (PDGFRA) gene, which are normally confined to separate loop domains [53]. This allows the constitutive enhancer FIP1L1 to interact aberrantly with PDGFRA, a prominent glioma oncogene. This has not been so far demonstrated in hematological malignancies. Importantly, epigenetic integrity itself can be disrupted in two different ways. Epigenetic regulators can be directly mutated, or they can be epigenetically modified,leading to a positive feedback and a drift from a tightly regulated epigenetic set point. This leads to a growth advan- tage of cancer cells [56]. In most solid tumors, epigenetic mutations are rather rare; they are mainly found in hematologic malignancies, rare childhood cancers, and highly aggressive solid tumors like glioblastoma multiforme [57]. Much of what we know about the epigenetic dysre- gulation in cancer has been elucidated by studying hematologic malignancies, because most direct epigenetic mutations (both in epigenetic writers/erasers and writers) are found in hema- tologic cancers. Role in Cancer Epigenetic dysregulation manifests in cancer with global DNA hypomethylation, causing geno- mic instability as well as silencing of specific tumor suppressor genes and of microRNA (miRNA) genes by hypermethylation [4,27–31]. Large-scale studies of DNA methylation have detected extensive hypomethylated genomic regions in gene-poor areas in cancer cells and demonstrated that the degree of hypomethylation 3 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 TET2 [65] • AML: 8%–23% • 18%–23% (CN-AML) • 19%–24.5% (AML > 60y) • 7%–10% (AML < 60y) • MDS: 20%–25% • MPN: 4%-13% TET2 possesses DNA dioxygenase activity, which leads to the conversion of the methyl group at the 5-position of cytosine of DNA 5- methylcytosine [5mC] to 5-hydroxy-methylcytosine [5hmC]. TET2 enzymes are dependent on Fe(II) and α-ketoglutarate (α-KG). Adverse risk in patients with CN-AML independent from FLT3-ITD mutational status [63,66]; no clear prognostic importance in MDS and MPN [67,87]. Mutations in histone modifying enzymes EZH2 [59,68,69] • AML: Rare • MDS: 6%–7% • MPN: 3%–13% EZH2 is the catalytic subunit of the PcG Repressor Complex 2 (PRC2), a highly conserved Histone H3K27 methyltransferase. EZH2 mutations have a complex role, as they result both in gain and loss of function. EZH2 may serve a dual purpose as an oncogene and tumor-suppressor gene. [69]. Biological effects of mutations unclear as EZH2 conditional knockout leads to minimal myeloid haematopoietic defects [11]. Adverse risk in all studies to date (AML, MDS, and MPD) [11,68]. ASXL1 [11,70] • AML: 6%–30% • 16.2%–25% (AML > 60y) • 3%–6.8% (AML < 60y) • MDS: 14% • MPN: 2%–23% Unclear whether ASXL1 mutations confer a loss or gain of function. Their role in mammalian haematopoietic-specific context is not known [74,75]. Adverse risk in patients with CN-AML, intermediate risk AML [88], and MDS [87]. Significantly associated with RUNX1 and NPM1 mutations. (Continued) Mutations in histone modifying enzymes EZH2 [59,68,69] • AML: Rare • MDS: 6%–7% • MPN: 3%–13% EZH2 is the catalytic subunit of the PcG Repressor Complex 2 (PRC2), a highly conserved Histone H3K27 methyltransferase. EZH2 mutations have a complex role, as they result both in gain and loss of function. EZH2 may serve a dual purpose as an oncogene and tumor-suppressor gene. [69]. Biological effects of mutations unclear as EZH2 conditional knockout leads to minimal myeloid haematopoietic defects [11]. Adverse risk in all studies to date (AML, MDS, and MPD) [11,68]. ASXL1 [11,70] • AML: 6%–30% • 16.2%–25% (AML > 60y) • 3%–6.8% (AML < 60y) • MDS: 14% • MPN: 2%–23% Unclear whether ASXL1 mutations confer a loss or gain of function. Their role in mammalian haematopoietic-specific context is not known [74,75]. Adverse risk in patients with CN-AML, intermediate risk AML [88], and MDS [87]. Significantly associated with RUNX1 and NPM1 mutations. (Continued) Table 1. Mutations in epigenetic regulators in myeloid malignancies. Mutations Involving Epigenetic Writers/Erasers Many mutations in epigenetic regulators have been described[10,11,58–60] (see Table 1 for detailed review of mutational frequency, mechanism, and prognostic relevance of these muta- tions). Mutations in regulators of DNA methylation/hydoxymethylation are found in the DNA (5-cytosine)-methytransferase 3A (DNMT3A) [11,61–64] and the TET family of proteins (Fig 1) [63,65–68]. Mutations affecting histone modification are found in the Methyltransferase Enhancer of Zeste Homologue 2 (EZH2) [59,68,69] and the additional sex combs such as 1 transcriptional regulator (ASXL1) [11,70]. Apart from playing a role in myeloid malignancies [68,69], EZH2 mutations (at codon 641) have been found to be common in follicular and diffuse large B-cell lymphomas of germinal center origin and are a promising target in these lymphomas [71–73]. The role of ASXL1 mutations in myeloid malignancies is less well understood. ASXL1 is not thought to possess enzymatic activity [74] but may be important for the recruitment of EZH2 4 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Gene Mutational frequency in myeloid malignancies Mechanism Impact on outcome Mutations in DNA modifying enzymes DNMT3A [61] • AML: 4%–22% • up to 36% (CN-AML) • 16%–22% (AML > 60y) • 17.8%–23% (AML < 60y) • high dose: -68.3% • MDS: 8% • MPN: 7%–15% DNMT3A possesses DNA methyltransferase activity, which leads to the addition of a methyl group at the 5-position of cytosine of DNA 5-methylcytosine [5mC]. DNMT3A mutations result either in premature truncation of the protein product (nonsense or frameshift mutations), or occur at a single amino acid, R882 (60% of mutations). In most cases, one DNMT3A allele remains wild-type, as haploinsufficiency seems sufficient to contribute to myeloid transformation. Adverse risk in patients with CN-AML and FLT3-ITD mutations [62]. Improved outcome with high dose daunorubicin [63]. Single study showed adverse prognosis of DNMT3A mutations in MDS [64]. There is no known prognostic importance, if any, in patients with MPN [11,67]. TET2 [65] • AML: 8%–23% • 18%–23% (CN-AML) • 19%–24.5% (AML > 60y) • 7%–10% (AML < 60y) • MDS: 20%–25% • MPN: 4%-13% TET2 possesses DNA dioxygenase activity, which leads to the conversion of the methyl group at the 5-position of cytosine of DNA 5- methylcytosine [5mC] to 5-hydroxy-methylcytosine [5hmC]. TET2 enzymes are dependent on Fe(II) and α-ketoglutarate (α-KG). Adverse risk in patients with CN-AML independent from FLT3-ITD mutational status [63,66]; no clear prognostic importance in MDS and MPN [67,87]. Mutations in histone modifying enzymes EZH2 [59,68,69] • AML: Rare • MDS: 6%–7% • MPN: 3%–13% EZH2 is the catalytic subunit of the PcG Repressor Complex 2 (PRC2), a highly conserved Histone H3K27 methyltransferase. EZH2 mutations have a complex role, as they result both in gain and loss of function. EZH2 may serve a dual purpose as an oncogene and tumor-suppressor gene. [69]. Biological effects of mutations unclear as EZH2 conditional knockout leads to minimal myeloid haematopoietic defects [11]. Adverse risk in all studies to date (AML, MDS, and MPD) [11,68]. ASXL1 [11,70] • AML: 6%–30% • 16.2%–25% (AML > 60y) • 3%–6.8% (AML < 60y) • MDS: 14% • MPN: 2%–23% Unclear whether ASXL1 mutations confer a loss or gain of function. Their role in mammalian haematopoietic-specific context is not known [74,75]. Adverse risk in patients with CN-AML, intermediate risk AML [88], and MDS [87]. Significantly associated with RUNX1 and NPM1 mutations. (Continued) PLOS Genetics | DOI 10 1371/jo rnal pgen 1006193 October 10 2016 Impact on outcome Table 1. Mutations in epigenetic regulators in myeloid malignancies. PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 5 / 21 and the stability of the PRC2 complex as demonstrated in co-immunoprecipitation experi- ments (Fig 1) [75]. Mutations in IDH have been discovered first in glioblastoma and then in AML (Fig 1) [14,76–78]. IDH2 mutations at the active enzyme site at position R172 and R140 confer a gain- of-function and result in a neomorphic enzymatic activity of the mutated IDH enzyme: mutant IDH1/2 catalyzes the conversion of alpha ketoglutarate to beta-hydroxyglutarate (2-HG) [79]. Supra-normal levels of intracellular 2-HG lead to competitive inhibition of α -KG dependent epigenetic regulators like TET2 and subsequently to hypermethylation of DNA as well as his- tones and a blockade of cellular differentiation [80,81]. Importantly, IDH and TET2 seem to be almost entirely mutually exclusive, supporting the common mechanism of action of both mutations [82]. The discovery of IDH mutations has led to the concept that “oncometabolites” like 2-HG play a major role in tumorigenesis, further underscored by the interaction of epigenetics and metabolomics in cancer. Apart from being mutated, epigenetic writers and erasers can be aberrantly recruited by fusion proteins, which are formed by chromosomal translocation. The fusion proteins PML-RARa and AML1-ETO found in patients with t(15;17) and t(8;21) AML are the two most prominent examples. Both fusion proteins recruit multiprotein complexes including both HDACs and DNMTs to alter transcription, repress differentiation genes, and drive leu- kemogenesis [83–85]. Recently, the ecotropic viral integration site 1 (EVI1), a DNA binding zinc-finger transcription factor, has been shown to direct a unique recurrent DNA methyla- tion signature in AML by specifically recruiting DNMTs and HDACs to target promoters [86]. Table 1. (Continued) Gene Mutational frequency in myeloid malignancies Mechanism Impact on outcome Mutations in enzymes regulating metabolic input IDH1/2 [14,76,77,89] • 5%–30% (all AML) • IDH-1 (mutations at Arg132) • 10%–16% (CN-AML) • 9.6%–14% (AML > 60y) • 7%–10.9% (AML < 60y) • IDH-2 (mutations at Arg140 or Arg172) • 10%–19% (CN-AML) • 8%–19% (AML > 60y) • 8%–12.1% (AML < 60y) IDH converts isocitrate to α-KG, which is essential for TET2 function and mutated IDH has neomorphic enzymatic activity, which converts α-KG to 2-HG (“oncometabolite”). IDH1/2 mutations share a mutual exclusivity with TET2 mutations. IDH1/2 mutations are significantly associated with NPM1 mutations. Conflicting studies about the prognostic relevance of IDH mutations [79]. doi:10.1371/journal.pgen.1006193.t001 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 6 / 21 Table 1. (Continued) Gene Mutational frequency in myeloid malignancies Mechanism Impact on outcome Mutations in enzymes regulating metabolic input IDH1/2 [14,76,77,89] • 5%–30% (all AML) • IDH-1 (mutations at Arg132) • 10%–16% (CN-AML) • 9.6%–14% (AML > 60y) • 7%–10.9% (AML < 60y) • IDH-2 (mutations at Arg140 or Arg172) • 10%–19% (CN-AML) • 8%–19% (AML > 60y) • 8%–12.1% (AML < 60y) IDH converts isocitrate to α-KG, which is essential for TET2 function and mutated IDH has neomorphic enzymatic activity, which converts α-KG to 2-HG (“oncometabolite”). IDH1/2 mutations share a mutual exclusivity with TET2 mutations. IDH1/2 mutations are significantly associated with NPM1 mutations. Conflicting studies about the prognostic relevance of IDH mutations [79]. 6 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Indirect Effects of Epigenetic Enzyme Mutations- Impact on Epigenetic Reader Domains Rearrangement of the Histone-lysine N-methyltransferase 2A/mixed-lineage leukemia gene (KTM2A/MLL1) is found in approximately 5% of ALL cases and around 5% to 10% of AML cases in adults. This rearrangement results in aggressive leukemia with poor prognosis and is often refractory to conventional therapies [90,91]. Central to each of the translocations seems to be abnormal transcriptional elongation involving abnormal recruitment of histone reader proteins [5,92]. Many of the common translocation partners of MLL (including AF9, ENL, AF4, and ELL) are critical members of the super elongation complex, which contains the posi- tive transcription elongation factor b (SEC-P-TEFb complex) (Fig 1) [93–95]. The SEC-P- TEFb complex phosphorylates RNA polymerase II, facilitating transcriptional elongation, lead- ing to transcription of crucial oncogenes like myc and bcl-2. The SEC-P-TEFb uses “reader domains” in the form of bromodomain and extraterminal proteins (BET proteins) in order to bind to acetylated histones on chromatin. BET reader pro- teins can be targeted by small molecule inhibitors (BETi) (Fig 1) [96–98]. In addition to the ability of MLL1 fusion proteins to recruit transcriptional machinery such as the SEC, MLL1 rearrangement promotes gene expression by elevating local H3K79me2 levels [99,100]. The only known enzyme in mammals that catalyzes methylation of H3K79 is DOT1L (disruptor of telomeric silencing 1-like), and the MLL1 fusion proteins may directly recruit DOT1L to MLL1 fusion target loci, leading to activation of homeobox A (HOXA) cluster genes, which induce leu- kemic transformation of hematopoietic progenitors. Their high expression is a hallmark in MLL1 rearranged leukemias [101–103]. Currently, there are several DOT1L inhibitors in devel- opment (Fig 1) [100,104,105]. 7 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Fig 2. Epigenetic dysregulation as a hallmark of cancer. Overlay graphic synthesizing concepts from “Hallmarks of cancer: the next generation” by Hannahan and Weinberg [1] and “A panoramic view of acute myeloid leukemia [106]. Mutational frequency of different functional classes of mutations in parenthesis. Indirect Effects of Epigenetic Enzyme Mutations- Impact on Epigenetic Reader Domains Abbreviations: FLT-3: Fms-like tyrosine kinase 3; KRAS: Kirsten rat sarcoma viral oncogene homolog; NRAS: Neuroblastoma rat sarcoma viral oncogene homolog; TF: Transcription factor; RUNX1: Runt-related transcription factor 1; CEBPa: CCAAT/ enhancer-binding protein alpha; PML-RARA: fusion of the promyelocytic leukemia (PML) gene on chromosome 15 to the retinoic acid receptor (RAR) gene on chromosome 17; CBFB-MYH11: chromosomal rearrangements involving the core-binding factor, beta subunit (CBFB) gene on chromosome 16p13.1 and the Myosin, heavy chain 11, smooth muscle (MYH11) gene on chromosome 16q22; RUNX1-RUNX1T1: chromosomal rearrangements involving the RUNX1 gene on chromosome 21 and the Runt-related transcription factor 1 translocated to 1 (RUNX1T1) gene on chromosome 8. TP53: Tumor protein 53; WT1: Wilms tumor protein 1; PHF6: PHD Finger Protein 6; NPM1: Nucleophosmin 1; ARF: ADP-ribosylation factor 1; TAA: Tumor associated antigen; MHC I/II: Major histocompatibility complex I/II; SAM: S-Adenosyl Methionine. Fig 2. Epigenetic dysregulation as a hallmark of cancer. Overlay graphic synthesizing concepts from “Hallmarks of cancer: the next generation” by Hannahan and Weinberg [1] and “A panoramic view of acute myeloid leukemia [106]. Mutational frequency of different functional classes of mutations in parenthesis. Abbreviations: FLT-3: Fms-like tyrosine kinase 3; KRAS: Kirsten rat sarcoma viral oncogene homolog; NRAS: Neuroblastoma rat sarcoma viral oncogene homolog; TF: Transcription factor; RUNX1: Runt-related transcription factor 1; CEBPa: CCAAT/ enhancer-binding protein alpha; PML-RARA: fusion of the promyelocytic leukemia (PML) gene on chromosome 15 to the retinoic acid receptor (RAR) gene on chromosome 17; CBFB-MYH11: chromosomal rearrangements involving the core-binding factor, beta subunit (CBFB) gene on chromosome 16p13.1 and the Myosin, heavy chain 11, smooth muscle (MYH11) gene on chromosome 16q22; RUNX1-RUNX1T1: chromosomal rearrangements involving the RUNX1 gene on chromosome 21 and the Runt-related transcription factor 1 translocated to 1 (RUNX1T1) gene on chromosome 8. TP53: Tumor protein 53; WT1: Wilms tumor protein 1; PHF6: PHD Finger Protein 6; NPM1: Nucleophosmin 1; ARF: ADP-ribosylation factor 1; TAA: Tumor associated antigen; MHC I/II: Major histocompatibility complex I/II; SAM: S-Adenosyl Methionine. doi:10.1371/journal.pgen.1006193.g002 signaling), and NPM1 (tumor suppressor). DNMT3A mutations may occur early in leukemo- genesis and cause genetic instability, which is prone to FLT3, NPM1 mutations. On the other hand, mutual exclusivity exists among transcription factor fusion genes, NPM1, RUNX1, TP53, and CEBPA. New research focuses on the intersection of cancer epigenetics and the newly characterized cancer hallmarks of cancer immunology [115], metabolism [116,117], and alternative m-RNA processing/splicing. Immunosuppressive microenvironment and epigenetic alterations are known to silence/downregulateall steps of antigen processing and presentation machinery (APM) in cancer cells, including tumor-associated antigens, human leukocyte antigens, and accessory/co-stimulatorymolecules [118]. Epigenetic drugs have shown to up-regulate all the elements in the antigen presenting machinery, e.g., the expression of tumor associated anti- gens (TAA), MHC I and MHC II molecules as well as co-stimulatory surface markers like CD40, CD80 and ICAM1 [115,119–122]. There is significant evidence based on preclinical in vitro and in vivo models supporting combination therapy using epigenetic modulators and immunotherapy. Prime examples of the interaction of the epigenome with metabolism are mutations in IDH/IDH2, which lead to accumulation of the oncometabolite 2-HG as well as acetyl CoA and S-adenosylmethionine(SAM), connecting nutritional status to gene expression through their role as donors/coenzymes for histone acetylation and DNA/histone methylation, respectively [123–125]. Mutations in splicing factors are observedin up to 85% of myeloid neoplasms with myelo- dysplastic features [126]. Chromatin structure and epigenetic histone modifications may act as key regulators of alternative splicing [127]. Histone marks are non-randomly distributed in the genome and are enriched specificallyin exons relative to their flanking intronic regions [128]. There might be direct physical crosstalk between chromatin and the splicing machinery via an adaptor complex [129]. Importantly, each splice gene mutation seems to be associated with one concomitant mutation in a gene involved in epigenetic regulation of transcription. SF3B1, SRSF2 /ZRSR2, and U2AF35 mutations are enriched in patients with DNMT3A, TET2, and ASXL1, respectively [130]. On the other hand, mutations in the splicing factor genes U2AF1 and SRSF2 cause dysfunctional processing of pre-mRNA and reduced EZH2 expression [69]. The cohesin complex is important in mediating proper sister chromatid cohesion and separation from S phase to M phase in mitosis as well as in regulating transcription through genome-wide chromatin organization. Mutations of proteins of the complex are frequently found in myeloid neoplasms [131]; they collectively occur in approximately 15% of AML cases and other myeloid malignancies [132]. Leukemia-associatedcohesion mutations have been found to impair differentiation and enforce stem cell programs in human stem and progenitor cells by demonstrating increased chromatin accessibility of stem cell regulators like the Runt- related transcription factor 1 (RUNX1) and GATA2 [133,134]. Interaction of Epigenetics with Other Hallmarks of Cancer Recurrent genetic alterations in AML can be functionally categorized matching the hallmarks of cancer describedby Hanahan and Weinberg [1,106]. Disruption in epigenetic regulation has been found to collaborate with these hallmarks in cancer development in many ways [56,107– 114]. Fig 2 shows an overlay of the proposed cancer hallmarks and the recurrent genetic muta- tions of AML with epigenetic dysregulation at the center of these complex interactions. Certain mutations seem to collaborate while others are mutually exclusive. For example, there is a strong association between mutations in the epigenetic regulator DNMT3A, FLT3 (activating 8 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Isoform Specific HDAC Inhibitors Current research focuses on developing specific therapy by using isoform-specificHDACi [141]. For example, the class I HDAC inhibitor entinostat was recently awarded by the FDA a breakthrough therapy status for patients with metastatic, estrogen receptor-positive breast can- cer based on data from the phase II ENCORE 301 study (NCT00676663) [142]. Current Therapies To date, epigenetic therapies have been limited to targeting epigenetic writers in the form of DNA methyltransferase inhibitors (DNMTi) and epigenetic erasers in the form of histone dea- cetylase inhibitors (HDACi) (Fig 1). The Food and Drug Administation (FDA) approved two DNMTi (azacitidine and decitabine) for the treatment of MDS [135] and several HDACi (vori- nostat, romidepsin, belinostat, and panobinostat) for the treatment of cutaneous T-cell lym- phoma and multiple myeloma [136–140], respectively. Due to their pleitropic effects, it has been difficultto confirm the mechanism of action of DNMTi and HDACi. PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 9 / 21 Table 2. Selection of ongoing clinical trials evaluating epigenetic targeted therapies in hematologic malignancies. Clinical Trial Intervention Malignancy studied EZH2 inhibitors NCT02395601 Phase 1 Study: EZH2 inhibitor CPI-1205 Progressive B-cell lymphomas NCT01897571 Phase 1/2 Study: EZH2 inhibitor E7438 B-cell lymphomas and advanced solid tumors KDM inhibitors NCT02261779 Phase 1/2 Study: ATRA + tranylcypromine (TCP) an irreversible monoamine-oxidase (MAO) and Lysin-specific demethylase (LSD) inhibitor Relapsed/refractory AML IDH2 inhibitors NCT01915498 Phase 1/2 Study: reversible inhibitor of mutant IDH2 AG-221 Advanced hematologic malignancies with IDH2 mutation NCT02273739 Phase 1/2 Study: reversible inhibitor of mutant IDH2 AG-221 Advanced solid tumors (glioma) and angioimmunoblastic T-cell lymphoma BET1/DOT1L inhibitors NCT01943851 Phase 1/2 Study: BET inhibitor GSK525762 Relapsed/refractory hematologic malignancies (leukemias, myeloproliferative neoplasms, lymphomas, and myelomas) NCT02158858 Phase 1 Study: BET inhibitor CPI-0610 AML, myelodysplastic syndromse, myeloproliferative neoplasms, myelofibrosis NCT02308761 Phase 1 Study: BET inhibitor TEN-010 AML, myelodysplastic syndromse NCT01684150 Phase 1 Study: second generation DOT1L inhibitor EPZ-5676 AML/ALL/MLL with MLL1 rearrangements (including 11q23 or partial tandem duplications) in adult patients NCT02141828 Phase 1 Study: second generation DOT1L inhibitor EPZ-5676 AML/ALL with MLL1 rearrangements (including 11q23 or partial tandem duplications) in pediatric patients Combination treatment with cancer vaccines NCT01483274 Phase 1 study: Decitabine + donor lymphocyte infusion + Vaccine (autologous dendritic cells) AML with relapse after allogeneic stem cell transplantation Combination treatment with immune checkpoint inhibitors NCT02281084 Phase 2 Study: Durvalumab (PD-L1 inhibitor) + CC-486 (oral azacitidine) Myelodysplastic syndromes NCT02530463 Phase 2 Study: Nivolumab (PD-1 inhibitor) and/or Ipilimumab (CTLA-4 inhibitor) + azacitidine Myelodysplastic syndromes This list is not complete but presents a selection of clinical trials by the authors of this manuscript meant to illustrate the different strategies. d i 10 1371/j l 1006193 t002 Combination Strategies In combining epigenetic agents with cytotoxic chemotherapy, the reactivation of tumor-sup- pressor genes and restoration of DNA-repair pathways by epigenetic drugs results in more chemo-sensitive cells. These cells can then be targeted by another type of therapy [151]. Initial studies combining epigenetic agents with chemotherapy showed disappointing results [152], though further studies suggest that the timing of epigenetic therapy matters and that it might be able to reverse resistance to chemotherapy [153–156]. Initial approaches focused on combining epigenetic agents with cytokine-basedimmuno- therapy and vaccination with tumor cells or peptide vaccines [118,157–159]. With the dawn of the checkpoint inhibitors CTLA-4 and PD-1/ PDL-1 to stimulate the immune system in solid malignancies, the combination of immune checkpoint inhibitors with epigenetic therapy has been promising in preclinical models [160–162]. Several ongoing phase-I/II clinical trials are dedicated to investigating the effect of combining epigenetic agents with immunotherapy (see Table 2) [115,161]. (see Table 2) [146,147]. Other compounds like pan IDH and IDH1 inhibitors (AG-120) are in development. Exciting new data also comes from drugs developed to target leukemias harboring MLL1 translocations, BET inhibitors, and DOT1L inhibitors (see Table 2) [5,100,104]. The MLL fusion protein can aberrantly recruit multiprotein complexes including SEC and DOT1L, acti- vating important oncogenic genes like HOXA cluster genes, c-myc, bcl-2, and others, but can be interrupted by targeting the reader proteins BET within SEC or DOT1L directly. Interest- ingly, BET inhibitors and DOT1L inhibitors are also effective in vitro in a variety of other leu- kemias [148] and hematologic malignancies such as multiple myeloma or Burkitt lymphoma [149,150], for which HOXA or c-myc activation are key drivers of the disease. Novel Epigenetic-Targeted Pharmacologic Agents Several new agents targeting epigenetic writers and erasers are in development, including EZH2 inhibitors [73,143], protein methyltransferase inhibitors (PMT inhibitors) [144], and histone lysine demethylases (KDM inhibitors) [145]. Several phase I/II clinical trials will be dedicated to studying the effect of these new agents in patients (see Table 2). IDH inhibitors seem to be particularly promising [77,89,144]. Early results of AG-221, an inhibitor of mutant IDH2, showed that from 48 patients with advanced AML/MDS with an IDH2 mutation, 20 patients had evidence of an objective response (eight complete remissions) PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 10 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 cytotoxic chemotherapy and small molecule tyrosine kinase inhibitors [58,73,104,143]. Second, can we develop biomarkers to predict response to epigenetic therapy? CpG island methylation signatures have only been mildly successful in predicting therapy response [56,165,166]. Fur- thermore, as is the case for the majority of tumors sensitive to BET and DOT1L inhibition, epi- genetic therapeutic targets are not necessarily mutated in sensitive tumor types. And although a central theme of BET inhibition seems to be c-myc downregulation, there are multiple cancer cell lines that overexpress c-myc but do not respond to BETi [5,150], as c-myc downregulation does not predict a response [96,98]. Therefore, simple mutational screening or gene expression profiling may not provide a predictor of response and might require large drug screening studies to test sensitivities [167,168]. Last, but not least, will there be a role for chemoprevention similar to using statins in heart disease? As epigenetics is recognizedas a very early driver for cancer progression, this holds promise for both improved early diagnosis and therapy of cancer [56]. Will it be possible in the future to identify patients early in the course of their disease and treat even before development of overt cancer based on their epigenetic profile? There is some evi- dence that this might be possible in colorectal and cervicalcancer [169,170]. As single mutations in epigenetic regulators cannot induce cancer on their own, single epigenetic agents will only be part of the cure. AML and cancer in general is a multi-step process, and targeting a single defect (as has been seen with Flt-3 inhibitors) is not sufficient to control cancer. In that context, one of the most promising approaches is using epigenetic therapy in combination with other therapies targeting different hallmarks of cancer, including traditional chemo- and radiation therapy as well as immunotherapy, which is currently changing the paradigm of therapy in solid malignan- cies [171]. Although most data are generated in vitro and in mouse models, there is evidence that the use of epigenetic drugs improves the antitumor activity of immune checkpoint inhibi- tors. As the overall effectiveness of immunotherapy is still far from optimal—only a minority of treated patients achieve long-term clinical benefit [172]—and there are poorly immunogenic tumors like AML, epigenetic therapy could serve as an essential part of future combination immune therapy [118]. There might be two sides to the coin in terms of pleiotropic effects of epigenetic agents: initially viewed as a weakness, it might prove to be an advantage in the light of combination therapy. Understanding the impact that epigenetics has on cancer biology, diagnosis, and therapy is complex and fascinating and holds great promise for the future. Discussion/Conclusion We have come a long way in understanding the epigenetic network from the initial model of epigenetic regulation: DNA gets methylated, recruits histone deacetylases,and the two systems button down chromatin and silence expression. With the availability of NSG, CHIP-Seq, and Crisp-cas9 technologies, we now understand that epigenetics involves a complex and dynamic interplay of writers, erasers, and readers, which act not only on promoters but on many regula- tory elements, including enhancers and repressors, forming a three-dimensional network of regulation. The impact of epigenetic alterations in cancer is also complex. Many mutations in epige- netic writers, erasers, and readers have been identified as promoting cancer development. Fur- thermore, epigenetics has been recognizedas lying at the heart of multiple hallmarks of cancer, interacting with cell cycle promotion, cancer metabolism, neo angiogenesis and the immune system. Although not able to induce leukemia [163,164], a dysregulated epigenome allows other mutations to occur, giving cancer cells a growth advantage over normal cells. Unsurpris- ingly, epigenetic mutations are starting to get used as biomarkers in hematologic malignancies and have been found to be associated with poor prognosis. The parallel development of epigenetic therapy mirrors the evolution of biology. The “histor- ical” DNMTi and HDACi are registered in hematological cancers, but their mode of action is not fully understood, while more recently, NGS-driven drug discovery has led to the develop- ment of real targeted agents focused on epigenetic writers and IDH. Several important questions have to be answered: What is the optimal duration of therapy with epigenetic agents? Several of these agents might require longer exposure to have a therapeutic effect compared to traditional 11 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 9. Jaenisch R, Bird A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nature genetics. 2003; 33 Suppl:245–54. doi: 10.1038/ng1089 PMID: 12610534 10. Woods BA, Levine RL. The role of mutations in epigenetic regulators in myeloid malignancies. Immu- nol Rev. 2015; 263(1):22–35. doi: 10.1111/imr.12246 PMID: 25510269 11. Shih AH, Abdel-Wahab O, Patel JP, Levine RL. The role of mutations in epigenetic regulators in mye- loid malignancies. Nature reviews Cancer. 2012; 12(9):599–612. doi: 10.1038/nrc3343 PMID: 22898539 12. West AC, Johnstone RW. New and emerging HDAC inhibitors for cancer treatment. The Journal of clinical investigation. 2014; 124(1):30–9.; PubMed Central PMCID: PMCPMC3871231. doi: 10.1172/ JCI69738 PMID: 24382387 13. Volkel P, Dupret B, Le Bourhis X, Angrand PO. Diverse involvement of EZH2 in cancer epigenetics. Am J Transl Res. 2015; 7(2):175–93.; PubMed Central PMCID: PMCPMC4399085. PMID: 25901190 14. Ward PS, Patel J, Wise DR, Abdel-Wahab O, Bennett BD, Coller HA, et al. The common feature of leukemia-associated IDH1 and IDH2 mutations is a neomorphic enzyme activity converting alpha- ketoglutarate to 2-hydroxyglutarate. Cancer cell. 2010; 17(3):225–34.; PubMed Central PMCID: PMC2849316. doi: 10.1016/j.ccr.2010.01.020 PMID: 20171147 15. Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A, et al. Leukemic IDH1 and IDH2 muta- tions result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differ- entiation. Cancer cell. 2010; 18(6):553–67.; PubMed Central PMCID: PMC4105845. doi: 10.1016/j. ccr.2010.11.015 PMID: 21130701 16. Pellegrini M, Ferrari R. Epigenetic analysis: ChIP-chip and ChIP-seq. Methods in molecular biology. 2012; 802:377–87. doi: 10.1007/978-1-61779-400-1_25 PMID: 22130894 17. Capell BC, Berger SL. Genome-wide epigenetics. The Journal of investigative dermatology. 2013; 133(6):e9. doi: 10.1038/jid.2013.173 PMID: 23673507 18. Zhang Y, Jeltsch A. The application of next generation sequencing in DNA methylation analysis. Genes (Basel). 2010; 1(1):85–101.; PubMed Central PMCID: PMCPMC3960863. doi: 10.3390/ genes1010085 PMID: 24710012 19. Zentner GE, Henikoff S. Epigenome editing made easy. Nature biotechnology. 2015; 33(6):606–7. doi: 10.1038/nbt.3248 PMID: 26057978 20. Rusk N. CRISPRs and epigenome editing. Nature methods. 2014; 11(1):28. doi: 10.1038/nmeth. 2775 PMID: 24524135 21. Rando OJ, Chang HY. Genome-wide views of chromatin structure. Annual review of biochemistry. 2009; 78:245–71.; PubMed Central PMCID: PMC2811691. doi: 10.1146/annurev.biochem.78. 071107.134639 PMID: 19317649 22. Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, et al. High-resolution profiling of histone methylations in the human genome. Cell. 2007; 129(4):823–37. doi: 10.1016/j.cell.2007.05.009 PMID: 17512414 23. Bannister AJ, Kouzarides T. Regulation of chromatin by histone modifications. Cell research. References 1. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011; 144(5):646–74. doi: 10.1016/j.cell.2011.02.013 PMID: 21376230 1. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011; 144(5):646–74. doi: 10.1016/j.cell.2011.02.013 PMID: 21376230 2. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000; 100(1):57–70. doi: 10.1016/S0092- 8674(00)81683-9 PMID: 10647931 3. Holliday R. The inheritance of epigenetic defects. Science. 1987; 238(4824):163–70. doi: 10.1126/ science.3310230 PMID: 3310230 4. Esteller M. Epigenetics in cancer. The New England journal of medicine. 2008; 358(11):1148–59. doi: 10.1056/NEJMra072067 PMID: 18337604 4. Esteller M. Epigenetics in cancer. The New England journal of medicine. 2008; 358(11):1148–59. doi: 10.1056/NEJMra072067 PMID: 18337604 5. Dawson MA, Kouzarides T, Huntly BJ. Targeting epigenetic readers in cancer. The New England journal of medicine. 2012; 367(7):647–57. doi: 10.1056/NEJMra1112635 PMID: 22894577 5. Dawson MA, Kouzarides T, Huntly BJ. Targeting epigenetic readers in cancer. The New England journal of medicine. 2012; 367(7):647–57. doi: 10.1056/NEJMra1112635 PMID: 22894577 6. Kouzarides T. Chromatin modifications and their function. Cell. 2007; 128(4):693–705. doi: 10.1016/ j.cell.2007.02.005 PMID: 17320507 6. Kouzarides T. Chromatin modifications and their function. Cell. 2007; 128(4):693–705. doi: 10.1016/ j.cell.2007.02.005 PMID: 17320507 7. Tan M, Luo H, Lee S, Jin F, Yang JS, Montellier E, et al. Identification of 67 histone marks and histone lysine crotonylation as a new type of histone modification. Cell. 2011; 146(6):1016–28.; PubMed Cen- tral PMCID: PMC3176443. doi: 10.1016/j.cell.2011.08.008 PMID: 21925322 7. Tan M, Luo H, Lee S, Jin F, Yang JS, Montellier E, et al. Identification of 67 histone marks and histone lysine crotonylation as a new type of histone modification. Cell. 2011; 146(6):1016–28.; PubMed Cen- tral PMCID: PMC3176443. doi: 10.1016/j.cell.2011.08.008 PMID: 21925322 8. Kim TK, Gore SD, Zeidan AM. Epigenetic Therapy in Acute Myeloid Leukemia: Current and Future Directions. Seminars in hematology. 2015; 52(3):172–83. doi: 10.1053/j.seminhematol.2015.04.003 PMID: 26111464 12 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 30. Saito Y, Liang G, Egger G, Friedman JM, Chuang JC, Coetzee GA, et al. Specific activation of micro- RNA-127 with downregulation of the proto-oncogene BCL6 by chromatin-modifying drugs in human cancer cells. Cancer cell. 2006; 9(6):435–43. doi: 10.1016/j.ccr.2006.04.020 PMID: 16766263 31. Lujambio A, Ropero S, Ballestar E, Fraga MF, Cerrato C, Setien F, et al. Genetic unmasking of an epigenetically silenced microRNA in human cancer cells. Cancer research. 2007; 67(4):1424–9. doi: 10.1158/0008-5472.CAN-06-4218 PMID: 17308079 32. Weber M, Davies JJ, Wittig D, Oakeley EJ, Haase M, Lam WL, et al. Chromosome-wide and pro- moter-specific analyses identify sites of differential DNA methylation in normal and transformed human cells. Nature genetics. 2005; 37(8):853–62. doi: 10.1038/ng1598 PMID: 16007088 33. Fraga MF, Herranz M, Espada J, Ballestar E, Paz MF, Ropero S, et al. A mouse skin multistage carci- nogenesis model reflects the aberrant DNA methylation patterns of human tumors. Cancer research. 2004; 64(16):5527–34. doi: 10.1158/0008-5472.CAN-03-4061 PMID: 15313885 34. Eden A, Gaudet F, Waghmare A, Jaenisch R. Chromosomal instability and tumors promoted by DNA hypomethylation. Science. 2003; 300(5618):455. doi: 10.1126/science.1083557 PMID: 12702868 35. Karpf AR, Matsui S. Genetic disruption of cytosine DNA methyltransferase enzymes induces chromo- somal instability in human cancer cells. Cancer research. 2005; 65(19):8635–9. doi: 10.1158/0008- 5472.CAN-05-1961 PMID: 16204030 36. Bestor TH. Transposons reanimated in mice. Cell. 2005; 122(3):322–5. doi: 10.1016/j.cell.2005.07. 024 PMID: 16096053 37. Feinberg AP. Imprinting of a genomic domain of 11p15 and loss of imprinting in cancer: an introduc- tion. Cancer research. 1999; 59(7 Suppl):1743s–6s. PMID: 10197590 38. Kaneda A, Feinberg AP. Loss of imprinting of IGF2: a common epigenetic modifier of intestinal tumor risk. Cancer research. 2005; 65(24):11236–40. doi: 10.1158/0008-5472.CAN-05-2959 PMID: 16357124 39. Holm TM, Jackson-Grusby L, Brambrink T, Yamada Y, Rideout WM 3rd, Jaenisch R. Global loss of imprinting leads to widespread tumorigenesis in adult mice. Cancer cell. 2005; 8(4):275–85. doi: 10. 1016/j.ccr.2005.09.007 PMID: 16226703 40. Greger V, Passarge E, Hopping W, Messmer E, Horsthemke B. Epigenetic changes may contribute to the formation and spontaneous regression of retinoblastoma. Hum Genet. 1989; 83(2):155–8. doi: 10.1007/BF00286709 PMID: 2550354 41. Herman JG, Latif F, Weng Y, Lerman MI, Zbar B, Liu S, et al. Silencing of the VHL tumor-suppressor gene by DNA methylation in renal carcinoma. Proceedings of the National Academy of Sciences of the United States of America. 1994; 91(21):9700–4.; PubMed Central PMCID: PMCPMC44884. doi: 10.1073/pnas.91.21.9700 PMID: 7937876 42. 2011; 21(3):381–95.; PubMed Central PMCID: PMC3193420. doi: 10.1038/cr.2011.22 PMID: 21321607 24. Taverna SD, Li H, Ruthenburg AJ, Allis CD, Patel DJ. How chromatin-binding modules interpret his- tone modifications: lessons from professional pocket pickers. Nature structural & molecular biology. 2007; 14(11):1025–40. doi: 10.1038/nsmb1338 PMID: 17984965 25. Ruthenburg AJ, Li H, Patel DJ, Allis CD. Multivalent engagement of chromatin modifications by linked binding modules. Nature reviews Molecular cell biology. 2007; 8(12):983–94. doi: 10.1038/nrm2298 PMID: 18037899 26. Bheda P, Schneider R. Epigenetics reloaded: the single-cell revolution. Trends in cell biology. 2014; 24(11):712–23. doi: 10.1016/j.tcb.2014.08.010 PMID: 25283892 27. Feinberg AP, Vogelstein B. Hypomethylation distinguishes genes of some human cancers from their normal counterparts. Nature. 1983; 301(5895):89–92. doi: 10.1038/301089a0 PMID: 6185846 28. Merlo A, Herman JG, Mao L, Lee DJ, Gabrielson E, Burger PC, et al. 5’ CpG island methylation is associated with transcriptional silencing of the tumour suppressor p16/CDKN2/MTS1 in human can- cers. Nature medicine. 1995; 1(7):686–92. doi: 10.1038/nm0795-686 PMID: 7585152 29. Sakai T, Toguchida J, Ohtani N, Yandell DW, Rapaport JM, Dryja TP. Allele-specific hypermethyla- tion of the retinoblastoma tumor-suppressor gene. American journal of human genetics. 1991; 48 (5):880–8.; PubMed Central PMCID: PMC1683063. PMID: 1673287 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 13 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 50. Wouters BJ, Jorda MA, Keeshan K, Louwers I, Erpelinck-Verschueren CA, Tielemans D, et al. Dis- tinct gene expression profiles of acute myeloid/T-lymphoid leukemia with silenced CEBPA and muta- tions in NOTCH1. Blood. 2007; 110(10):3706–14.; PubMed Central PMCID: PMCPMC2077318. doi: 10.1182/blood-2007-02-073486 PMID: 17671232 51. Hollink IH, van den Heuvel-Eibrink MM, Arentsen-Peters ST, Zimmermann M, Peeters JK, Valk PJ, et al. Characterization of CEBPA mutations and promoter hypermethylation in pediatric acute mye- loid leukemia. Haematologica. 2011; 96(3):384–92.; PubMed Central PMCID: PMCPMC3046269. doi: 10.3324/haematol.2010.031336 PMID: 21134981 52. Figueroa ME, Wouters BJ, Skrabanek L, Glass J, Li Y, Erpelinck-Verschueren CA, et al. Genome- wide epigenetic analysis delineates a biologically distinct immature acute leukemia with myeloid/T- lymphoid features. Blood. 2009; 113(12):2795–804.; PubMed Central PMCID: PMCPMC2945920. doi: 10.1182/blood-2008-08-172387 PMID: 19168792 53. Flavahan WA, Drier Y, Liau BB, Gillespie SM, Venteicher AS, Stemmer-Rachamimov AO, et al. Insu- lator dysfunction and oncogene activation in IDH mutant gliomas. Nature. 2016; 529(7584):110–4.; PubMed Central PMCID: PMCPMC4831574. doi: 10.1038/nature16490 PMID: 26700815 54. Kang JY, Song SH, Yun J, Jeon MS, Kim HP, Han SW, et al. Disruption of CTCF/cohesin-mediated high-order chromatin structures by DNA methylation downregulates PTGS2 expression. Oncogene. 2015; 34(45):5677–84. doi: 10.1038/onc.2015.17 PMID: 25703332 55. Hnisz D, Weintraub AS, Day DS, Valton AL, Bak RO, Li CH, et al. Activation of proto-oncogenes by disruption of chromosome neighborhoods. Science. 2016; 351(6280):1454–8. doi: 10.1126/science. aad9024 PMID: 26940867 56. Timp W, Feinberg AP. Cancer as a dysregulated epigenome allowing cellular growth advantage at the expense of the host. Nature reviews Cancer. 2013; 13(7):497–510.; PubMed Central PMCID: PMC4636434. doi: 10.1038/nrc3486 PMID: 23760024 57. Ichimura K, Pearson DM, Kocialkowski S, Backlund LM, Chan R, Jones DT, et al. IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas. Neuro Oncol. 2009; 11(4):341–7.; PubMed Central PMCID: PMCPMC2743214. doi: 10.1215/15228517-2009-025 PMID: 19435942 58. Abdel-Wahab O, Levine RL. Mutations in epigenetic modifiers in the pathogenesis and therapy of acute myeloid leukemia. Blood. 2013; 121(18):3563–72.; PubMed Central PMCID: PMC3643757. doi: 10.1182/blood-2013-01-451781 PMID: 23640996 59. Conway O’Brien E, Prideaux S, Chevassut T. The epigenetic landscape of acute myeloid leukemia. Adv Hematol. 2014; 2014:103175.; PubMed Central PMCID: PMCPMC3980839. doi: 10.1155/2014/ 103175 PMID: 24778653 60. Shen H, Laird PW. Interplay between the cancer genome and epigenome. Cell. 2013; 153(1):38–55.; PubMed Central PMCID: PMCPMC3648790. doi: 10.1016/j.cell.2013.03.008 PMID: 23540689 61. Ley TJ, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE, et al. Esteller M, Silva JM, Dominguez G, Bonilla F, Matias-Guiu X, Lerma E, et al. Promoter hypermethyla- tion and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl Cancer Inst. 2000; 92 (7):564–9. doi: 10.1093/jnci/92.7.564 PMID: 10749912 43. Herman JG, Baylin SB. Gene silencing in cancer in association with promoter hypermethylation. The New England journal of medicine. 2003; 349(21):2042–54. doi: 10.1056/NEJMra023075 PMID: 14627790 44. Costello JF, Fruhwald MC, Smiraglia DJ, Rush LJ, Robertson GP, Gao X, et al. Aberrant CpG-island methylation has non-random and tumour-type-specific patterns. Nature genetics. 2000; 24(2):132–8. doi: 10.1038/72785 PMID: 10655057 45. Esteller M, Corn PG, Baylin SB, Herman JG. A gene hypermethylation profile of human cancer. Can- cer research. 2001; 61(8):3225–9. PMID: 11309270 46. Esteller M. Cancer epigenomics: DNA methylomes and histone-modification maps. Nature reviews Genetics. 2007; 8(4):286–98. doi: 10.1038/nrg2005 PMID: 17339880 47. Figueroa ME, Lugthart S, Li Y, Erpelinck-Verschueren C, Deng X, Christos PJ, et al. DNA methylation signatures identify biologically distinct subtypes in acute myeloid leukemia. Cancer cell. 2010; 17 (1):13–27.; PubMed Central PMCID: PMCPMC3008568. doi: 10.1016/j.ccr.2009.11.020 PMID: 20060365 48. Frohling S, Schlenk RF, Stolze I, Bihlmayr J, Benner A, Kreitmeier S, et al. CEBPA mutations in youn- ger adults with acute myeloid leukemia and normal cytogenetics: prognostic relevance and analysis of cooperating mutations. Journal of clinical oncology: official journal of the American Society of Clini- cal Oncology. 2004; 22(4):624–33. doi: 10.1200/JCO.2004.06.060 PMID: 14726504 49. Pabst T, Mueller BU, Zhang P, Radomska HS, Narravula S, Schnittger S, et al. Dominant-negative mutations of CEBPA, encoding CCAAT/enhancer binding protein-alpha (C/EBPalpha), in acute mye- loid leukemia. Nature genetics. 2001; 27(3):263–70. doi: 10.1038/85820 PMID: 11242107 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 14 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 69. Lund K, Adams PD, Copland M. EZH2 in normal and malignant hematopoiesis. Leukemia. 2014; 28 (1):44–9. doi: 10.1038/leu.2013.288 PMID: 24097338 70. Abdel-Wahab O, Dey A. The ASXL-BAP1 axis: new factors in myelopoiesis, cancer and epigenetics. Leukemia. 2013; 27(1):10–5. doi: 10.1038/leu.2012.288 PMID: 23147254 71. Ryan RJ, Nitta M, Borger D, Zukerberg LR, Ferry JA, Harris NL, et al. EZH2 codon 641 mutations are common in BCL2-rearranged germinal center B cell lymphomas. PLoS ONE. 2011; 6(12):e28585.; PubMed Central PMCID: PMCPMC3237460. doi: 10.1371/journal.pone.0028585 PMID: 22194861 72. Morin RD, Johnson NA, Severson TM, Mungall AJ, An J, Goya R, et al. Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin. Nature genetics. 2010; 42(2):181–5.; PubMed Central PMCID: PMCPMC2850970. doi: 10.1038/ng.518 PMID: 20081860 73. McCabe MT, Ott HM, Ganji G, Korenchuk S, Thompson C, Van Aller GS, et al. EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2-activating mutations. Nature. 2012; 492(7427):108– 12. doi: 10.1038/nature11606 PMID: 23051747 74. Aravind L, Iyer LM. The HARE-HTH and associated domains: novel modules in the coordination of epigenetic DNA and protein modifications. Cell cycle. 2012; 11(1):119–31.; PubMed Central PMCID: PMCPMC3272235. doi: 10.4161/cc.11.1.18475 PMID: 22186017 75. Abdel-Wahab O, Adli M, LaFave LM, Gao J, Hricik T, Shih AH, et al. ASXL1 mutations promote mye- loid transformation through loss of PRC2-mediated gene repression. Cancer cell. 2012; 22(2):180– 93.; PubMed Central PMCID: PMCPMC3422511. doi: 10.1016/j.ccr.2012.06.032 PMID: 22897849 76. Gross S, Cairns RA, Minden MD, Driggers EM, Bittinger MA, Jang HG, et al. Cancer-associated metabolite 2-hydroxyglutarate accumulates in acute myelogenous leukemia with isocitrate dehydro- genase 1 and 2 mutations. The Journal of experimental medicine. 2010; 207(2):339–44.; PubMed Central PMCID: PMCPMC2822606. doi: 10.1084/jem.20092506 PMID: 20142433 77. Stein EM. IDH2 inhibition in AML: Finally progress? Best practice & research Clinical haematology. 2015; 28(2–3):112–5. doi: 10.1016/j.beha.2015.10.016 PMID: 26590767 78. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, et al. IDH1 and IDH2 mutations in gliomas. The New England journal of medicine. 2009; 360(8):765–73.; PubMed Central PMCID: PMC2820383. doi: 10.1056/NEJMoa0808710 PMID: 19228619 79. Green CL, Evans CM, Zhao L, Hills RK, Burnett AK, Linch DC, et al. The prognostic significance of IDH2 mutations in AML depends on the location of the mutation. Blood. 2011; 118(2):409–12. doi: 10.1182/blood-2010-12-322479 PMID: 21596855 80. Wang F, Travins J, DeLaBarre B, Penard-Lacronique V, Schalm S, Hansen E, et al. DNMT3A mutations in acute myeloid leukemia. The New England journal of medicine. 2010; 363(25):2424–33.; PubMed Central PMCID: PMCPMC3201818. doi: 10.1056/NEJMoa1005143 PMID: 21067377 62. Thiede C. Mutant DNMT3A: teaming up to transform. Blood. 2012; 119(24):5615–7. doi: 10.1182/ blood-2012-04-423905 PMID: 22700693 63. Patel JP, Gonen M, Figueroa ME, Fernandez H, Sun Z, Racevskis J, et al. Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. The New England journal of medicine. 2012; 366(12):1079–89.; PubMed Central PMCID: PMCPMC3545649. doi: 10.1056/NEJMoa1112304 PMID: 22417203 64. Walter MJ, Ding L, Shen D, Shao J, Grillot M, McLellan M, et al. Recurrent DNMT3A mutations in patients with myelodysplastic syndromes. Leukemia. 2011; 25(7):1153–8.; PubMed Central PMCID: PMCPMC3202965. doi: 10.1038/leu.2011.44 PMID: 21415852 65. Abdel-Wahab O, Mullally A, Hedvat C, Garcia-Manero G, Patel J, Wadleigh M, et al. Genetic charac- terization of TET1, TET2, and TET3 alterations in myeloid malignancies. Blood. 2009; 114(1):144–7.; PubMed Central PMCID: PMCPMC2710942. doi: 10.1182/blood-2009-03-210039 PMID: 19420352 66. Chou WC, Chou SC, Liu CY, Chen CY, Hou HA, Kuo YY, et al. TET2 mutation is an unfavorable prog- nostic factor in acute myeloid leukemia patients with intermediate-risk cytogenetics. Blood. 2011; 118 (14):3803–10. doi: 10.1182/blood-2011-02-339747 PMID: 21828143 67. Tefferi A, Pardanani A, Lim KH, Abdel-Wahab O, Lasho TL, Patel J, et al. TET2 mutations and their clinical correlates in polycythemia vera, essential thrombocythemia and myelofibrosis. Leukemia. 2009; 23(5):905–11.; PubMed Central PMCID: PMCPMC4654629. doi: 10.1038/leu.2009.47 PMID: 19262601 68. Xu F, Li X. The role of histone methyltransferase EZH2 in myelodysplastic syndromes. Expert review of hematology. 2012; 5(2):177–85. doi: 10.1586/ehm.12.5 PMID: 22475286 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 15 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 the ELN Favorable genetic category. Blood. 2011; 118(26):6920–9.; PubMed Central PMCID: PMCPMC3245212. doi: 10.1182/blood-2011-08-368225 PMID: 22031865 89. Levis M. Targeting IDH: the next big thing in AML. Blood. 2013; 122(16):2770–1. doi: 10.1182/blood- 2013-09-522441 PMID: 24136078 90. Krivtsov AV, Armstrong SA. MLL translocations, histone modifications and leukaemia stem-cell development. Nature reviews Cancer. 2007; 7(11):823–33. doi: 10.1038/nrc2253 PMID: 17957188 91. Meyer C, Schneider B, Jakob S, Strehl S, Attarbaschi A, Schnittger S, et al. The MLL recombinome of acute leukemias. Leukemia. 2006; 20(5):777–84. doi: 10.1038/sj.leu.2404150 PMID: 16511515 92. Mohan M, Lin C, Guest E, Shilatifard A. Licensed to elongate: a molecular mechanism for MLL-based leukaemogenesis. Nature reviews Cancer. 2010; 10(10):721–8. doi: 10.1038/nrc2915 PMID: 20844554 93. Mueller D, Garcia-Cuellar MP, Bach C, Buhl S, Maethner E, Slany RK. Misguided transcriptional elongation causes mixed lineage leukemia. PLoS Biol. 2009; 7(11):e1000249.; PubMed Central PMCID: PMC2774266. doi: 10.1371/journal.pbio.1000249 PMID: 19956800 94. Lin C, Smith ER, Takahashi H, Lai KC, Martin-Brown S, Florens L, et al. AFF4, a component of the ELL/P-TEFb elongation complex and a shared subunit of MLL chimeras, can link transcription elon- gation to leukemia. Molecular cell. 2010; 37(3):429–37.; PubMed Central PMCID: PMC2872029. doi: 10.1016/j.molcel.2010.01.026 PMID: 20159561 95. Yokoyama A, Lin M, Naresh A, Kitabayashi I, Cleary ML. A higher-order complex containing AF4 and ENL family proteins with P-TEFb facilitates oncogenic and physiologic MLL-dependent transcription. Cancer cell. 2010; 17(2):198–212.; PubMed Central PMCID: PMC2824033. doi: 10.1016/j.ccr.2009. 12.040 PMID: 20153263 96. Dawson MA, Prinjha RK, Dittmann A, Giotopoulos G, Bantscheff M, Chan WI, et al. Inhibition of BET recruitment to chromatin as an effective treatment for MLL-fusion leukaemia. Nature. 2011; 478 (7370):529–33.; PubMed Central PMCID: PMC3679520. doi: 10.1038/nature10509 PMID: 21964340 97. Filippakopoulos P, Qi J, Picaud S, Shen Y, Smith WB, Fedorov O, et al. Selective inhibition of BET bromodomains. Nature. 2010; 468(7327):1067–73.; PubMed Central PMCID: PMC3010259. doi: 10. 1038/nature09504 PMID: 20871596 98. Zuber J, Shi J, Wang E, Rappaport AR, Herrmann H, Sison EA, et al. RNAi screen identifies Brd4 as a therapeutic target in acute myeloid leukaemia. Nature. 2011; 478(7370):524–8.; PubMed Central PMCID: PMC3328300. doi: 10.1038/nature10334 PMID: 21814200 99. Krivtsov AV, Feng Z, Lemieux ME, Faber J, Vempati S, Sinha AU, et al. H3K79 methylation profiles define murine and human MLL-AF4 leukemias. Cancer cell. 2008; 14(5):355–68.; PubMed Central PMCID: PMC2591932. doi: 10.1016/j.ccr.2008.10.001 PMID: 18977325 100. Chen CW, Armstrong SA. Targeting DOT1L and HOX gene expression in MLL-rearranged leukemia and beyond. Experimental hematology. Targeted inhibi- tion of mutant IDH2 in leukemia cells induces cellular differentiation. Science. 2013; 340(6132):622– 6. doi: 10.1126/science.1234769 PMID: 23558173 81. Xu W, Yang H, Liu Y, Yang Y, Wang P, Kim SH, et al. Oncometabolite 2-hydroxyglutarate is a com- petitive inhibitor of alpha-ketoglutarate-dependent dioxygenases. Cancer cell. 2011; 19(1):17–30.; PubMed Central PMCID: PMC3229304. doi: 10.1016/j.ccr.2010.12.014 PMID: 21251613 82. Gaidzik VI, Paschka P, Spath D, Habdank M, Kohne CH, Germing U, et al. TET2 mutations in acute myeloid leukemia (AML): results from a comprehensive genetic and clinical analysis of the AML study group. Journal of clinical oncology: official journal of the American Society of Clinical Oncology. 2012; 30(12):1350–7. doi: 10.1200/JCO.2011.39.2886 PMID: 22430270 83. Falkenberg KJ, Johnstone RW. Histone deacetylases and their inhibitors in cancer, neurological dis- eases and immune disorders. Nature reviews Drug discovery. 2014; 13(9):673–91. doi: 10.1038/ nrd4360 PMID: 25131830 84. Hug BA, Lazar MA. ETO interacting proteins. Oncogene. 2004; 23(24):4270–4. doi: 10.1038/sj.onc. 1207674 PMID: 15156183 85. Rice KL, de The H. The acute promyelocytic leukaemia success story: curing leukaemia through tar- geted therapies. Journal of internal medicine. 2014; 276(1):61–70. doi: 10.1111/joim.12208 PMID: 24635409 86. Lugthart S, Figueroa ME, Bindels E, Skrabanek L, Valk PJ, Li Y, et al. Aberrant DNA hypermethyla- tion signature in acute myeloid leukemia directed by EVI1. Blood. 2011; 117(1):234–41.; PubMed Central PMCID: PMCPMC3037746. doi: 10.1182/blood-2010-04-281337 PMID: 20855866 87. Bejar R, Stevenson K, Abdel-Wahab O, Galili N, Nilsson B, Garcia-Manero G, et al. Clinical effect of point mutations in myelodysplastic syndromes. N Engl J Med. 2011; 364(26):2496–506. Epub 2011/ 07/01.; PubMed Central PMCID: PMC3159042. doi: 10.1056/NEJMoa1013343 PMID: 21714648 88. Metzeler KH, Becker H, Maharry K, Radmacher MD, Kohlschmidt J, Mrozek K, et al. ASXL1 muta- tions identify a high-risk subgroup of older patients with primary cytogenetically normal AML within PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 16 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 stem cells and fibroblasts. Nature genetics. 2009; 41(12):1350–3.; PubMed Central PMCID: PMC2958040. doi: 10.1038/ng.471 PMID: 19881528 stem cells and fibroblasts. Nature genetics. 2009; 41(12):1350–3.; PubMed Central PMCID: PMC2958040. doi: 10.1038/ng.471 PMID: 19881528 108. Timp W, Levchenko A, Feinberg AP. A new link between epigenetic progenitor lesions in cancer and the dynamics of signal transduction. Cell cycle. 2009; 8(3):383–90. doi: 10.4161/cc.8.3.7542 PMID: 19177016 109. Fuks F, Milner J, Kouzarides T. BRCA2 associates with acetyltransferase activity when bound to P/ CAF. Oncogene. 1998; 17(19):2531–4. doi: 10.1038/sj.onc.1202475 PMID: 9824164 110. Esteve PO, Chin HG, Pradhan S. Human maintenance DNA (cytosine-5)-methyltransferase and p53 modulate expression of p53-repressed promoters. Proceedings of the National Academy of Sciences of the United States of America. 2005; 102(4):1000–5.; PubMed Central PMCID: PMC544618. doi: 10.1073/pnas.0407729102 PMID: 15657147 111. Campbell PM, Szyf M. Human DNA methyltransferase gene DNMT1 is regulated by the APC path- way. Carcinogenesis. 2003; 24(1):17–24. doi: 10.1093/carcin/24.1.17 PMID: 12538344 112. Coomber BL, Yu JL, Fathers KE, Plumb C, Rak JW. Angiogenesis and the role of epigenetics in metastasis. Clinical & experimental metastasis. 2003; 20(3):215–27. PMID: 12741680 113. Lujambio A, Esteller M. How epigenetics can explain human metastasis: a new role for microRNAs. Cell cycle. 2009; 8(3):377–82. doi: 10.4161/cc.8.3.7526 PMID: 19177007 114. Jones PA, Baylin SB. The epigenomics of cancer. Cell. 2007; 128(4):683–92.; PubMed Central PMCID: PMC3894624. doi: 10.1016/j.cell.2007.01.029 PMID: 17320506 115. Maio M, Covre A, Fratta E, Di Giacomo AM, Taverna P, Natali PG, et al. Molecular Pathways: At the Crossroads of Cancer Epigenetics and Immunotherapy. Clinical cancer research: an official journal of the American Association for Cancer Research. 2015; 21(18):4040–7. doi: 10.1158/1078-0432. CCR-14-2914 PMID: 26374074 116. Carrer A, Wellen KE. Metabolism and epigenetics: a link cancer cells exploit. Current opinion in bio- technology. 2015; 34:23–9. doi: 10.1016/j.copbio.2014.11.012 PMID: 25461508 117. Kaelin WG Jr., McKnight SL. Influence of metabolism on epigenetics and disease. Cell. 2013; 153 (1):56–69.; PubMed Central PMCID: PMC3775362. doi: 10.1016/j.cell.2013.03.004 PMID: 23540690 118. Sigalotti L, Fratta E, Coral S, Maio M. Epigenetic drugs as immunomodulators for combination thera- pies in solid tumors. Pharmacology & therapeutics. 2014; 142(3):339–50. doi: 10.1016/j.pharmthera. 2013.12.015 PMID: 24384533 119. Coral S, Parisi G, Nicolay HJ, Colizzi F, Danielli R, Fratta E, et al. Immunomodulatory activity of SGI- 110, a 5-aza-2’-deoxycytidine-containing demethylating dinucleotide. Cancer immunology, immuno- therapy: CII. 2013; 62(3):605–14. doi: 10.1007/s00262-012-1365-7 PMID: 23138873 120. Srivastava P, Paluch BE, Matsuzaki J, James SR, Collamat-Lai G, Karbach J, et al. 2015; 43(8):673–84.; PubMed Central PMCID: PMC4540610. doi: 10.1016/j.exphem.2015.05.012 PMID: 26118503 101. Bernt KM, Zhu N, Sinha AU, Vempati S, Faber J, Krivtsov AV, et al. MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1L. Cancer cell. 2011; 20(1):66–78.; PubMed Central PMCID: PMC3329803. doi: 10.1016/j.ccr.2011.06.010 PMID: 21741597 102. Milne TA, Martin ME, Brock HW, Slany RK, Hess JL. Leukemogenic MLL fusion proteins bind across a broad region of the Hox a9 locus, promoting transcription and multiple histone modifications. Can- cer research. 2005; 65(24):11367–74. doi: 10.1158/0008-5472.CAN-05-1041 PMID: 16357144 103. Ferrando AA, Armstrong SA, Neuberg DS, Sallan SE, Silverman LB, Korsmeyer SJ, et al. Gene expression signatures in MLL-rearranged T-lineage and B-precursor acute leukemias: dominance of HOX dysregulation. Blood. 2003; 102(1):262–8. doi: 10.1182/blood-2002-10-3221 PMID: 12637319 104. Daigle SR, Olhava EJ, Therkelsen CA, Majer CR, Sneeringer CJ, Song J, et al. Selective killing of mixed lineage leukemia cells by a potent small-molecule DOT1L inhibitor. Cancer cell. 2011; 20(1):53–65.; PubMed Central PMCID: PMC4046888. doi: 10.1016/j.ccr.2011.06.009 PMID: 21741596 105. Daigle SR, Olhava EJ, Therkelsen CA, Basavapathruni A, Jin L, Boriack-Sjodin PA, et al. Potent inhi- bition of DOT1L as treatment of MLL-fusion leukemia. Blood. 2013; 122(6):1017–25.; PubMed Cen- tral PMCID: PMC3739029. doi: 10.1182/blood-2013-04-497644 PMID: 23801631 106. Chen SJ, Shen Y, Chen Z. A panoramic view of acute myeloid leukemia. Nature genetics. 2013; 45 (6):586–7. doi: 10.1038/ng.2651 PMID: 23715324 107. Doi A, Park IH, Wen B, Murakami P, Aryee MJ, Irizarry R, et al. Differential methylation of tissue- and cancer-specific CpG island shores distinguishes human induced pluripotent stem cells, embryonic PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 17 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 127. Luco RF, Allo M, Schor IE, Kornblihtt AR, Misteli T. Epigenetics in alternative pre-mRNA splicing. Cell. 2011; 144(1):16–26.; PubMed Central PMCID: PMCPMC3038581. doi: 10.1016/j.cell.2010.11. 056 PMID: 21215366 128. Kolasinska-Zwierz P, Down T, Latorre I, Liu T, Liu XS, Ahringer J. Differential chromatin marking of introns and expressed exons by H3K36me3. Nature genetics. 2009; 41(3):376–81.; PubMed Central PMCID: PMCPMC2648722. doi: 10.1038/ng.322 PMID: 19182803 129. Luco RF, Pan Q, Tominaga K, Blencowe BJ, Pereira-Smith OM, Misteli T. Regulation of alternative splicing by histone modifications. Science. 2010; 327(5968):996–1000.; PubMed Central PMCID: PMCPMC2913848. doi: 10.1126/science.1184208 PMID: 20133523 130. Damm F, Kosmider O, Gelsi-Boyer V, Renneville A, Carbuccia N, Hidalgo-Curtis C, et al. Mutations affecting mRNA splicing define distinct clinical phenotypes and correlate with patient outcome in mye- lodysplastic syndromes. Blood. 2012; 119(14):3211–8. doi: 10.1182/blood-2011-12-400994 PMID: 22343920 131. Leeke B, Marsman J, O’Sullivan JM, Horsfield JA. Cohesin mutations in myeloid malignancies: underlying mechanisms. Exp Hematol Oncol. 2014; 3:13.; PubMed Central PMCID: PMCPMC4046106. doi: 10.1186/2162-3619-3-13 PMID: 24904756 132. Kon A, Shih LY, Minamino M, Sanada M, Shiraishi Y, Nagata Y, et al. Recurrent mutations in multiple components of the cohesin complex in myeloid neoplasms. Nature genetics. 2013; 45(10):1232–7. doi: 10.1038/ng.2731 PMID: 23955599 133. Mazumdar C, Shen Y, Xavy S, Zhao F, Reinisch A, Li R, et al. Leukemia-Associated Cohesin Mutants Dominantly Enforce Stem Cell Programs and Impair Human Hematopoietic Progenitor Differentia- tion. Cell stem cell. 2015; 17(6):675–88.; PubMed Central PMCID: PMCPMC4671831. doi: 10.1016/ j.stem.2015.09.017 PMID: 26607380 134. Mullenders J, Aranda-Orgilles B, Lhoumaud P, Keller M, Pae J, Wang K, et al. Cohesin loss alters adult hematopoietic stem cell homeostasis, leading to myeloproliferative neoplasms. The Journal of experimental medicine. 2015; 212(11):1833–50.; PubMed Central PMCID: PMCPMC4612095. doi: 10.1084/jem.20151323 PMID: 26438359 135. Garcia-Manero G, Fenaux P. Hypomethylating agents and other novel strategies in myelodysplastic syndromes. Journal of clinical oncology: official journal of the American Society of Clinical Oncology. 2011; 29(5):516–23.; PubMed Central PMCID: PMC3056493. doi: 10.1200/JCO.2010.31.0854 PMID: 21220589 136. Duvic M, Talpur R, Ni X, Zhang C, Hazarika P, Kelly C, et al. Phase 2 trial of oral vorinostat (suberoy- lanilide hydroxamic acid, SAHA) for refractory cutaneous T-cell lymphoma (CTCL). Blood. 2007; 109 (1):31–9. doi: 10.1182/blood-2006-06-025999 PMID: 16960145 137. Whittaker SJ, Demierre MF, Kim EJ, Rook AH, Lerner A, Duvic M, et al. Final results from a multicen- ter, international, pivotal study of romidepsin in refractory cutaneous T-cell lymphoma. Immunomodula- tory action of SGI-110, a hypomethylating agent, in acute myeloid leukemia cells and xenografts. Leukemia research. 2014; 38(11):1332–41. doi: 10.1016/j.leukres.2014.09.001 PMID: 25260825 121. Fratta E, Sigalotti L, Colizzi F, Covre A, Nicolay HJ, Danielli R, et al. Epigenetically regulated clonal heritability of CTA expression profiles in human melanoma. Journal of cellular physiology. 2010; 223 (2):352–8. doi: 10.1002/jcp.22040 PMID: 20127705 122. Fonsatti E, Nicolay HJ, Sigalotti L, Calabro L, Pezzani L, Colizzi F, et al. Functional up-regulation of human leukocyte antigen class I antigens expression by 5-aza-2’-deoxycytidine in cutaneous mela- noma: immunotherapeutic implications. Clinical cancer research: an official journal of the American Association for Cancer Research. 2007; 13(11):3333–8. doi: 10.1158/1078-0432.CCR-06-3091 PMID: 17545540 123. Cairns RA, Mak TW. Oncogenic isocitrate dehydrogenase mutations: mechanisms, models, and clin- ical opportunities. Cancer discovery. 2013; 3(7):730–41. doi: 10.1158/2159-8290.CD-13-0083 PMID: 23796461 124. Cai L, Sutter BM, Li B, Tu BP. Acetyl-CoA induces cell growth and proliferation by promoting the acet- ylation of histones at growth genes. Molecular cell. 2011; 42(4):426–37.; PubMed Central PMCID: PMC3109073. doi: 10.1016/j.molcel.2011.05.004 PMID: 21596309 125. Kaochar S, Tu BP. Gatekeepers of chromatin: Small metabolites elicit big changes in gene expres- sion. Trends in biochemical sciences. 2012; 37(11):477–83.; PubMed Central PMCID: PMC3482309. doi: 10.1016/j.tibs.2012.07.008 PMID: 22944281 126. Yoshida K, Sanada M, Shiraishi Y, Nowak D, Nagata Y, Yamamoto R, et al. Frequent pathway muta- tions of splicing machinery in myelodysplasia. Nature. 2011; 478(7367):64–9. doi: 10.1038/ nature10496 PMID: 21909114 18 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 145. Rotili D, Mai A. Targeting Histone Demethylases: A New Avenue for the Fight against Cancer. Genes & cancer. 2011; 2(6):663–79.; PubMed Central PMCID: PMC3174264. doi: 10.1177/ 1947601911417976 PMID: 21941621 146. Stein Eytan M, Altman Jessica K, Collins Robert, DeAngelo Daniel J. 115 AG-221, an Oral, Selective, First-in-Class, Potent Inhibitor of the IDH2 Mutant Metabolic Enzyme, Induces Durable Remissions in a Phase I Study in Patients with IDH2 Mutation Positive Advanced Hematologic Malignancies 2014. 147. DiNardo C. S EM, Altman J.K., Collins R., DeAngelo D.J., Fathi A.T., et al. AG-221, an oral, selective, first-in-class, potent inhibitor of the IDH2 mutant enzyme, induced durable responses in a phase 1 study of IDH2 mutation-positive advanced hematologic malignancies. Haematol Eur Hematol Assoc Annu Meet, 100 (s1) (2015), p. 569 [abstr)]. 148. Deshpande AJ, Deshpande A, Sinha AU, Chen L, Chang J, Cihan A, et al. AF10 regulates progres- sive H3K79 methylation and HOX gene expression in diverse AML subtypes. Cancer cell. 2014; 26 (6):896–908.; PubMed Central PMCID: PMC4291116. doi: 10.1016/j.ccell.2014.10.009 PMID: 25464900 149. Delmore JE, Issa GC, Lemieux ME, Rahl PB, Shi J, Jacobs HM, et al. BET bromodomain inhibition as a therapeutic strategy to target c-Myc. Cell. 2011; 146(6):904–17.; PubMed Central PMCID: PMC3187920. doi: 10.1016/j.cell.2011.08.017 PMID: 21889194 150. Mertz JA, Conery AR, Bryant BM, Sandy P, Balasubramanian S, Mele DA, et al. Targeting MYC dependence in cancer by inhibiting BET bromodomains. Proceedings of the National Academy of Sciences of the United States of America. 2011; 108(40):16669–74.; PubMed Central PMCID: PMC3189078. doi: 10.1073/pnas.1108190108 PMID: 21949397 151. Yoo CB, Jones PA. Epigenetic therapy of cancer: past, present and future. Nature reviews Drug dis- covery. 2006; 5(1):37–50. doi: 10.1038/nrd1930 PMID: 16485345 152. Schwartsmann G, Schunemann H, Gorini CN, Filho AF, Garbino C, Sabini G, et al. A phase I trial of cisplatin plus decitabine, a new DNA-hypomethylating agent, in patients with advanced solid tumors and a follow-up early phase II evaluation in patients with inoperable non-small cell lung cancer. Inves- tigational new drugs. 2000; 18(1):83–91. PMID: 10830142 153. Fillmore CM, Xu C, Desai PT, Berry JM, Rowbotham SP, Lin YJ, et al. EZH2 inhibition sensitizes BRG1 and EGFR mutant lung tumours to TopoII inhibitors. Nature. 2015; 520(7546):239–42.; PubMed Central PMCID: PMC4393352. doi: 10.1038/nature14122 PMID: 25629630 154. Balch C, Nephew KP. Epigenetic targeting therapies to overcome chemotherapy resistance. Advances in experimental medicine and biology. 2013; 754:285–311. doi: 10.1007/978-1-4419- 9967-2_14 PMID: 22956507 155. Journal of clin- ical oncology: official journal of the American Society of Clinical Oncology. 2010; 28(29):4485–91. doi: 10.1200/JCO.2010.28.9066 PMID: 20697094 138. Foss FM, Zinzani PL, Vose JM, Gascoyne RD, Rosen ST, Tobinai K. Peripheral T-cell lymphoma. Blood. 2011; 117(25):6756–67. doi: 10.1182/blood-2010-05-231548 PMID: 21493798 139. Laubach JP, Moreau P, San-Miguel JF, Richardson PG. Panobinostat for the Treatment of Multiple Myeloma. Clinical cancer research: an official journal of the American Association for Cancer Research. 2015; 21(21):4767–73. doi: 10.1158/1078-0432.CCR-15-0530 PMID: 26362997 140. O’Connor OA, Horwitz S, Masszi T, Van Hoof A, Brown P, Doorduijn J, et al. Belinostat in Patients With Relapsed or Refractory Peripheral T-Cell Lymphoma: Results of the Pivotal Phase II BELIEF (CLN-19) Study. Journal of clinical oncology: official journal of the American Society of Clinical Oncol- ogy. 2015; 33(23):2492–9. doi: 10.1200/JCO.2014.59.2782 PMID: 26101246 141. Balasubramanian S, Verner E, Buggy JJ. Isoform-specific histone deacetylase inhibitors: the next step? Cancer letters. 2009; 280(2):211–21. doi: 10.1016/j.canlet.2009.02.013 PMID: 19289255 142. Yardley D. A. I-K R, Klein P.. Results of ENCORE 301, a randomized, phase II, double-blind, pla- cebo-controlled study of exemestane with or without entinostat in postmenopausal women with locally recurrent or metastatic estrogen receptor-positive (ER+) breast cancer progressing on a non- steroidal aromatase inhibitor (AI). J Clin Oncol. 2011; 29(suppl 27; abstr 268). 143. Knutson SK, Wigle TJ, Warholic NM, Sneeringer CJ, Allain CJ, Klaus CR, et al. A selective inhibitor of EZH2 blocks H3K27 methylation and kills mutant lymphoma cells. Nature chemical biology. 2012; 8(11):890–6. doi: 10.1038/nchembio.1084 PMID: 23023262 144. Kaniskan HU, Konze KD, Jin J. Selective inhibitors of protein methyltransferases. Journal of medici- nal chemistry. 2015; 58(4):1596–629.; PubMed Central PMCID: PMC4345896. doi: 10.1021/ jm501234a PMID: 25406853 19 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 Candelaria M, Gallardo-Rincon D, Arce C, Cetina L, Aguilar-Ponce JL, Arrieta O, et al. A phase II study of epigenetic therapy with hydralazine and magnesium valproate to overcome chemotherapy resistance in refractory solid tumors. Annals of oncology: official journal of the European Society for Medical Oncology / ESMO. 2007; 18(9):1529–38. doi: 10.1093/annonc/mdm204 PMID: 17761710 156. Appleton K, Mackay HJ, Judson I, Plumb JA, McCormick C, Strathdee G, et al. Phase I and pharma- codynamic trial of the DNA methyltransferase inhibitor decitabine and carboplatin in solid tumors. Journal of clinical oncology: official journal of the American Society of Clinical Oncology. 2007; 25 (29):4603–9. doi: 10.1200/JCO.2007.10.8688 PMID: 17925555 157. Kozar K, Kaminski R, Switaj T, Oldak T, Machaj E, Wysocki PJ, et al. Interleukin 12-based immuno- therapy improves the antitumor effectiveness of a low-dose 5-Aza-2’-deoxycitidine treatment in L1210 leukemia and B16F10 melanoma models in mice. Clinical cancer research: an official journal of the American Association for Cancer Research. 2003; 9(8):3124–33. PMID: 12912964 158. Simova J, Pollakova V, Indrova M, Mikyskova R, Bieblova J, Stepanek I, et al. Immunotherapy aug- ments the effect of 5-azacytidine on HPV16-associated tumours with different MHC class I-expres- sion status. British journal of cancer. 2011; 105(10):1533–41.; PubMed Central PMCID: PMC3242529.doi: 10.1038/bjc.2011.428 PMID: 22015556 159. Shen L, Ciesielski M, Ramakrishnan S, Miles KM, Ellis L, Sotomayor P, et al. Class I histone deacety- lase inhibitor entinostat suppresses regulatory T cells and enhances immunotherapies in renal and prostate cancer models. PLoS ONE. 2012; 7(1):e30815.; PubMed Central PMCID: PMC3267747. doi: 10.1371/journal.pone.0030815 PMID: 22303460 160. Covre A, Coral S, Nicolay H, Parisi G, Fazio C, Colizzi F, et al. Antitumor activity of epigenetic immu- nomodulation combined with CTLA-4 blockade in syngeneic mouse models. Oncoimmunology. 2015; 4(8):e1019978.; PubMed Central PMCID: PMC4570131. doi: 10.1080/2162402X.2015. 1019978 PMID: 26405573 161. Covre A, Coral S, Di Giacomo AM, Taverna P, Azab M, Maio M. Epigenetics meets immune check- points. Seminars in oncology. 2015; 42(3):506–13. doi: 10.1053/j.seminoncol.2015.02.003 PMID: 25965370 20 / 21 PLOS Genetics | DOI:10.1371/journal.pgen.1006193 October 10, 2016 162. Christiansen AJ, West A, Banks KM, Haynes NM, Teng MW, Smyth MJ, et al. Eradication of solid tumors using histone deacetylase inhibitors combined with immune-stimulating antibodies. Proceed- ings of the National Academy of Sciences of the United States of America. 2011; 108(10):4141–6.; PubMed Central PMCID: PMC3054015. doi: 10.1073/pnas.1011037108 PMID: 21368108 163. Challen GA, Sun D, Jeong M, Luo M, Jelinek J, Berg JS, et al. Dnmt3a is essential for hematopoietic stem cell differentiation. Nature genetics. 2012; 44(1):23–31.; PubMed Central PMCID: PMC3637952. doi: 10.1038/ng.1009 PMID: 22138693 164. Sasaki M, Knobbe CB, Munger JC, Lind EF, Brenner D, Brustle A, et al. IDH1(R132H) mutation increases murine haematopoietic progenitors and alters epigenetics. Nature. 2012; 488(7413):656– 9.; PubMed Central PMCID: PMC4005896. doi: 10.1038/nature11323 PMID: 22763442 165. Silber JR, Bobola MS, Blank A, Chamberlain MC. O(6)-methylguanine-DNA methyltransferase in gli- oma therapy: promise and problems. Biochimica et biophysica acta. 2012; 1826(1):71–82.; PubMed Central PMCID: PMC3340514. doi: 10.1016/j.bbcan.2011.12.004 PMID: 22244911 166. Rodriguez-Paredes M, Esteller M. Cancer epigenetics reaches mainstream oncology. Nature medi- cine. 2011; 17(3):330–9. doi: 10.1038/nm.2305 PMID: 21386836 167. Garnett MJ, Edelman EJ, Heidorn SJ, Greenman CD, Dastur A, Lau KW, et al. Systematic identifica- tion of genomic markers of drug sensitivity in cancer cells. Nature. 2012; 483(7391):570–5.; PubMed Central PMCID: PMC3349233. doi: 10.1038/nature11005 PMID: 22460902 168. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, et al. The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 2012; 483 (7391):603–7.; PubMed Central PMCID: PMC3320027. doi: 10.1038/nature11003 PMID: 22460905 169. Cui H, Cruz-Correa M, Giardiello FM, Hutcheon DF, Kafonek DR, Brandenburg S, et al. Loss of IGF2 imprinting: a potential marker of colorectal cancer risk. Science. 2003; 299(5613):1753–5. doi: 10. 1126/science.1080902 PMID: 12637750 170. Teschendorff AE, Widschwendter M. Differential variability improves the identification of cancer risk markers in DNA methylation studies profiling precursor cancer lesions. Bioinformatics. 2012; 28 (11):1487–94. doi: 10.1093/bioinformatics/bts170 PMID: 22492641 171. Sharma P, Allison JP. Immune checkpoint targeting in cancer therapy: toward combination strategies with curative potential. Cell. 2015; 161(2):205–14. doi: 10.1016/j.cell.2015.03.030 PMID: 25860605 172. Wolchok JD, Weber JS, Maio M, Neyns B, Harmankaya K, Chin K, et al. Four-year survival rates for patients with metastatic melanoma who received ipilimumab in phase II clinical trials. Annals of oncol- ogy: official journal of the European Society for Medical Oncology / ESMO. 2013; 24(8):2174–80.; PubMed Central PMCID: PMC4081656. doi: 10.1093/annonc/mdt161 PMID: 23666915 21 / 21

https://openalex.org/W4220992355

https://hal-insu.archives-ouvertes.fr/insu-03672053/file/Dresing_2022_ApJL_925_L21.pdf

English

The role of coronal shocks for accelerating solar energetic electrons

cc-by

To cite this version: Nina Dresing, Athanasios Kouloumvakos, Rami Vainio, Alexis Rouillard. On the Role of Coronal Shocks for Accelerating Solar Energetic Electrons. The Astrophysical journal letters, 2022, 925, ￿10.3847/2041-8213/ac4ca7￿. ￿insu-03672053￿ Distributed under a Creative Commons Attribution 4.0 International License 1. Introduction the usual presence of flares during CME-associated events make it difficult to separate potential flare-related and shock-related components in particle acceleration (e.g., Kouloumvakos et al. 2015). Interplanetary shocks crossing spacecraft located near 1 au have been shown to be nearly inefficient in accelerating electrons to energies of several tens to hundreds of keV (Tsurutani & Lin 1985; Dresing et al. 2016; Yang et al. 2019). However, exceptional electron-shock spikes of hundreds of keV or even MeV energies were found in Ulysses (Simnett 2003) and Voyager (Sarris & Krimigis 1985) observations at distances of 1.35 and 1.9 au, respectively. Furthermore, Masters et al. (2013) reported that Saturn’s high-Mach-number bow shock accelerates electrons up to MeV energies. It seems therefore possible that specific conditions can be favorable for efficient electron acceleration at shocks (e.g., Mann et al. 2018), and these might also be met close to the Sun. The classic two-type picture of solar energetic particle (SEP) events (Reames 1999, 2021a) assumed two distinct accelera- tion processes for impulsive and gradual events. Impulsive, electron-rich events, which are characterized by their special elemental composition (e.g., rich in 3He; Cane et al. 1986; Reames 2021b), their impulsive time history, and small angular extent, are attributed to particle acceleration from magnetic reconnection in solar jets and flares. Gradual, proton-rich events are believed to be generated by coronal mass ejection (CME) driven shocks that accelerate and inject SEPs over longer times and that produce events with larger angular extents corresponding to the larger acceleration region. It is accepted that energetic electrons, which cause the hard X-Ray (HXR) flare emission, when precipitating back into the solar chromosphere, are accelerated within magnetic reconnections in closed field regions (Mann 2015) to tens of keV. They can also be injected into open magnetic field lines connecting the sites of acceleration in the corona with the heliosphere. These propagating electron beams cause type III radio bursts and can eventually be observed in situ. The connection between the flare and these in situ electrons is supported by strong correlations between the spectral indices of the HXR flare and the corresponding in situ electron event (Krucker et al. 2007; Dresing et al. 2021; Wang et al. 2021). The classic dichotomy of SEP events between impulsive and gradual is, however, not present in all events. Abstract We study the role of coronal mass ejection (CME) driven shocks in the acceleration of solar energetic electrons. Using observations by the two STEREO spacecraft, we correlate electron peak intensities of solar energetic particle events measured in situ with various parameters of the associated coronal shocks. These shock parameters were derived by combining 3D shock reconstructions with global modeling of the corona. This modeling technique provides also shock properties in the specific shock regions that are magnetically connected to the two STEREO spacecraft. We find significant correlations between the peak intensities and the Mach number of the shock with correlation coefficients of about 0.7, which are similar for electrons at ∼1 MeV and protons at >60 MeV. Lower- energy electrons with <100 keV show a smaller correlation coefficient of 0.47. The causal relationship between electron intensities and the shock properties is supported by the vanishing correlations when peak intensities at STEREO A are related with the Alfvénic Mach number at the magnetic footpoint of STEREO B and vice versa, which yields correlation coefficients of 0.03 and −0.13 for ∼1 MeV and <100 keV electron peak intensities, respectively. We conclude that the high-energy electrons are accelerated mainly by the shock, while the low-energy electrons are likely produced by a mixture of flare and shock-related acceleration processes. Unified Astronomy Thesaurus concepts: Active sun (18); Solar coronal mass ejection shocks (1997); Interplanetary particle acceleration (826); Solar flares (1496); Solar energetic particles (1491); Solar particle emission (1517) Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. HAL Id: insu-03672053 https://insu.hal.science/insu-03672053v1 Submitted on 19 May 2022 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License https://doi.org/10.3847/2041-8213/ac4ca7 The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 © 2022. The Author(s). Published by the American Astronomical Society. 13/ac4ca7 On the Role of Coronal Shocks for Accelerating Solar Energetic Electrons Nina Dresing1,2 , Athanasios Kouloumvakos3 , Rami Vainio4 , and Alexis Rouillard3 1 Department of Physics and Astronomy, Turku Collegium for Science, Medicine and Technology, University of Turku, Finland; nina.dresing@utu.fi 2 Institut für Experimentelle und Angewandte Physik, Universität Kiel, D-24118, Kiel, Germany 3 IRAP, Université Toulouse III—Paul Sabatier, CNRS, CNES, Toulouse, France 4 Department of Physics and Astronomy, University of Turku, Finland Received 2021 November 25; revised 2022 January 17; accepted 2022 January 17; published 2022 February 2 pp y p // g/ / / 022. The Author(s). Published by the American Astronomical Society. 1. Introduction 2020, 2021) and therefore prevent a clear shock association to SEE events so far. Long-lasting electron anisotropies marking temporally extended electron injections close to the Sun might be a strong indication for electron-shock acceleration as suggested also by Kahler et al. (2007). However, even such features can be produced by alternative processes, e.g., ongoing acceleration or trapping in post-flare loops (e.g., Klein et al. 2005) or CME−CME interaction regions (e.g., Gopalswamy et al. 2004; Dresing et al. 2018), where continuous leakage from the trap would cause the ongoing electron injection. Dresing et al. (2020) analyzed a large sample of near- relativistic SEE events observed with the two STEREO spacecraft and found evidence for the presence of at least two types of electron events. They found long rise-time events associated with hard spectral indices as well as with the presence of higher-energy (>0.7 MeV) electrons, which could not be explained by a purely flare-related scenario. They, therefore, suggested an additional acceleration mechanism to be involved for these events that provides a prolonged particle injection. In the following section we analyze the relation between electron peak intensities and various shock parameters determined by Kouloumvakos et al. (2019) at the point where field lines passing the observer connect with the shock surface (termed the “cob-point”). For each event, the 3D structure of the shock wave is triangulated from images taken by STEREO and the Solar and Heliospheric Observatory (SOHO; e.g., Rouillard et al. 2016), then a global 3D megnetohydrodynamic (MHD) coronal model (PSI/MAST model; Lionello et al. 2009; Riley et al. 2011) is used to obtain the 3D distribution of basic shock parameters such as Mach numbers, compression ratios, and shock geometry ΘBn, which is defined as the magnetic field obliquity with respect to the shock normal (Kouloumvakos et al. 2019). The global MHD model is also used to derive the time-varying position of the cob-point specifically for each spacecraft measuring SEPs. To investigate the link between the SEP peak intensity and the shock parameters, Kouloumvakos et al. (2019) used the maximum values of the Mach numbers and compression ratios, while, for the shock geometry, they used the ΘBn value at the time when the Alfvén Mach number is maximum (see further details in Section 3 of Kouloumvakos et al. 2019). In our analysis we use the same shock parameter values. 1. Introduction Previous statistical studies highlight the possibility that both flare- or shock-related acceleration processes can contribute to the acceleration of energetic electrons and protons (e.g., Kouloumvakos et al. 2015; Trottet et al. 2015; Papaioannou et al. 2016). Strong correlations between relativistic electron and deka-MeV proton flux increases (Posner 2007) suggest a common, possibly shock-related source of the two species. The velocity dispersion analyses of the 2011 March 21 SEP event, measured along magnetic field lines rooted at an estimated 90° in heliolongitude away from the flare site, gave similar particle release times for electrons and protons, suggesting a common shock origin (see Figure 11 of Rouillard et al. 2012). Also other features of solar energetic electron (SEE) events led authors to attribute them to shocks. These were, for example, gradual rising phases in well-connected events, or the frequently observed onset delays, suggesting a continuous or The role of shocks in accelerating solar energetic electrons, especially to energies of more than ∼100 keV, is, however, still unclear. The close temporal relationship of flares and CMEs, and 1 The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. strophysical Journal Letters, 925:L21 (7pp), 2022 Febru electrons provided by the STEREO High Energy Telescope (HET; von Rosenvinge et al. 2008). For comparison, Table 1 also includes the 60–100 MeV proton peak intensities that were used by Kouloumvakos et al. (2019). The bottom row of the table shows the number of events observed by each of the two STEREO spacecraft in the different channels, showing that the number of associated proton events (in total 54 events) is always higher than the corresponding electron events, which sum up to 50 (47) events at ∼1 MeV (<100 keV). There are various reasons for missing events. In most cases, no corresponding event could be identified. However, a few SEPT electron peak intensities were corrupted by ion contamination (IC), and some events were likely masked by a high pre-event background (High BG). In some other cases, no unambiguous corresponding peak intensity could be determined because of the mixing of events. delayed electron injection at the Sun with respect to the flare, respectively (Haggerty & Roelof 2002, 2009; Kouloumvakos et al. 2015). However, also processes related to particle transport, the electron injection at the Sun, and instrumental effects can be involved in producing such features (e.g., Dresing et al. 1. Introduction The clear shock association of large energetic proton events was underlined by the strong correlation between 20 and 100 MeV proton peak intensities and shock parameters at magnetically well-connected regions to the observing space- craft that Kouloumvakos et al. (2019) determined from 3D reconstructions and modeling of the observed pressure fronts. The strongest correlations were found with the strength of the shock as quantified by its Mach number. In this work we make use of the two latter studies and combine their results. Namely we compare shock parameters for 33 large SEP events modeled by Kouloumvakos et al. (2019) with key observations of SEE events provided by Dresing et al. (2020) to determine the potential role of shocks for electron acceleration. 5 http://www2.physik.uni-kiel.de/stereo/downloads/sept_electron_ events.pdf 2. Data Analysis 2.2. Relating Solar Energetic Electron Intensities to Shock Parameters 2.1. Data Selection Figures 1 and 2 show the correlations between energetic particle peak intensities and shock parameters at the cob-point. Each column shows the correlations of 55–85 keV electrons (left), 0.7–1.4 MeV electrons (center), and 60–100 MeV protons (right). Each row shows from top to bottom the correlations with the Alfvénic Mach Number MA, the shock speed Vsh, and the shock speed at the leading edge of the CME Vsh (LE). Figure 2 shows correlations of the shock geometry ΘBn (top) and the shock connection height (center). Both shock parameters have been determined at the time when the MA is maximum. The bottom row of Figure 2 shows the shock density compression ratio X. Our study is based on the 33 coronal pressure waves modeled in 3D by Kouloumvakos et al. (2019) that occurred between 2011 and 2017. The modeled events were associated with energetic protons clearly detected in at least two locations in the heliosphere, at energies greater than 50 MeV. We compare this data set with the large sample5 (781 events) of near-relativistic (55–85 keV) SEE events observed with the two STEREO spacecraft (Dresing et al. 2020). The two lists were synchronized based on temporal coincidence of the events, and false correlations, e.g., due to mixing of events observed closely in time, were excluded through manual inspection. Table 1 presents the particle peak intensities for each of the 33 SEP events. Peak intensities of the near-relativistic electrons measured by the Solar Electron and Proton Telescope (SEPT; Müller-Mellin et al. 2007) are provided by Dresing et al. (2020). To complement these with relativistic electron measurements, we also determined the peak intensities of 0–7–1.4 MeV Following Kouloumvakos et al. (2019), we determine the Pearson correlation coefficients of the logarithms of the peak intensities with the shock parameters. The logarithm was also used for the Alfvénic Mach Numbers. The legend of each panel in Figure 1 provides the correlation coefficient (cc) together with a determination of its uncertainty using the Fisher r-to-z transformation providing a 90% confidence interval. The lower-most number of each legend provides the p-value of 2 The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. Table 1 Events and Corresponding Peak Intensities (in cm−2 sr−1 s−1 MeV−1) of 55–85 keV Electrons, 0.7–1.4 MeV Electrons, and 60–100 MeV Protons for Events Analyzed by Kouloumvakos et al. 2.1. Data Selection (2019) STA Peak Intensities STB Peak Intensities Date SEPT e HET e HET p SEPT e HET e HET p 2011-02-15 L L L 1.64E+04 3.46E-01 1.20E-02 2011-03-07 8.32E+02 1.32E+00 L 2.36E+04 1.51E+00 3.10E-02 2011-03-21 2.70E+05 1.71E+01 1.15E+00 L L L 2011-08-04 6.58E+02 2.10E-01 L 3.43E+02 1.94E-01 5.00E-03 2011-09-06 2.39E+02 L 3.00E-03 L 1.52E-01 5.00E-03 2011-09-22 IC 5.06E-01 1.10E-02 3.66E+05 3.16E+01 5.15E-01 2011-10-04 1.19E+04 7.90E-01 6.00E-03 4.84E+04 1.13E+00 1.50E-02 2011-11-03 9.55E+04 4.69E+01 2.81E-01 4.25E+03 7.97E-01 1.30E-02 2012-01-23 3.08E+04 2.15E+00 1.00E-02 1.89E+04 1.59E+00 1.60E-02 2012-01-27 IC 9.11E+00 5.10E-02 L L L 2012-03-05 L L L 1.21E+05 1.02E+01 3.70E-02 2012-03-07 2.33E+03 3.19E+00 3.00E-02 3.04E+05 3.37E+02 3.04E+00 2012-03-24 6.17E+04 2.15E+01 2.46E-01 2.75E+03 1.11E+00 8.00E-03 2012-05-17 IC 3.61E-01 8.00E-03 9.43E+01 9.95E-02 3.00E-03 2012-07-23 4.76E+06 7.92E+04 1.35E+01 4.92E+04 6.86E-01 2.00E-02 2012-09-20 2.87E+05 5.31E+01 2.47E-01 3.86E+05 8.73E+00 1.10E-02 2012-09-28 3.16E+05 1.67E+01 1.20E-02 1.75E+04 2.06E+00 1.20E-02 2013-03-05 4.54E+05 4.28E+01 1.73E+00 9.59E+03 2.14E+00 1.90E-02 2013-05-22 1.96E+04 L L 2.53E+03 0.38 L 2013-10-05 4.61E+04 7.17E+00 2.90E-02 IC 2.23E-01 3.00E-03 2013-10-11 9.55E+04 3.00E+01 6.17E-01 9.19E+03 6.02E+00 8.50E-02 2013-10-25 1.38E+02 L L 1.83E+04 1.32E+00 3.50E-02 2013-10-28 8.83E+02 1.23E-01 3.00E-03 8.83E+02 1.14E+00 2.40E-02 2013-11-02 6.48E+04 7.27E+00 2.30E-01 1.15E+03 2.30E-01 8.00E-03 2013-11-07 7.84E+04 4.81E+01 2.72E-01 1.06E+05 4.63E+00 1.14E-01 2013-12-28 IC High BG 1.10E-02 High BG High BG 1.50E-02 2014-01-06 2.35E+03 L 6.00E-03 2.21E+02 1.02E-01 6.00E-03 2014-01-07 L DG 9.00E-03 3.55E+03 8.94E-01 8.00E-03 2014-02-25 5.49E+04 1.72E+01 9.30E-02 9.83E+04 3.76E+01 3.83E-01 2014-03-05 L L 2.90E-02 L L 7.60E-02 2014-09-01 DG DG DG 8.35E+05 1.91E+03 1.89E+01 2014-09-10 DG DG DG 1.63E+03 1.12E+00 8.00E-03 2017-09-10 4.53E+05 2.94E+01 4.50E-02 L L L Total number 21 22 25 26 28 29 Note. DG: data gap; IC: ion contamination; High BG: high pre-event background. Table 1 Events and Corresponding Peak Intensities (in cm−2 sr−1 s−1 MeV−1) of 55–85 keV Electrons, 0.7–1.4 MeV Electrons, and 60–100 MeV Protons for Events Analyzed by Kouloumvakos et al. (2019) Note. DG: data gap; IC: ion contamination; High BG: high pre-event background. Note. DG: data gap; IC: ion contamination; High BG: high pre-event background. shock speed, the correlation with the low-energy electrons gets stronger, when only considering events with peak intensities above ∼104 cm−2 sr−1 s−1 MeV−1 (see Figure 1, left column). The correlations with the shock speed at the magnetic footpoints of the spacecraft (Figure 1, center row) are clearly stronger than the correlations with global value of the shock speed at the leading edge of the CME (Figure 1, bottom row). the statistics. 2.1. Data Selection All correlation parameters are summarized in Table 2, which also includes the correlations with the fast magnetosonic Mach number Mfm. To allow for a qualitative comparison of the correlations of the different particle energies and species, we reduced the near- relativistic and the proton event sample to those events that were observed in the relativistic electron channel. Because the lower-energy electrons show fewer events in correspondence to the analyzed sample, the statistics are slightly worse. The number of associated proton events is the largest among the three analyzed groups, so that the sample used for the correlation was reduced. p g g ( g , ) No correlation is found between the SEP peak intensities and the shock obliquity ΘBn (Figure 2, top) and also no differing trend between electrons and protons can be identified; both species show a tendency of the highest peak intensities of the sample to occur quasi-parallel to oblique shocks (0 < ΘBn <40). The center and bottom rows of Figure 2 show that the correlations with the shock connection height and with the density compression ratio are weak, with even smaller numbers for the 55–85 keV electrons. The bottom rows of Table 2 show that the correlations of protons with electron peak intensities are stronger than those with the shock parameters. While for the near-relativistic electrons the correlations weaken with increasing proton energy, it is the other way around for the relativistic electrons. No correlation is found between the SEP peak intensities and the shock obliquity ΘBn (Figure 2, top) and also no differing trend between electrons and protons can be identified; both species show a tendency of the highest peak intensities of the sample to occur quasi-parallel to oblique shocks (0 < ΘBn <40). The center and bottom rows of Figure 2 show that the correlations with the shock connection height and with the density compression ratio are weak, with even smaller numbers for the 55–85 keV electrons. Figure 1 and Table 2 show that the strongest correlations (cc = 0.72) are found for protons and the Alfvénic Mach number, followed by the fast magnetosonic Mach Number of cc = 0.65. While the correlation coefficients of high-energy electrons and protons are similar within the limits of uncertainties, the correlations of shock parameters with near- relativistic electron intensities are always weaker. 2.1. Data Selection However, in case of the magnetosonic and Alfvénic Mach Numbers and the The bottom rows of Table 2 show that the correlations of protons with electron peak intensities are stronger than those with the shock parameters. While for the near-relativistic electrons the correlations weaken with increasing proton energy, it is the other way around for the relativistic electrons. 3 The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Figure 1. Correlation of SEP peak intensities with various shock parameters. From left to right: peak intensities of <100 keVelectrons, ∼1 MeV electrons, and 60–100 MeV protons. From top to bottom: correlations with Alfvénic Mach number MA, shock speed Vsh, and the shock speed at the leading edge of the CME Vsh (LE). The Pearson correlation coefficient cc is provided in each figure legend together with its uncertainty range corresponding to a 90% confidence interval (in square brackets) and the corresponding p-value (in round brackets). The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. Dresing et al. Figure 1. Correlation of SEP peak intensities with various shock parameters. From left to right: peak intensities of <100 keVelectrons, ∼1 MeV electrons, and 60–100 MeV protons. From top to bottom: correlations with Alfvénic Mach number MA, shock speed Vsh, and the shock speed at the leading edge of the CME Vsh LE). The Pearson correlation coefficient cc is provided in each figure legend together with its uncertainty range corresponding to a 90% confidence interval (in square brackets) and the corresponding p-value (in round brackets). Figure 1. Correlation of SEP peak intensities with various shock parameters. From left to right: peak intensities of <100 keVelectrons, ∼1 MeV electrons, and 60–100 MeV protons. From top to bottom: correlations with Alfvénic Mach number MA, shock speed Vsh, and the shock speed at the leading edge of the CME Vsh (LE). The Pearson correlation coefficient cc is provided in each figure legend together with its uncertainty range corresponding to a 90% confidence interval (in square brackets) and the corresponding p-value (in round brackets). 2.3. The Relation of Electron Event Spectra with Shock Parameters figure shows that the events analyzed here belong to those with the hardest energy spectra in the whole sample of electron events observed with STEREO during solar cycle 24. Almost all of the near-relativistic events are accompanied by relativistic electrons (see Table 1), proving the presence of efficient electron acceleration. Dresing et al. (2020) focused on electron spectra in the near- relativistic range (45–425 keV), which often show broken- power-law shapes, and determined the spectral index at two reference energies (70 and 200 keV) for all events in their sample. No clear correlations were found between these spectral indices and the shock parameters (not shown), for example, a spectral hardening related with stronger shocks, which would explicitly point to shock acceleration. However, we suspect that this missing correlation is caused by a selection effect, as illustrated by Figure 3. It presents the histograms of the two reference spectral indices for the whole sample as in Dresing et al. (2020; blue) compared to the distribution of spectral indices of the 33 events analyzed in this study. The 3. Discussion and Conclusions Our analysis shows a clear correlation between the peak intensities of SEEs and key parameters of modeled shocks at the relevant cob-points. We analyzed the peak intensities of near-relativistic (55–85 keV) and relativistic (0.7–1.4 MeV) electrons and compared their correlations with several modeled shock parameters, such as Alfvénic and magnetosonic Mach numbers, shock speed (at the spacecraft magnetic footpoint and 4 The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. Figure 2. As in Figure 1 but for the magnetic field obliquity with respect to the shock normal ΘBn (top), the shock connection height above the photosphere (center), and the density compression ratio X (bottom). The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. Figure 2. As in Figure 1 but for the magnetic field obliquity with respect to the shock normal ΘBn (top), the shock connection height above the photosphere (center), and the density compression ratio X (bottom). the CME leading edge), magnetic field obliquity with respect to the shock normal (ΘBn), the height of shock at the time of maximum Mach number, and the density compression ratio. expected from shock acceleration theory. Electrons should be preferentially accelerated by shock-drift acceleration at quasi- perpendicular shocks and protons by diffusive-shock accelera- tion at quasi-parallel shocks, respectively (e.g., Jokipii 1987; Burgess 2006). A rapidly changing shock geometry could induce important time-dependent effects of the acceleration process that are hard to assess without a detailed modeling effort, as indeed for most events the shock transits from an initially quasi-perpendicular to a quasi-parallel in the upper corona. y p As in Kouloumvakos et al. (2019) for protons, the strongest correlations occur with the Mach number; the Alfvénic Mach number yields the strongest correlations of cc = 0.69 for relativistic and cc = 0.47 for near-relativistic electrons, respec- tively. The correlations are weaker between near-relativistic electrons, with all other shock parameters as seen for 60–100 MeV protons. No correlation is found between peak intensities and shock obliquity, although there is a tendency for the highest electron peak intensities to occur during quasi- parallel to oblique shock. Our results show that the role of the shock for relativistic electrons is significant and of the same importance as for the high-energy protons, suggesting a common, shock-related acceleration process. Table 2 Table 2 Pearson Correlation Coefficients between SEP Peak Intensities and Different Shock Parameters Pearson Correlation Coefficients between SEP Peak Intensities and Different Shock Parameters Shock Parameters CCs SEPT e HET e HET p (55–85 keV) (0.7–1.4 MeV) (60–100 MeV) (1) (2) (3) (4) MA 0.47 (1.46E-03) 0.69, (2.29E-08) 0.72, (1.64E-08) [0.34, 0.57] [0.62, 0.76] [0.64, 0.78] Mfm 0.43 (3.88E-03) 0.61, (2.48E-06) 0.65, (1.00E-06) [0.3, 0.54] [0.52, 0.69] [0.56, 0.73] Vsh 0.33 (2.84E-02) 0.53, (7.08E-05) 0.57, (2.98E-05) [0.19, 0.46] [0.43, 0.62] [0.47, 0.66] Vsh (LE) 0.24 (1.20E-01) 0.35, (1.36E-02) 0.28, (5.53E-02) [0.09, 0.37] [0.22, 0.46] [0.15, 0.41] ΘBn −0.01 (9.26E-01) 0.04, (7.98E-01) −0.06, (6.90E-01) [−0.16, 0.13] [−0.1, 0.17] [−0.2, 0.08] Hc 0.38 (1.07E-02) 0.47, (6.58E-04) 0.54, (1.07E-04) [0.25, 0.5] [0.35, 0.57] [0.43, 0.63] X 0.24 (1.21E-01) 0.4, (3.94E-03) 0.47, (8.97E-04) [0.09, 0.37] [0.28, 0.51] [0.35, 0.58] HET p 0.86 (5.87E-14) 0.84, (4.22E-14) L (20 MeV) [0.82, 0.9] [0.79, 0.87] L HET p 0.82 (1.47E-11) 0.92, (2.55E-21) L (40 MeV) [0.77, 0.86] [0.9, 0.94] L HET p 0.76 (1.28E-08) 0.91, (5.17E-18) L (60 MeV) [0.69, 0.82] [0.88, 0.93] L Note. P-values of the correlation coefficients are provided in parentheses, uncertainties of cc corresponding to a 90% confidence interval are shown below in square brackets. The three rows at the bottom show the correlation of proton peak intensities above 20, 40, and 60 MeV with those of the near- relativistic and relativistic electrons. One might argue that the observed correlations with the shock parameters are only apparent because of the big flare syndrome (Kahler 1982), which postulates that, statistically, various solar activity phenomena are more intense during large flares. Therefore, a correlation with the CME or the CME- driven shock parameters might be found only because they correlate with the strength of the flare. We note, however, that this possibility is highly unlikely. The correlations found in this study are determined by the specific shock parameters at the cob-point. Correlating, for example, peak intensities observed at STEREO A with the Alfvénic Mach number determined at the magnetic footpoint of STEREO B yields completely vanishing correlations with cc = 0.03 and cc = −0.13 for ∼1 MeV and <100 keV electron peak intensities, respectively (not shown). Table 2 Similarly, the shock parameters of STEREO A and STEREO B of the same events show only very weak correlations with each other (e.g., cc = 0.17 for the Alfvénic Mach number), and correlating the peak intensities with the global parameter of the shock speed at the CME leading edge (Figure 1, bottom row) yields smaller correlation coefficients than the correlations with the shock speeds at the respective magnetic footpoints (Figure 1, mid-row). In addition, we find a very weak correlation between the analyzed peak intensities and the strength of the flare. Correlation of the peak flare intensity measured in soft X-rays by the GOES spacecraft yields correlation coefficients of 0.26, 0.19, and 0.12 for 60–100 MeV protons, ∼1 MeV electrons, and <100 keV SEPT electrons (not shown). However, the statistical sample for these correlations is lower because the flare was only observed during 20 out of the 33 analyzed events. Note. P-values of the correlation coefficients are provided in parentheses, uncertainties of cc corresponding to a 90% confidence interval are shown below in square brackets. The three rows at the bottom show the correlation of proton peak intensities above 20, 40, and 60 MeV with those of the near- relativistic and relativistic electrons. Note. P-values of the correlation coefficients are provided in parentheses, uncertainties of cc corresponding to a 90% confidence interval are shown below in square brackets. The three rows at the bottom show the correlation of proton peak intensities above 20, 40, and 60 MeV with those of the near- relativistic and relativistic electrons. (Table 2). Because of the weaker correlations between the shock parameters and the near-relativistic electrons, we conclude that the role of the shock is smaller at these energies and another process must be involved in accelerating these lower-energy electrons, which is most likely flare-related (Dresing et al. 2021). The stronger correlation of near-relativistic electron intensities with the lower-energy proton intensities (e.g., HET-protons at 20 MeV; see Table 2) suggests that this other process could possibly contribute to those lower-energy proton fluxes as well. Figure 1 shows that the correlation of the near-relativistic (Table 2). Because of the weaker correlations between the shock parameters and the near-relativistic electrons, we conclude that the role of the shock is smaller at these energies and another process must be involved in accelerating these lower-energy electrons, which is most likely flare-related (Dresing et al. 2021). 3. Discussion and Conclusions This is also supported by the strong correlations between electron and proton peak intensities that are stronger for the relativistic than for the near-relativistic electrons Surprisingly, no difference between electrons and protons can be observed with respect to shock obliquity that would be 5 Figure 3. Histograms of spectral indices at reference energies of 70 keV (left) and 200 keV (right). Blue histograms correspond to the distribution of all STEREO/ SEPT SEE events (Dresing et al. 2020), while magenta histograms represent the sample under study here. The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. Figure 3. Histograms of spectral indices at reference energies of 70 keV (left) and 200 keV (right). Blue histograms correspond to the distribution of all STEREO/ SEPT SEE events (Dresing et al. 2020), while magenta histograms represent the sample under study here. Figure 3. Histograms of spectral indices at reference energies of 70 keV (left) and 200 keV (right). Blue histograms corr SEPT SEE events (Dresing et al. 2020), while magenta histograms represent the sample under study here. electrons with the Alfvénic Mach Number seems to improve if only peak intensities above ∼104 cm−2 sr−1 s−1 MeV−1 are considered. electrons with the Alfvénic Mach Number seems to improve if only peak intensities above ∼104 cm−2 sr−1 s−1 MeV−1 are considered. References Burgess, D. 2006, ApJ, 653, 316 Cane, H. V., McGuire, R. E., & von Rosenvinge, T. T. 1986, ApJ, 301, 448 Dresing, N., Effenberger, F., Gómez-Herrero, R., et al. 2020, ApJ, 889, 143 Dresing, N., Effenberger, F., Gómez-Herrero, R., et al. 2020, ApJ, 889, 143 g g p Dresing, N., Gómez-Herrero, R., Heber, B., et al. 2018, A&A, 613, A21 esing, N., Theesen, S., Klassen, A., & Heber, B. 2016, A&A, Dresing, N., Warmuth, A., Effenberger, F., et al. 2021, A&A, 654, A92 Gopalswamy, N., Yashiro, S., Krucker, S., Stenborg, G., & Howard, R. A 2004, JGRA, 109, A12105 Haggerty, D. K., & Roelof, E. C. 2002, ApJ, 579, 841 Haggerty, D. K., & Roelof, E. C. 2009, in AIP Conf. Ser. 1183, SHOCK WAVES IN SPACE AND ASTROPHYSICAL ENVIRONMENTS: 18th Annual International Astrophysics Conference, Probing SEP Acceleration Processes With Near-relativistic Electrons, ed. X. Ao & G. Z. R. Burrows (New York: AIP), 3 gy p We conclude that the relativistic SEE events in the MeV range, studied here, are mainly produced by coronal shock acceleration, with the shock being as important as it is for solar energetic proton events in the >60 MeV range. However, we do not exclude the possibility that a flare-related process can also contribute to the acceleration with a smaller importance. For the SEEs at lower energies, we conclude that this population is probably a mixture of flare and shock-accelerated electrons with a larger contribution of the flare component compared to the higher-energy electrons. Jokipii, J. R. 1987, ApJ, 313, 842 Kahler, S. W. 1982, JGR, 87, 3439 Kahler, S. W. 2007, SSRv, 129, 359 Kahler, S. W., Aurass, H., Mann, G., & Klassen, A. 2007, ApJ, 656, 567 Klein, K. L., Krucker, S., Trottet, G., & Hoang, S. 2005, A&A, 431, 1047 Kouloumvakos, A., Rouillard, A. P., Wu, Y., et al. 2019, ApJ, 876, 80 Krucker, S., Kontar, E. P., Christe, S., & Lin, R. P. 2007, ApJL, 663, L109 Lionello, R., Linker, J. A., & Mikić, Z. 2009, ApJ, Mann, G. 2015, JPlPh, 81, 475810601 Mann, G. 2015, JPlPh, 81, 475810601 Mann, G., Melnik, V. N., Rucker, H. O., Konovalenko, A. A., & Brazhenko, A. I. 2018, A&A, 609, A41 This study was funded by the European Union’s Horizon 2020 research and innovation program under grant agreement No. 101004159 (SERPENTINE). Table 2 The stronger correlation of near-relativistic electron intensities with the lower-energy proton intensities (e.g., HET-protons at 20 MeV; see Table 2) suggests that this other process could possibly contribute to those lower-energy proton fluxes as well. Figure 1 shows that the correlation of the near-relativistic No clear correlations were found between the shock parameters and the spectral indices of the near-relativistic (45–425 keV) electron events that are assumed to represent the efficiency of the underlying acceleration process. This missing correlation could be caused, on the one hand, by transport effects that modify the energy spectrum, e.g., particle scattering at magnetic field irregularities as described by Strauss et al. (2020). On the other hand, the analyzed event sample might suffer a 6 The Astrophysical Journal Letters, 925:L21 (7pp), 2022 February 1 Dresing et al. strophysical Journal Letters, 925:L21 (7pp), 2022 Febru Rami Vainio https://orcid.org/0000-0002-3298-2067 Alexis Rouillard https://orcid.org/0000-0003-4039-5767 Rami Vainio https://orcid.org/0000-0002-3298-2067 Alexis Rouillard https://orcid.org/0000-0003-4039-5767 selection effect, which is demonstrated when comparing the spectral indices of the events studied here with the distribution of spectral indices of all SEE events observed with the STEREO spacecraft in solar cycle 24 (Dresing et al. 2020). This comparison (Figure 3) shows that the events studied here are among the most energetic events of the cycle, shown by the hardest electron spectra observed in the whole sample. This is also underlined by the presence of relativistic electrons for almost all events studied here. For comparison, only about 20% of all STEREO/SEPT electron events were accompanied by relativis- tic electrons (Dresing et al. 2020). New spacecraft measurements with higher energy resolution as provided, for example, by the Solar Orbiter Energetic Particle Detector suite (Rodríguez- Pacheco et al. 2020) might help to identify specific footprints of the shock in the electron energy spectra. References Work in the University of Turku was performed under the umbrella of Finnish Centre of Excellence in Research of Sustainable Space. N.D. is grateful for support by the Turku Collegium for Science, Medicine and Technology of the University of Turku, Finland. A.K. acknowledges financial support from the SERPENTINE project and the ANR COROSHOCK project (ANR-17-CE31-0006-01). We also acknowledge support from the French space agency Centre National des Etudes Spatiales (CNES, https://cnes.fr/fr) that supports the Solar-Terrestrial Observations and Modeling Service (STORMS, http://storms- service.irap.omp.eu/) exploited for the modeling part of this study. We also thank Predictive Sciences Inc. for making their 3D MHD model continuously available through their website. Masters, A., Stawarz, L., Fujimoto, M., et al. 2013, NatPh, 9, 164 Masters, A., Stawarz, L., Fujimoto, M., et al. 2013, NatPh, 9, 164 Müller-Mellin, R., Böttcher, S., Falenski, J., et al. 2007, SSSRv, 136, 371 Müller-Mellin, R., Böttcher, S., Falenski, J., et al. 2007, SSSRv, 136, 371 Papaioannou, A., Sandberg, I., Anastasiadis, A., et al. 2016, JSWSC, 6, A42 Posner, A. 2007, SpWea, 5, 05001 Reames, D. V. 1999, SSRv, 90, 413 Reames, D. V. 2021a, Solar Energetic Particles. A Modern Primer on Understanding Sources, Acceleration and Propagation, 978 (Cham: Springer) Reames, D. V. 2021b, FrASS, 8, 164 Reames, D. V. 2021b, FrASS, 8, 164 Riley, P., Lionello, R., Linker, J. A., et al. 2011, SoPh, 274, 361 Rodríguez-Pacheco, J., Wimmer-Schweingruber, R. F., Mason, G. M., et al. 2020, A&A, 642, A7 Rouillard, A. P., Plotnikov, I., Pinto, R. F., et al. 2016, ApJ, 833, 45 Rouillard, A. P., Sheeley, N. R., Tylka, A., et al. 2012, ApJ, 752, 44 y y p Sarris, E. T., & Krimigis, S. M. 1985, ApJ, 298, 676 Sarris, E. T., & Krimigis, S. M. 1985, ApJ, Simnett, G. M. 2003, SoPh, 213, 387 Strauss, R. D., Dresing, N., Kollhoff, A., & Brüdern, M. 2020, ApJ, 897, 24 Trottet, G., Samwel, S., Klein, K. L., Dudok de Wit, T., & Miteva, R. 2 S Ph 290 819 Trottet, G., Samwel, S., Klein, K. L., Dudok de Wit, T., & Miteva, R. 2015, SoPh, 290, 819 SoPh, 290, 819 Tsurutani, B. T., & Lin, R. P. 1985, JGR, 90, 1 von Rosenvinge, T. T., Reames, D. V., Baker, R., et al. 2008, SSRv, 136, 391 Wang, W., Wang, L., Krucker, S., et al. 2021, ApJ, 913, 89 Yang, L., Wang, L., Li, G., et al. 2019, ApJ, 875, 104 Nina Dresing https://orcid.org/0000-0003-3903-4649 Athanasios Kouloumvakos https://orcid.org/0000-0001- 6589-4509 g g p Yang, L., Wang, L., Li, G., et al. 2019, ApJ, 875, 104

https://openalex.org/W2898938897

https://repositorio.uam.es/bitstream/10486/706215/1/collective_gordo_SLLERJ_2018.pdf

English

Collective transnational bargaining: practical implementation experiences from European Works Councils in Spain

Spanish labour law and employment relations journal

cc-by

1 In 1983, one year after the second Auroux Law came into force in France, the transnational company Saint-Gobain in­ formally extended its group works council to workers in other workplaces located in other States. The companies BSN, Bull and Scansped helped consolidate this process between 1984 and 1989. However, in 1985 the company Thomson constituted what is considered to be the first European Works Council. Later would come councils at Volkswagen, with the German ID Metall union, and that of ELF Aquitaine, through an agreement with French unions. Regarding the evolution of this first phase of European Works Councils, see: KÖLER, Holm-Detlev, GONZÁLEZ BEGEGA, Sergio, “¿Hacia un sistema de relaciones industriales europeo? La experiencia de los Comités de Empresa Europeos (CEUs)”, in Cuadernos de Relaciones Laborales, no. 1, 2004, p. 9 to 10; BOGONI, Milena, El Comité de Empresa Europeo, Albacete, Bomarzo, 2010, p. 9; ANTONELLO BENITES, Flavio, “Los Comités de Empresa Europeos y los cambios organizativos en los centros de trabajo”, in AA.VV. (Coord. Rodríguez Piñero-Royo, Miguel C.), El empleador en el derecho del trabajo. XVI Jornadas Universitarias Andaluzas de Derecho del Trabajo y Relaciones Laborales, Tecnos/Consejo Andaluz de Relaciones Laborales, Madrid, 1999, p. 257. For an in-depth analysis of the European Works Council at Saint-Gobain, see: GONZÁLEZ BEGEGA, Sergio, Empresa transna­ cional y nuevas relaciones laborales: La experiencia de los comités de empresa europeos, Catarata, Madrid, 2011, p.197–234. 2 Draft Directive dated 24 October 1980, on procedures for informing and consulting employees, known as the “Vredeling Directive” in honour of the Dutch Commissioner who proposed it (OJ no C 297 of 15 11 80 p 3) The aim of this proposal * This article forms part of the research project “El dialogo social en un derecho del trabajo para la recuperación: competi­ tividad empresarial y cohesión social” (DER2015-64676-C2-2-P). Labour Law, Economic Changes and New Society http://www.dtcens.com Labour Law, Economic Changes and New Society http://www.dtcens.com Labour Law, Economic Changes and New Society http://www.dtcens.com Labour Law, Economic Changes and New Society http://www.dtcens.com Collective transnational bargaining: practical implementation experiences from European Works Councils in Spain* Luis Gordo González Assistant Professor of Labour and Social Security Law Universidad Autónoma de Madrid Recibido: 10.6.2018 Aceptado: 7.7.2018 DOI: https://doi.org/10.20318/sllerj.2018.4435 Abstract: Council Directives 94/45/EC and 2009/38/EC impose transnational collective bargain­ ing in Community-scale undertakings so as to create procedures for informing and consulting employees in said undertakings. More than twenty years after the first Directive was passed, this article examines the agreements reached by European-scale companies with headquarters in Spain and tries to construct a typical model for European Works Councils among companies in Spain. Keywords: European Works Council, EWC, collective transnational bargaining. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 y p p p 2 Draft Directive dated 24 October 1980, on procedures for informing and consulting employees, known as Directive” in honour of the Dutch Commissioner who proposed it. (OJ no. C 297, of 15.11.80, p. 3). The aim * This article forms part of the research project “El dialogo social en un derecho del trabajo para la recuperación: competi­ tividad empresarial y cohesión social” (DER2015-64676-C2-2-P). I. Introduction The first European experiences with regulating workers' transnational representation dates back to the 1980s. These initial practices, which were developed in France and Germany as a result of the in­ ternationalization of their companies,1 extended worker representation to the workforce in other States, using the regulations for determining workers' representation in company groups on the national scale.i The French and German initiatives encouraged debate in the European Union, which, after a first at­ tempt at a Directive (the Vredeling proposal of 20 October, 19802), finally passed Council Directive 94/45/ EC of 22 September on the establishment of a European Works Council or a procedure in Community- 41 Collective transnational bargaining: practical implementation experiences from European Gordo González scale undertakings and Community-scale groups of undertakings for the purposes of informing and con­ sulting employees. This regulation would later be updated by Council Directive 2009/38/EC, of 6 May.3 As is well known, both Directive 94/45 and the subsequent Directive 2009/38/EC stipulate that Community-scale undertakings must approve mechanisms for informing and consulting their workers,4 either through specific procedures or through the creation of a European Works Council (hereinafter EWC). As is well known, both Directive 94/45 and the subsequent Directive 2009/38/EC stipulate that Community-scale undertakings must approve mechanisms for informing and consulting their workers,4 either through specific procedures or through the creation of a European Works Council (hereinafter EWC). European regulations are characterised by giving a central role to collective autonomy and to the member States: the central role given to collective autonomy is because national and European regulations will only be applied if no agreement can be reached between the parties. The importance of national legis­ lation arises from the fact that it determines important issues such as the concept of worker, the formula used to calculate the number of workers, the guarantees for worker representatives, Council operation costs, or sanctions for non-compliance.5 Thus, we can say that the Directives have attempted to establish a minimum baseline that can be improved upon by agreements between the parties or by national legislation. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 I. Introduction Analysing the success or failure of the implementation of information and consultation processes in Europe is a difficult task given that, although unions have demanded the creation of a public Communi­ ty registry on several occasions, we do not have official data on a European level regarding collective bar­ gaining processes that lead to the approval of ad hoc information and consultation processes or an EWC. i The only unofficial data is provided by the European Trade Union Institution (hereinafter ETUI) database.6 The problem with this database is that, precisely due to its unofficial nature, it may contain agreements that have since been terminated by one of the parties, or may omit current agreements if the ETUI is not aware of their approval.7 While this data must be considered with a certain degree of caution, it may be able to show how information and consultation processes are evolving in Europe. At this time, 1,127 companies have an agreement in force that approves the creation of an EWC,8 which represents just over 50% of the 2,492 Community-scale undertakings identified by the ETUI. The was to achieve coordination between the national legislatures so as to limit or prevent cases of social dumping. When blocked by business, a new draft Directive, known as “Vredeling-Richard”, was presented on 13 July, 1983. This draft was much less ambi­ tious, and did not have the support of management (the modified proposal was published in OJ no. C 217, of 12/12/1983, p. 3). 3 The modifications made under Directive 2009/38/EC have been, without a doubt, the most significant. However, Directive 94/45/EC was also modified by Directive 97/74/EC, of 15 December, 1997 (OJ L 10, of 16 January 1998), which expanded the legal scheme set out in Directive 94/45 to include the United Kingdom of Great Britain and Northern Ireland, States which had been excluded from application until that time.i pp 4 Determining whether a European undertaking or group of undertakings has a Community scale is not without difficulties. However, as the aim of this article is not to provide a detailed analysis of the problems arising from determining European scale, we can succinctly summarize that Community-scale undertakings are those undertakings or groups of undertakings that employ 1,000 workers or more across all member States and which provide employment for at least 150 or more workers in two group workplaces or companies located in two different States. I. Introduction Furthermore, in order to consider that several undertakings are related within one group, one of these undertakings must exercise a position of control over the others. An undertaking exercises control over another when it has developed a dominating influence, for example, for reasons of ownership, financial participation or articles of association. Unless otherwise proven, it shall be assumed that one undertaking exercises control over another when: it holds the majority of the undertaking's subscribed capital, it has the majority of the votes corresponding to the shares issued by the undertaking, or it can name more than half of the members of the undertaking's board of directors or management board. If, in accordance with the above rules, several undertakings could be said to exercise control over another, it will be assumed, unless otherwise proven, that the undertaking that exercises control over the group is that which can name more than half of the members of the board of directors or management board. Control may be exercised directly or indirectly, through other controlled companies. i 4 Determining whether a European undertaking or group of undertakings has a Community scale is not without difficulties. However, as the aim of this article is not to provide a detailed analysis of the problems arising from determining European scale, we can succinctly summarize that Community-scale undertakings are those undertakings or groups of undertakings that employ 1,000 workers or more across all member States and which provide employment for at least 150 or more workers in two group workplaces or companies located in two different States. Furthermore, in order to consider that several undertakings are related within one group, one of these undertakings must exercise a position of control over the others. An undertaking exercises control over another when it has developed a dominating influence, for example, for reasons of ownership, financial participation or articles of association. Unless otherwise proven, it shall be assumed that one undertaking exercises control over another when: it holds the majority of the undertaking's subscribed capital, it has the majority of the votes corresponding to the shares issued by the undertaking, or it can name more than half of the members of the undertaking's board of directors or management board. was to achieve coordination between the national legislatures so as to limit or prevent cases of social dumping. When blocked by business, a new draft Directive, known as “Vredeling-Richard”, was presented on 13 July, 1983. This draft was much less ambi­ tious, and did not have the support of management (the modified proposal was published in OJ no. C 217, of 12/12/1983, p. 3). I. Introduction If, in accordance with the above rules, several undertakings could be said to exercise control over another, it will be assumed, unless otherwise proven, that the undertaking that exercises control over the group is that which can name more than half of the members of the board of directors or management board. Control may be exercised directly or indirectly, through other controlled companies. i 5 This double subordination has been defined graphically as vertical subordination (European legislation - national legisla­ tion) and horizontal subordination (European legislation - agreements between the parties), see: the report prepared by the Eco­ nomic and Social Committee (ESC) on the “Draft Law on workers' rights to information and consultation in Community-scale undertakings and groups of undertakings” (EESC plenary session of 24 January 1996), p. 6. 5 This double subordination has been defined graphically as vertical subordination (European legislation - national legisla­ tion) and horizontal subordination (European legislation - agreements between the parties), see: the report prepared by the Eco­ nomic and Social Committee (ESC) on the “Draft Law on workers' rights to information and consultation in Community-scale undertakings and groups of undertakings” (EESC plenary session of 24 January 1996), p. 6. 6 The datebase could be found at: http://www.ewcdb.eu. 7 One of the changes included in Directive 2009/38/EC was that the parties must notify European workers' and employer organizations of the start and composition of bargaining commissions (Directive art. 5.2.c). However, this obligation is limited to formally communicating the beginning of negotiations, but not the result. Furthermore, it should be kept in mind that nego­ tiations can last for a maximum of three years and thus, over such a long period of time, the conclusion of the negotiations may go unnoticed by the ETUI. 7 One of the changes included in Directive 2009/38/EC was that the parties must notify European workers' and employer organizations of the start and composition of bargaining commissions (Directive art. 5.2.c). However, this obligation is limited to formally communicating the beginning of negotiations, but not the result. Furthermore, it should be kept in mind that nego­ tiations can last for a maximum of three years and thus, over such a long period of time, the conclusion of the negotiations may go unnoticed by the ETUI. 3 The modifications made under Directive 2009/38/EC have been, without a doubt, the most significant. However, Directive 94/45/EC was also modified by Directive 97/74/EC, of 15 December, 1997 (OJ L 10, of 16 January 1998), which expanded the legal scheme set out in Directive 94/45 to include the United Kingdom of Great Britain and Northern Ireland, States which had been excluded from application until that time.i was to achieve coordination between the national legislatures so as to limit or prevent cases of social dumping. When blocked by business, a new draft Directive, known as “Vredeling-Richard”, was presented on 13 July, 1983. This draft was much less ambi­ tious, and did not have the support of management (the modified proposal was published in OJ no. C 217, of 12/12/1983, p. 3). 3 The modifications made under Directive 2009/38/EC have been, without a doubt, the most significant. However, Directive 94/45/EC was also modified by Directive 97/74/EC of 15 December 1997 (OJ L 10 of 16 January 1998) which expanded the I. Introduction 8 Barely 1% of the total has chosen to constitute specific agreements instead of an EWC, which is why we will generally speak about agreements constituting an EWC. 8 Barely 1% of the total has chosen to constitute specific agreements instead of an EWC, which is why we will generally speak about agreements constituting an EWC. 42 Collective transnational bargaining: practical implementation experiences from European Gordo González majority of these agreements have been reached in the metal and chemical sectors, which is not surpri­ sing given that these sectors have a strong history of union activity and are characterised by large-scale companies. If we look at EWCs according to the headquarters' country of origin, it is also not surprising to see that the countries with the greatest number of EWCs are, in decreasing order: Germany, USA, France, and the UK. They are, clearly, countries with a large number of Community-scale undertakings. 10 In the implementation of Directives 94/45 and 2009/38/EC, Spanish legislators introduced an additional requirement. Law 10/1997 of 24 April, article 13, referring to article 90 of the Workers' Statute, establishes that the agreements must be presented to the competent labour authority for their registration, filing, and publication in the BOE. This requirement is fully consistent with the regulations regarding Spain's collective agreements for general efficiency or ergea omnes. However, four companies have reached agreements to constitute an EWC which have not been published in the BOE: Grupo Gestamp (29 January 2015), Grupo Ferroatlántica (4 February 2014), Grupo AMADEUS (25 May 2013) and Grupo Campofrío (4 March 2009). Publishing the collective agreements, as established in article 90 of the Workers' Statute, aims to guarantee that general knowledge of the terms agreed upon by the parties is made available to third parties, without official publication depriving the agreement of effectiveness. Several legal rulings have been made along these lines: Ruling of the National Court of Spain dated 29 October 1998 (Proc. 77/1998) Court Opinion no. 4; Ruling of the High Court of Justice of Madrid dated 14 September 2012 (R. 4089/2012) Court Opinion no. 25; Ruling of the High Court of Justice of Asturias dated 3 June 2003 (R. 2968/2001) Court Opinion no. 2. 11 Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 9 Law 10/1997, of 24 April, was subsequently amended by Law 44/1999, of 29 November, so as to adapt its content to the extension of Directive 94/45/EC to the United Kingdom; and again by Law 10/2011, of 19 May, to incorporate the amendments set out in Directive 2009/38/EC. II. The Spanish model of European Works Council The content of Directive 94/45 was implemented in Spain through Law 10/1997, of 24 April, regarding workers' rights to information and consultation in Community-scale companies and groups of companies.9 From that time on, representatives for workers and management in European-scale compa­ nies with their headquarters in Spain or with headquarters in a non-EU member state, but designating Spain as their headquarters in the EU, were able to create an EWC in accordance with Community regulations.i Of the 56 Community-scale companies in Spain identified by the ETUI, only fourteen have rea­ ched agreements to incorporate an EWC that have been published in the Official State Gazette (BOE, using its acronym in Spanish).10 Without a doubt, the results are disappointing. Such a small number of agreements may be explained by the low number of Community-scale companies in Spain. Spanish companies tend to internationalise their activity towards Latin America due to reasons of cultural affi­ nity. However, it does not explain why large Spanish companies with offices in European States have taken so long to constitute an EWC. Perhaps it is due to the mistrust and lack of knowledge regarding this institution.11 Of the fourteen agreements published in the BOE, the first was approved almost at the same time the Directive was implemented, signed by Repsol on 16 July 199712 for a period of two years and to be automatically renewed at the end of each period. The creation of this EWC is of particular interest, as it is the only one to be included within a Spanish framework agreement. This means that this agreement was only negotiated by the most representative unions on the national level, who also have majority re­ presentation in the different Spanish works councils. Section 17 of this agreement contains regulations regarding information and consultation procedures on the European level. 9 Law 10/1997, of 24 April, was subsequently amended by Law 44/1999, of 29 November, so as to adapt its content to the extension of Directive 94/45/EC to the United Kingdom; and again by Law 10/2011, of 19 May, to incorporate the amendments set out in Directive 2009/38/EC. 10 In the implementation of Directives 94/45 and 2009/38/EC, Spanish legislators introduced an additional requirement. Collective transnational bargaining: practical implementation experiences from European… Luis Gordo González Collective transnational bargaining: practical implementation experiences from European Gordo González Later came agreements at GE Power Controls Ibérica,13 Praxair Group,14 Tafisa Group,15 Altadis,16 Banco Bilbao Vizcaya Argentaria Group,17 Saica Group,18 Santander Group,19 AXA Group,20 Roca Cor­ poración Empresarial,21 Abertis Group,22 and, the most recent, Prosegur Group,23 Schreiber Foods24 and, the most recent, IAG International Group.25 Additionally, six other EWCs have been constituted but not published in the BOE at Gestamp, Ferroatlántica, Amadeus, Campofrío Groups., NH Hoteles and Coca- Cola Iberian Partners.26 Each of these agreements was influenced by the context in which it was negotiated, and by each company's own business culture. This means that some have incorporated interesting solutions or alter­ natives to problems related to articulating worker representation on a European scale, while others have simply limited themselves to transcribing the content of the Directives and Law 10/1997. However, by analysing all of these agreements, some general conclusions can be drawn and we can even formulate a general paradigm of Spanish EWCs. The analysis of the agreements has taken into consideration twelve variables: the type of EWC; the number of representatives who comprise it; the criteria for distributing representation among the different States where the company operates; the existence (or lack thereof) of a select committee; the formula cho­ sen for designating representatives to form part of the EWC; the requirements that representatives must meet to be on the EWC; the appointment (or lack thereof) of substitutes; the possibility of the company's management determining representation beforehand; the EWC's operational costs that are to be assumed by the company; the number and types of meetings to be held; if there is a confidentiality obligation; and, finally, the topics subject to information and consultation. These variables will allow us to perform an in- depth examination of the reality of EWCs in Spain and determine their common characteristics. 13 Agreement dated 26 November 1999 (BOE of 21 December, no. 304). The aforementioned agreement is no longer in effect as a new regulation was agreed upon between the workers' representatives and headquarters, now located in Hungary. The group, now called GE Consumer & Industrial, updated the EWC for the last time on 14 September 2007. III. Content and characteristics of the EWC constitution agreements in Spain In accordance with the methodology described above, we will examine below the agreements subject to Spanish law, both those published in the Official State Gazette and those not.27 13 Agreement dated 26 November 1999 (BOE of 21 December, no. 304). The aforementioned agreement is no longer in effect as a new regulation was agreed upon between the workers' representatives and headquarters, now located in Hungary. The group, now called GE Consumer & Industrial, updated the EWC for the last time on 14 September 2007. 14 The Praxair Group's headquarters are located in the US. The US parent company designated Praxair Spain as the central office for the purposes of constituting the EWC. The EWC was finally constituted through an agreement dated 4 April 2000 (BOE of 21 June, no. 148). It was then updated through an agreement dated 4 May 2004 (BOE of 1 June, no. 158). g ( p , , ) 16 Agreement dated 14 November 2002 (BOE of 10 January 2003, no. 9). This Agreement went out of effect after the ac­ quisition of Altadis by Imperial Tobacco in 2010. The new group that arose from the acquisition renegotiated the information and consultation procedures based on the pre-existing agreement at Imperial Tobacco and created a new agreement, which went into effect on 30 March 2011, effective retroactively from the date of purchase, 14 November 2010. However, as the Altadis agreement was published in the Spanish Official State Gazette and it contains interesting advances compared to other Spanish EWCs, we will analyse this agreement. 18 Agreement dated 21 June 2004 (BOE of 24 August, no. 204). 19 Agreement dated 16 March 2005 (BOE of 30 May, no. 128). Although the AXA Group's central headquarters are located in France, the EWC constitution agreement was published Agreement dated 6 October 2005 (BOE of 7 June 2006, no. 135). p g ( ) 21 Agreement dated 31 October 2000 (BOE of 19 December, no. 303). Subsequently renegotiated via Agreement dated 2 July 2007 (BOE of 3 September, no. 211) and Agreement dated 9 July 2014 (BOE of 18 August, no. 200). 22 Agreement dated 23 July 2012 (BOE of 21 September, no. 228). 23 Agreement dated 15 November 2013 (BOE of 23 January 2014, no. 20). ed 23 November de 2016 (BOE 30 of December 2016, no. 25 Agreement dated 27 de April de 2017 (BOE 16 of June 2017, no. Spanish Labour Law and Employment Relations Journal. EISSN 2255 2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.20 Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N º 1 2 Vol 7 noviembre 2018 pp 41 55 DOI: https://doi org/10 20318/s g p, , p p 14 The Praxair Group's headquarters are located in the US. The US parent company designated Praxair Spain as the central office for the purposes of constituting the EWC. The EWC was finally constituted through an agreement dated 4 April 2000 (BOE of 21 June, no. 148). It was then updated through an agreement dated 4 May 2004 (BOE of 1 June, no. 158). ( , ) p g g 15 Agreement dated 13 March 2002 (BOE of April 11, 2002, no. 87). 44 Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 13 Agreement dated 26 November 1999 (BOE of 21 December, no. 304). The aforementioned agreement is no longer in effect as a new regulation was agreed upon between the workers' representatives and headquarters, now located in Hungary. The group, now called GE Consumer & Industrial, updated the EWC for the last time on 14 September 2007. 14 The Praxair Group's headquarters are located in the US. The US parent company designated Praxair Spain as the central office for the purposes of constituting the EWC. The EWC was finally constituted through an agreement dated 4 April 2000 (BOE of 21 June, no. 148). It was then updated through an agreement dated 4 May 2004 (BOE of 1 June, no. 158). 15 Agreement dated 13 March 2002 (BOE of April 11, 2002, no. 87). 16 Agreement dated 14 November 2002 (BOE of 10 January 2003, no. 9). This Agreement went out of effect after the ac­ quisition of Altadis by Imperial Tobacco in 2010. The new group that arose from the acquisition renegotiated the information and consultation procedures based on the pre-existing agreement at Imperial Tobacco and created a new agreement, which went into effect on 30 March 2011, effective retroactively from the date of purchase, 14 November 2010. However, as the Altadis agreement was published in the Spanish Official State Gazette and it contains interesting advances compared to other Spanish EWCs, we will analyse this agreement. 17 Agreement dated 3 June 2004 (BOE of 3 August 2004, no. 186). 18 Agreement dated 21 June 2004 (BOE of 24 August, no. 204). 19 Agreement dated 16 March 2005 (BOE of 30 May, no. 128). 20 Although the AXA Group's central headquarters are located in France, the EWC constitution agreement was published in Spain. Agreement dated 6 October 2005 (BOE of 7 June 2006, no. 135). 21 Agreement dated 31 October 2000 (BOE of 19 December, no. 303). Subsequently renegotiated via Agreement dated 2 July 2007 (BOE of 3 September, no. 211) and Agreement dated 9 July 2014 (BOE of 18 August, no. 200). 22 Agreement dated 23 July 2012 (BOE of 21 September, no. 228). 23 Agreement dated 15 November 2013 (BOE of 23 January 2014, no. 20). 24 Agreement dated 23 November de 2016 (BOE 30 of December 2016, no. 315). 25 Agreement dated 27 de April de 2017 (BOE 16 of June 2017, no. 143). 26 Grupo Gestamp (29 January 2015), Grupo Ferroatlántica (4 February 2014), Grupo AMADEUS (25 May 2013) and Gru­ po Campofrío (4 March 2009), NH Hoteles (9 October 2015) y Coca-Cola Iberian Parther (25 May 2016). See ETUI database. (http://www.ewcdb.eu/). Last access: June 2018. 27 It has already been mentioned that publishing the agreement in the BOE is not a requirement for said agreement to be valid (see footnote 10). In that regard, based on this analysis of the EWCs, the agreements' publication in the BOE or lack g ( p , , ) 16 Agreement dated 14 November 2002 (BOE of 10 January 2003, no. 9). This Agreement went out of effect after the ac­ quisition of Altadis by Imperial Tobacco in 2010. The new group that arose from the acquisition renegotiated the information and consultation procedures based on the pre-existing agreement at Imperial Tobacco and created a new agreement, which went into effect on 30 March 2011, effective retroactively from the date of purchase, 14 November 2010. However, as the Altadis agreement was published in the Spanish Official State Gazette and it contains interesting advances compared to other Spanish EWCs, we will analyse this agreement. 18 Agreement dated 21 June 2004 (BOE of 24 August, no. 204). II. The Spanish model of European Works Council Law 10/1997 of 24 April, article 13, referring to article 90 of the Workers' Statute, establishes that the agreements must be presented to the competent labour authority for their registration, filing, and publication in the BOE. This requirement is fully consistent with the regulations regarding Spain's collective agreements for general efficiency or ergea omnes. However, four companies have reached agreements to constitute an EWC which have not been published in the BOE: Grupo Gestamp (29 January 2015), Grupo Ferroatlántica (4 February 2014), Grupo AMADEUS (25 May 2013) and Grupo Campofrío (4 March 2009). Publishing the collective agreements, as established in article 90 of the Workers' Statute, aims to guarantee that general knowledge of the terms agreed upon by the parties is made available to third parties, without official publication depriving the agreement of effectiveness. Several legal rulings have been made along these lines: Ruling of the National Court of Spain dated 29 October 1998 (Proc. 77/1998) Court Opinion no. 4; Ruling of the High Court of Justice of Madrid dated 14 September 2012 (R. 4089/2012) Court Opinion no. 25; Ruling of the High Court of Justice of Asturias dated 3 June 2003 (R. 2968/2001) Court Opinion no. 2. 11 Other factors that may explain low numbers of EWCs in Spain are: the structure of Spanish business, with a high number of SMEs that have fewer than the 1,000 worker cut-off established by the Directive; the strong reticence of Spanish manage­ ment to give up, even indirectly, some control over its companies to the workers' representatives; and the lack of tradition and initiative of Spanish unions in defending the creation of EWCs in Spain. While these factors seem to be confirmed by the actors involved in EWC constitution themselves (unions, management, and institutional representatives), others, such as the economic cost of implementing this measure, are not supported by empirical data. In this regard, see: ALBALATE, Joaquín Juan, “La implantación de los Comités de Empresa Europeos en España”, en Revista Española de Investigaciones Sociológicas, núm. 124, 2008, pp. 177-207, p. 193 et seq. 12 Agreement published in the BOE on 10 December 1997 (BOE 295). Several updates have been published since then: Agreement dated 12 November 2001 (BOE of 29 November, no. 286) and Agreement dated 30 May 2011 (BOE of 18 July, no. 171). 43 Collective transnational bargaining: practical implementation experiences from European… Luis Gordo González , ) p g g y ( , ) 15 Agreement dated 13 March 2002 (BOE of April 11, 2002, no. 87). III. Content and characteristics of the EWC constitution agreements in Spain Those companies with EWCs inspired in the German model, in general, have more complex regulations that provide a greater guarantee in terms of worker representation. On the contrary, those agreements that call for mixed councils have chosen more conventional regulations, which cover the necessary minimum, but whose usefulness is doubtful. The decision of the parties to choose the French model or the German (EWC comprised solely of workers) is not insignificant, as the French model has more similarities to the dominant representation model in Spain. Those companies with EWCs inspired in the German model, in general, have more complex regulations that provide a greater guarantee in terms of worker representation. On the contrary, those agreements that call for mixed councils have chosen more conventional regulations, which cover the necessary minimum, but whose usefulness is doubtful. b) Number of representatives: Unlike Spanish companies' national councils, in Community-sca­ le companies the number of worker representatives is small. The number of representatives comprising the EWCs barely exceeds a dozen members on average, in companies where the workforce often exceeds 5,000 workers.28 On the contrary, in Spain, for the national works councils, a company with a workforce of 5,000 workers should have a council consisting of 29 representatives; the number of representatives could be as many as 75 for companies with very large workforces.29 The average number of representatives on the Spanish EWCs is even far from the mini­ mum of 30 representatives set out in the subsidiary provisions in the Annex of Directive 94/45, to be applied in lieu of an agreement between the parties. The formula that Spanish companies have chosen to use more closely follows the criteria that were set out in the Annex of Directive 2009/38/EC. These criteria stipulate that one representative be chosen per portion of emplo­ yees employed in that Member State amounting to 10%, or a fraction thereof, of the number of employees employed in the entire workforce. This regulation tacitly reduces representation to a mere dozen representatives30.i The significant difference between the number of participants on the national and Commu­ nity councils may perhaps be explained by the goals of each entity. The national councils are a negotiating entity, designed to serve as a counterweight to the employer's decisions. Whereas the EWCs, in principle, only aim to correctly transmit information and consultations on a transna­ tional level. III. Content and characteristics of the EWC constitution agreements in Spain 143). ated 27 de April de 2017 (BOE 16 of June 2017, no. 143). g p ( ) 26 Grupo Gestamp (29 January 2015), Grupo Ferroatlántica (4 February 2014), Grupo AMADEUS (25 May 2013) and Gru­ po Campofrío (4 March 2009), NH Hoteles (9 October 2015) y Coca-Cola Iberian Parther (25 May 2016). See ETUI database. (http://www.ewcdb.eu/). Last access: June 2018. 26 Grupo Gestamp (29 January 2015), Grupo Ferroatlántica (4 February 2014), Grupo AMADEUS (25 May 2013) and Gru­ po Campofrío (4 March 2009), NH Hoteles (9 October 2015) y Coca-Cola Iberian Parther (25 May 2016). See ETUI database. (http://www.ewcdb.eu/). Last access: June 2018. ( p ) 27 It has already been mentioned that publishing the agreement in the BOE is not a requirement for said agreement to be valid (see footnote 10). In that regard, based on this analysis of the EWCs, the agreements' publication in the BOE or lack 27 It has already been mentioned that publishing the agreement in the BOE is not a requirement for said agreement to be valid (see footnote 10). In that regard, based on this analysis of the EWCs, the agreements' publication in the BOE or lack 44 Collective transnational bargaining: practical implementation experiences from European… do González a) Type of EWC: The majority of Spanish EWCs have been based on a French model of repre­ sentation. That is, almost all of the agreements include a mixed composition of company and worker representatives, with the exception of GF Power, Roca, Abertis, Prosegur, Campofrío and Ferroatlántica. a) Type of EWC: The majority of Spanish EWCs have been based on a French model of repre­ sentation. That is, almost all of the agreements include a mixed composition of company and worker representatives, with the exception of GF Power, Roca, Abertis, Prosegur, Campofrío and Ferroatlántica. a) Type of EWC: The majority of Spanish EWCs have been based on a French model of repre­ sentation. That is, almost all of the agreements include a mixed composition of company and worker representatives, with the exception of GF Power, Roca, Abertis, Prosegur, Campofrío and Ferroatlántica. The decision of the parties to choose the French model or the German (EWC comprised solely of workers) is not insignificant, as the French model has more similarities to the dominant representation model in Spain. III. Content and characteristics of the EWC constitution agreements in Spain The nature of this second task, much more limited and simple than the first, is what allows the EWCs to have a smaller number of members. In reality, just one member in each State would be sufficient to communicate Community-level information to the national councils. thereof does not imply differences in terms of content and characteristics, with one exception: training. While of the published agreements, only five of twelve state that the company will assume the training costs for representatives, in the unpublished agreements the majority (three of four) state that the company will assume said costs. However, this sole difference between the published and unpublished agreements is easier to explain if we keep in mind the date the agreements were adopted. Almost all of the most modern agreements include the representatives' right to training; this can be explained by the fact that as the EWCs have evolved, it has been seen that the representatives require specific technical, legal, and economic training, as well as language training, in order to be able to satisfactorily complete their work. Therefore, the most modern agreements have adapted to this need. thereof does not imply differences in terms of content and characteristics, with one exception: training. While of the published agreements, only five of twelve state that the company will assume the training costs for representatives, in the unpublished agreements the majority (three of four) state that the company will assume said costs. However, this sole difference between the published and unpublished agreements is easier to explain if we keep in mind the date the agreements were adopted. Almost all of the most modern agreements include the representatives' right to training; this can be explained by the fact that as the EWCs have evolved, it has been seen that the representatives require specific technical, legal, and economic training, as well as language training, in order to be able to satisfactorily complete their work. Therefore, the most modern agreements have adapted to this need. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 thereof does not imply differences in terms of content and characteristics, with one exception: training. While of the published agreements, only five of twelve state that the company will assume the training costs for representatives, in the unpublished agreements the majority (three of four) state that the company will assume said costs. However, this sole difference between the published and unpublished agreements is easier to explain if we keep in mind the date the agreements were adopted. Almost all of the most modern agreements include the representatives' right to training; this can be explained by the fact that as the EWCs have evolved, it has been seen that the representatives require specific technical, legal, and economic training, as well as language training, in order to be able to satisfactorily complete their work. Therefore, the most modern agreements have adapted to this need. III. Content and characteristics of the EWC constitution agreements in Spain Only one agre­ ement, for the company Saica, opts for a purely proportional distribution; that is, dividing the number of representatives by the workforce and assigning transnational representatives based on that calculation. The rest have adopted a proportional distribution based on a scale of the number of workers or a percentage of workers employed in each State.31 Proportional distribution makes representation of States with a smaller number of workers difficult; they may not obtain representation on the EWC. This problem does not exist in those agreements that determine the States' representation based on a linear formula, assigning the same number of representatives to all States, or discretionally,32 assigning a different number of representatives to each State. However, linear and discretionary amount representation over- represents States with fewer workers, at the expense of States with larger workforces.33 Proportional systems attempt to prevent under- and over-representation by reserving a fixed representation for all States. This minimum representation may be establishing one repre­ sentative for each State in which the company operates, or setting minimum joint representa­ tion for all States that do not reach the first threshold on the distribution scale.34 The AXA group's agreement offers an interesting solution somewhere between the others, with the advantages of the two models, a limitation on the number of workers, and a correct articulation of the mechanisms of workforce information and consultation which, although in the minority, should be extended to other agreements. The agreement stipulates that only those States in which the group employs more than 150 workers will have the right to name represen­ tatives. For those States that do not reach this cut-off, they have created the figure of indirect representative. That is, a direct representative, elected by another State, is in charge of indi­ rectly informing the workers in another State who do not have representation. The interesting part of this solution is that the agreement itself establishes that these indirect representatives are entitled to meet with the national representatives of the countries that they indirectly represent twice a year. Therefore, this agreement includes effective, directly enactable mechanisms that allow the entire company workforce to be adequately represented. Lastly, only two agreements have established additional representation for those States in which the company's headquarters is located. Axa's agreement assigns three additional worker representatives to the State where central headquarters is located,35 while Abertis' agreement only assigns one. 31 Agreements that use a distribution based on the percentage of workers: Abertis, Roca, BBVA, Campofrío, Gestamp and Tafisa. On the other hand, agreements that distribute representation based on the number of workers: AXA, Santander, Praxair, Amadeus, Schreiber Foods and IAG International Group. III. Content and characteristics of the EWC constitution agreements in Spain p 28 For example, the BBVA group, with more than 32,000 workers, has an EWC consisting of 8 representatives; the Santand­ er group, with more than 68,000 workers, has an EWC with 10 representatives; Prosegur, with more than 17,900 workers, also has a council of 10 workers; Praxair has more than 2,900 workers and its EWC consists of 12 members; Ferroatlántica has a 12-member EWC; Saica, with more than 5,400 workers, has an EWC with 14 representatives; the Abertis group's EWC consists of 17 workers who represent a total of 8,700; the agreement no longer in effect at Altadis provided for 17 representatives for 13,800 workers; the Amadeus group's agreement establishes an EWC with 20 members for a workforce of 6,205; Gestamp's agreement stipulates that 21 members must represent 18,840 workers. As an exception, AXA group's EWC has 50 members to represent a workforce of just over 57,500. 29 The number of members on the works councils is determined in accordance with the scale set out in article 66.1 of the Workers' Statute: workplaces with more than 3,000 workers shall have 25 council members; 27 representatives for 4,000 work­ ers; 31 for 6,000; 33 for 7,000; 35 for 8,000, etc. 30 If the EWC had one representative for every 10%, the maximum number of representatives on the council would be 10. However, since each fraction may also increase the number of council members, it may be the case that a company has small workforces and workplaces in several States, so they would have to choose more than 10 representatives since they had 5% of their workforce in different States. 30 If the EWC had one representative for every 10%, the maximum number of representatives on the council would be 10. However, since each fraction may also increase the number of council members, it may be the case that a company has small workforces and workplaces in several States, so they would have to choose more than 10 representatives since they had 5% of their workforce in different States. 45 Collective transnational bargaining: practical implementation experiences from European Gordo González c) Criteria for distributing the number of representatives among the different States: In this case, the Spanish EWC prototype favours proportionality in the distribution of council members. The agreements perform a proportional distribution using three formulas. 32 The agreements made by Prosegur, Altadis, GE Power, Repsol and Ferroatlántica set the number of representatives assigned to each State without referencing any criteria used for the distribution. However, a higher or lower number of repre­ sentatives seems to coincide with those States in which the company has a higher or lower number of workers employed, or in which the unions are more organised in the company. g p y 33 This imbalance in favour of States that employ a lower number of workers is diminished in discretionary assignment of representatives. Although not explicitly stated, distribution involving a discretionary number of members from each State for the EWC tends to assign more representatives to States in which more workers are employed and, therefore, establishes a de facto hidden proportional distribution.i Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 34 Abertis, Gestamp and Roca recognise one representative from each State. BBVA, Tafisa, Praxair and Amadeus recog­ nise one representative from each State that employs at least 100, 100, 40, and 20 workers respectively. Finally, Santander's agreement establishes a scale with a first threshold of 500 workers. In principle, the States in which the group employs fewer workers would not have representation. However, the agreement stipulates that between all of them, they shall designate a common representative. III. Content and characteristics of the EWC constitution agreements in Spain However, while these types of agreements are the exception, it should be no­ ted at this time that they are examples of what should not be done. In proportional agreements, there is no reason that justifies additional representation for the State where the headquarters is located. 32 The agreements made by Prosegur, Altadis, GE Power, Repsol and Ferroatlántica set the number of representatives assigned to each State without referencing any criteria used for the distribution. However, a higher or lower number of repre­ sentatives seems to coincide with those States in which the company has a higher or lower number of workers employed, or in which the unions are more organised in the company. 33 This imbalance in favour of States that employ a lower number of workers is diminished in discretionary assignment of representatives. Although not explicitly stated, distribution involving a discretionary number of members from each State for the EWC tends to assign more representatives to States in which more workers are employed and, therefore, establishes a de facto hidden proportional distribution.i 34 Abertis, Gestamp and Roca recognise one representative from each State. BBVA, Tafisa, Praxair and Amadeus recog­ nise one representative from each State that employs at least 100, 100, 40, and 20 workers respectively. Finally, Santander's agreement establishes a scale with a first threshold of 500 workers. In principle, the States in which the group employs fewer workers would not have representation. However, the agreement stipulates that between all of them, they shall designate a common representative. 34 Abertis, Gestamp and Roca recognise one representative from each State. BBVA, Tafisa, Praxair and Amadeus recog­ nise one representative from each State that employs at least 100, 100, 40, and 20 workers respectively. Finally, Santander's agreement establishes a scale with a first threshold of 500 workers. In principle, the States in which the group employs fewer workers would not have representation. However, the agreement stipulates that between all of them, they shall designate a common representative. p 35 Despite Axa's central headquarters being located in France, their agreement was published in the Spanish BOE. There­ fore, the three additional representatives for the State where the headquarters are located are awarded to France. 35 Despite Axa's central headquarters being located in France, their agreement was published in the Spanish BOE. There­ fore, the three additional representatives for the State where the headquarters are located are awarded to France. 36 Article 6.2.e of Directive 2009/38/EC: “Without prejudice to the autonomy of the parties, the agreement referred to in paragraph 1 [regarding the implementation of information and consultation procedures in the company] and effected in writing between the central management and the special negotiating body shall determine: […] (e) where necessary, the composition, the appointment procedure, the functions and the procedural rules of the select committee set up within the European Works Council.” Directive 94/45 did not contain any reference to the select committee in its articles, however it did in the subsidiary regulations set out in the Annex of the Directive; these regulations were to be used by States when drawing up their own reg­ ulations and by companies that wished to directly apply the Directive's provisions. Section 1.c of the Annex established that if the size of the EWC so justified, a select committee would be elected, consisting of three members maximum. III. Content and characteristics of the EWC constitution agreements in Spain In this way, the workers' representatives, chosen in compliance with the general rules set out in Spain, will be chosen in a second election from among the Spanish representatives on workers' committees, or from among the trade union delegates. The choice will be made by an agreement between those union representations that together make up the majority of the work council or councils and personnel delegates, as applicable, or by majority agreement between said members and delegates.40 e) Choosing EWC representatives: Almost unanimously, the process for choosing representatives for Spanish EWCs refers to the subsidiary regulation in each State. In this way, the workers' representatives, chosen in compliance with the general rules set out in Spain, will be chosen in a second election from among the Spanish representatives on workers' committees, or from among the trade union delegates. The choice will be made by an agreement between those union representations that together make up the majority of the work council or councils and personnel delegates, as applicable, or by majority agreement between said members and delegates.40 38 The select committee is not included in the agreements made by Repsol, Praxair, Altadis, Santander and BBVA. 39 The Roca agreement (revised first generation), and the Abertis, Amadeus, Gestamp, Tafisa and Ferroatlántica, Scheiber Foods, and IAG Group agreements (second generation) include a select committee. 40 In Spain, the representatives that should make up the special negotiating body and the European Works Council shall be designated in accordance with those union representations that together make up the majority of the work council or councils and personnel delegates, as applicable, or by majority agreement between said members and delegates. Substitute represen­ tatives will be chosen in the same manner in the case of resignations or loss of condition of national worker representatives (art. 27 Law 10/1997). Other States, on the other hand, have chosen alternative methods. For example, in Germany, represen­ tatives shall be designated by the Central Workers' Committee [what in Spain would be called an Intercentres Committee] in Community-scale companies or by the combined committee in European-scale groups of companies; if a combined committee does not exist, the representatives shall be designated through a joint meeting of the central committees or of the work coun­ cils if there are no central committees, which will be called by the Chairman of the committee of the company that employs the largest number of workers. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 38 The select committee is not included in the agreements made by Repsol, Praxair, Altadis, Santander and BBVA. 39 The Roca agreement (revised first generation), and the Abertis, Amadeus, Gestamp, Tafisa and Ferroatlántica, Scheiber Foods, and IAG Group agreements (second generation) include a select committee. III. Content and characteristics of the EWC constitution agreements in Spain Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 46 Collective transnational bargaining: practical implementation experiences from European… Luis Gordo González Collective transnational bargaining: practical implementation experiences from European… González d) Select committee: Directive 2009/38/EC included for the first time the provision that the agree­ ments constituting the EWCs could include the creation of a select committee.36 The incorpora­ tion of this body intended to guarantee that the regular activity of the European Works Councils was coordinated and efficient, and that consultations could be performed as quickly as possible in exceptional circumstances.37 The subsidiary requirements set out in the Annex of the Direc­ tive stipulate that the select committee should not consist of more than five members. Spanish legislators developed this measure through article 19.2 of Law 10/1997. The subsidiary regu­ lations applicable in Spain establish that if the number of European Works Council members is greater than twelve, they must choose a select committee consisting of three members. This rule seems to be counter-intuitive, as in EWCs with fewer than 12 members it should not be necessary to constitute a select committee. As this characteristic was incorporated in Directive 2009/38/EC, its presence in the Spa­ nish EWCs is inconsistent. In general, the earlier agreements subject to Directive 94/45 do not include a select committee.38 However, the majority of earlier agreements that were later revi­ sed and the second-generation agreements, which are subject to the regulations arising from Directive 2009/38/EC, do include this entity.39 i Five agreements break this general rule: the first-generation agreements made by GE Power, Saica, Axa and Campofrío include the figure of a select committee, even before the EC regulation expressly included it in its articles. This measure was not impossible, as the EC re­ gulation was only of subsidiary application. On the other hand, the second-generation Prosegur agreement does not include a select committee, probably due to the low number of members of the plenary EWC, comprised of just ten members. Therefore, the majority of the EWCs in Spain have chosen to include this body so as to bolster the effectiveness of transnational worker representation. e) Choosing EWC representatives: Almost unanimously, the process for choosing representatives for Spanish EWCs refers to the subsidiary regulation in each State. 37 See Whereas 30 Directive 2009/38/EC. e s e o e WC so jusi ed, se ec co ee wou d be e ec ed, co s s g o ee e be s u . 37 See Whereas 30 Directive 2009/38/EC. 38 The select committee is not included in the agreements made by Repsol, Praxair, Altadis, Santander and BBVA. 39 The Roca agreement (revised first generation), and the Abertis, Amadeus, Gestamp, Tafisa and Ferroatlántica, Scheiber Foods, and IAG Group agreements (second generation) include a select committee. 40 In Spain, the representatives that should make up the special negotiating body and the European Works Council shall be designated in accordance with those union representations that together make up the majority of the work council or councils and personnel delegates, as applicable, or by majority agreement between said members and delegates. Substitute represen­ tatives will be chosen in the same manner in the case of resignations or loss of condition of national worker representatives (art. 27 Law 10/1997). Other States, on the other hand, have chosen alternative methods. For example, in Germany, represen­ 40 In Spain, the representatives that should make up the special negotiating body and the European Works Council shall be designated in accordance with those union representations that together make up the majority of the work council or councils and personnel delegates, as applicable, or by majority agreement between said members and delegates. Substitute represen­ tatives will be chosen in the same manner in the case of resignations or loss of condition of national worker representatives (art. 27 Law 10/1997). Other States, on the other hand, have chosen alternative methods. For example, in Germany, represen­ tatives shall be designated by the Central Workers' Committee [what in Spain would be called an Intercentres Committee] in Community-scale companies or by the combined committee in European-scale groups of companies; if a combined committee does not exist, the representatives shall be designated through a joint meeting of the central committees or of the work coun­ cils if there are no central committees, which will be called by the Chairman of the committee of the company that employs the largest number of workers. If a work council were not represented in the central committees, their Chairman would also be invited to this meeting (§11 Gesetz über Europäische Betriebsräte). In Italy, the designation will be performed by unions that have signed the national collective agreement that applies to the European-scale company or group of companies and the 38 The select committee is not included in the agreements made by Repsol, Praxair, Altadis, Santander and BBVA.ii union confederations in the company or group. When there is no union representation, the union organisations that signed the national collective agreement that applies to the company and the company's central management will come to an agreement over how to create a system that will allow workers to be represented (art. 6 of Italian Legislative Decree no. 74, dated 2 April 2002, regarding all'istituzione di un comitato aziendale europeo o di una procedura per l'informazione e la consultazione dei lavoratori nelle imprese e nei gruppi di imprese di dimensioni comunitarie). In France, the representatives shall be chosen by union organizations from among the workers' representatives in the company or workplaces (arts. L 439-19 Loi n° 96-985 du 12 novembre 1996 relative à l'information et à la consultation des salariés dans les entreprises et les groupes d'entreprises de dimension communautaire, ainsi qu'au développement de la négociation collective). 41 Representatives from other States shall be chosen by unions or by personnel representation bodies, depending on the laws in each State Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 III. Content and characteristics of the EWC constitution agreements in Spain If a work council were not represented in the central committees, their Chairman would also be invited to this meeting (§11 Gesetz über Europäische Betriebsräte). In Italy, the designation will be performed by unions that have signed the national collective agreement that applies to the European-scale company or group of companies and the 47 Collective transnational bargaining: practical implementation experiences from European… Luis Gordo González Collective transnational bargaining: practical implementation experiences from European… do González Interestingly, only one agreement does not follow this general rule: the Abertis agreement. This agreement states that the Spanish representatives shall be elected by the most represen­ tative union organisations in the Spanish national sphere that have a minimum of 20% of the representation in the Abertis group, from among those members elected to the work council, the personnel delegates or the union representatives.41 The Campofrío and IAG Group agreements also include an exception to the general rule, although of less importance than the exception in the Abertis agreement in terms of union par­ ticipation. This agreement states that the representatives shall be chosen by the representatives on the national works councils and by union organisations established in the company. If there is no agreement in place, the representatives will be chosen in proportion to the distribution of the national councils, or by the union organisation with the greatest representation in the sector in that State if there are no works councils. The IAG agreement asserts that the national workers' representatives have to elect the European representatives. If there are no national representatives, it is the workers who have to elect these European representatives. So, the Campofrío and IAG agreements are exceptions to the general rule and they create a general fixed norm with regards to elections for all States. This fixed norm for regulating the election of workers' representatives helps to stimulate corporate culture within the group and boosts the visibility of the workers' transnational representatives. These regulations are one of the few examples in which, to a lesser or greater extent, unions are allowed to directly intervene. This general exclusion is due to the fact that European management organisations, despite the unions' demands, refused to give them a more promi­ nent role. This stance has been followed by companies when drawing up their agreements, giving the unions no role or recognition. dimension communautaire, ainsi qu au développement de la négociation collective). 41 Representatives from other States shall be chosen by unions or by personnel representation bodies, depending on the laws in each State. 42 The agreements thus follow the subsidiary rule set out in Art. 17.1 of Law 10/1997. 43 The eligibility requirements differ from one State to another. For example: Greece requires workers to be at the compa­ ny for a minimum of two months; Belgium, Austria and Germany require six months; France, the Netherlands, Romania and Luxembourg require at least one year of service. The periods, as well as additional requirements set out by some legislatures, such as the worker's minimum age or type of contract that they must have to be eligible, can be found in a table prepared by the ETUI, see http://www.ewcdb.eu/show_pdf.php?document=9963ori_EN.pdf [January 2015].i III. Content and characteristics of the EWC constitution agreements in Spain 48 Collective transnational bargaining: practical implementation experiences from European Gordo González Therefore, in the case of Tafisa, Spanish workers must have worked at the company for three years, and not the six months required by Spanish legislation.43 Therefore, in the case of Tafisa, Spanish workers must have worked at the company for years, and not the six months required by Spanish legislation.43 g) Substitute representatives: almost all of the agreements provide for the election of substitu­ te representatives for cases in which the original representative cannot fulfil his or her main obligations,44 which are usually limited to attending the annual EWC meeting.i g) Substitute representatives: almost all of the agreements provide for the election of substitu­ te representatives for cases in which the original representative cannot fulfil his or her main obligations,44 which are usually limited to attending the annual EWC meeting.i The figure of a substitute in worker representation bodies is a strange institution in Spa­ nish regulation, but is very useful for a transnational body. Including a substitute is beneficial in a representation model as limited as the EWC. Essentially, the activity of the transnational wor­ kers' representatives is limited to the annual meeting with the company's central management. Therefore, if one of the representatives were missing at these meetings, it would endanger the effectiveness of informing and consulting the workers. This situation does not occur in worker representation on a national scale, where the representative's activity is carried out in a less concentrated period of time. In almost all of the agreements, the substitute's purpose is limited to substituting the main representative in those cases in which he or she cannot perform his or her functions. Thus, we should ask ourselves what requirements the substitutes must meet and what guarantees pro­ tect them. Some agreements explicitly require the substitutes to meet the same requirements as the main representatives, but even these agreements are silent on their guarantees. Given the nature of the substitute institution, the most appropriate approach would be to grant them the same guarantees as the main representatives, while they must also meet the same requirements. Otherwise, it could be the case that those who are designated as substitutes, who would be seen as potential transnational representatives for the workers, would not be protected equally. Only one agreement confers an additional function upon the substitute beyond subs­ tituting the main representative. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 y p p gi g 45 See Art. 4.2 Schreiber Foods, Agreement SL, 23 of November 2016, (BOE 30 November 2016, No. 315). III. Content and characteristics of the EWC constitution agreements in Spain ) Representative requirements: Regardless of how the EWC members are chosen, all of the agre­ ements have established that the candidates must be employees at the company or group where the transnational worker representation body is to be established.42 However, almost all of the agreements are silent with regards to additional requirements that the workers must meet to be eligible. Therefore, in these cases, each national delegation must respect the regulations in their home State that govern the requirements that workers must meet in order to be elected as representatives by their colleagues.i ) Representative requirements: Regardless of how the EWC members are chosen, all of the agre­ ements have established that the candidates must be employees at the company or group where the transnational worker representation body is to be established.42 However, almost all of the agreements are silent with regards to additional requirements that the workers must meet to be eligible. Therefore, in these cases, each national delegation must respect the regulations in their home State that govern the requirements that workers must meet in order to be elected as representatives by their colleagues.i However, some agreements have included some specific requirements. The agreements for the AXA and Prosegur groups have expressly stipulated that employees must have worked at the company for a minimum of six months in order to be elected to the EWC. With this rule, the companies are extending a requirement from Spanish law, article 69.2 of the Workers' Statute, to all of the States in which the group operates, even if there is no such requirement in these other States. Similarly, the Campofrío group agreement stipulates that an employee must have worked for the company for a year in order to be chosen for the EWC, while Praxair requires two years. The regulations regarding worker requirements to be eligible for the EWC are much stricter at the Tafisa group. Tafisa requires workers who wish to be part of the group's EWC to have worked at group companies continuously for at least three years. By virtue of the collective autonomy that governs the agreement between the parties, the inclusion of a minimum time at the company impedes application of the subsidiary rules in each country. , p _p p p _ p [ 44 Only the Repsol and Campofrío agreements do not mention this figure. III. Content and characteristics of the EWC constitution agreements in Spain We would like to highlight it as a recommendable practice that should be included in all Spanish agreements. The AXA agreement establishes that for meetings which take place outside of the country in which the headquarters is located, the substitutes may attend the annual meetings held in that State and participate, but not vote. This is a simple way to facilitate coordination between main representatives and substitutes and to strengthen the EWC's activity. Other agreements only ensure that the substitutes receive copies of the minutes from the meetings. 45 This obligation is interesting because it means that the substitutes are informed and familiar with all that they need to know in case they one day need to attend a meeting, but it would have been better to allow the substitutes to participate in all the meetings as listeners. h) Composition of employer representation: it would be expected that in a transnational body crea­ ted to provide information to workers and seek their opinion, which generally meets annually, that the agreements constituting said bodies would also contain rules or regulations governing the presence of employer representation at these meetings. However, the reality is that, regard­ less of whether the council follows the French model, with simultaneous worker and employer participation, or whether it follows the German model and consists exclusively of the workers, the agreements do not contain specific rules regarding the composition of the company's mana­ gerial representation. The only reference that is generally made in these agreements is that the number of management representatives cannot exceed the number of worker representatives. Two agreements have included provisions regarding managerial representation, but they are not very ambitious. The Amadeus group ambiguously stipulates that the management shall be represented by members of the management with the appropriate decision-making 49 Collective transnational bargaining: practical implementation experiences from European Gordo González and representation level within the company's organisation chart and who are duly authorised. The Ferroatlántica group establishes that the managerial representation will consist of the Ge­ neral Production Manager and/or the Managing Director. The Director may be assisted in all meetings or may delegate his/her presence to three people who he or she considers to be fully authorised to participate in the meeting. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 46 The Abertis, Santander and Saica agreements limit translation to three languages. The Roca agreement mentions trans­ lation to the languages that are strictly necessary. The IAG agreement establishes English as the main language, but with the possibility of translating into Spanish when necessary. III. Content and characteristics of the EWC constitution agreements in Spain Although setting minimum guarantees to regulate the presence of the managerial representation at the EWC is a good approach, the Ferroatlántica group does it in such a broad way that the regulation is essentially meaningless, since it allows the company total freedom in designating their representatives. In addition to the Ferroatlántica and Amadeus group EWCs, three other exceptions can be mentioned as more suitable formulas that should be adopted by future EWCs. First, the GE Power agreement stipulates that the company's representatives shall be the CEO and the Euro­ pean Director of Human Resources. Both may be substituted, and both may be accompanied by other company employees who have technical knowledge related to the agenda of each meeting. Second, the agreement that regulates Roca's EWC goes even further and states that the managerial representatives shall include, at least, the Group Manager, who will also preside over joint meetings, three members of human resources management, and two members of ma­ nagement who oversee operations in the production or sales departments. Finally, the Schreiber Foods agreement designates the CEO and two Regional Directors of Human Resources as the enterprise representatives. Moreover, the European Director of Human Resources will chair the meetings along with the President of the EWC. Given the control that the EWC may exercise, it is a very positive step that the company specify which managerial representatives will attend council meetings. This ensures that the meetings will always have sufficient participation and that the management is not making it more difficult for the EWC to receive the appropriate information by sending second- or third- tier managers to the annual meetings. i) ) EWC operational costs: the distribution of the costs associated with the EWC is one of the as­ pects that should be regulated by the constitution agreements. This may be one area of the agre­ ements where the greatest dissimilarities can be found. However, there may be some overlap. ) EWC operational costs: the distribution of the costs associated with the EWC is one of the as­ pects that should be regulated by the constitution agreements. This may be one area of the agre­ ements where the greatest dissimilarities can be found. However, there may be some overlap. In Spain, almost all companies' central management assumes the costs of simultaneous interpretation at meetings between worker representatives and central management, and even covers the translation of working documents. III. Content and characteristics of the EWC constitution agreements in Spain However, some agreements limit the number of languages for interpretation/translation.46 In Spain, almost all companies' central management assumes the costs of simultaneous interpretation at meetings between worker representatives and central management, and even covers the translation of working documents. However, some agreements limit the number of languages for interpretation/translation.46 Furthermore, although when the Directives were being negotiated business organiza­ tions did everything in their power to avoid including as an imperative right the payment for external consultants, in practice, the agreements that do not include a clause that allows wor­ kers' representatives to appoint an advisor are in the minority; only three agreements (Campo­ frío, Ferroatlántica and Roca) do not mention this issue. In the rest, they generally recognise the right of the parties to name at least one external consultant, provided management is informed in advance, with the management assuming the economic cost. However, some agreements, such as that of Altadis, have gone even further, stipulating that the council shall have a maximum budget of 18,000 euros for experts. Up to four consultants may be appointed, two for Spain and two for France. Santander's agreement is also interesting, as it provides for the appointment of an external advisor, but also includes the possibility of naming up to three internal consultants from the na­ tional representatives, who would participate in the EWC meetings but would not be able to vote. Generally speaking, expenses arising from EWC activity, such as expenses for meals, daily allowances, etc., are dealt with similarly in the different agreements. The common practice is for the company to assume the costs of transportation and lodging. With regard 50 Collective transnational bargaining: practical implementation experiences from European Gordo González to other expenses, several agreements state they will cover reasonable expenses. This is a nonspecific legal concept that will allow all ordinary expenses for the correct functioning of the EWC to be covered by the company. The AXA agreement is the only one that differs from this general criterion. The parties set an annual budget of €85,000 to cover the costs associated with the EWC.47 Although not in the majority, ten agreements recognise the right of the workers' repre­ sentatives to have training paid for by company management.48 This training may be on legal/ economic knowledge, languages, or both. III. Content and characteristics of the EWC constitution agreements in Spain The management of a transnational company entails highly complex issues that may not be easy to understand or master for the average represen­ tative. For example, dealing with a transnational business group's financial statements may re­ quire a high level of accounting knowledge that perhaps not all representatives possess. To this we must add a lack of knowledge of the company's native language, which can make the task of the workers' representative even more difficult. Therefore, including regulations regarding training should be considered a good practice and it should be included in other agreements. j) EWC meetings: The regulation of the EWC meetings is also an important issue for the effecti­ veness of the worker representation bodies. If more meetings are held and workers' representa­ tives are more involved, workers will have greater control and the information and consultation work will be more effective. j) EWC meetings: The regulation of the EWC meetings is also an important issue for the effecti­ veness of the worker representation bodies. If more meetings are held and workers' representa­ tives are more involved, workers will have greater control and the information and consultation work will be more effective. Across the board, the negotiators have opted for annual council meetings; the most sparing agreements have chosen to hold a single annual meeting, while others have chosen two sessions.49 In some cases, additionally, the agreement includes the possibility of holding an extraordinary meeting if the parties so agree under extraordinary circumstances.50 As a mecha­ nism for increasing flexibility regarding the necessary agreement between the parties in order to call an extraordinary EWC meeting, some instruments have opted for allowing the workers' representatives to meet without the presence of company management when the latter do not consider an extraordinary meeting to be necessary.51 A minority of agreements have set the interval in which a meeting must be held bet­ ween the workers' representation and the company management. This requirement certainly eliminates the uncertainty regarding the meeting and makes it impossible for the party calling 50 In this sense, art. 4.3 of the Repsol Agreement “[...] Exceptionally, a meeting separate from the annual meeting may be held, provided that Management and workers’ representatives agree that it is necessary.” and art. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 47 Article 5.7 of the Agreement stipulates: “The G.E.W.C. Secretariat shall have an annual budget of € 85,000 to cover all costs (excluding contributions covering 50% of the compensation of the Secretary’s assistant) relating to the Secretariat’s oper­ ations and meetings (meetings, travel and interpretation expenses, committee expenses and experts)[...]”. g ( g p p p p )[ ] 48 The agreements that expressly recognise training to be paid for by company management are: Prosegur, Abe Altadis, GE Power, Amadeus, Gestamp, Ferroatlántica, Schereiber foods and IAG Group. 49 The Repsol, GE Power, Praxair, Saica, Santander, BBVA, Roca, Abertis, Prosegur and Schreiber Foods agreements call for a single meeting. III. Content and characteristics of the EWC constitution agreements in Spain 4.b) of the Saica group Agreement “Exceptionally, a meeting other than the annual meeting may be held, provided that both parties agree that there is an objective need to hold one. This extraordinary meeting may be held either during the European Works Council plenary session or between the Group Management and the Select Committee, as circumstances dictate.” The Roca group Agreement goes a step further and allows for an extraordinary meeting whenever one of the parties requests it, art. 6.4.1.2: “The Forum, to be convened by its Chairman at the venue appointed by the latter, shall meet once per year and on an extraordinary basis whenever one of the parties, Management or Coordinators [the workers' representatives name three coordinators as permanent spokespeople with the company management], so request should exceptional circumstances arise. [...].” Some agreements, such as that of the Praxair Group, have established that extraordinary meetings may be held via teleconference, and that in-per­ son meetings may then be held to follow up on the subject that originated in the extraordinary teleconference meeting, art. 6.8: “In the case of Exceptional Circumstances that are not included on the agenda for the annual meeting, the Management Representatives will communicate with Employee Representatives. This communication, Information and Consultation will be effective either through a meeting of the Employee Representatives or by a teleconference call. In the case of a teleconference each party will have the right to decide whether a physical follow-up meeting is required.” 51 “[...] Another meeting may be held by prior agreement of the parties, if circumstances exist that make it advisable. If no such meeting is envisaged, the trade union party may, after informing the Central Management, hold a second annual meeting in Madrid attended only by its members. [...].” This clause appears literally in art. 3 of the BBVA group agreement and in art. 3 of the Santander Group agreement. 51 Collective transnational bargaining: practical implementation experiences from European… do González the meeting, usually the management or the management after having consulted the workers' representatives, to do so when it is more convenient to their interests.52 The Abertis, Altadis and AXA group agreements provide improvements on the general regulations regarding the annual meetings mentioned above. 52 Art. 5 Prosegur Group Agreement: “The European Works Council will meet annually in the last quarter of the year on a call notice agreed by the Company, chair and secretary, who together will act as a standing point of contact”; art. 3 of the BBVA Group Agreement: “During the first four months of the year, the Works Council shall hold an annual meeting [...]”; art. 6 of the Amadeus agreement states that the annual meeting will be held in November; art. 3.1 of the Tafisa agreement stipulates that the meeting will be held on the second Wednesday of October every year. 53 Abertis Agreement art. 6.1: “The meeting shall be convened by the Central Management at least 1 (one) month in ad­ vance, hereby the notice convening the meeting shall be accompanied by a report on the development and prospects of the activities of the Abertis group [...]. The Central Management shall inform the Select Committee on a semi-annual basis, and the EWC on an annual basis”; art. 6.2.d) “When exceptional circumstances occur or when decisions may be taken that have a con­ siderable effect on the interests of the workers (whereby, such an exceptional situation includes, but is not limited to, relocation, closure of companies or collective redundancies of a transnational nature), which does not augur well for the annual meeting of the EWC, the Select Committee shall be entitled to be informed. The Select Committee shall moreover be entitled to meet, with the Central Management or any other more appropriate management, with prior request to the Central Management, so as to be informed and consulted about the aforementioned circumstances of an exceptional nature or which have a considerable effect on the employees. [...]”; and art. 8 “If situations of an exceptional nature arise (as defined in Section 6.2.d) which can have important consequences for the interests of the employees, the Central Management shall initiate immediately a process of information and consultation for the EWC on the matter. An extraordinary meeting of the EWC shall accordingly be convened within 1 (one) month, in accordance with the procedure provided under Section 6.8 of this Agreement”. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 III. Content and characteristics of the EWC constitution agreements in Spain The first states that the EWC will meet yearly with company management and that management must inform the Select Com­ mittee every semester and the EWC every year of all operational plans that may have an effect on group companies and their staff. Furthermore, in the event of exceptional circumstances, company management may call a meeting with the Select Committee or, depending on the im­ portance of the issue to be discussed, the entire EWC.53 The Altadis Agreement establishes two annual meetings, one each semester.54 Finally, the AXA Group stipulates two annual meetings and the possibility of the workers' representatives requesting an extraordinary meeting in the event of exceptional circumstances. Company management will decide if this additional mee­ ting should be held, explaining its reasoning in writing. Additionally, the agreement includes the possibility of holding an annual preparatory meeting in which the council members would visit the different States in which the group operates.55 Finally, just one agreement, that of GE Power, does not include the possibility of hol­ ding preparatory meetings for the delegation of the workers' representatives without the presen­ ce of company management. All the other agreements, both those using the French and German models, establish that the workers' representatives can meet behind closed doors immediately before meeting with central management in order to prepare for the meeting, agree upon stan­ ces, or answer questions before sharing information. Three practices are exceptions to the general rule, those of the Abertis, AXA and Schrei­ ber Foods groups, which allow for the holding of a private meeting attended by the representa­ tives of the workers. The former includes the possibility of the workers' representation holding a private meeting after the meeting with company management. This is an interesting option, as it allows the workers' delegation to evaluate the objectives achieved during the meeting with management. The second exception is that set out in the AXA group's agreement. In this case, in addition to the preparatory meetings of the workers' representation before the meeting, a preparatory visit is established. This visit is not to be immediately before the meeting with management, but consists of a meeting during the quarter prior to the meeting with manage­ ment. This visit allows for representative training and to promote knowledge of the reality of the company in the country in which the meeting will be held.56 52 Art. III. Content and characteristics of the EWC constitution agreements in Spain 5 Prosegur Group Agreement: “The European Works Council will meet annually in the last quarter of the year on a call notice agreed by the Company, chair and secretary, who together will act as a standing point of contact”; art. 3 of the BBVA Group Agreement: “During the first four months of the year, the Works Council shall hold an annual meeting [...]”; art. 6 of the Amadeus agreement states that the annual meeting will be held in November; art. 3.1 of the Tafisa agreement stipulates that the meeting will be held on the second Wednesday of October every year. 55 The AXA group Committee establishes the existence of an EWC chair, composed of members of company management and of workers. Art. 5.2 of the Agreement. 55 The AXA group Committee establishes the existence of an EWC chair, composed of members of company management and of workers. Art. 5.2 of the Agreement. g 56 The AXA group agreement contains some of the most exhaustive regulations regarding meetings of the workers' repre­ sentation. It includes monthly meetings of the committee chair, which is essentially the select committee; preparatory meetings 56 The AXA group agreement contains some of the most exhaustive regulations regarding meetings of the workers' repre­ sentation. It includes monthly meetings of the committee chair, which is essentially the select committee; preparatory meetings 52 Collective transnational bargaining: practical implementation experiences from European… González k) Confidentiality: All of the agreements contain specific provisions regarding the confidentiality obligations of the committee members. Additionally, several agreements contain provisions in which they explicitly state that the EWC's work will in no way limit the company's decision- making power. In other words, the agreements expressly recognise that the responsibility to provide information and consultation does not imply a responsibility to reach agreements. III. Content and characteristics of the EWC constitution agreements in Spain Both provisions were strongly promoted by Community business organisations, dis­ trusting the limiting effects that the EWCs could have on their actions.i Both provisions were strongly promoted by Community business organisations, dis­ trusting the limiting effects that the EWCs could have on their actions.i This obligation of confidentiality will persist even after the members' term on the com­ mittee ends.57 This obligation of confidentiality will persist even after the members' term on the com­ mittee ends.57 l) Topics subject to information and consultation: Directive 2009/38/EC innovated with regard to earlier regulation and included a definition of the transnational issues with regard to which the EWC could be informed and consulted. The definition was very generic,58 but was set up as a minimum that could be improved upon through negotiation between the parties. However, the parties have not broadened the scope of the issues that could be the topic of a negotiation in the framework of the EWC. In fact, some agreements expressly limit this possibility.59 The Directive states that transnational information and consultation with the EWC shall refer especially to the Community-scale company or group of companies' structure, economic and financial state, probable evolution of activity, production and sales, as well as the probable evolution of employment, investments, substantial changes that affect the organisation, the implementation of new work or production methods, production moves, mergers, reduction in size or closures of companies, establishments, or important parts thereof, and collective redun­ dancies. This list, which does not contain a numerus clausus, has been included without hardly any modification in the majority of the Spanish agreements. Only a minority of the agreements have created a somewhat more specific list, although in practice this does not change much, as in no case is it a closed list.60 It should be mentioned that not all of the matters listed above, or those included in each agreement, shall be handled by the EWC, as they must also be matters of transnational scope. Otherwise, they may not be dealt with in the framework of the EWC. In conclusion, if we had to describe the typical Spanish EWC, we could say that it follows the French model and has a small number of members and, therefore, differs from the habitual practices regarding the quantitative composition of workers' representation in Spain. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 III. Content and characteristics of the EWC constitution agreements in Spain EWC members are distri­ buted proportionally among all States in which the undertaking or group of undertakings operates; the Spanish representatives, the result of this proportional distribution, are elected from among the Spanish representatives on workers' committees or union representatives and must meet the requirements set out in article 69.2 of the Workers' Statute in order to be selected. The same requirements and formula will be used to designate the substitutes for the main representatives. for the EWC plenary meetings; two annual plenary meetings, which are held alternately in the company's headquarters and in the different States in which the group has companies; extraordinary meetings in the event of circumstances that greatly affect the interests and rights of the workforce; and finally, a preparatory meeting before meetings that take place outside the country where headquarters is located, so that representatives can better understand the company's productive reality in that country. 57 Art. 8 Directives 94/45 and 2009/38/EC, art. 22 of Law 10/1997. for the EWC plenary meetings; two annual plenary meetings, which are held alternately in the company's headquarters and in the different States in which the group has companies; extraordinary meetings in the event of circumstances that greatly affect the interests and rights of the workforce; and finally, a preparatory meeting before meetings that take place outside the country where headquarters is located, so that representatives can better understand the company's productive reality in that country. 57 A t 8 Di ti 94/45 d 2009/38/EC t 22 f L 10/1997 58 Article 1.4 Directive 2009/38/EC: “Matters shall be considered to be transnational where they concern the Communi­ ty-scale undertaking or Community-scale group of undertakings as a whole, or at least two undertakings or establishments of the undertaking or group situated in two different Member States”. 59 Along these lines, for example, the Prosegur agreement expressly states that the EWC shall not deal with matters that affect a single State or States outside of Europe. The BBVA agreement stipulates that matters that affect a single State must be resolved within the national structures and procedures. for the EWC plenary meetings; two annual plenary meetings, which are held alternately in the company's headquarters and in the different States in which the group has companies; extraordinary meetings in the event of circumstances that greatly affect the interests and rights of the workforce; and finally, a preparatory meeting before meetings that take place outside the country where headquarters is located, so that representatives can better understand the company's productive reality in that country. 57 Art. 8 Directives 94/45 and 2009/38/EC, art. 22 of Law 10/1997. IV. Conclusion The constitution of EWCs in Spain has met with modest success. Only 17.8% of companies that could have established an EWC have effectively done so. Therefore, we are far from the European average, where half of companies have already implemented mechanisms to inform and consult with their workers. Additionally, these agreements can hardly be called innovative; in general, they limit themselves to following the subsidiary scheme established in Community Directives and in Law 10/1997. Almost all of the texts are succinct, and in some cases only formally constitute the EWC without implementing any provisions beyond its establishment. The lack of ambition in transnational bargaining between the parties can be evidenced by the fact that no agreement has tried to make use of the EWC as a space for supranational bargaining.i Thus, no agreement has chosen to configure the European area as a negotiation unit where the company's work conditions can be discussed and agreed upon. In fact, it is just the opposite; some agre­ ements expressly prohibit this possibility, as in the Saica Group agreement.61 The Community Directives and their national implementations only set subsidiary regulations for the parties, but said regulations could be improved upon or regulated in a different manner. Therefore, company management and wor­ kers' representatives could have taken advantage of the negotiation for these agreements to implement Community-scale negotiation mechanisms, however rudimentary. The agreements have also not included States outside of the European Union. Nothing prevents the negotiating parties from agreeing to a scope of application that extends beyond European borders. In this regard, Directive 94/45 could have been used to constitute information and consultation procedures in consonance with the company or group of companies' real structure. That is, information and consul­ tation procedures or EWCs could have been established that included all States in which the company or group of companies operated, regardless of whether they were European or not. A less ambitious, although more reasonable option would have been to take into consideration at least those States that are candidates for European Union membership. However, not even this step was taken by the negotiation commissions. It is true that the European Union is not expected to grow in the short term.62 However, in the past, at times when it was easier to envision the incorporation of certain States, the negotiators did not choose to include a foresighted, expansive regulation. 61 Article 1 of the Saica group's EWC constitution Agreement states: “The SAICA European Works Council is a body that brings together the Group’s Management with its employee representatives in European Union Member States where the Group has companies in whose capital it has a 51% stake or more. […] Under no circumstances shall the SAICA European Works Council act as a negotiating body”. Collective transnational bargaining: practical implementation experiences from European… Luis Gordo González Collective transnational bargaining: practical implementation experiences from European Gordo González Despite the low number of representatives, the Spanish EWC prototype has a select committee so as to facilitate the effectiveness of its activity. The ordinary operational expenses of the select com­ mittee, as well as the plenary session of the EWC will be covered by central management. Ordinary expenses include simultaneous interpretation in meetings and translation of working documents, expen­ ses for travel and daily stipends for the workers' representatives at the annual EWC meeting and any extraordinary meetings that are called, as well as the costs of an external consultant, who may attend said meetings but may not vote. EWC information and consultation shall be limited exclusively to transnational matters that affect the group or company. The workers' representatives are subject to confidentiality. 62 Candidate States are: Albania, the Former Yugoslav Republic of Macedonia, Iceland, Montenegro, Serbia and Turkey. Bosnia and Herzegovina and Kosovo are recognised as potential candidates. Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 III. Content and characteristics of the EWC constitution agreements in Spain Finally, the Santander agreement states that the EWC shall not have au­ thority over nor shall it deal with local or national matters subject to national legislation or Collective Agreements, nor over the rights of the Works Councils or union delegates, over co-management, compensation, work hours, wages or other benefits, nor any other matter regarding individual employees, as all of these matters should be dealt with within the structures, procedures and regulations that apply in each Group Company. 59 Along these lines, for example, the Prosegur agreement expressly states that the EWC shall not deal with matters that affect a single State or States outside of Europe. The BBVA agreement stipulates that matters that affect a single State must be resolved within the national structures and procedures. Finally, the Santander agreement states that the EWC shall not have au­ thority over nor shall it deal with local or national matters subject to national legislation or Collective Agreements, nor over the rights of the Works Councils or union delegates, over co-management, compensation, work hours, wages or other benefits, nor any other matter regarding individual employees, as all of these matters should be dealt with within the structures, procedures and regulations that apply in each Group Company. 60 In this regard, the Prosegur agreement includes health and the group's environmental policy as transnational matters. The Abertis group includes work hours and health. The Saica group includes the group's strategic plan. Finally, GE Power includes environmental policy and training. 60 In this regard, the Prosegur agreement includes health and the group's environmental policy as transnational matters. The Abertis group includes work hours and health. The Saica group includes the group's strategic plan. Finally, GE Power includes environmental policy and training. 53 Spanish Labour Law and Employment Relations Journal. EISSN 2255-2081 N.º 1-2, Vol. 7, noviembre 2018, pp. 41-55. DOI: https://doi.org/10.20318/sllerj.2018.4435 IV. Conclusion Along these lines, for example, in 2007 Roca had to adapt its EWC to the incorporation of Bulgaria and Romania to the EU, which had occurred that same year. In conclusion, there is still room to encourage and incentivise Spanish companies to develop Community-scale information and consultation mechanisms. Without a doubt, the most interesting pos­ sibility would be for social agents to take advantage of the transnational bargaining made possible by an EWC in order to agree upon a common minimum baseline for the work conditions applicable in their 54 Collective transnational bargaining: practical implementation experiences from European… do González workplaces. However, a first and commendable step would be for negotiators, instead of choosing an agreement model that limits itself to transcribing the content of the Law and Directives, to choose to incorporate and develop the more progressive practices mentioned here within the agreements analysed – those experiences which have been highlighted by this study as the most inspirational and ambitious in constructing the most mature and effective transnational model of worker representation. 55 55

https://openalex.org/W4321462602

https://iqan.com.pk/index.php/iqan/article/download/398/160

Urdu

Family System of Islam and its Significance in Reformation of Society

Īqān

cc-by

< [email protected] > < [email protected] > Ghazala Bashir1 < [email protected] > Mrs. Shazia2 Sonia Bari3 1Visiting Lecturer, Government College Women University, Faisalabad 2Assistant Professor, Government College Women University, Faisalabad 3Visiting Lecturer, Government College Women University, Faisalabad Received: 10-Oct-22 Accepted: 08-Jan-22 Online: 10-Jan-22 KEYWORDS Family System, Islam, Reformation, Society, Mutual Cooperation, Ummah, Harakia ĪQĀN (9) Vol. 05, No. 01, Dec-2022 eISSN: 2617-3336 pISSN: 2617-3700 االسم اک اخدناین اظنم اور االصح اعمرشہ ںیم اس یک ونعمتی Family System of Islam and its Significance in Reformation of Society ĪQĀN (9) Vol. 05, No. 01, Dec-2022 eISSN: 2617-3336 pISSN: 2617-3700 االسم اک اخدناین اظنم اور االصح اعمرشہ ںیم اس یک ونعمتی Family System of Islam and its Significance in Reformation of Society ĪQĀN (9) Vol. 05, No. 01, Dec-2022 تعنوییںرعالاظایداا Family System of Islam and its Significance in Reformation of Society ABSTRACT Received: 10-Oct-22 Accepted: 08-Jan-22 Online: 10-Jan-22 10-Oc Accep 08-Jan Onlin 10-Jan KEYW Family Islam, Reform Societ Mutua Coope Umma Harak Islam gives much importance to the individual, along with family, tribe, and community; Muslim Ummah is formed by mutual cooperation and brotherhood. It protects the rights and determines the duties of every section of society. Human life begins with the individual and spreads from one circle to another. The importance of family relationships in human relations is not in need of explanation. This family basically consists of parents, children, spouses, and siblings. Islam exhorts to pay the rights of all, the Prophet (peace and blessings of Allah be upon him) presented the best example of love to his family, his wives, and other relatives and exercised before the Ummah how the family should be organized and how to take full care of their rights. In the same way, Islam has instructed both men and women to respect each other's rights. Harakia is an essential part of Muslim society. The concept of the reformation of society is interrelated with the strong family system. In conclusion, the reformation of the family creates a reformed society. رصع احرض ںیم اعمرشے یک امتم رخاویبں یک اینبد مظن اخدنان ںیم رخایب ےہ۔ اخدنان اعمرشے یک ااکیئ بج کت اس ںیم رخایب وہ یگ اعمرشہ ونسر ںیہن اتکس۔ اینبد ڑیٹیھ وہ وت امعرت ات رثای یھب اجےئ وت ڑیٹیھ یہ رےہ یگ۔ االسیم اعمرشے یک وصخًاص االصح اک رتہبنی رطہقی ہی 43 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. کہا لبال اجات ےہ۔ ABSTRACT 01, Dec-2022 44 ےہ ہک اخدناین مظن ںیم رتہبی الیئ اجےئ وج وادلنی، زونیج اور اوخنی رپ لمتشم وہات ےہ۔ االسم وہ اک لم دنی ےہ سج ےناس اعمرشے یک ااک یئ یک االص ح ےک ذرےعی وپرے اعمرشے یک االصح اک رطفی اور یقطنم اقونن دای ےہ۔ اخدنان ےک امتم ارفاد اگل اگل اور اامتجیع وقحق االسم ےن تہب لیصفت ےک اسھت ایبن ےیک ںیہ۔ اس رصع وموجد ںیم اعمرش دن دبن رخایب اور اسفد یک رطف اجراہ ےہ اور االصح یک وکیئ لیبس ہجیتن زیخ ںیہن وہ ر یہ۔ سج یک وہج ہی ےہ ہک اینبدی یک درتس رکےن ےک اجبےئ ابیق امعرت وک درتس رکےن یک ریغ رطفت وکشش یک اجریہ ےہ ہک نیحلصم اورذہم داران یک وتہج اس رطف ذبمول رکایئ اجےئ اور االسم ےک اخدناین مظن یک الیصفتت وقحق و رفاضئ اور ذہم دارویں اک نیعت اور ان یک رتہب رطےقی ےس ادایگیئ ےک ابرے ںیم وپرے لیصفت ےس یقیقحت ادناز ںیم شیپ ایک اجےئ ات ہک افمدسےس ےتچب وہےئ اصمحل ےس اعمرشے وک ٓاراہتس ایک اج ےکس۔ االسم ںیم اخدناین مظن اک وصتر ااسنن یک ادتباء اخدنان ےس وہیت ےہ رعیب زابن ںیم اخدنان وک اعہلئ اور ریشعہ اہک اجات ےہ۔ ہکبج اردو ںیم زونیج و اوالد رپ لمتشم و ارفاد وک اخدنان اور ہبنک ےس ریبعت ایک اجات ےہ۔ :اوملوسہع ہیقف ںیم اخدنان یک رعتفی ان اافلظ ںیم یک یئگ ےہ ‘‘’’والارسۃ عشریۃ الرجل واھل بیتہ(1) ‘‘’’ااسنین اعمرشے ےک اخدنان اور الہ تیب وک اخدنان اہک اجات ےہ ہکبج اتج:ارعلوس ںیم ےہ ‘‘ ’’الارسۃ اقارب الرجل من قبل ابیہ(2) ‘‘’’ارسہ ینعی اخدنان اکی رمد ےک ریشعہ ینعی رھگ ےک ارفاد وک ےتہک ںیہ۔ وت ولعمم وہا ہک وہ ارفاد وج رہتش داری ںیم ابمہ اےھٹک ںیہ، ینعی زونیج، اوبنی، اوالد، نہب اھبیئ واےل رہتش ںیہ اجکی وہں ان وک اخ دنان کہا اجات ےہ۔ لبق از االسم اخدنان ملظ و متس ےک اوصولں رپ لمتشم اھت، سج ںیم رصف رمدوں وک یہ رہ مسق اک رشف و اشن و رمہبت احلص اھت ای دورسے ینعم ںیم ہی اہک اجاتکس ےہ ہک رصف ذمرک وک یہ اخص تیثیح احلص یھت۔ 1 ۔ ،ث ہ اوملوسۃع اہہیقفل، ومجمہع نم اابلحث ی ن ، (ومعق ادلررالسی ۱۴۳۳ھ)، ج۴، ص۲۲۳ 2 ۔ ،زدیبی، دمحم نب دبعارلزاق، اتج ارعلوس، (دارادہلاہی۱۴۲۳ھ)، ج۱۰، ص۹ ۵۱ ےہ ہک اخدناین مظن ںیم رتہبی الیئ اجےئ وج وادلنی، زونیج اور اوخنی رپ لمتشم وہات ےہ۔ االسم وہ اک لم دنی ےہ سج ےناس اعمرشے یک ااک یئ یک االص ح ےک ذرےعی وپرے اعمرشے یک االصح اک رطفی اور یقطنم اقونن دای ےہ۔ اخدنان ےک امتم ارفاد اگل اگل اور اامتجیع وقحق االسم ےن تہب لیصفت ےک اسھت ایبن ےیک ںیہ۔ اس رصع وموجد ںیم اعمرش دن دبن رخایب اور اسفد یک رطف اجراہ ےہ اور االصح یک وکیئ لیبس ہجیتن زیخ ںیہن وہ ر یہ۔ سج یک وہج ہی ےہ ہک اینبدی یک درتس رکےن ےک اجبےئ ابیق امعرت وک درتس رکےن یک ریغ رطفت وکشش یک اجریہ ےہ ہک نیحلصم اورذہم داران یک وتہج اس رطف ذبمول رکایئ اجےئ اور االسم ےک اخدناین مظن یک الیصفتت وقحق و رفاضئ اور ذہم دارویں اک نیعت اور ان یک رتہب رطےقی ےس ادایگیئ ےک ابرے ںیم وپرے لیصفت ےس یقیقحت ادناز ںیم شیپ ایک اجےئ ات ہک افمدسےس ےتچب وہےئ اصمحل ےس اعمرشے وک ٓاراہتس ایک اج ےکس۔ الںیادیمظاوص 44 ĪQĀN (9) Vol. ABSTRACT 05, No. ABSTRACT 01, Dec-2022 وعرت ای ڑلیک اکی ولظمم اور ذلیل یس زیچ یھت، اس یک اثمل ہی ےہ ہک ارگ رمد وفت وہ اجات اور اےنپ ےھچیپ اس ےن ویبی وھچڑی وہیت وت رمد یک دورسی ویبی ےک ےچب وک ہی قح احلص اھت ہک وہ اس ےس اشدی رک ےل اور اس رپ اانپ مکح الچےئ، ای رھپ اےس اشدی رکےن ےس یہ عنم رک دے۔ اور وراتث ےک دقحار رصف رمد یہ وہےت ےھت اور وعروتں اور وھچ ےٹ وچبں وک ھچک یھب احلص ںیہن وہات اھت، ر وت وعرت اچےیہ وہ امں وہ ای یٹیب رہ احتل ںیم اےس اقحرت یک اگنہ ےس داھکی اجات اھت۔ اس ےیل ہک وہ اتکس ےہ وہ دیقی نب اجےئ اےنپ اخدنان واولں ےک ےیل ذتل و اعر اک ابثع نب اجےئ وت ایس انب رپ ٓادیم اینپ یٹیب وک زدنہ دروگر:رک دای رکےت ےھت۔ اہلل ابترک و اعتٰیل ےن ایس یک رطف ااشرہ رکےت وہےئ رفامای ’’وَاِذَا بُِشّ َِ َ اَََُُھُْ اِ ا ُْ ى ََل وَْ ُُہ مُّْ دَا ا وَھُدَكَظِ ي مo َ َِّيَتَدَار ى مِنَ ال قَد مِ مِن سُّ د ءِِ مَا بُِش بِہٖ اَيُم ِْ كُہ عَل ھُد نٍ اَم يَُُسُّ ہ‘‘َِفِ الُّتَُّابِ اَاَ سَ اءَِ مَا َيَ ُكُ ُد ن1 ’’ان ںیم ےس بج یسک وک ڑلیک وہےن یک ربخ دی اجےئ وت اس اک رہچہ ایسہ وج اجات ےہ اور دل یہ دل ںیم ےنٹھگ اتگل ےہ، اس ربی ربخ یک وہج ےس وہ ولوگں ےس اتپھچ رھپات ےہ وساتچ ےہ ہک ایک اس ذتل وک اسھت ےیل وہےئ یہ رےہ ای اےس یٹم ںیم داب دے، ٓاہ وہ ایک یہ ربے ےلصیف رکےت ںیہ؟ اخدنان ےک ازجاےئ رتیبیک ںیم اکی رمد، اکی وعرت ینعی وادلنی اور ان ےک ےچب اشلم ںیہ وج ٓاسپ ںیم نہب اھبیئ ںیہ اخدنان یک اینبد اور اقب ےک ےیل رمد و وعرت اک ابیمہ قلعت رضوری ےہ ۔ اخدنان ےک دعب وعیس وہفمم ںیم ہلیبق ےہ وج اکی دورسے یک دمد و اعتون یک ااسس رپ اقمئ وہان اچےیہ ہی یھب زامہن اجتیلہ ںیم اوصل ملظ رپاقمئ اھت، االسم ےن ان بس طلغ اوصولں وک اٹم رک دعم و ااصنف رکےت وہےئ رہ دقحار وک اس اک قح دای یتح ہک دودھ ےتیپ ےچب وک یھب اس اک قح دالای اور ااج ےک وموجد ہ دور ںیم ویرپ ےک اخدنان وک دےنھکی اور اس رپ رظن دوڑاےن واال اےس ابلکل وٹاٹ وھپاٹ اور دجادجا دےھکی اگ وادلنی وک یسک مسق اک وکیئ قح ںیہن ہک وہ اوالد رپ ڑنکول رک ںیکس ہن رکفی اور ہن یہ االخیق اابتعر ےس۔ ویرپ ںیم ےٹیب وک ہی قح احلص ےہ ہک وہ اہجں اچےہ اور وج اچ ےہ رکات رھپے اےس وکیئ روےنک واال ںیہن اور یٹیب وک یھب ہی ٓازادی ےہ ہک وہ اہجں اور سج ےک اسھت رمیض ےھٹیب اور ٓازادی اور وقحق یک ادایگیئ ےک انم ےس سج ےک اسھت رہتش اقمئ رکے احتل ہی ےہ ریغ وملسمں ےک اہں اخدنان ابتہ احیل اک اکشر وہ اکچ ےہ ہکبج االسم وادلنی وک ہشیمہ زعت داتی ےہ اور اوالد یک رتتیب اک ذہم دار یھب۔ االسم اخدنان وک اکی دورسے ےک اسھت المےن یک ابت رکات ےہ اور اس ےک رہ رفد وک زدنیگ ںیم امہ رکدار داتی ےہ۔ ےسج ادا رک ےک اکی ااھچ اخدنان نب اتکس ےہ۔ 1الحنل۱۶ : ۵۸ ۵۹ ĪQĀN (9) Vol. 1 ۔ الحنل۱۶ : ۵۸۔۵۹ :ےنہک اگل ABSTRACT 05, No. 01, Dec-2022 45 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 االسم ےن وعرت وک امں، یٹیب اور نہب ےک روپ ںیم زع ت دی۔ امں ےک روپ ںیم اےس زعت دی اس ےک ابرے ںیم اوبرھریہ ریض اہلل ہنع ایبن رکےت ںیہ ہک یبنﷺ ےک اپس اکی صخش ٓارک :ےنہک اگل ’’اے اہلل اعتٰیل ےک روسل ﷺ ریمے نسح ولسک اک بس ےس زایدہ وکن قحتسم ےہ؟ یبن ﷺ ےن رفامای: ’’ ریتی امں اس ےن اہک اس ےک دعب رھپ وکن؟ یبنﷺ رفامےن ےگل: ریتی امں اس ےن اہک اس ےک دعب رھپ وکن؟ یبنﷺ رفامےن ےگل: ریتی امں اس ےن اہک ہک اس ےک دعب رھپ وکن؟ یبن ﷺ ےن رفامای: رھپ ریتا ‘‘ابپ۔( 1) سج اک ینعم ہی ےہ ہک ابپ ےس نیت انگ زایدہ امں ےک اسھت نسح ولسک اک مکح دای رفامای ایگ ویکہکن یہی امں اکی رعہص کت امتم وقحق ےس رحموم ریہ یھت۔یٹیب ےک روپ ںیم االسم ےن اےس ھچک اس رطح زعت دی۔ :یبن رکمیﷺ ےن رفامای ’’سج یک یھب نیت ایٹیبں ای نیت ںینہب، ای رھپ دو ایٹیبں ای دو ںینہب وہں اور وہ ان یک ایھچ رتتیب رکے اور ان ےک اعمامت ںیم اہلل اعتٰیل ےس ڈرات راہ وہ تنج ںیم‘‘اجےئ اگ۔(2) ویبی ےک روپ ںیم االسم ےن وعرت وک ھچک اس رطح زعت ےس ونازا۔ دیسہ اعہشئ ریض اہلل اعتٰیل اہنع ایبن رکیت ںیہ ہک :یبن رکمی ﷺ ےن رفامای ،’’مت ںیم بس ےس رتہب اور ااھچ وہ صخش ےہ وج اےنپ رھگ واولں ےک اسھت ااھچ رباتؤ رکات ےہ اور ںیم اےنپ رھگ واولں ےک ‘‘اسھت مت بس ںیم ےس رتہب رباتؤ رکات وہں۔ (3 ) االسم ےن ویبی ےک ابرہ ںیم وتیص یک اور وعرت وک اخودن ےک اایتخر ںیم یھب ٓازادی دی اور اس رپ رتتیب اوالد یک مسؤتیل اک اکی ڑبا 1 ۔ اوبداؤد، ننس ایب داؤد، امیلسن نب اثعش، اتکب االدب ، ابب یف رباولادلنی، (ریبوت: المث ک ث ہ ارصعلۃی، س ن)، رمق ادحلثی۵۱۴۲ ĪQĀN (9) Vol. 05, No. ABSTRACT 01, Dec-2022 االسم ےن وعرت وک امں، یٹیب اور نہب ےک روپ ںیم زع ت دی۔ امں ےک روپ ںیم اےس زعت دی اس ےک ابرے ںیم اوبرھریہ ریض اہلل ہنع ایبن رکےت ںیہ ہک یبنﷺ ےک اپس اکی صخش ٓارک :ےنہک اگل ’’اے اہلل اعتٰیل ےک روسل ﷺ ریمے نسح ولسک اک بس ےس زایدہ وکن قحتسم ےہ؟ یبن ﷺ ےن رفامای: ’’ ریتی امں اس ےن اہک اس ےک دعب رھپ وکن؟ یبنﷺ رفامےن ےگل: ریتی امں اس ےن اہک اس ےک دعب رھپ وکن؟ یبنﷺ رفامےن ےگل: ریتی امں اس ےن اہک ہک اس ےک دعب رھپ وکن؟ یبن ﷺ ےن رفامای: رھپ ریتا ‘‘ابپ۔( 1) سج اک ینعم ہی ےہ ہک ابپ ےس نیت انگ زایدہ امں ےک اسھت نسح ولسک اک مکح دای رفامای ایگ ویکہکن یہی امں اکی رعہص کت امتم وقحق ےس رحموم ریہ یھت۔یٹیب ےک روپ ںیم االسم ےن اےس ھچک اس رطح زعت دی۔ :یبن رکمیﷺ ےن رفامای ’’سج یک یھب نیت ایٹیبں ای نیت ںینہب، ای رھپ دو ایٹیبں ای دو ںینہب وہں اور وہ ان یک ایھچ رتتیب رکے اور ان ےک اعمامت ںیم اہلل اعتٰیل ےس ڈرات راہ وہ تنج ںیم‘‘اجےئ اگ۔(2) ویبی ےک روپ ںیم االسم ےن وعرت وک ھچک اس رطح زعت ےس ونازا۔ دیسہ اعہشئ ریض اہلل اعتٰیل اہنع ایبن رکیت ںیہ ہک :یبن رکمی ﷺ ےن رفامای ،’’مت ںیم بس ےس رتہب اور ااھچ وہ صخش ےہ وج اےنپ رھگ واولں ےک اسھت ااھچ رباتؤ رکات ےہ اور ںیم اےنپ رھگ واولں ےک ‘‘اسھت مت بس ںیم ےس رتہب رباتؤ رکات وہں۔ (3 ) االسم ےن ویبی ےک ابرہ ںیم وتیص یک اور وعرت وک اخودن ےک اایتخر ںیم یھب ٓازادی دی اور اس رپ رتتیب اوالد یک مسؤتیل اک اکی ڑبا ĪQĀN (9) Vol. 05, No. 01, Dec-2022 46 1 ۔ اوبداؤد، ننس ایب داؤد، امیلسن نب اثعش، اتکب االدب ، ابب یف رباولادلنی، (ریبوت: المث ک ث ہ ارصعلۃی، س ن)، رمق ادحلثی۵۱۴۲ 2 ۔ ،رتذمی، دمحم نب ٰیسیع، اجعم رتذمی، اوباب اربلواۃلصل، (ریبوت: دارارغلب االالسیم۱۹۹۸ ء)، رمق ادحلثی۱۹۱۲ 3 ۔ رتذمی، دمحم نب ٰیسیع، اجعم رتذمی، اوباب اانملبق، ابب لضف ازواج ایبنلﷺ، رمق ادحلثی۳۸۹۵ 46 ĪQĀN (9) Vol. 05, No. ABSTRACT 01, Dec-2022 ہصح راھک۔ االسم ےن امں اور ابپ رپ اوالد یک رتتیب ےک ابرہ ںیم تہب ڑبی مسؤتیل اور ذہم دا ری ریھک ےہ۔ دبعاہلل نب رمع ؓ ایبن رکےت ںیہ ہک :اوہنں ےن یبن ﷺ وک ہی رفامےت وہےئ انس مت ںیم ےس رہ اکی رسرباہ ےہ اور رہ اکی ےس اس یک راعای ےک قلعتم وسال وہاگ، اریم رایع ےہ وہ اےنپ اموتحتں ےک ابرہ ںیم وجاب دہ ےہ، اور ٓادیم اےنپ رھگ واولں رپ رسرباہ ےہ وہ ان ےک ،قلعتم وجاب دہ وہ اگ وعرت اخودن ےک رھگ رپ راہیع ےہ اےس اس ےک ابرہ ںیم وسال وہ اگ، اور الغم اےنپ امکل ےک امل اک رایع ےہ ‘‘اےس اس ےک ابرہ ںیم وسال وہ اگ۔(1 ) اور اخدنان ےک رہرفد ےک وقحق ےک اسھت ان ےک رفاضئ اک نیقی رکواای ایگ ےہ ینعی اعمرشے یک اس ااکیئ وک ےب اگلن ںیہن وھچڑا ہکلب وقادع و وضاطب اور اوصل ےک اسھت کلسنم و رموبط ایک ایگ ےہ ات یک ےب رہ اور دعم وت ازن اک اکشر ہن وہ۔ االسم ےن وادلنی ےک ادب و ارتح ام اور ان ےک وفیگت کت ااطتع رکےن اور ان اک ایخل اور ان ےک اسھت نسح ولسک رکےن اک مکح دای ےہ۔ اہلل احبسہن و اعت ٰیل اک ایس ہلسلس ںیم ھچک اس رطح :رفامن ےہ ٓ ٰ ًَِِّ ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. 1 ۔ ،ربتزیی، دمحم نب دبعاہلل، ٰوکشمۃ ااصملحیب، (ریبوت: المکت ب االالسیم۱۹۸۵ ء)، رمق ادحلثی۳۶۸۵ را ABSTRACT 01, Dec-2022 رف انےہ َ ’’وَقَٰض رَبُّكَ اَا تَع بُُُو ْٓا اِل َ َاَِّي ہُ وَاِ ل دَاِلِ َي نِ اِح َْ اًنا اِم ا يَب لُغَن عِن َُكَ ال كِب اَََُُھُمَآَ اَو ِكِ ُُّّّمَا فََلَ تَـقُل ل ُُمَآَ اُفٍَ و اَ تَن َُر ھُمَا وَقُل‘‘ل ُُمَا قَد اا كَرِي ماا(2) ’’اور ٓاپ ےک رب ےن اصف اصف ہی مکح دے راھک ےہ ہک مت اس ےک وسا یسک اور یک ابعدت ہن رکان اور امں ابپ ےک اسھت ااسحن رکان، ارگ اہمترے وموجدیگ ںیم ان ںیم ےس اکی ای وہ دوونں ڑباھےپ وک چنہپ اجںیئ وت ان ےک ٓاےگ اف کت ہن انہک، اور ہن اںیہن ڈاٹن ڈٹپ رکان ہکلب ان ےک اسھت ادب و ارتحام ےس ‘‘ابت تیچ رکان۔ وادلنی وج تقیقح ںیم اخدنان یک اینبد ںیہ ینعی ےلہپ وہ ابمہ ایمں ویبی ےک رےتش ںیم کلسنم وہےت ںیہ اور اب اوالد ےک ےیل وادلنی یک وصرت ںیم اس مظن ںیم رڑیھ یک ڈہی یک تیثیح ےس وموجد ںیہ۔ اس ےک دعب اوالد اک ابمہ رہتش ان وادلنی یک وہج ےس اقمئ وہا وت ہی نہب اھبیئ الہکےئ سج اک ینعم ہی وہا ہک مظن اخدنان ےک امتم رےتش ایس ااکیئ ینعی وادلنی ےک رموہن تنم ںیہ۔ اس ےیل رقٓان میکح ان ےک وقحق وج زع ت امں ابپ ےک اسھت ااسحن رکان، ارگ اہمترے وموجدیگ ںیم ان ںیم ےس اکی ای وہ دوونں ڑباھےپ وک چنہپ اجںیئ وت ان ےک ٓاےگ اف کت ہن انہک، اور ہن اںیہن ڈاٹن ڈٹپ رکان ہکلب ان ےک اسھت ادب و ارتحام ےس ‘‘ابت تیچ رکان۔ وادلنی وج تقیقح ںیم اخدنان یک اینبد ںیہ ینعی ےلہپ وہ ابمہ ایمں ویبی ےک رےتش ںیم کلسنم وہےت ںیہ اور اب اوالد ےک ےیل وادلنی یک وصرت ںیم اس مظن ںیم رڑیھ یک ڈہی یک تیثیح ےس وموجد ںیہ۔ اس ےک دعب اوالد اک ابمہ رہتش ان وادلنی یک وہج ےس اقمئ وہا وت ہی نہب اھبیئ الہکےئ سج اک ینعم ہی وہا ہک مظن اخدنان ےک امتم رےتش ایس ااکیئ ینعی وادلنی ےک رموہن تنم ںیہ۔ اس ےیل رقٓان میکح ان ےک وقحق وج زع ت وادلنی وج تقیقح ںیم اخدنان یک اینبد ںیہ ینعی ےلہپ وہ ابمہ ایمں ویبی ےک رےتش ںیم کلسنم وہےت ںیہ اور اب اوالد ےک ےیل وادلنی یک وصرت ںیم اس مظن ںیم رڑیھ یک ڈہی یک تیثیح ےس وموجد ںیہ۔ اس ےک دعب اوالد اک ابمہ رہتش ان وادلنی یک وہج ےس اقمئ وہا وت ہی نہب اھبیئ الہکےئ سج اک ینعم ہی وہا ہک مظن اخدنان ےک امتم رےتش ایس ااکیئ ینعی وادلنی ےک رموہن تنم ںیہ۔ اس ےیل رقٓان میکح ان ےک وقحق وج زع ت 47 47 ĪQĀN (9) Vol. ABSTRACT 05, No. ABSTRACT 01, Dec-2022 و ارتحام ےک اسھت ادا رکےن اک مکح دای ےہ ویکہکن ان ےک ووجد ےس وپرے اخدنان اک ووجد اور اس یک اینبد ےہ اخدنان ےک رہ رفد یک اچہپن اب اںیہن ےک وحاےل ےس وہیت ےہ اور بسن و بسح ںیم یھب اںیہن اک وحاہل اکم رکات ےہ۔ ارگ ان اک وحاہل متخ وہ اجےئ وت ااسنن ےب بسن و ےب بسح وہ اجات ےہ۔ ےکملسادںیرایویےچیالںیہوےہکلزدرالاداعیرنوصاتھ 48 و ارتحام ےک اسھت ادا رکےن اک مکح دای ےہ ویکہکن ان ےک ووجد ےس وپرے اخدنان اک ووجد اور اس یک اینبد ےہ اخدنان ےک رہ رفد یک اچہپن اب اںیہن ےک وحاےل ےس وہیت ےہ اور بسن و بسح ںیم یھب اںیہن اک وحاہل اکم رکات ےہ۔ ارگ ان اک وحاہل متخ وہ اجےئ وت ااسنن ےب بسن و ےب بسح وہ اجات ےہ۔ زمدی رباں االسم اخدنان اک وعیس رتنی وصتر راتھک ےہ۔ اکی ملسم اخدنان ںیم رصف ایمں ویبی اور ےچب یہ اشلم ںیہن وہےت ہکلب وادلنی، دادا، دادی، انان، اچچ، یچچ، وھپایھپں، امومں، اخہل وریغہ یھب اشلم وہےت ںیہ۔ االسم اےسی اخدنان اک اکی وصتر شیپ رکات ےہ وج وقحق و رفاضئ اور ولخص و تبحم، ااثیر ورقابین ےک اٰیلع رتنی یبلق ااسحاست اور ذجابت یک وبضمط ڈورویں ےس دنباھ وہا وہ۔ االسم اخدنان ےس ےننب واےل اعمرشے ےک ہلمج اعمالمت یک ااسس االخق وک انبات ےہ۔ وادلنی ےک رمےن ےک دعب یھب مظن اک اقتاض ےہ ہک ان اک الھبیئ ےک اسھت ذرک ایک اجےئ ان وک ااصیل وثاب ایک اجےئ اور ان ےک ےیل داع رفغمت یک اجےئ۔ نکیل رصف ہی اکیف ںیہن ہکلب ان ےک رےتش داروں ےک اسھت الھبیئ یک ابت اور ان ےک اسھت نسح ولسک ایک اجےئ ہک اخدنان ںیم ان یک وافت ےک دعب یھب ان اک اضیفن اجری ےہ اور ان اک رکشت اجری رےہ۔ وادلنی یک وافت ےک دعب ان ےس کین رضحت ادقس دمحم یفطصمﷺ ےن اتدیک رفامیئ ےہ ہک اوالد اےنپ وادلنی یک زدنیھ ںیم ان ےس نسح ولسک اک رباتؤ رکےن ےک العوہ ان یک وافت ےک دعب یھب رکشاہن ےک وطر رپ اس قلعت وک اقمئ رےتھک وہےئ کین اک ااہظر رکےت رںیہ انچہچن دحثی روسل وبقملﷺ ںیم اس رطح ایبن رفامایں ایگ ےہ۔ ہک اکی صخش:ونب ہملس اک احرض دختم وہا اور رعض رکےن اگل ’’َّيرسّّّّّدل ب ھّل بقر من یریشءبد رِ یشءیرھع بّہ بعُّ مدت عع قّالال ْعْ ال،َّّّّّلۃ علی ع والاسّّّّاتغلار ل ع واذْلاذ ع ُھع من بعُھع واّّّّلل الرو الإ اتداّّّّل اذا ب ع واذ رام ‘‘اُيق ع(1) ’’ ای روسل اہلل! ABSTRACT وادلنی یک وافت ےک دعب وکیئ ایسی کین ےہ وج ںیم ان ےک ےیل رک وکسں؟ ٓاپ ےن رفامای۔ اہں ویکں ںیہن۔ مت ان ےک ےیل داعںیئ رکو، ان ےک ےیل ششخب بلط رکو، اوہنں ےن وج ودعے یسک ےس رک رےھک ےھت اںیہن وپرا رکو۔ ان ےک زعزی ااقرب ےس ایس رطح ہلص ریمح اور نسح ولسک رکو سج رطح وہ اینپ زدنیگ ںیم ان ےک اسھت ایک رکےت ےھت اور ان ےک دووتسں ےک اسھت زعت و ارکام ےک 1 ۔ اتسجسین، ننس ایب داؤد، اتکب االدب، ابب یف رباولادلنی، رمق ادحلثی۵۱۴۲ یفطصےادیرضددمحوادلنی یک وافت ےک دعب ان ےس کین رضحت ادقس دمحم یفطصمﷺ ےن اتدیک رفامیئ ےہ ہک اوالد اےنپ وادلنی یک زدنیھ ںیم ان ےس نسح ولسک اک رباتؤ رکےن ےک العوہ ان یک وافت ےک دعب یھب رکشاہن ےک وطر رپ اس قلعت وک اقمئ رےتھک وہےئ کین اک ااہظر رکےت رںیہ انچہچن دحثی روسل وبقملﷺ ںیم اس رطح ایبن رفامایں ایگ ےہ۔ ہک اکی صخش:ونب ہملس اک احرض دختم وہا اور رعض رکےن اگل ’’ ای روسل اہلل! وادلنی یک وافت ےک دعب وکیئ ایسی کین ےہ وج ںیم ان ےک ےیل رک وکسں؟ ٓاپ ےن رفامای۔ اہں ویکں ںیہن۔ مت ان ےک ےیل داعںیئ رکو، ان ےک ےیل ششخب بلط رکو، اوہنں ےن وج ودعے یسک ےس رک رےھک ےھت اںیہن وپرا رکو۔ ان ےک زعزی ااقرب ےس ایس رطح ہلص ریمح اور نسح ولسک رکو سج رطح وہ اینپ زدنیگ ںیم ان ےک اسھت ایک رکےت ےھت اور ان ےک دووتسں ےک اسھت زعت و ارکام ےک 48 1 ۔ اتسجسین، ننس ایب داؤد، اتکب االدب، ابب یف رباولادلنی، رمق ادحلثی۵۱۴۲ 48 ثحد ĪQĀN (9) Vol. 05, No. ABSTRACT 01, Dec-2022 ‘‘اسھت شیپ ٓاؤ۔ دورِ احرض ںیم اخدناین مظن اور االصح اعمرشہ رصعِ احرض ےب امشر رخاویبں ےک ابثع الکشمت، اصمبئ اور ےب وکسین ےس دو اچر ےہ سج ےک تہب اسرے اابسب ںیہ ان ںیم اکی امہ ببس اعمرشے یک بس ےس اینبدی ااکیئ اخدنان ےک مظن ںیم دعم ااکحتسم، اشمورت یک یمک اور ابیمہ دعم اامتعد ےہ۔ ارفاد اخدنان اےنپ وقحق و رفاضئ ےس اغلف ںیہ۔ وقحق و رفاضئ اک اظنم سج ےن اعمرشے وک اینپ ذہم دارویں ےس ٓااگہ ارفاد ایہم رکان ےھت اور وخد ےب رطب ہکلب اینپ اینپ اافدتی وھک اکچ ےہ۔ سج یک اینبدی وہج االسم ےک اخدناین مظن ےس انواتیفق ےہ۔ ہکبج املسمونں ےن االسم وک رصف زابن کت دحمود رک ےک لمع ںیم اایغر یک ریپوی اک رکموہ دددنھہ رشوع ایک ےہ۔ بت ےس اس ےک امتم اعمالمت ااشتنر اک اکشر وہ رک رہ ےئگ ںیہ ۔رضورت وتق ہی ےہ ہک اےنپ اخدنان وک االسیم وطخط رپ اوتسار رک ےک وقحق و رفاضئ ےس ٓااگیہ یک اکی ایسی رحتکی اک ٓااغز ایکاجےئ وج لمع یک وقت ےس زمنی و ٓارہتس وہ ویکہکن ضحم ٓااگیہ اکیف ںیہن وہیت بج کت لمع یک وجادبیہ اک وصتر ہن وہ۔ اخدنان یک اینبد اخدنان یک اینبد الےاکوا نیایدا االسم ےن اکنح وک اخدنان یک اینبد انبای ےہ۔ االسم ےن دباکری وک رحام اور اکنح وک دنسپدیہ رقار دای ےہ۔ ویکہکن االسم اکی اخدنان وک وبضمط اینبدںی رفامہ رکان اچاتہ ےہ۔االسم یک روےس اخدنان یک وبضمط اینبد اکنح ےک ذرےعی ڑپیت ےہ۔ انچہچن وہ اس اینبد وک اخًاتصل ولخ ، ص، تبحم اپزیکیگ، دایدتناری اور وبضمط اعمدہے ےسیج:وھٹس امدے ےس ڈاےنل اک مکح داتی ےہ۔وسرۃ روم ںیم اراشد ابری اعتٰیل ےہ ’’وَمِن ا ي تِہَٖ اَن خَلَقَ لَُكُ مَِن اَْ لُِْ ُكُ اَز وَاجاا لَِتَْ كُنُد ْٓا اِلَی َُا وََْعَلَ بَی نَُكُ م دَا ۃا و رَ ْح َلا‘‘ ( 1) ’’اور اس یک اشنوینں ںیم ےس اکی ہی ےہ ہک اس ےن اہمترے ےیل وخد مت یہ ںیم ےس وجڑے دیپا ےیک ات ‘‘ہک مت ان ےک اپس وکسن احلص رکو اور اس ےن اہمترے درایمن تبحم اور رتمح دیپا رکدی ےہ۔ االسم یک روےس یسک وقم یک اقبء و السیتم اکنح ےسیج اپزیکہ دنبنھ یک رموہن تنم ےہ۔ اس داین ںیم ااسنن یک قیلخت اک بس ےس الہپ رمہلح اہلل اور اس ےک روسل ےک رطےقی ےکاطمقب رہتش ازدواج اک ایقم ےہ وج اخودن و ویبی ںیم ابیمہ اعتون، ایپر، اسن اور تبحم اک ومنہن وہ ، نج ےک اقمدص و ادہاف اسکیں وہےت ںیہ اور وہ ہی ہک ضحم نیکست امسجین ےک اجبےئ رواحین اپزیکیگ اقبء لسن اور ریمعت اعمرشہ اقمدص ںیم اشلم وہں۔ دوونں ایمں ویبی اےنپ رفاضئ یک ادایگیئ وک ںیھجمس اور ان یک ادا رکےن ںیم االخص اک اظمرہہ رکںی۔ اکی دورسے ےک وقحق یک 1 ۔ ارلوم۳۰ : ۲۱ 49 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 رپاوہ رکںی۔ اور رھپ رھگ ےک اموحل وک ااسی رپوکسن رںیھک ہک اس دحہقی تبحم ںیم رپوان ڑچےنھ وایل اوالد ےک وھپل لھک رک اعمرشے وک اینپ وخوبش ےس رطعم رک ںیکس۔ اس ےلسلس ںیم اایبنء رکام علی ھم االسلم وصخًاص اتدجار دمہنی انجب دمحم ﷺ اور ٓاپ یک ازدواج رہطمات ےک ابیمہ اقلعتت ےس روینش احلص رک ےک اےنپ رھگوں وک ونمر رکان اس دصقم ںیم اکایمیب یک دلیل وہ۔ رقٓان ےن ایمں ویبی اور اوالد ںیم رہتش تبحم وک ایبن :رکےت وہےئ اصنیحل یک اس داع وک وطبر اخص ذرک ایک ےہ َ ’’وَاَّل ِي نَ يَقُد لُد نَ رَب نَا ھَب لَنَا مِن اَز وَاِْنَا وَذُرَِيَ تِنَا قُر ۃَ اَع ُيُ ٍ و اْ عَل نَا لِل مُت قُِي‘‘ اِمَاماا(1) ’’ اور وہ ولھ وج ےتہک ںیہ: اے امہرے رپورداگر! اخدنان یک اینبد الےاکوا ںیمہ ویبویں او ر اوالد ےس ٓاوھکنں یک ڈنھٹک اطع رفام اور ‘‘ںیمہ یقتم ولوگں اک وشیپا انب۔ ٰٓاتی ںیم وقتیٰ و رپزیہ اگری وک ولوگں ای اخدناین ایپر و تبحم اک ہجیتن رقار دای ےہ اور اس ںیم وکیئ کش ںیہن ہک رھگ اک اموحل وقتی یک تمعن احلص رکےن ںیم دمد داتی ےہ۔ اور دورسی ہگج ویبی یک: قیلخت اک دصقم ایبن رکےت وہےئ رفامای ‘‘’’ھُدَاَّل ِي خَلَقَُكُ مَِن ْ ل سٍ و اََُِۃٍ و َْعَلَ مِن َُا زَو ََُْا لِیَْ كُنَ اِلَی َُا(2) ’’ ویہ وت ےہ سج ےن ںیہمت اکی اجن ےس دیپا ایک اور اس ےس اس یک ویبی انبیئ ات ہک اس ےک اہں وکسن ‘‘احلص رکے۔ وت ویبی یک رمد یک تیثیح ےس اور اس یک الص ےس قیلخت رکان یہ نیکست ابیمہ ںیم اینبد ےہ نکیل دعب ںیم اس تقیقح وک وھکل رک ایبن رفام دای ہک ویبی رمد ےک ےیل اوررمد ویبی ےک ےیل وکسن اک ابثع وہں وت زدنیگ اک اکروابر ےلچ اگ ورہن ان یک ےب وکسین اعمرشے یک ےب وکسین یک بنیا د ےہ۔ رتشمہک اخدناین اظنم رھگ ااسنن اک اینبدی قح ےہ۔ االسم ےن اس قح یک راعتی سج دقر یک ےہ اشدی یسک ذمبہ ںیم اس اک وصتر اس رطح ہن وہ، اس ےلسلس ںیم ہی ابت یھب ذنہ ںیم رینک اچےہ ہک االسیم اعمرشے ںیم رمد وچہکن اخدنان اک رسرباہ اور مظتنم ےہ۔ ذہلا اےس اس ابت اک اپدنب ایک ایگ ےہ ہک وہ اینپ دعتمد ازدواج ےک ےیل اگل اگل راہشئ ےک قح وک یتح ادقملور وپرا رکے۔ االسیم امیلعتت اک اطمہعل رکےن ےس ہی ابت واحض وہیت ےہ ہک االسم ابیمہ اعمالمت وک ارتحام و واقر ےس الچےن ےک ےیل اگل رطز راہشئ اک وصتر داتی ےہ۔ اراشد ابری اعتٰیل ےہ: رپاوہ رکںی۔ اور رھپ رھگ ےک اموحل وک ااسی رپوکسن رںیھک ہک اس دحہقی تبحم ںیم رپوان ڑچےنھ وایل اوالد ےک وھپل لھک رک اعمرشے وک اینپ وخوبش ےس رطعم رک ںیکس۔ اس ےلسلس ںیم اایبنء رکام علی ھم االسلم وصخًاص اتدجار دمہنی انجب دمحم ﷺ اور ٓاپ یک ازدواج رہطمات ےک ابیمہ اقلعتت ےس روینش احلص رک ےک اےنپ رھگوں وک ونمر رکان اس دصقم ںیم اکایمیب یک دلیل وہ۔ رقٓان ےن ایمں ویبی اور اوالد ںیم رہتش تبحم وک ایبن :رکےت وہےئ اصنیحل یک اس داع وک وطبر اخص ذرک ایک ےہ مظایداہشتر رھگ ااسنن اک اینبدی قح ےہ۔ االسم ےن اس قح یک راعتی سج دقر یک ےہ اشدی یسک ذمبہ ںیم اس اک وصتر اس رطح ہن وہ، اس ےلسلس ںیم ہی ابت یھب ذنہ ںیم رینک اچےہ ہک االسیم اعمرشے ںیم رمد وچہکن اخدنان اک رسرباہ اور مظتنم ےہ۔ ذہلا اےس اس ابت اک اپدنب ایک ایگ ےہ ہک وہ اینپ دعتمد ازدواج ےک ےیل اگل اگل راہشئ ےک قح وک یتح ادقملور وپرا رکے۔ االسیم امیلعتت اک اطمہعل رکےن ےس ہی ابت واحض وہیت ےہ ہک االسم ابیمہ اعمالمت وک ارتحام و واقر ےس الچےن ےک ےیل اگل رطز راہشئ اک وصتر داتی ےہ۔ اراشد ابری اعتٰیل ےہ: 1 ۔ ارلدع۳۸ : ۱۳ افر 2 ۔ ارفلاقن۲۵ : ۷۴ 50 ĪQĀN (9) Vol. اخدنان یک اینبد الےاکوا 05, No. 01, Dec-2022 ‘‘ ’’وَاذ كُر نَ مَا يُت ٰل ِفِ بُیُد تِكُن(1) ‘‘’’اور اہمترے رھگ ںیم وج دخا یک ٓاںیتی ڑپیھ اجیت ںیہ ان وک ایدروھک۔ اس ٓاتی ےس ولعمم وہات ےہ ہک امتم ازدواج رہطماتؓ ےک اگل اگل رھگ ےھت۔ ایس ےیل ویبت عمج ہغیص اامعتسل وہا ےہ۔ ااحدثی و ریست یک بتک ںیم ازو اج رہطماتؓ ےک رھگوں اکذرک اکیف لیصفت ےس اتلم ےہ۔ وشرہ اک رھگ یہ ویبی اک رھگ وہات ےہ۔ نکیل اکی ویبی اک رھگ دورسی ویبی اک رھگ ںیہن وہات۔ اس ےیل یبن رکمیﷺ ےن اینپ امتم ازواج ےک رھگدجادجاانبےئ۔ ہکلب ارگدوران رفس اکی ےس زادئ ازواج ٓاپﷺ یک تیعم ںیم وہںیت وت رفس یک وساری اک یھب اگل اگل ااظتنم اک اجات۔ ریثک االزواج اخدناین مظن اور رھگولی زدنیگ اک ااکحتسم اس ںیم ےہ ہک زونیج ےک ےیل اگل رھگ اک اخرط وخاہ ااظتنم ایک اجےئ ات ہک وہ اینپ تبحم رھبی زدنیگ اک ٓازادی ےس فطل اُاھٹ ںیکس۔ رتتیب اوالد ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 52 یٹیب یک دیپاشئ رپامےھت رپ نکش اک ٓاان، اےس اندنسپدییگ ےس وبقل رک ان اہلل اعتٰیل ےک ےلصیف ےس رورگداین ےہ اور تمکح اہیہٰل وک امےنن ےس اگای ااکنر ےہ۔ ہی زامہن اجتیلہ یک ابایقت ںیم ےس ےہ ہک یٹیب وک وحنمس اھجمس اجےئ۔ رصع دوراں یک ہی تہب ڑبی رخایب ےہ ہک ےٹیب ےک اقمےلب ںیم یٹیب وک ضعب ولگ ااھچ ایخل ںیہن رکےت اور اےسی رھگوں اور اخدنان ںیم اسملئ دیپا وہےت ںیہ اور ایسی یہ ایٹیبں اعمرشے یک االصح ںیم اانپ رکدار ادا ںیہن رک ںیتکس اور سج اعمرشے ںیم یٹیب یک رتتیب رپ یھب وکیئ وتہج ںیہن دی اجیت اور ہجیتن ہی اتلکن ےہ ہک ریغ رتتیب ایہتف یٹیب بج امں اک بصنم احلص رکیت ےہ وت ٓاےن وایل لسن اینبدی رتتیب: ےس رحموم وہ اجیت ےہ۔دحثی اپک ںیم ےہ ’’ رضحت اوبدیعس دخریؓ ےس رواتی ےہ ہک یبن رکمی ےن رفامای سج صخش یک نیت ایٹیبں ای نیت ںینہب وہں ای دوایٹیبں ای دو ںینہب وہ اوروہ ان ےک اسھت تہب اےھچ رطہقی ےس زگارے ان ےک وقحق ادا رکے وج رشتعی ےن رقمر ےئک ںیہ اور ان ےک وقحق یک ادایگیئ ںیم اہلل ےس ڈرات ےہ وت اہلل اعتٰیل اس یک دبوتل اس ‘‘وک تنج ںیم دالخ رفامںیئ ےگ۔( 1) وت وعروتں ےک ھتس ریغ انمبس ولسک اور ان یک قح یفلت ریغ االسیم ارثات اک ہجیتن ںیہ اور ہی تہب ڑبی اعمرشیت رخایب نب یئگ ںیہ۔ اس ےیل االسیم امیلعتت وک اسےنم رںیھک ےگ وت یسک مسق یک رپاشیین اور قح یفلت وک ونتب ہن ٓاےئ یگ۔ دشتد و یتخس اک روہی ضعب رضحات دشتمد اور تخس زماج ےس وہےت ںیہ اور ویبی وچبں رپ ےب اجیتخس رکےت ںیہ سج یک وہج ےس ہن ویبی یک االصح وہ یتکس ےہ ہن اوالد یک ہکلب رمد ےک الخف اکی االتخف ےک وطر رپ اباقدعہ وادلہ یک رسرپیتس ںیم اکی رگوہ منج اتیل ےہ سج اک اکم رسرباہ اخہن یک رہب وصرت اخمتفل رکان وہات ےہ۔ نکیل اس ےک ےھچیپ رحمک رسرباہ اخہن یک ےب اجیتخس وہیت ےہ وج وپرے رھگولی اموحل وک ےب وکسین و ےب اانیمط ین یک تیفیک ےس دواچر رک دیتی ےہ۔ یسب اواقت اس تخس یک :ےک ھچک اابسب یھب وہ ےتکس ںیہ ےسیج ۱۔ رغتب و االفس ۲۔ زایدہ وچبں اک وہان ۳۔ ایمں ویبی ےک امنیب دعم ٓایگنہ ۵۔ وعروتں وک رفامربندار انبےن یک ومرویث وخاشہ 1 ۔ :رتذمی، اننسل ارتلذمی، اوباب اویبلع، ابب اماجء یف ااجتلر، رمق ادحلثی۱۹۲۳ ĪQĀN (9) Vol. اخدنان یک اینبد الےاکوا 01, Dec-2022 رہتش ازدواج اک اکی اینبدی او رامہ دصقم اوالد اک وصحل اوررھپ اس یک رتتیب ےہ اور ہی دوونں ایمں ویبی یک رتشمہک وکشش ےس نکمم وہ اتکس ےہ امں اک رکدار اس ںیم تہب امہ ےہ سج یک وگد ااسنتین یک یلہپ دراگسہ رقار اپیت ےہ۔ نکیل ااحدثی ںیم رتتیب یک تبسن رم د یک رطف یک یئگ ےہ اسیج ہک رفامن وبنی ےہ’’سج ےن نیت ویٹیبں یک رتتیب یک اور رھپ اےس ایبہ دای وت وہ ایقتم ےک دن ریمے اسھت وہاگ۔ میلعت و ملعت اور رتتیب، االصح اعمرشہ اک ٓااغز ںیہی ےس وہات ےہ ہک وادلنی اینپ اوالد ےک ےلسلس ںیم اس ذہم داری وک وپری رکںی اور اعمرشے وک ایسی لسن ایہم رکںی وج ااسفد ےک اجبےئ ا الصح ےس، یفنم وسچ ےک اجبےئ تبثم رکف ےس اورجک روی ےک اجبےئ ااقتستم راہ ےس امونس وہ۔ اوالد ںیم ےٹیب اور ایٹیبں دوونں اشلم ںیہ اور ہی اہلل اعتٰیل یک تیشم رپ وموقف ےہ ہک وہ سک وک وکن ےس تمعن داتی ےہ۔انچ ںیےہچن رقٓان :لِْمیکح ںیم ےہ َ ’’لِلِ َ ِ مُْل ُ الْ م د تِ وَاا ر ضِ َيَ لُقُ مَا يَشَ اُِ يََُبُ لِمَن َي شَ اُِ اًِنَ ًثا و يََُبُ لِمَن ي شَ اُِ اَّلُّكُد رo ُِاَو يُزَوَُُُِْْ ذُك رَاًنا و اًِنَ ًثا وََيَ عَلُ مَن ي شَ ا ‘‘عَــقِـی ماا اِْ ہ عَلِي م قَُِي رم (2) ’’ٓاامسونں اور زنیم یک ابداشتہ رصف اہلل ےک ےیل ےہ وہ وج اچےہ دیپا رکات ےہ سج وک اچاتہ ےہ ایٹیبں داتی ےہ اور سج وک اچاتہ ےہ ےٹیب داتی ےہ اور سج وک اچاتہ ےہ ےٹیب اور ایٹیبں دوونں اطع رکات ےہ اور سج وک ‘‘اچاتہ ےہ ابھجن رک داتی ےہ۔ 51 ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 05, No. 01, Dec-2022 52 ثحد ĪQĀN (9) Vol. 05, No. 01, Dec-2022 53 ۶۔ وکشک و ابہشت االسم دنی رطفت وہےت وہےئ وقحق ااسنین یک زعت و ارتحام یک میلعت داتی ےہ، ات ہک اعمرشے ںیم انم و تبحم وک رفوغ ےلم، اور وقحق یک اپامیل ہن وہ۔ االسم ےن اس یقیلخت اایتمز وک اٹمای اور ہی اتبای ہک رمد اور وعرت یک قیلخت یک اینبد اکی یہ ےہ۔ وطبر ااسنین یسک وک یسک رپ فض لت ب ای :وفتیق احلص ںیہن ےہ۔ اسیج ہک اراشد ابری اعتٰیل ےہ ’’َّي ََيُّ َا الن اسُ ات قُد ا رَی ُكُ ُ اَّل ِي خَلَقَُكُ مَِن ْ ل سٍ و اََُِۃٍ و خَلَقَ مِن َا زَو ََُْا وَبَث مِن ُُمَارِجَااا كَثِری اا‘‘و نَِْ اِا(1) ’’اے ولوگ! اخدنان یک اینبد الےاکوا اےنپ رب ےس ڈرو، سج ےن اہمتری دیپاشئ(یک ادتبا) اکی اجن ےس یک اور ایس ےس اس اک وجڑا دیپا رفامای رھپ ان دو‘‘ونں ںیم ےسرثکبت رمدوں اور وعروتں(یک قیلخت)وک الیھپ دای۔ بج کت ایمں ویبی ےک درایمن ابیمہ ارتحام، اامتعد و نیقی اور االخص ہن وہ وت رھگ اک اموحل درتس ںیہن وہ اتکس اور رھگ اک اموحل درتس ہن وہ وت وچبں یک رتتیب ًانیقی اتمرث وہ یگ۔ ایمں ویبی ےک درایمن االتخف ےس ہی ہجیتن ینیقی وہات ےہ ہک ےچب یسک اکی ےک اسھت واہتسب وہ اجںیئ ای ھچک ابپ ےک اسھت اورھچک امں ےک اسھت سج ےس دو احتمرب رگوہ ووجد ںیم ٓااجےت ںیہ اور اسری زدنیگ زونیج یھب اور رھپ اوالد یھب شکمشک یک ز دنیگ زگارےن رپ وبجمر وہ اجیت ےہ۔ ایمہن روی اک دقفان ارگزونیج ںیم ےس وکیئ اکی اریم اور دورسا رغبی رھگاےن ےس قلعت راتھک ےہ وت یھب اخدناین مظن ںیم اسملئ دیپا وہےت ںیہ۔ ویکہکن ان ےک اخدناونں ےک رنہ نہس، رطز زدنیگ ںیم رفق وہ اگ وج اسب اواقت ومعمیل انتزاعت ےس ڑباتھ وہا دشت اایتخر رک اجات ےہ اور ونتب لط ق کت اج یتچنہپ ےہ۔ :وعسمد ادمح اڈیوٹیک ررطمقاز ںیہ ’’اعمیش دعام وتازن اخدناونں یک تسکش و رتخی ںیم امہ رکدار ادا رکات ےہ اور وصخًاص وعرںیت اس ےس زایدہ اتمرث وہیت ںیہ وج اعمرشیت رتیق ںیم اینپ اور اوالد ےک لبقتسم یک رتہب وجتسج ےک ےیل رپاشین ریتہ (‘‘ںیہ۔2) دورسی رطف االسم ےن اخودن ےک ذہم ویبی ےک وقحق ںیم رہم، ہقفن، اور راہشئ وک اشلم ایک ےہ اور اس ںیم اابتعر رمد یک تیثیح ایک 1 ۔ ااسنلء۴ :۱ 2 ۔ ،ہٹھب، وعسمد ادمح، ایمں، ایحت ااسنلء، (ٓانہ ادارہ ااشتع و قیقحت۲۰۱۵ ء)، ص۳۵۲ ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ۶۔ وکشک و ابہشت االسم دنی رطفت وہےت وہےئ وقحق ااسنین یک زعت و ارتحام یک میلعت داتی ےہ، ات ہک اعمرشے ںیم انم و تبحم وک رفوغ ےلم، اور وقحق یک اپامیل ہن وہ۔ االسم ےن اس یقیلخت اایتمز وک اٹمای اور ہی اتبای ہک رمد اور وعرت یک قیلخت یک اینبد اکی یہ ےہ۔ وطبر ااسنین یسک وک یسک رپ فض لت ب ای :وفتیق احلص ںیہن ےہ۔ اسیج ہک اراشد ابری اعتٰیل ےہ ’’َّي ََيُّ َا الن اسُ ات قُد ا رَی ُكُ ُ اَّل ِي خَلَقَُكُ مَِن ْ ل سٍ و اََُِۃٍ و خَلَقَ مِن َا زَو ََُْا وَبَث مِن ُُمَارِجَااا كَثِری اا‘‘و نَِْ اِا(1) ’’اے ولوگ! اخدنان یک اینبد الےاکوا 01, Dec-2022 ےہ بج ویبی اک ایعمر زدنیگ اٰیلع اور اخودن اک ایعمر تسپ وہ وت اس ےس دعم وتازن دیپا وہات ےہ۔ وصخًاص ویبی اےنپ اور اخودن اےنپ ایعم ر رپ اقمئ وہ اور ارگ اخو دن اینپ تیثیح ےس ڑبھ رکانن و ن قفہ یک وکشش رکے اگ وت ہی زایدہ دری کت ںیہن لچ ےکس اگ اور ابالٓرخ ونتب االتخف کت اج ےچنہپ یگ۔ رصع احرض ےک اعمرشے ںیم اسب اواقت رمد اینپ تیثیح ےس اٰیلع تیثیح یک وعرت ےس اشدی رک اتیل ےہ اور وبتق دقع اینپ تیثیح وھچیٹ اظرہ رکات ےہ ۔ دقع اکنح اور اشدی ےک دعب یلمع زدنیگ ےک ادتبایئ دونں ںیم وہ اانپ رھبم اھبنےن ےک ےیل اینپ یقیقح تیثیح ےس زایدہ رخچ رکات ےہ اور ویبی اس رطز زدنیگ رپ اعدی وہ اجیت ےہ۔ نکیل اس ایعمر وک زایدہ دری کت ربرقار رانھک رمد ےک سب ںیم ںیہن وہات سج یک وہج ےس تہب وہ تہب دلج ےب سب وہ اجات ےہ اور اب ہن وعرت یسک تحلصم رپ رایض وہیت ےہ اور ہن یہ رمد اس یک رضورایت ےلہپ ایعمر رپ وپرا رکاتکس ےہ سج یک وہج ےس االتخف ےس رھگ ںیم ےب وکسین اک وہان الزیم ارم ےہ۔ ہکبج رقٓان میکح ےن ایمہن روی اور ادتعال اک مکح دای ےہ۔ :انچہچن رفامای ‘‘’’وَاَّل ِي نَ اِذَیشَ اَْ لَقُد ا لَْ يُْس ِفُد ا وَلَْ يَـق ُّتُ ُو ا وََكَ نَ بَُي َ ذ ِلِ َ قَدَاماا (1 ) ’’ الہ اامین بج رخچ رکےت ںیہ وت ہن ارساف رکےت ںیہ ہن یمک رکےت ںیہ ہکلب وہ اس ےک درایمن اقمئ ‘‘رےتہ ںیہ۔ ارگ رھگ ےک ارخااجت ںیم ایمہن روی ےس اکم ایل اجےئ وت زدنیگ ںیم ٓااسین دیپا وہ اجیت ےہ۔ اس ےیلرشتعی ںیم وفک اےنپ ربارب یک تیثیح یک وعرت ےک اسھت اشدی اک مکح دای ےہ سج ںیم اکی تمکح ہی یھب ےہ ہک دعب ںیم انن ہقفن ںیم االتخف ہن وہ۔ رقٓان میکح ےن رمد ےک ےیل وج اوصل اک انن و ہقفن دای ےہ وہ ذلی یک ٓاتی ںیم واحض ےہ۔ ‘‘ ِ’’وَعَلَ ال مَد لُد اِ لَہ رِز قُُُن وَكِْ دَتُُُن اِ ل مَع رُو ف (2 ) ‘‘’’اور نج ےک ےچب ںیہ ان ےک ذہم ان وعروتں اک رویٹ ڑپکا اور راہشئ دوتسر ےک اطمقب ےہ۔ :اور اکی دورسے اقمم رپ اہلل اعتٰیل ےن ھچک اس رطح رفامای ’’لِیُن ‘‘ُلِق ذُو سَ عَلٍ مَِن سَ عَتِہٖ وَمَن قُُِرَ عَلَی ہِ رِز قُہ فَل یُن لِق مِم آَ ا ت ىہُ ب (3 ) ĪQĀN (9) Vol. 1 ۔ ارفلاقن۲۵ : ۶۷ 2 ۔ ارقبلۃ۲ : ۲۳۳ 3 ۔ االطلق۶۵ :۷ اخدنان یک اینبد الےاکوا اےنپ رب ےس ڈرو، سج ےن اہمتری دیپاشئ(یک ادتبا) اکی اجن ےس یک اور ایس ےس اس اک وجڑا دیپا رفامای رھپ ان دو‘‘ونں ںیم ےسرثکبت رمدوں اور وعروتں(یک قیلخت)وک الیھپ دای۔ بج کت ایمں ویبی ےک درایمن ابیمہ ارتحام، اامتعد و نیقی اور االخص ہن وہ وت رھگ اک اموحل درتس ںیہن وہ اتکس اور رھگ اک اموحل درتس ہن وہ وت وچبں یک رتتیب ًانیقی اتمرث وہ یگ۔ ایمں ویبی ےک درایمن االتخف ےس ہی ہجیتن ینیقی وہات ےہ ہک ےچب یسک اکی ےک اسھت واہتسب وہ اجںیئ ای ھچک ابپ ےک اسھت اورھچک امں ےک اسھت سج ےس دو احتمرب رگوہ ووجد ںیم ٓااجےت ںیہ اور اسری زدنیگ زونیج یھب اور رھپ اوالد یھب شکمشک یک ز دنیگ زگارےن رپ وبجمر وہ اجیت ےہ۔ ایمہن روی اک دقفان ارگزونیج ںیم ےس وکیئ اکی اریم اور دورسا رغبی رھگاےن ےس قلعت راتھک ےہ وت یھب اخدناین مظن ںیم اسملئ دیپا وہےت ںیہ۔ ویکہکن ان ےک اخدناونں ےک رنہ نہس، رطز زدنیگ ںیم رفق وہ اگ وج اسب اواقت ومعمیل انتزاعت ےس ڑباتھ وہا دشت اایتخر رک اجات ےہ اور ونتب لط ق کت اج یتچنہپ ےہ۔ :وعسمد ادمح اڈیوٹیک ررطمقاز ںیہ ’’اعمیش دعام وتازن اخدناونں یک تسکش و رتخی ںیم امہ رکدار ادا رکات ےہ اور وصخًاص وعرںیت اس ےس زایدہ اتمرث وہیت ںیہ وج اعمرشیت رتیق ںیم اینپ اور اوالد ےک لبقتسم یک رتہب وجتسج ےک ےیل رپاشین ریتہ (‘‘ںیہ۔2) دورسی رطف االسم ےن اخودن ےک ذہم ویبی ےک وقحق ںیم رہم، ہقفن، اور راہشئ وک اشلم ایک ےہ اور اس ںیم اابتعر رمد یک تیثیح ایک 1 ۔ ااسنلء۴ :۱ 2تحقیقتانٓنسایایامدسعوھٹہ ĪQĀN (9) Vol. 05, No. 01, Dec-2022 53 حقیق ĪQĀN (9) Vol. 05, No. 1 ۔ ،ث ہ وأایہم (احیحصل ااخبلری)، اتکب المظ ا مل وابصغل اخبری، دمحم نب اامسلیع، ااجلعم ادنسمل احیحصل ارصتخمل نم أومر روسل اہللﷺ و سی ابب اصقس ،المظ لوؤم اذاودجامل اظہمل، (داروطق ااجنلۃ۱۴۲۲ ھ)، رمق ادحلثی۲۴۶۰ نو 2 ۔ اونلر۲۴:۴ اخدنان یک اینبد الےاکوا 05, No. 01, Dec-2022 رصع احرض ےک اعمرشے ںیم اسب اواقت رمد اینپ تیثیح ےس اٰیلع تیثیح یک وعرت ےس اشدی رک اتیل ےہ اور وبتق دقع اینپ تیثیح وھچیٹ اظرہ رکات ےہ ۔ دقع اکنح اور اشدی ےک دعب یلمع زدنیگ ےک ادتبایئ دونں ںیم وہ اانپ رھبم اھبنےن ےک ےیل اینپ یقیقح تیثیح ےس زایدہ رخچ رکات ےہ اور ویبی اس رطز زدنیگ رپ اعدی وہ اجیت ےہ۔ نکیل اس ایعمر وک زایدہ دری کت ربرقار رانھک رمد ےک سب ںیم ںیہن وہات سج یک وہج ےس تہب وہ تہب دلج ےب سب وہ اجات ےہ اور اب ہن وعرت یسک تحلصم رپ رایض وہیت ےہ اور ہن یہ رمد اس یک رضورایت ےلہپ ایعمر رپ وپرا رکاتکس ےہ سج یک وہج ےس االتخف ےس رھگ ںیم ےب وکسین اک وہان الزیم ارم ےہ۔ ہکبج رقٓان میکح ےن ایمہن روی اور ادتعال اک مکح دای ےہ۔ انہچراا 3 ۔ االطلق۶۵ :۷ 54 54 ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 ’’اور اشکدیگ واال اینپ اشکدیگ ںیم ےس رخچ رکے اور سج رپ رزق یک یگنت وہ اےس وج ھچک اہلل اعتٰیل ےن دای ‘‘ےہ اس ںیم ےس رخچ رکان اچےیہ۔ دنھتنب ہبتعؓ وج ہک اوبایفسنؓ ہنع یک ویبی ےن بج یبن رکمیﷺ ےس اکشتی یک ہک اوبایفسنؓ اس رپ رخہچ ںیہن رکات وت یبنﷺ ےن :اںیہن رفامای اھت ہک ‘‘’’ٓاپ اےنپ اور اینپ اوالد ےک ےیل وج اکیف وہ رعموف رطےقی ےس ےل ایل رکو۔( 1) کش اور دبامگین کش اور دبامگین ایسی امیبرایں ںیہ ارگ روتشں ںیم ٓا اجںیئ وت ان ںیم وٹٹ وھپٹ اک ابثع نب اجیت ںیہ اور اکی دورسے ےک اسھت انتؤ یک تیفیک دیپا وہ اجیت ےہ۔ ایمں ویبی ےک درایمن ارگ ابیمہ اامتعد وہ وت ہی رہتش اپدیئار و مکحتسم اینبدوں رپ اقمئ راتہ ےہ۔ نکیل ا رگ اس اامتعد ںیم یمک ٓااجےئ ای رسے ےس متخ وہ اجےئ اور اس یک ہگج کش ٓااجےئ وت رھپ اخدنان اور ازدوایج زدنیگ اک وکسن اقمئ رانہ لکشم وہ اجات ےہ۔ :العہم انب دقاہم رفامےت ںیہ ’’بج اخودن اینپ اپابکزویبی رپ تمہت اگلےئ وت اخودن رپ دحوابج وہ یگ اور اس وک افقس اک مکح دای اجاگیئ اور اس یک وگایہ وبقل ںیہن وہ یگ، االہی ہک وہ اس تمہت یک وکیئ دلیل شیپ رکے ای رھپ اعلن رکے اور ارگوہ قمی س ن ااھٹےن ےس یھب ارتحاز رکات ےہ وت اس رپ ہی بس الزم اچر وگاہ شیپ ںیہن رکات ای رھپ اعلن رکےن ینعی :اور الوگ وہاگ، اس یک دلیل اہلل احبسہن و اعتٰیل اک ہی رفامن ےہ ِ ٰ َّ’’وَاَّل ِي نَ يَر مُد نَ ال مُح ص َُُْ ُثُ لَْ َّيَ تُد ا اِ َر بَعَلِ شُّ ََُُاَِ فَاج دِ ُو ھُْ ََم نُِي َ جَد َۃا و اَ تَــّّّّّق بَلُد ا ل ‘‘َشَ َُااَۃا اَبَُاا وَاُول ىِٕكَ ھُُْ ال ل ِْ قُد ن(2) ’’اور وج ولگ اپدکانم وعروتں رپ زان یک تمہت اگلںیئ رھپ اچر وگاہ شیپ ہن رک ںیکس وت اںیہن ایس وکڑے اگلؤ ‘‘اور یھبک ان یک وگایہ یھب وبقل ہن رکو ہی افقس ولگ ںیہ۔ 1 ۔ ،ث ہ وأایہم (احیحصل ااخبلری)، اتکب المظ ا مل وابصغل اخبری، دمحم نب اامسلیع، ااجلعم ادنسمل احیحصل ارصتخمل نم أومر روسل اہللﷺ و سی ابب اصقس داملاظہمل ( ا وطقااجنلۃ المظ ل ماذا۱۴۲۲) مقادحلثی۲۴۶۰ ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 ’’بج اخودن اینپ اپابکزویبی رپ تمہت اگلےئ وت اخودن رپ دحوابج وہ یگ اور اس وک افقس اک مکح دای اجاگیئ اور اس یک وگایہ وبقل ںیہن وہ یگ، االہی ہک وہ اس تمہت یک وکیئ دلیل شیپ رکے ای رھپ اعلن رکے اور ارگوہ قمی س ن ااھٹےن ےس یھب ارتحاز رکات ےہ وت اس رپ ہی بس الزم اچر وگاہ شیپ ںیہن رکات ای رھپ اعلن رکےن ینعی :اور الوگ وہاگ، اس یک دلیل اہلل احبسہن و اعتٰیل اک ہی رفامن ےہ ہےار یہایٰعاہبحالہیلیاووا ِ ٰ َّ’’وَاَّل ِي نَ يَر مُد نَ ال مُح ص َُُْ ُثُ لَْ َّيَ تُد ا اِ َر بَعَلِ شُّ ََُُاَِ فَاج دِ ُو ھُْ ََم نُِي َ جَد َۃا و اَ تَــّّّّّق بَلُد ا ل ‘‘َشَ َُااَۃا اَبَُاا وَاُول ىِٕكَ ھُُْ ال ل ِْ قُد ن(2) ’’اور وج ولگ اپدکانم وعروتں رپ زان یک تمہت اگلںیئ رھپ اچر وگاہ شیپ ہن رک ںیکس وت اںیہن ایس وکڑے اگلؤ ‘‘اور یھبک ان یک وگایہ یھب وبقل ہن رکو ہی افقس ولگ ںیہ۔ 1 ۔ ،ث ہ وأایہم (احیحصل ااخبلری)، اتکب المظ ا مل وابصغل اخبری، دمحم نب اامسلیع، ااجلعم ادنسمل احیحصل ارصتخمل نم أومر روسل اہللﷺ و سی ابب اصقس ،المظ لوؤم اذاودجامل اظہمل، (داروطق ااجنلۃ۱۴۲۲ ھ)، رمق ادحلثی۲۴۶۰ نو 55 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 امہرے اعمرشے ںیم ریغ االسیم روسم و رواج وک اانت دلخ ےہ ہک ان ےس انکرہ یشک رکان اغبوت ےک رتمادف ےہ اور وج ان روسم ورواج وک رتک رکے اس ےک الخف نعط و عینشت اک اکی وطافن رباپ وہ اجات ےہ۔ ان روسم و رواج یک وہج ےس وقحق یک قح یفلت وہیت ےہ۔ رقٓان ےس اشدی، وراتث ےس رحمویم ، ریغت ےک انم رپ لتق ، وہٹ ہٹس یک اشدی وریغہ۔ ہی بس اجالہہن روسم ںیہ اور االسیم امیلعتت ےس انواتیفق اس اک ڑبابسےہ اشدی زدنیگ ںیم اکی امہ ومڑ ےہ سج ےس وادلنی ، اھبویئں، ونہبں اور دووتس ں ےن تہب اسری رومسں وک کلسنم رک راھک وہا ےہ اور دواہل ارگ اچےہ یھب وت ان ےس اجن ںیہن ڑھچا اتکس۔ ینگنم، دنہمی، ابرات، السیم اور زیہج وریغہ یسیج رومسں وک فلتخم العوقں ںیم ڑبی اتیمہ دی اجیت ےہ اور ان یک دایگیئ ںیم ریطخ رسامہی رصف ایک اجات ےہ ان رومسں اک بس ےس حیبق ولہپ ہی ےہ ہک ہی بس ٖریغ االسیم رںیمس ںیہ۔ ان اک االسیم ذہتبی ےس وکیئ واہطس ںیہن ےہ۔ 56 1 ۔ انب دقاہم، دبعاہلل نب ادمح، اینغمل النب ، دقاہم، (ۃبتکم ااقلرہہ۱۳۸۸ھ) ج۹، ص۳۰ 56 56 ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 57 اشدی اکی اخدناین اور امسیج رضورت اور اسھت یہ اکی ابعدت یھب۔ االسم ےن اشدی یک رتتیغ دی ےہ اور اس ومعق رپ وخیش انمےن یک ااجزت یھب دی ےہ۔ رگم اشدی ےک وج وطر رطےقی امہ رے اعمرشے ںیم راجئ ںیہ اور نج روسامت یک یج اجن ےس اپدساری یک اجیت ےہ ان ںیم ارثک امیلعتت االسم ےس اصتمدم ںیہ۔ االسم روتشں ےک ااختنابت ںیم دنی داری وک ایعمر انبےن، عمجم اعم ںیم اکنح رکےن اور اکن ح ےک وتق االسیم ااکحم یک نیقلت رکےن ، رہم ادا رکےن اور وہمیل رک ےن اکمکح داتی ےہ۔ اوران بس ںیم اسدیگ اورافکتی اعشری یک اخص وطر رپ اتدیک :یک ےہ ’’ اذن یشءعظَْ الناكحِ یركةا یشَي ْسَ ُهُ مُؤ َْةا‘‘(1) ‘‘’’ بس ےس ابربتک اکنح وہ ےہ وج بس ےس مک رخچ ںیم وہ اس ایعمر رپ روسلِ ارکمﷺ ےک دعتمد اکنح وہےئ ۔ ٓاپ یک اچر ویٹیبں یک اشدی وہ یئ۔ امتم احصہب ایس اُوسہ رپ اقمئ رےہ۔ اس زامےن یک یسک یھب اشدی ںیم دوھم داھم، وضفل رخیچ، ےب اج امنشئ یک اکی ریظن یھب ںیہن یتلم۔ اس اوسہ یک دیلقت رپ ٓاج یھب رشاف ےک اہی ں اسدیگ اک ایخل راھک اجات ےہ۔ اور ان اُومر ںیم نج رواایت یک اپس داری اک االسم ےن اخص وطر ےس مکح دای ےہ وہ مظن ادناز وہ اجیت ںیہ۔ ًالثم روتشں ےک ااختنب ںیم دنی داری وک ایعمر انبان، ااجیب و وبقل ےک وتق اینبدی ااکحم یک اید داہین ینعی ہبطخ اکنح اک اامتہم اور انمبس دقمار ںیم رہک یک نییعت اور وفری ادایگیئ یک دتاریب۔ وموجدہ زامےن ںیم ان ابوتں وک ومعم یل ھجمس رک رظن ادناز رکدای اجات ےہ بج ہک روتشں ںیم ااکحتسم یک یہی الص اینبد ںیہ۔ ان وموجد رومسں ےس رغبی ولگ یسلفم اک اکشر وہےت ںیہ، رومسں ےک اہوھتں اینپ انک رےنھک ےک ےیل رقض کت ےل ایل اجات ےہ اور رھپ اسری زدنیگ رقہض وتق رپ واسپ ہن رکےن یک وصرت ںیم یئک اب ر یتٹک ےہ۔ دعم ربداش اخیگن زدنیگ ںیم اسفد و رختبی اک اکی ڑبا ابثع دعم ربداےش اک اس دح کت ڑبھ اجان ےہ ہک ونتب ڑلایئ ڑگھجے، لتق و اغرت رگی اور ایمں ویبی ےک درایمن لطق اور دجایئ کت اج یتچنہپ ےہ۔ احالہکن وفع ودرزگر، لمحت و ربداش اور رنم زمایج ہن رصف ہی ہک رقٓان میکح اک مکح ےہ ہکلب ریست ہبیط وبنہی اک اغبل ولہپ یھب ےہ۔ ارگ اخدناین مظن ںیم لمحت ربداش یک رتبیت ہن دی اجےئ اور اس اک لمع اظمرہہ ہن ایک اجےئ وت یہی ارفاد بج اعمرشے ںیم اجےت ںیہ وت واہں یھب اس دعم ربداش یک وہج ےس ےب وکسین اور دعم اانیمطین اک اموحل دیپا وہات ےہ، سج اک ہجیتن ہی اتلکن ےہ ہک اخیگن اظنم ےلہپ ابتہ وہ اکچ وہات 1 ۔ ربتزیی، ٰوکشمۃ ااصملحیب، اتکب ااکنلح، الصفل ااثلثل، رمق ادحلثی۳۰۹۷ ĪQĀN (9) Vol. اخدنان یک اینبد الےاکوا 05, No. 01, Dec-2022 ےہ اور اب ریبون اخہن یھب یہی اموحل نب رک ارفاد اعمرشہ ےک ےیل انیج لکشم انب داتی ےہ سج ےس رصف اکی اخدنان یہ ںیہن ہن رصف اکی رف د ہکلب یئک اخدنان اور ان ےس کلسنم ےنتک ارفاد ےب وکسین یک زدنیگ زگارےن رپ وبجمر وہ اجےت ںیہ ای رھپ اےنپ یہ اہوھتں اےنپ ایپروں وک ومت ےک اھگٹ اُاتر رک ہشیمہ ےک ےیل ذعاب ےک قحتسم رہھٹےت ںیہ۔ ربداش ےس اکی صخش رقابین دے رک یئک ارفاد اور یئک رھگاونں وک ربابدی ےس اچب اتکس ےہ۔ الخۃص احبل االسم اک اخدناین مظن اکی اہلل اعتٰیل یک تمعن ےہ سج اک اایح رضوری ےہ نج ولوگں اور وقومں ےن اس یک دقر یک وہ وقںیم اور اخدنان رتیق ایہتف وہےئ اور وہنجں ےن اس وک رخاب ایک وہ وپرے اعمرشے یک رخایب اک ابثع وہےئ، روتشں اک دقتس اور احلظ سج اکاالسم ےن مکح دای ےہ ہی اس مظن یک اینبد ےہ۔ ٓاج رغمب ںیم ان دقمس روتشں یک دقر ںیہن ےہ سج ےک ےجیتن ںیم اوڈل وہزم لھک ےئگ۔ وادلنی وک دختم رکےن وایل اور اوالد دایتسب ںیہن اور اوالد ےک اپس اےنپ وادلنی، ادجاد ےک ےیل وتق ںیہن۔ االسم ان اخدناین ادقار یک اچہپن رکواات ےہ ان ےک وقحق ایبن رکات ےہ، ان یک زعت و رکتمی وک ارج اک ابثع رقار داتی ےہ۔ رقٓان اور دحثی اور یبنﷺ یک زدنیگ رھپ ااہمت اور الہِ تیب یک زدنیگ اور احصہب و احصایبت یک ایحت ہبیط اور رقونِ اوٰیل اس ےک ےیل انیمرۂ دہا تی ےہ۔ یئن رںیمس اور دجدی رواج مظن اخدنان ےک الخف اکی راکوٹ ںیہ نج ےک وہےت وہےئ ہن رصف اخدناین اظنم اپرہ اپرہ وہ راہ ےہ ہکلب وپرے اعمرشے وک ال العج رمض ااحہط رک راہ ےہ سج اکالعج رصف اور رصف االسیم امیلعتت ںیہ۔ اتنجئ اے اکووصالموصت نظمےدا ئیانیا کیرعاےلا ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 ےہ اور اب ریبون اخہن یھب یہی اموحل نب رک ارفاد اعمرشہ ےک ےیل انیج لکشم انب داتی ےہ سج ےس رصف اکی اخدنان یہ ںیہن ہن رصف اکی رف د ہکلب یئک اخدنان اور ان ےس کلسنم ےنتک ارفاد ےب وکسین یک زدنیگ زگارےن رپ وبجمر وہ اجےت ںیہ ای رھپ اےنپ یہ اہوھتں اےنپ ایپروں وک ومت ےک اھگٹ اُاتر رک ہشیمہ ےک ےیل ذعاب ےک قحتسم رہھٹےت ںیہ۔ ربداش ےس اکی صخش رقابین دے رک یئک ارفاد اور یئک رھگاونں وک ربابدی ےس اچب اتکس ےہ۔ الخۃص احبل االسم اک اخدناین مظن اکی اہلل اعتٰیل یک تمعن ےہ سج اک اایح رضوری ےہ نج ولوگں اور وقومں ےن اس یک دقر یک وہ وقںیم اور اخدنان رتیق ایہتف وہےئ اور وہنجں ےن اس وک رخاب ایک وہ وپرے اعمرشے یک رخایب اک ابثع وہےئ، روتشں اک دقتس اور احلظ سج اکاالسم ےن مکح دای ےہ ہی اس مظن یک اینبد ےہ۔ ٓاج رغمب ںیم ان دقمس روتشں یک دقر ںیہن ےہ سج ےک ےجیتن ںیم اوڈل وہزم لھک ےئگ۔ وادلنی وک دختم رکےن وایل اور اوالد دایتسب ںیہن اور اوالد ےک اپس اےنپ وادلنی، ادجاد ےک ےیل وتق ںیہن۔ االسم ان اخدناین ادقار یک اچہپن رکواات ےہ ان ےک وقحق ایبن رکات ےہ، ان یک زعت و رکتمی وک ارج اک ابثع رقار داتی ےہ۔ رقٓان اور دحثی اور یبنﷺ یک زدنیگ رھپ ااہمت اور الہِ تیب یک زدنیگ اور احصہب و احصایبت یک ایحت ہبیط اور رقونِ اوٰیل اس ےک ےیل انیمرۂ دہا تی ےہ۔ یئن رںیمس اور دجدی رواج مظن اخدنان ےک الخف اکی راکوٹ ںیہ نج ےک وہےت وہےئ ہن رصف اخدناین اظنم اپرہ اپرہ وہ راہ ےہ ہکلب وپرے اعمرشے وک ال العج رمض ااحہط رک راہ ےہ سج اکالعج رصف اور رصف االسیم امیلعتت ںیہ۔ اتنجئ • االسم ےن اعمرشے یک اینبدی ااکیئ ’’اخدنان‘‘ ےک مظن اک اکی وخوصبرت اور اکلم وصتر دای ےہ۔ • یبن رکمیﷺ اک اینپ ازواج رہطمات اور درگی ااقرب ےک اسھت قلعت اور روہی مظن اخدنان یک اکی رتہبنی اثمل ےہ سج یک ریپوی ےس رصع احرض ےک اعمرشے اینپ امتم زمکورویں وک دور رک ےک عیمج وخایبں اےنپ ادنر عمج رک ےتکس ںیہ۔ • رھگ ےک ارفاد، زونیج اور اوالد ےک وقحق و رفاضئ اک وج وصترِ االسم ےن دای ےہ ارگ ان وک لمع ےس ٓاراہتس رک ایل اجےئ وت داین اک اعمرشہ تنج اک ہشقن شیپ رک اتکس ےہ۔ ĪQĀN (9) Vol. اخدنان یک اینبد الےاکوا 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 اشدی اکی اخدناین اور امسیج رضورت اور اسھت یہ اکی ابعدت یھب۔ االسم ےن اشدی یک رتتیغ دی ےہ اور اس ومعق رپ وخیش انمےن یک ااجزت یھب دی ےہ۔ رگم اشدی ےک وج وطر رطےقی امہ رے اعمرشے ںیم راجئ ںیہ اور نج روسامت یک یج اجن ےس اپدساری یک اجیت ےہ ان ںیم ارثک امیلعتت االسم ےس اصتمدم ںیہ۔ االسم روتشں ےک ااختنابت ںیم دنی داری وک ایعمر انبےن، عمجم اعم ںیم اکنح رکےن اور اکن ح ےک وتق االسیم ااکحم یک نیقلت رکےن ، رہم ادا رکےن اور وہمیل رک ےن اکمکح داتی ےہ۔ اوران بس ںیم اسدیگ اورافکتی اعشری یک اخص وطر رپ اتدیک :یک ےہ ‘‘’’ بس ےس ابربتک اکنح وہ ےہ وج بس ےس مک رخچ ںیم وہ اس ایعمر رپ روسلِ ارکمﷺ ےک دعتمد اکنح وہےئ ۔ ٓاپ یک اچر ویٹیبں یک اشدی وہ یئ۔ امتم احصہب ایس اُوسہ رپ اقمئ رےہ۔ اس زامےن یک یسک یھب اشدی ںیم دوھم داھم، وضفل رخیچ، ےب اج امنشئ یک اکی ریظن یھب ںیہن یتلم۔ اس اوسہ یک دیلقت رپ ٓاج یھب رشاف ےک اہی ں اسدیگ اک ایخل راھک اجات ےہ۔ اور ان اُومر ںیم نج رواایت یک اپس داری اک االسم ےن اخص وطر ےس مکح دای ےہ وہ مظن ادناز وہ اجیت ںیہ۔ ًالثم روتشں ےک ااختنب ںیم دنی داری وک ایعمر انبان، ااجیب و وبقل ےک وتق اینبدی ااکحم یک اید داہین ینعی ہبطخ اکنح اک اامتہم اور انمبس دقمار ںیم رہک یک نییعت اور وفری ادایگیئ یک دتاریب۔ وموجدہ زامےن ںیم ان ابوتں وک ومعم یل ھجمس رک رظن ادناز رکدای اجات ےہ بج ہک روتشں ںیم ااکحتسم یک یہی الص اینبد ںیہ۔ ان وموجد رومسں ےس رغبی ولگ یسلفم اک اکشر وہےت ںیہ، رومسں ےک اہوھتں اینپ انک رےنھک ےک ےیل رقض کت ےل ایل اجات ےہ اور رھپ اسری زدنیگ رقہض وتق رپ واسپ ہن رکےن یک وصرت ںیم یئک اب ر یتٹک ےہ۔ در 57 ث ĪQĀN (9) Vol. 05, No. اخدنان یک اینبد الےاکوا 01, Dec-2022 ےہ اور اب ریبون اخہن یھب یہی اموحل نب رک ارفاد اعمرشہ ےک ےیل انیج لکشم انب داتی ےہ سج ےس رصف اکی اخدنان یہ ںیہن ہن رصف اکی رف د ہکلب یئک اخدنان اور ان ےس کلسنم ےنتک ارفاد ےب وکسین یک زدنیگ زگارےن رپ وبجمر وہ اجےت ںیہ ای رھپ اےنپ یہ اہوھتں اےنپ ایپروں وک ومت ےک اھگٹ اُاتر رک ہشیمہ ےک ےیل ذعاب ےک قحتسم رہھٹےت ںیہ۔ ربداش ےس اکی صخش رقابین دے رک یئک ارفاد اور یئک رھگاونں وک ربابدی ےس اچب ے سکتا • االسم ےن اعمرشے یک اینبدی ااکیئ ’’اخدنان‘‘ ےک مظن اک اکی وخوصبرت اور اکلم وصتر دای ےہ۔ • یبن رکمیﷺ اک اینپ ازواج رہطمات اور درگی ااقرب ےک اسھت قلعت اور روہی مظن اخدنان یک اکی رتہبنی اثمل ےہ سج یک ریپوی ےس رصع احرض ےک اعمرشے اینپ امتم زمکورویں وک دور رک ےک عیمج وخایبں اےنپ ادنر عمج رک ےتکس ںیہ۔ • رھگ ےک ارفاد، زونیج اور اوالد ےک وقحق و رفاضئ اک وج وصترِ االسم ےن دای ےہ ارگ ان وک لمع ےس ٓاراہتس رک ایل اجےئ وت داین اک اعمرشہ تنج اک ہشقن شیپ رک اتکس ےہ۔ • رھگ ےک امتم ارفاد ںیم ااسحسِ ذہم داری، اےنپ رفاضئ ےس واتیفق اور لمحت ربص و رکش اور درگی االخیق افصت اک اپای اجان رضوری ےہ۔ 58 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 • اخدنان یک اینبد وچہکن دقع زونیج ےہ اس ےیل اس دقع ےک رفنیقی ںیم انم تبس، وتازن اور وصخًاص رکف مہ ٓایگنہ وک وپری رہگایئ ےک اسھت اجانچن رضوری ےہ ورہن اخدنان ےک لحم یک تشخ اوّل یہ ارگ ڑیٹیھ وہیئ وت اترث ای دویار یجک اک اکشر وہ یگ۔ • اعمرشہ فلتخم اخدناونں ےس رتبیت اپاتےہ اس ےک اعمرشے یک االصح ےک ےیل ان اینبدوں اک اصحل وہان رضوری ےہ اور ان ںیم الصتیح یھبت ٓاےئ یگ بج رقٓان میکح اور ریست ہبیط ںیم دےی ےئگ مظن اخدنان ےک اوصولں ےس روینش احلص یک اجےئ یگ۔ افسراشت :اخدناین مظن وک ربرقار رےنھک ےک ےیل دنچ زیچںی رضوری ںیہ 1. روتشں ےک دقتس وک رھپ ےس احبل ایک اجےئ۔ 2. اامتعد یک اضف ےس رھگولی زدنیگ وک ٓاراہتس ایک اجےئ۔ 3. اینپ زدنویگں وک اسدہ اور رشتعی ےک اطمقب انبای اجےئ۔ 4. اےسی رواوجں وک علق عمق ایک اجےئ وج رشتعی ےس اصتمدم وہں۔ 5. وسلش ڈیمای اور درگی ذراعئ ےس احبیل دقتس رہتش یک اباقدعہ رحتکی الچیئ اجےئ۔ 6. 1. روتشں ےک دقتس وک رھپ ےس احبل ایک اجےئ۔ ٓہتایاےامتیاضےرھولدیو 2. اامتعد یک اضف ےس رھگولی زدنیگ وک ٓاراہتس ایک اجےئ۔ یندویوارتعےاطقانااے اخدنان یک اینبد الےاکوا وادلنی، املعء اور ااسذتہ وج رتتیب ےک ذہم دار ںیہ، وہ ا س ےلسلس ںیم اانپ رھبوپر رکدار ادا رکںی۔ 7. مظن اخدنان یک مظن اعمرشہ ںیم اتیمہ وک امتم یمیلعت و رتیتیب اداروں ںیم اباقدعہ اور رھبوپر رطےقی ےس اعتمرف رکواای اجےئ۔ ĪQĀN (9) Vol. 05, No. 01, Dec-2022 ĪQĀN (9) Vol. 05, No. 01, Dec-2022 • اخدنان یک اینبد وچہکن دقع زونیج ےہ اس ےیل اس دقع ےک رفنیقی ںیم انم تبس، وتازن اور وصخًاص رکف مہ ٓایگنہ وک وپری رہگایئ ےک اسھت اجانچن رضوری ےہ ورہن اخدنان ےک لحم یک تشخ اوّل یہ ارگ ڑیٹیھ وہیئ وت اترث ای دویار یجک اک اکشر وہ یگ۔ • اعمرشہ فلتخم اخدناونں ےس رتبیت اپاتےہ اس ےک اعمرشے یک االصح ےک ےیل ان اینبدوں اک اصحل وہان رضوری ےہ اور ان ںیم الصتیح یھبت ٓاےئ یگ بج رقٓان میکح اور ریست ہبیط ںیم دےی ےئگ مظن اخدنان ےک اوصولں ےس روینش احلص یک اجےئ یگ۔ افس اش :اخدناین مظن وک ربرقار رےنھک ےک ےیل دنچ زیچںی رضوری ںیہ وتےدقورھےاحایاے 1. روتشں ےک دقتس وک رھپ ےس احبل ایک اجےئ۔ ٓہتایاےامتیاضےرھولدیو 2. اامتعد یک اضف ےس رھگولی زدنیگ وک ٓاراہتس ایک اجےئ۔ یندویوارتعےاطقانااے 3. اینپ زدنویگں وک اسدہ اور رشتعی ےک اطمقب انبای اجےئ۔ ےسووعلعمایاےورتعےاصتو ےااناقطاےعتراویودنی 4. اےسی رواوجں وک علق عمق ایک اجےئ وج رشتعی ےس اصتمدم وہں۔ 5. وسلش ڈیمای اور درگی ذراعئ ےس احبیل دقتس رہتش یک اباقدعہ رحتکی الچیئ اجےئ۔ 6. وادلنی، املعء اور ااسذتہ وج رتتیب ےک ذہم دار ںیہ، وہ ا س ےلسلس ںیم اانپ رھبوپر رکدار ادا رکںی۔ 7. مظن اخدنان یک مظن اعمرشہ ںیم اتیمہ وک امتم یمیلعت و رتیتیب اداروں ںیم اباقدعہ اور رھبوپر رطےقی ےس اعتمرف رکواای اجےئ۔ 59

https://openalex.org/W1970747000

https://journals.iucr.org/e/issues/2008/07/00/hb2744/hb2744.pdf

English

Diazidobis(2,2′-biimidazole)manganese(II)

Acta crystallographica. Section E

cc-by

Table 1 Table 1 Details of articles to be retracted, in order of publication. Details of articles to be retracted, in order of publication. Title Reference DOI Refcode [N,N0-Bis(2-hydroxynaphthylmethylene)-1,2-ethanediaminato]zinc(II) Chen et al. (2005) 10.1107/S1600536805026796 YAWZOM Diazidobis(2,20-biimidazole)copper(II) Liu et al. (2007) 10.1107/S1600536807047873 SILZIX Dichlorido(1,10-phenanthroline)copper(II) Liu (2007) 10.1107/S1600536807056735 MISSAJ Diazidobis(2,20-biimidazole)cobalt(II) Li et al. (2008) 10.1107/S1600536807062873 MIRYAO Diazidobis(2,20-biimidazole)manganese(II) Zhang et al. (2008) 10.1107/S1600536808017984 MODBUD Diazidobis(2,20-biimidazole)iron(II) Hao et al. (2008a) 10.1107/S1600536808018539 MODFOB Bis(pentane-2,4-dionato)bis[2-(4-pyridyl)-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide]nickel(II) Hao et al. (2008b) 10.1107/S1600536808018552 MODFUH Bis(pentane-2,4-dionato-2O,O0)bis[4,4,5,5-tetramethyl-2-(4-pyridyl)imidazoline- 1-oxyl 3-oxide-N2]manganese(II) Liu, Zhang et al. (2008) 10.1107/S1600536808022952 MODLUN Bis[2,4-pentanedionato(1)]bis[4,4,5,5-tetramethyl-2-(4-pyridyl)imidazoline- 1-oxyl 3-oxide]manganese(II) Liu, He et al. (2008) 10.1107/S1600536808038440 MODLUN01 Di--chlorido-bis[chlorido(1,10-phenanthroline-2N,N0)zinc(II)] Yang et al. (2009) 10.1107/S1600536809014482 JOLBOC Tris(ethylenediamine)manganese(II) sulfate Lu (2009) 10.1107/S1600536809034874 YUCZEC IUCr Editorial Office 5 Abbey Square, Chester CH1 2HU, England This article reports the retraction of 11 articles published in Acta Crystallographica Section E between 2005 and 2009. After further thorough investigation (see Harrison et al., 2010), 11 additional articles are retracted by the authors or by the journal as a result of problems with the data sets or incorrect atom assignments. Full details of all the articles are given in Table 1. Table 1 Details of articles to be retracted, in order of publication. Acta Crystallographica Section E Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 References Chen, G., Zhao, B., Sun, M. & Qi, W. (2005). Acta Cryst. E61, m1869–m1870. Hao, L., Mu, C. & Kong, B. (2008a). Acta Cryst. E64, m956. Hao, L., Mu, C. & Kong, B. (2008b). Acta Cryst. E64, m957. Harrison, W. T. A., Simpson, J. & Weil, M. (2010). Acta Cryst. E66, e1–e2. Li, S., Wang, S.-B., Zhang, F.-L. & Tang, K. (2008). Acta Cryst. E64, m76. Liu, Y.-Q. (2007). Acta Cryst. E63, m2991. Liu, Y., Dou, J., Li, D. & Zhang, X. (2007). Acta Cryst. E63, m2661. Liu, Y., He, Q., Zhang, X., Xue, Z. & Lv, C. (2008). Acta Cryst. E64, m1604. Liu, Y., Zhang, X., Xue, Z., He, Q. & Zhang, Y. (2008). Acta Cryst. E64, m1077. Lu, J. (2009). Acta Cryst. E65, m1187. Yang, X.-M., Leng, Q.-B., Chen, Y., He, Y.-G. & Luo, S.-W. (2009). Acta Cryst. E65, m567. Zhang X Wei P & Li B (2008) Acta Cryst E64 m934 Yang, X.-M., Leng, Q.-B., Chen, Y., He, Y.-G. & Luo, S.-W. (2009). Acta Cryst. E65, m567. Zhang, X., Wei, P. & Li, B. (2008). Acta Cryst. E64, m934. e14 Acta Cryst. (2011). E67, e14 # 2011 International Union of Crystallography e14 # 2011 International Union of Crystallography doi:10.1107/S1600536810053882 metal-organic compounds Xiutang Zhang, Peihai Wei* and Bin Li Department of Chemistry and Chemical Engineering, ShanDong Institute of Education, Jinan 250013, People’s Republic of China Correspondence e-mail: [email protected] Received 9 June 2008; accepted 12 June 2008 Key indicators: single-crystal X-ray study; T = 293 K; mean
(C–C) = 0.005 A˚; R factor = 0.039; wR factor = 0.131; data-to-parameter ratio = 11.5. In the title compound, [Mn(N3)2(C6H6N4)2], the Mn atom (site symmetry 1) is bonded to two azide ions and two bidentate biimidizole ligands, resulting in a slightly distorted octahedral MnN6 geometry for the metal ion. In the crystal structure, N— H


N hydrogen bonds help to consolidate the packing. metal-organic compounds retracted e e l — Monoclinic, C2=c a = 12.5097 (10) A˚ b = 8.9728 (5) A˚ c = 14.1416 (10) A˚  = 91.883 (10) V = 1586.50 (19) A˚ 3 Z = 4 Mo K radiation  = 0.87 mm1 T = 293 (2) K 0.40  0.26  0.20 mm Data collection Bruker APEXII CCD diffractometer Absorption correction: multi-scan (SADABS; Bruker, 2004) Tmin = 0.723, Tmax = 0.846 1966 measured reflections 1505 independent reflections 1250 reflections with I > 2
(I) Rint = 0.022 Refinement R[F 2 > 2
(F 2)] = 0.038 wR(F 2) = 0.131 S = 1.00 1505 reflections 131 parameters 2 restraints H atoms treated by a mixture of independent and constrained refinement
max = 0.49 e A˚ 3
min = 0.25 e A˚ 3 Table 1 Selected bond lengths (A˚ ). Mn1—N2 2.094 (3) Mn1—N3 2.114 (3) Mn1—N5 2.138 (3) Table 2 Hydrogen-bond geometry (A˚ , ). D—H


A D—H H


A D


A D—H


A N1—H1A


N7i 0.966 (18) 2.26 (3) 3.031 (4) 136 (3) N1—H1A


N5ii 0.966 (18) 2.33 (4) 3.021 (4) 127 (3) N4—H4


N7i 0.952 (19) 1.92 (2) 2.834 (4) 160 (4) Symmetry codes: (i) x þ 3 2; y þ 1 2; z þ 1; (ii) x  1 2; y  1 2; z. Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Related literature For a related structure, see: Hester et al. (1997). Article r Experimental Crystal data [Mn(N3)2(C6H6N4)2] Mr = 407.30 Data collection: APEX2 (Bruker, 2004); cell refinement: SAINT- Plus (Bruker, 2004); data reduction: SAINT-Plus; program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: SHELXTL. The authors thank the National Ministry of Science and Technology of China (grant No. 2001CB6105-07). Supplementary data and figures for this paper are available from the IUCr electronic archives (Reference: HB2744). References Bruker (2004). APEX2, SADABS and SAINT-Plus. Bruker AXS Inc., Madison, Wisconsin, USA. Hester, C. A., Baughman, R. G. & Collier, H. L. (1997). Polyhedron, 16, 2893– 2895. Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122. S1. Comment Article retracted S1. Comment The study of coordination compounds including one-, two- and three-dimensional infinite frameworks has been expanding rapidly because of their fascinating structural diversity and potential application as functional materials. To date, much of the work has been focused on coordination polymers with semi-rigid ligands, such as 4,4′-bipyridine, pyrazine and their analogues. In this paper, we report the structure of the molecular title compound, (I), with the use of the 2,2′-biimidazole bridging ligand (Hester et al., 1997). As shown in Fig. 1, the Mn ion in (I) occupies an inversion centre, and is hexacoordinated by six N atoms from two chelating ligands of H2bim (biimidizole; C6H6N4) and two azide ions, showing a slightly distorted MnN6 octahedral geometry (Table 1). In the crystal of (I), N—H···N hydrogen bonds, one of which is bifurcated (Table 2), help to consolidate the packing. S2. Experimental A mixture of manganese(II) perchlorate hexahydrate (1 mmol), 2,2′-biimidazoline (2 mmol) and Na3N3 (2 mmol) in 20 ml ethanol was reflued for several hours. The cooled solution was filtered and the filtrate was kept in an ice box for about one week. Yellow blocks of (I) were obtained with a yield of 10%. Anal. Calc. for C12H12MnN14: C 35.35, H 2.95, N 48.12%; Found: C 35.31, H 2.92, N 48.06%. S3. Refinement The N-bound H atoms were located in a difference map and their positions were freely refined with Uiso(H) = 1.2Ueq(N). The C-bound H atoms were placed in calculated positions (C—H = 0.93 Å) and refined as riding with Uiso(H) = 1.2Ueq(C). cted The study of coordination compounds including one-, two- and three-dimensional infinite frameworks has been expanding rapidly because of their fascinating structural diversity and potential application as functional materials. To date, much of the work has been focused on coordination polymers with semi-rigid ligands, such as 4,4′-bipyridine, pyrazine and their analogues. In this paper, we report the structure of the molecular title compound, (I), with the use of the 2,2′-biimidazole bridging ligand (Hester et al., 1997). tract As shown in Fig. 1, the Mn ion in (I) occupies an inversion centre, and is hexacoordinated by six N atoms from two chelating ligands of H2bim (biimidizole; C6H6N4) and two azide ions, showing a slightly distorted MnN6 octahedral geometry (Table 1). tract As shown in Fig. Acta Cryst. (2008). E64, m934 [doi:10.1107/S1600536808017984] Acta Cryst. (2008). E64, m934 [doi:10.1107/S1600536808017984] Acta Cryst. (2008). E64, m934 [doi:10.1107/S1600536808017984] References Experimental Crystal data Mr = 407.30 Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122. [Mn(N3)2(C6H6N4)2] m934 Zhang et al. Acta Cryst. (2008). E64, m934 doi:10.1107/S1600536808017984 supporting information S2. Experimental cle re A mixture of manganese(II) perchlorate hexahydrate (1 mmol), 2,2′-biimidazoline (2 mmol) and Na3N3 (2 mmol) in 20 ml ethanol was reflued for several hours. The cooled solution was filtered and the filtrate was kept in an ice box for about one week. Yellow blocks of (I) were obtained with a yield of 10%. Anal. Calc. for C12H12MnN14: C 35.35, H 2.95, N 48.12%; Found: C 35.31, H 2.92, N 48.06%. cle re A mixture of manganese(II) perchlorate hexahydrate (1 mmol), 2,2′-biimidazoline (2 mmol) and Na3N3 (2 mmol) in 20 ml ethanol was reflued for several hours. The cooled solution was filtered and the filtrate was kept in an ice box for about one week. Yellow blocks of (I) were obtained with a yield of 10%. Anal. Calc. for C12H12MnN14: C 35.35, H 2.95, N 48.12%; Found: C 35.31, H 2.92, N 48.06%. S1. Comment 1, the Mn ion in (I) occupies an inversion centre, and is hexacoordinated by six N atoms from two chelating ligands of H2bim (biimidizole; C6H6N4) and two azide ions, showing a slightly distorted MnN6 octahedral geometry (Table 1). etra In the crystal of (I), N—H···N hydrogen bonds, one of which is bifurcated (Table 2), help to consolidate the packing. S2 Experimental retra In the crystal of (I), N—H···N hydrogen bonds, one of which is bifurcated (Table 2), help to consolidate the packing. S2. Experimental etra In the crystal of (I), N—H···N hydrogen bonds, one of which is bifurcated (Table 2), hel Acta Cryst. (2008). E64, m934 Figure 1 Article retract sup-2 Acta Cryst. (2008). E64, m934 Figure 1 The molecular structure of (I), drawn with 30% probability displacement ellipsoids for the non-hydrogen atoms. Symmetry code: (i) 3/2-x, 3/2-y, 1-z. Diazidobis(2,2′-biimidazole)manganese(II) Crystal data [Mn(N3)2(C6H6N4)2] Mr = 407.30 Monoclinic, C2/c Hall symbol: -C 2yc a = 12.5097 (10) Å b = 8.9728 (5) Å c = 14.1416 (10) Å β = 91.883 (10)° V = 1586.50 (19) Å3 Z = 4 F(000) = 828 Dx = 1.705 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 1505 reflections θ = 2.8–25.9° µ = 0.87 mm−1 T = 293 K Block, yellow 0.40 × 0.26 × 0.20 mm Data collection Bruker APEXII CCD diffractometer Radiation source: fine-focus sealed tube Graphite monochromator φ and ω scans Absorption correction: multi-scan (SADABS; Bruker, 2004) Tmin = 0.723, Tmax = 0.846 1966 measured reflections 1505 independent reflections 1250 reflections with I > 2σ(I) Rint = 0.022 θmax = 25.9°, θmin = 2.8° h = −1→15 k = −1→10 l = −17→17 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.038 wR(F2) = 0.131 S = 1.00 1505 reflections 131 parameters 2 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map tract The molecular structure of (I), drawn with 30% probability displacement ellipsoids for the non-hydrogen atoms. Symmetry code: (i) 3/2-x, 3/2-y, 1-z. Diazidobis(2,2′-biimidazole)manganese(II) S3. Refinement Arti The N-bound H atoms were located in The C-bound H atoms were placed in 1.2Ueq(C). Artic The N-bound H atoms were located in a difference map and their positions were freely refined with Uiso(H) = 1.2Ueq(N). The C-bound H atoms were placed in calculated positions (C—H = 0.93 Å) and refined as riding with Uiso(H) = 1.2Ueq(C). Arti toms were placed in sup-1 Acta Cryst. (2008). E64, m934 supporting information Article retracted s A t C t (2008) E64 934 Figure 1 The molecular structure of (I), drawn with 30% probability displacement ellipsoids for the non-hydrogen atoms. Symmetry code: (i) 3/2-x, 3/2-y, 1-z. Diazidobis(2,2′-biimidazole)manganese(II) Crystal data [Mn(N3)2(C6H6N4)2] Mr = 407.30 Monoclinic, C2/c Hall symbol: -C 2yc a = 12.5097 (10) Å b = 8.9728 (5) Å c = 14.1416 (10) Å β = 91.883 (10)° V = 1586.50 (19) Å3 Z = 4 F(000) = 828 Dx = 1.705 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 1505 reflections θ = 2.8–25.9° µ = 0.87 mm−1 T = 293 K Block, yellow 0.40 × 0.26 × 0.20 mm Data collection Bruker APEXII CCD diffractometer Radiation source: fine-focus sealed tube Graphite monochromator φ and ω scans Absorption correction: multi-scan (SADABS; Bruker, 2004) Tmin = 0.723, Tmax = 0.846 1966 measured reflections 1505 independent reflections 1250 reflections with I > 2σ(I) Rint = 0.022 θmax = 25.9°, θmin = 2.8° h = −1→15 k = −1→10 l = −17→17 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.038 wR(F2) = 0.131 S = 1.00 1505 reflections 131 parameters 2 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map acted Figure 1 Special details d p Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. Refinement. Refinement of F2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F2, conventional R-factors R are based on F, with F set to zero for negative F2. The threshold expression of F2 > σ(F2) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F2 are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger. Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. d Refinement. Refinement of F2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F2, conventional R-factors R are based on F, with F set to zero for negative F2. The threshold expression of F2 > σ(F2) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F2 are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger. Diazidobis(2,2′-biimidazole)manganese(II) Article r Crystal data [Mn(N3)2(C6H6N4)2] Mr = 407.30 Monoclinic, C2/c Hall symbol: -C 2yc a = 12.5097 (10) Å b = 8.9728 (5) Å c = 14.1416 (10) Å β = 91.883 (10)° V = 1586.50 (19) Å3 Z = 4 Data collection Bruker APEXII CCD diffractometer Radiation source: fine-focus sealed tube Article r N4)2] /c 2yc ) Å Å ) Å ° 9) Å3 CCD r e: fine focus sealed tube 1966 measured reflections 1505 independent reflections 1250 reflections with I > 2σ(I) Rint = 0.022 θmax = 25.9°, θmin = 2.8° h = −1→15 k = −1→10 l = −17→17 sup-2 Acta Cryst. (2008). E64, m934 supporting information supporting information w = 1/[σ2(Fo2) + (0.081P)2 + 1.7249P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.024 Δρmax = 0.49 e Å−3 Δρmin = −0.25 e Å−3 Hydrogen site location: inferred from neighbouring sites H atoms treated by a mixture of independent and constrained refinement Acta Cryst. (2008). E64, m934 Special details Article retracted Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso*/Ueq Mn1 0.7500 0.7500 0.5000 0.0486 (5) C1 0.5842 (3) 0.5617 (4) 0.3628 (2) 0.0572 (8) H1 0.5711 0.6280 0.3132 0.069* C2 0.5366 (3) 0.4245 (4) 0.3721 (2) 0.0585 (9) H2 0.4865 0.3819 0.3303 0.070* C3 0.6462 (2) 0.4631 (4) 0.4891 (2) 0.0482 (7) C4 0.7106 (2) 0.4562 (4) 0.5744 (2) 0.0489 (7) C5 0.8223 (3) 0.5322 (4) 0.6783 (2) 0.0580 (8) H5 0.8713 0.5915 0.7118 0.070* C6 0.7940 (3) 0.3903 (4) 0.7030 (2) 0.0610 (9) H6 0.8195 0.3369 0.7554 0.073* N1 0.5763 (2) 0.3633 (3) 0.45300 (18) 0.0534 (7) H1A 0.546 (3) 0.270 (3) 0.474 (3) 0.064* N2 0.6529 (2) 0.5851 (3) 0.43706 (17) 0.0520 (7) N3 0.7688 (2) 0.5733 (3) 0.59802 (18) 0.0525 (7) N4 0.7220 (2) 0.3432 (3) 0.63640 (19) 0.0551 (7) H4 0.680 (3) 0.255 (3) 0.634 (3) 0.066* N5 0.8833 (2) 0.6634 (3) 0.4268 (2) 0.0529 (7) N6 0.8963 (2) 0.5320 (3) 0.4212 (2) 0.0552 (7) N7 0.9109 (3) 0.4018 (3) 0.4143 (2) 0.0723 (9) Atomic displacement parameters (Å2) U11 U22 U33 U12 U13 U23 Mn1 0.0504 (12) 0.0400 (13) 0.0550 (13) 0.0096 (10) −0.0016 (10) −0.0034 (10) C1 0.0542 (18) 0.069 (2) 0.0481 (17) −0.0156 (16) −0.0081 (14) 0.0038 (15) C2 0.0551 (18) 0.070 (2) 0.0495 (17) −0.0171 (16) −0.0056 (14) −0.0031 (15) C3 0.0485 (16) 0.0500 (18) 0.0462 (15) −0.0087 (13) 0.0012 (12) −0.0003 (13) C4 0.0481 (15) 0.0480 (17) 0.0503 (16) −0.0051 (13) 0.0012 (13) 0.0026 (13) C5 0.0598 (19) 0.061 (2) 0.0527 (17) −0.0041 (16) −0.0095 (14) 0.0033 (15) C6 0.064 (2) 0.067 (2) 0.0510 (18) −0.0005 (18) −0.0087 (15) 0.0092 (16) Article retracted Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso*/Ueq Mn1 0.7500 0.7500 0.5000 0.0486 (5) C1 0.5842 (3) 0.5617 (4) 0.3628 (2) 0.0572 (8) H1 0.5711 0.6280 0.3132 0.069* C2 0.5366 (3) 0.4245 (4) 0.3721 (2) 0.0585 (9) H2 0.4865 0.3819 0.3303 0.070* C3 0.6462 (2) 0.4631 (4) 0.4891 (2) 0.0482 (7) C4 0.7106 (2) 0.4562 (4) 0.5744 (2) 0.0489 (7) C5 0.8223 (3) 0.5322 (4) 0.6783 (2) 0.0580 (8) H5 0.8713 0.5915 0.7118 0.070* C6 0.7940 (3) 0.3903 (4) 0.7030 (2) 0.0610 (9) H6 0.8195 0.3369 0.7554 0.073* N1 0.5763 (2) 0.3633 (3) 0.45300 (18) 0.0534 (7) H1A 0.546 (3) 0.270 (3) 0.474 (3) 0.064* N2 0.6529 (2) 0.5851 (3) 0.43706 (17) 0.0520 (7) N3 0.7688 (2) 0.5733 (3) 0.59802 (18) 0.0525 (7) N4 0.7220 (2) 0.3432 (3) 0.63640 (19) 0.0551 (7) H4 0.680 (3) 0.255 (3) 0.634 (3) 0.066* N5 0.8833 (2) 0.6634 (3) 0.4268 (2) 0.0529 (7) N6 0.8963 (2) 0.5320 (3) 0.4212 (2) 0.0552 (7) N7 0.9109 (3) 0.4018 (3) 0.4143 (2) 0.0723 (9) cted ctional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) U */U Atomic displacement parameters (Å2) U11 U22 U33 U12 U13 U23 Mn1 0.0504 (12) 0.0400 (13) 0.0550 (13) 0.0096 (10) −0.0016 (10) −0.0034 (10) C1 0.0542 (18) 0.069 (2) 0.0481 (17) −0.0156 (16) −0.0081 (14) 0.0038 (15) C2 0.0551 (18) 0.070 (2) 0.0495 (17) −0.0171 (16) −0.0056 (14) −0.0031 (15) C3 0.0485 (16) 0.0500 (18) 0.0462 (15) −0.0087 (13) 0.0012 (12) −0.0003 (13) C4 0.0481 (15) 0.0480 (17) 0.0503 (16) −0.0051 (13) 0.0012 (13) 0.0026 (13) C5 0.0598 (19) 0.061 (2) 0.0527 (17) −0.0041 (16) −0.0095 (14) 0.0033 (15) C6 0.064 (2) 0.067 (2) 0.0510 (18) −0.0005 (18) −0.0087 (15) 0.0092 (16) Atomic displacement parameters (Å2) sup-3 Acta Cryst. Acta Cryst. (2008). E64, m934 Special details (2008). E64, m934 Article retracted supporting information N1 0.0539 (15) 0.0539 (16) 0.0525 (15) −0.0158 (13) 0.0020 (12) −0.0012 (12) N2 0.0512 (14) 0.0569 (17) 0.0477 (14) −0.0117 (13) −0.0035 (11) 0.0045 (12) N3 0.0519 (15) 0.0543 (16) 0.0508 (14) −0.0088 (13) −0.0063 (11) 0.0042 (12) N4 0.0600 (16) 0.0524 (16) 0.0524 (14) −0.0079 (13) −0.0021 (12) 0.0077 (12) N5 0.0585 (16) 0.0552 (17) 0.0444 (14) −0.0006 (14) −0.0080 (12) −0.0089 (12) N6 0.0525 (15) 0.0540 (18) 0.0585 (16) −0.0129 (13) −0.0071 (12) 0.0069 (13) N7 0.073 (2) 0.0512 (18) 0.092 (2) −0.0078 (15) −0.0108 (17) 0.0073 (16) Geometric parameters (Å, º) Mn1—N2 2.094 (3) C3—C4 1.430 (4) Mn1—N2i 2.094 (3) C4—N3 1.315 (4) Mn1—N3i 2.114 (3) C4—N4 1.345 (4) Mn1—N3 2.114 (3) C5—N3 1.350 (4) Mn1—N5 2.138 (3) C5—C6 1.370 (5) Mn1—N5i 2.138 (3) C5—H5 0.9300 C1—N2 1.351 (4) C6—N4 1.350 (4) C1—C2 1.375 (5) C6—H6 0.9300 C1—H1 0.9300 N1—H1A 0.966 (18) C2—N1 1.349 (4) N4—H4 0.952 (19) C2—H2 0.9300 N5—N6 1.193 (4) C3—N2 1.323 (4) N6—N7 1.187 (4) C3—N1 1.340 (4) N2—Mn1—N2i 180.0) N3—C4—N4 113.0 (3) N2—Mn1—N3i 101.59 (10) N3—C4—C3 118.2 (3) N2i—Mn1—N3i 78.41 (10) N4—C4—C3 128.8 (3) N2—Mn1—N3 78.41 (10) N3—C5—C6 110.1 (3) N2i—Mn1—N3 101.59 (10) N3—C5—H5 125.0 N3i—Mn1—N3 180.0 C6—C5—H5 125.0 N2—Mn1—N5 89.31 (11) N4—C6—C5 106.5 (3) N2i—Mn1—N5 90.69 (11) N4—C6—H6 126.7 N3i—Mn1—N5 91.53 (11) C5—C6—H6 126.7 N3—Mn1—N5 88.47 (11) C3—N1—C2 105.6 (3) N2—Mn1—N5i 90.69 (11) C3—N1—H1A 135 (3) N2i—Mn1—N5i 89.31 (11) C2—N1—H1A 119 (3) N3i—Mn1—N5i 88.47 (11) C3—N2—C1 104.7 (3) N3—Mn1—N5i 91.53 (11) C3—N2—Mn1 113.26 (19) N5—Mn1—N5i 180.0 C1—N2—Mn1 141.7 (2) N2—C1—C2 109.3 (3) C4—N3—C5 104.5 (3) N2—C1—H1 125.3 C4—N3—Mn1 112.6 (2) C2—C1—H1 125.3 C5—N3—Mn1 142.9 (2) N1—C2—C1 107.3 (3) C4—N4—C6 105.8 (3) N1—C2—H2 126.4 C4—N4—H4 124 (3) C1—C2—H2 126.4 C6—N4—H4 130 (3) N2—C3—N1 113.1 (3) N6—N5—Mn1 120.1 (2) supporting information N1 0.0539 (15) 0.0539 (16) 0.0525 (15) −0.0158 (13) 0.0020 (12) −0.0012 (12) N2 0.0512 (14) 0.0569 (17) 0.0477 (14) −0.0117 (13) −0.0035 (11) 0.0045 (12) N3 0.0519 (15) 0.0543 (16) 0.0508 (14) −0.0088 (13) −0.0063 (11) 0.0042 (12) N4 0.0600 (16) 0.0524 (16) 0.0524 (14) −0.0079 (13) −0.0021 (12) 0.0077 (12) N5 0.0585 (16) 0.0552 (17) 0.0444 (14) −0.0006 (14) −0.0080 (12) −0.0089 (12) N6 0.0525 (15) 0.0540 (18) 0.0585 (16) −0.0129 (13) −0.0071 (12) 0.0069 (13) N7 0.073 (2) 0.0512 (18) 0.092 (2) −0.0078 (15) −0.0108 (17) 0.0073 (16) sup-4 Acta Cryst. (2008). E64, m934 supporting information supporting information Article retracted pp g N2—C3—C4 117.4 (3) N7—N6—N5 178.6 (4) N1—C3—C4 129.5 (3) Symmetry code: (i) −x+3/2, −y+3/2, −z+1. Hydrogen-bond geometry (Å, º) D—H···A D—H H···A D···A D—H···A N1—H1A···N7ii 0.97 (2) 2.26 (3) 3.031 (4) 136 (3) N1—H1A···N5iii 0.97 (2) 2.33 (4) 3.021 (4) 127 (3) N4—H4···N7ii 0.95 (2) 1.92 (2) 2.834 (4) 160 (4) Symmetry codes: (ii) −x+3/2, −y+1/2, −z+1; (iii) x−1/2, y−1/2, z. N2—C3—C4 117.4 (3) N7—N6—N5 178.6 (4) N1—C3—C4 129.5 (3) Symmetry code: (i) −x+3/2 −y+3/2 −z+1 Symmetry code: (i) −x+3/2, −y+3/2, −z+1. d Hydrogen-bond geometry (Å, º) D—H···A D—H H···A D···A D—H···A N1—H1A···N7ii 0.97 (2) 2.26 (3) 3.031 (4) 136 (3) N1—H1A···N5iii 0.97 (2) 2.33 (4) 3.021 (4) 127 (3) N4—H4···N7ii 0.95 (2) 1.92 (2) 2.834 (4) 160 (4) Symmetry codes: (ii) −x+3/2, −y+1/2, −z+1; (iii) x−1/2, y−1/2, z. Hydrogen-bond geometry (Å, º) Hydrogen-bond geometry (Å, º) Article re sup-5 Acta Cryst. (2008). E64, m934

https://openalex.org/W1994867863

https://europepmc.org/articles/pmc3737412?pdf=render

English

Species-Specific and Cross-Reactive IgG1 Antibody Binding to Viral Capsid Protein 1 (VP1) Antigens of Human Rhinovirus Species A, B and C

PloS one

cc-by

Abstract Competing Interests: The authors have declared that no competing interest exist. * E-mail: [email protected] Species-Specific and Cross-Reactive IgG1 Antibody Binding to Viral Capsid Protein 1 (VP1) Antigens of Human Rhinovirus Species A, B and C Jua Iwasaki1,2, Wendy-Anne Smith1, Shane R. Stone1, Wayne R. Thomas1, Belinda J. Hales1,2* 1 Telethon Institute for Child Health Research, Centre for Child Health Research, University of Western Australia, Perth, Australia, 2 School of Paediatrics and Child Health, University of Western Australia, Perth, Australia Abstract Background: Human rhinoviruses (HRV) are associated with upper and lower respiratory illnesses, including severe infections causing hospitalization in both children and adults. Although the clinical significance of HRV infections is now well established, no detailed investigation of the immune response against HRV has been performed. The purpose of this study was to assess the IgG1 antibody response to the three known HRV species, HRV-A, -B and -C in healthy subjects. Methods: Recombinant polypeptides of viral capsid protein 1 (VP1) from two genotypes of HRV-A, -B and -C were expressed as glutathione S-transferase (GST) fusion proteins and purified by affinity and then size exclusion chromatography. The presence of secondary structures similar to the natural antigens was verified by circular dichroism analysis. Total and species-specific IgG1 measurements were quantitated by immunoassays and immunoabsorption using sera from 63 healthy adults. Results: Most adult sera reacted with the HRV VP1 antigens, at high titres. As expected, strong cross-reactivity between HRV genotypes of the same species was found. A high degree of cross-reactivity between different HRV species was also evident, particularly between HRV-A and HRV-C. Immunoabsorption studies revealed HRV-C specific titres were markedly and significantly lower than the HRV-A and HRV-B specific titres (P,0.0001). A truncated construct of HRV-C VP1 showed greater specificity in detecting anti-HRV-C antibodies. Conclusions: High titres of IgG1 antibody were bound by the VP1 capsid proteins of HRV-A, -B and -C, but for the majority of people, a large proportion of the antibody to HRV-C was cross-reactive, especially to HRV-A. The improved specificity found for the truncated HRV-C VP1 indicates species-specific and cross-reactive regions could be defined. Editor: Ellen R. Goldman, Naval Research Laboratory, United States of America Received April 23, 2013; Accepted June 23, 2013; Published August 7, 2013 Copyright:
2013 Iwasaki et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by a National Health and Medical Research Council of Australia grant (458513). J. Iwasaki was supported by an Australian Postgraduate Award, a University of Western Australia Scholarship, a Stan and Jean Perron Scholarship and an Asthma Foundation of Western Australia Scholarship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction Neutralising antibodies provided the key observations that first distinguished HRV-A from HRV-B but, because they are directed to a limited number of capsid structures that frequently mutate [8], they have limited application for sero-epidemiology and profiling the history of immune responses to repeated infections. The neutralising antibodies induced by experimental HRV inoculation appear 2–3 weeks after infection, well after the clearance of virus and resolution of the symptoms [9,10], and at least for experimental infection can no longer be detected after 9 months [11]. As found for other respiratory virus infections, it is likely that innate immunity and T-cell responses are the most important responses for the resolution of infection, while neutralising antibodies protect against short-term reinfection [12]. Antibodies have been shown to interact with T-cells in adaptive immune responses to clear viruses, but interestingly these were non- neutralising antibodies [13,14] that are also likely to be more abundant. Such interaction between non-neutralising antibodies Human rhinovirus (HRV) infection is not only the main cause of the common cold [1] but is strongly associated with exacerbations of asthma, acute bronchitis, chronic obstructive pulmonary disease in adults, as well as pneumonia in immunocompromised patients [2–4]. The biologically and molecularly distinct HRV-C species, which has been found in hospitalised patients, has attracted attention for a role in asthma exacerbation of children [5] and for serious disease in adults and the elderly, including transplant patients [4]. HRV-C has been difficult to grow in vitro [6], so its detection has relied on the sequencing of viral RNA in tissue samples, the method now most commonly used to detect HRV-A and HRV-B. The prevalence of RNA detection for all species has been higher in hospitalized patients [3] but has also been high in healthy subjects often in similar abundance [7]. August 2013 | Volume 8 | Issue 8 | e70552 1 PLOS ONE | www.plosone.org August 2013 | Volume 8 | Issue 8 | e70552 Antibody Binding to Human Rhinovirus Antigens Table 1. Amino acid sequence identity for six HRV VP1 and HPV Sabin 1 VP1. HRV-A HRV-B HRV-C HPV 34 1B 14 69 3 5 Sabin HRV-A 34 77.7 35.2 35.5 41.4 44.6 35.2 1B 36.9 37.2 39.4 43.6 33.9 HRV-B 14 71.1 32.9 33.1 40.5 69 33.8 36.2 42.2 HRV-C 3 61.9 28.3 5 32.0 HPV Sabin doi:10.1371/journal.pone.0070552.t001 doi:10.1371/journal.pone.0070552.t001 and HRV-C3 (EF186077 [20]) and HRV-C5 (EF582386 [2]) of HRV-C species (Table S1). Ethics Statement Written informed consent was obtained from all participants. Ethics approval was provided by The Princess Margaret Hospital Human Ethics Committee. Introduction The VP1 of another enterovirus, human poliovirus (HPV) Sabin VP1 (AY184219.1) was produced as a control to determine specificity in antibody binding to HRV. The amino acid sequence identities of the VP1 proteins are shown in Table 1. and T-cells might also be expected to affect immunopathology, for example in aiding the capture of virus by monocytes and thereby enhancing inflammatory cytokine production [15–17]. The objectives of this study were to use antibody-binding to defined HRV antigens to determine whether immune responses to the different HRV species could be distinguished, and further, if the antibody responses to the different species had different characteristics. The antigen examined was the VP1 capsid protein that exhibits the highest surface exposure of the four HRV capsid proteins and forms part of neutralising antigenic determinants [18]. It, along with VP2 and VP3, exhibits the amino acid sequence variation found amongst genotypes of each species. Comparing the VP1 of HRV-A and HRV-B, most isolates form a cluster with 58–99% amino acid sequence identity within each species, contrasting with the 35–44% identity found between HRV-A and HRV-B [19]. Similarly, the VP1 of HRV-C genotypes has 59–90% amino acid sequence identity within HRV-C, and about 45% and 36% identity between HRV-C and HRV-A and HRV-B respectively [2]. Our studies here show high IgG1 antibody binding to recombinant full length VP1 antigen constructs for all species, with high titres of species-specific antibody for HRV-A and HRV-B, in contrast to a low prevalence and titre for HRV-C. A construct of HRV-C made without the N- terminal alpha-helical domain showed a lower degree of antibody binding but, significantly, was restricted to sera from subjects with species-specific antibodies. Study Population For expression of VP1, an overnight culture diluted 1:20 was grown to OD600 nm 0.6 and induced with 0.1 mM IPTG at 30uC for 2 hours. The E.coli pellets were resuspended in 5 ml/g Buffer A (150 mM NaCl, 50 mM NaH2PO4, 1% Tween-20, 1 mM PMSF, pH 8) with the addition of lysozyme (1 mg/ml, Sigma-Aldrich, St Louis, MO), sonicated and clarified at 18,000 rpm for 60 min. The soluble supernatant was then purified in accordance with the manufacturer’s protocols (Sigma, USA) with modifications. Briefly, glutathione agarose was pre-equilibrated with Buffer B (150 mM NaCl, 50 mM NaH2PO4, 0.1% Tween-20, pH 8). The clarified lysate was bound to the matrix and the column was washed with 106column volume with Buffer B. Bound protein was eluted with Buffer C (Buffer B +10 mM reduced glutathione). Fractions collected from the column containing recombinant protein were pooled, concentrated and passed over a high resolution S300 26/ 60 column (GE Healthcare, Uppsala, Sweden). Sera from 63 adults with no evidence of current respiratory illness from the general population in Perth, Western Australia were examined (40 women and 23 men; age ranged from 20–53; median age 31). Expression and Purification of Recombinant HRV VP1 The nucleotide sequences encoding VP1 cDNAs were synthe- sized with codon optimization for expression in E.coli by GenScript (Piscataway, NJ). They were subsequently engineered for expres- sion as fusion proteins with glutathione S-transferase (GST) at the N-terminus and a hexa-histidine tag on the C-terminus. The genes were amplified by PCR from cDNA in pUC57 as a template. Specific PCR primers were designed to amplify the VP1 coding sequence and the addition of six histidine residues. PCR was performed using high-fidelity Pfu DNA polymerase (Promega, Madison, WI) using the following conditions: 1 cycle at 95uC for 5 min; 35 cycles at 95uC for 1 min, 55uC for 30 s, and 74uC for 3 min; and finally 74uC for 7 min. The PCR products were extracted from a 1% agarose gel using the Gel Purification Kit (Qiagen, Hilden, Germany). The amplified DNA fragment was digested with BamHI and EcoRI, and ligated into a BamHI/EcoRI digested pGEX-2T expression vector (GE Healthcare). After transformation into chemically competent TOP 10 E.coli, positive clones were verified by DNA sequencing (AGRF, Australia). The recombinant plasmid was subsequently transformed into the E.coli expression strain BL21. A GST control was produced directly from pGEX-2T. HRV Antigens VP1 antigens from two genotypes of HRV-A, -B and -C were produced in Escherichia coli (E.coli) as they represent genetically disparate variants within each species. The following HRV VP1 proteins were produced: HRV-A34 (GenBank accession number FJ445189.1) and HRV-A1B (D00239.1) of HRV-A species; HRV- B14 (NC001490) and HRV-B69 (FJ445151) of HRV-B species; The purity of the recombinant proteins were analysed by size exclusion chromatography and SDS-PAGE analysis using a 12.5% August 2013 | Volume 8 | Issue 8 | e70552 August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 2 Antibody Binding to Human Rhinovirus Antigens Figure 1. Purification and SDS-PAGE analysis of recombinant HRV-C3 VP1 pGEX-2T. (A) Chromatograph of HRS300 size exclusion chromatography immediately following affinity chromatography (Run 1). The void volume and the peak at 660 kDa are indicated above each of the main peaks. Each fraction has a 5 ml elution volume. Fractions containing the 660 kDa peak from Run 1 were pooled and concentrated for a second run and third run on the HRS300. The bar indicates fractions used for subsequent runs and for binding analysis. (B) A 10 mL aliquot of the pooled fractions containing the 660 kDa peak collected from the first run of gel filtration (as indicated by *) was analysed by SDS-PAGE. The MW of the protein standard is indicated. doi:10.1371/journal.pone.0070552.g001 Figure 1. Purification and SDS-PAGE analysis of recombinant HRV-C3 VP1 pGEX-2T. (A) Chromatograph of HRS300 size exclusion chromatography immediately following affinity chromatography (Run 1). The void volume and the peak at 660 kDa are indicated above each of the main peaks. Each fraction has a 5 ml elution volume. Fractions containing the 660 kDa peak from Run 1 were pooled and concentrated for a second run and third run on the HRS300. The bar indicates fractions used for subsequent runs and for binding analysis. (B) A 10 mL aliquot of the pooled fractions containing the 660 kDa peak collected from the first run of gel filtration (as indicated by *) was analysed by SDS-PAGE. The MW of the protein standard is indicated. doi:10 1371/journal pone 0070552 g001 (NH4)2SO4 buffer (pH 8) and circular dichroism (CD) spectrosco- py performed as outlined in Hales et al. [21]. Briefly, CD spectral measurements were performed on an OLIS DSM-1000 spectro- photometer over the wavelength range 260–190 nm operating at 20uC in a 1-mm path length cell (Starna). Circular Dichroism Analysis Purified protein preparations were diluted to a final concentra- tion of 3 mM in 10 mM potassium phosphate, 100 mM Table 2. Percentages of the secondary structure of the circular dichroism (CD) spectra of HRV VP1 antigens following subtraction of GST*. a-helix b-sheet b-turns Disordered HRV-A34 VP1 10 30 24 36 HRV-B14 VP1 10 36 25 27 HRV-C3VP1 13 33 25 29 Truncated HRV-C3 VP1 12 33 26 30 *Analysed using DiChroWeb Server: CDSSTR algorithm, reference set 4. doi:10.1371/journal.pone.0070552.t002 HRV Antigens The analysis of protein secondary structure from CD spectra was performed using the DiChroWeb Server [22] and the CDSSTR algorithm and reference sets 4, 7 and SP175 (short) [23]. electrophoretic gel and GelCode Blue Safe Protein Stain (Thermo Scientific). Protein concentrations were calculated using OD280 nm and extinction coefficients calculated for each fusion protein. August 2013 | Volume 8 | Issue 8 | e70552 Results Expression of Recombinant GST-fusion HRV VP1 Antigens Recombinant GST-fusion HRV and HPV Sabin VP1 with a C- terminal histidine tag was produced in E.coli as a soluble protein. Following glutathione-agarose affinity chromatography and se- quential gel exclusion chromatography, the final gel filtration isolated a multimeric protein estimated to be 660 kDa (Figure 1A) distinct from the aggregated protein that was excluded from the gel. Under denaturing conditions, the SDS-PAGE revealed a single band of 59 kDa (Figure 1B). IgG1 Competitive Inhibition Studies The plates were read using a Wallac Victor 3 plate reader (Wallac, Oy). blocking buffer (0.5% bovine serum albumin in 50 mM Tris-HCl, 0.9% NaCl, 0.05% sodium azide (pH 7.4) buffer with 0.01% Tween-20) and incubated overnight at 4uC with shaking. The coated plates were washed 5 times with 50 mM Tris-HCl, 0.9% NaCl (pH 7) buffer with 0.01% Tween-20. This washing step was conducted after each incubation step. The plates were blocked with 200 mL of blocking buffer for 2 hours at room temperature with shaking. 100 mL of sera was then added to the antigen-coated wells and incubated for 2 hours at room temperature with shaking. The wells were then incubated with 100 mL of biotinylated anti- human IgG1 (BD Pharmingen, USA) diluted 1:2000 in assay buffer (Wallac, Oy) for 2 hours at room temperature with shaking. This was then followed by incubation for 30 minutes with 100 mL Europium-labelled streptavidin diluted 1:1000 in assay buffer, with shaking. The final washing step involved washing the plates 8 times before the addition of 100 mL enhancement solution. The plates were read using a Wallac Victor 3 plate reader (Wallac, Oy). Antibody Binding to Human Rhinovirus Antigens Antibody Binding to Human Rhinovirus Antigens Figure 2. Total IgG1 antibody binding (ng/ml) to HRV and HPV Sabin VP1 antigens in 63 healthy adults. The geometric mean and 95% confidence interval are indicated. Human myeloma IgG1 was used as a negative control to determine non-specific binding and identification of negative control sera. Three negative control sera for each antigen and a titration of reference sera were included on every plate for the quantitation of IgG1 binding and to assess reproducibility of the assay. doi:10.1371/journal.pone.0070552.g002 determine non-specific binding and the identification of negative control sera (Figure S1). Three negative control sera for each antigen were included on every plate. The absolute quantity of antigen-specific IgG1 (ng/ml) in the reference sera was interpo- lated from a titration curve constructed with equivalent concen- trations of recombinant GST-fusion Der p 2 and a standardised humanised anti-Der p 2 chimeric IgG1 (Indoor Biotechnologies, Charlottesville, USA) [25,26]. Wells coated with the same quantity of antigen were used, as determined by the binding of anti- glutathione-S-transferase antibody. The lower limit of detection for IgG1 binding was 500 ng/ml and thus negative values were assigned 50% of the lower limit of detection (250 ng/ml). Structural Characterization of HRV VP1 Antigens Structural Characterization of HRV VP1 Antigens Purified GST-fusion proteins were subjected to subtractive CD analysis in which the signal from the GST control was subtracted from the fusion proteins. The CD spectrum of the GST dimer showed predominantly a-helical structure analogous to published data [28]. The CD spectra of HRV-A34, HRV-B14 and HRV-C3 VP1 demonstrated the presence of well-folded secondary struc- ture, with broad minima between 208–220 nm indicative of mixed b-sheet and a-helical structure. Minima at 194 nm, close to the expected 198 nm minimum, also suggested random coil confor- mation (Figure S2). Secondary structure calculations of HRV-A34 Statistical Analysis The absolute antibody binding to each VP1 antigen was log- transformed to approximate a normal distribution. Differences in antibody binding to different antigens were compared by the paired t-test. A P value ,0.05 was considered significant. The geometric mean and 95% confidence intervals were calculated. Correlations between antibody responses were analysed by the Spearman correlation. All analyses were performed using GraphPad Prism Software (La Jolla, USA). IgG1 Competitive Inhibition Studies Sera from individuals with high IgG1 binding to two HRV species were diluted 1:50 with blocking buffer and then mixed 1:2 with different dilutions of purified VP1 antigens to give final concentrations ranging from 0.004 to 8 mg/ml of the inhibiting protein. Following overnight incubation at 4uC, the mixtures were centrifuged at 14,000 g for 10 min and 100 mL aliquots of the supernatant were added to the antigen-coated plates for the DELFIAH assays. The assay was then developed as per the IgG1 immunoassay. An irrelevant recombinant antigen, either Fel d 3 or the bacterial protein pneumococcal surface protein (Psp)-C, was used as a negative control to illustrate that any cross-reactivity that existed was specific to HRV species and, where present to HPV Sabin. Recombinant Fel d 3 and Psp-C were both expressed in an E.coli expression system and purified by multiple steps of purification. Details of the antigen preparations have been described previously [24,27]. Figure 2. Total IgG1 antibody binding (ng/ml) to HRV and HPV Sabin VP1 antigens in 63 healthy adults. The geometric mean and 95% confidence interval are indicated. Human myeloma IgG1 was used as a negative control to determine non-specific binding and identification of negative control sera. Three negative control sera for each antigen and a titration of reference sera were included on every plate for the quantitation of IgG1 binding and to assess reproducibility of the assay. doi:10.1371/journal.pone.0070552.g002 doi:10.1371/journal.pone.0070552.g002 blocking buffer (0.5% bovine serum albumin in 50 mM Tris-HCl, 0.9% NaCl, 0.05% sodium azide (pH 7.4) buffer with 0.01% Tween-20) and incubated overnight at 4uC with shaking. The coated plates were washed 5 times with 50 mM Tris-HCl, 0.9% NaCl (pH 7) buffer with 0.01% Tween-20. This washing step was conducted after each incubation step. The plates were blocked with 200 mL of blocking buffer for 2 hours at room temperature with shaking. 100 mL of sera was then added to the antigen-coated wells and incubated for 2 hours at room temperature with shaking. The wells were then incubated with 100 mL of biotinylated anti- human IgG1 (BD Pharmingen, USA) diluted 1:2000 in assay buffer (Wallac, Oy) for 2 hours at room temperature with shaking. This was then followed by incubation for 30 minutes with 100 mL Europium-labelled streptavidin diluted 1:1000 in assay buffer, with shaking. The final washing step involved washing the plates 8 times before the addition of 100 mL enhancement solution. Immunoassays to Determine Total IgG1 Antibody Binding Table 2. Percentages of the secondary structure of the circular dichroism (CD) spectra of HRV VP1 antigens following subtraction of GST*. g The DELFIAH assay is described in detail elsewhere [24] and was performed with modifications. In brief, microtitre plates (Nunc, Roskilde, Denmark) were coated with 100 mL of purified VP1 antigens diluted to 0.125–4 mg/ml in 50 mM Na2CO3/ NaHCO3 (pH 9.6) at 4uC overnight. The concentration of antigen used for coating was standardised by titrating the coating concentrations using monoclonal anti-glutathione-S-transferase antibody (Sigma-Aldrich, St Louis, MO) to measure the amount adhering to the well. Optimal concentrations were determined by referencing to a recombinant GST-fusion Der p 2 standard that was later used for quantitation of antibody binding (see below). Each test sera was diluted to a final concentration of 1:100 in *Analysed using DiChroWeb Server: CDSSTR algorithm, reference set 4. doi:10.1371/journal.pone.0070552.t002 August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 3 Immunoabsorption Assays to Determine Species-Specific IgG1 Antibody Binding Immunoabsorption assays were conducted to determine the species-specific IgG1 antibody binding and were performed as per the immunoassay above, except that each test sera was pre- incubated in a lysate mixture of E.coli producing the other two HRV species and HPV Sabin VP1 to absorb out cross-reactive binding. The lysates (produced using soluble supernatants following sonication as described above) were used at a final concentration of 1:250 shown by pilot experiments to be an excess amount to ensure complete inhibition where present. Quantitation of IgG1 Antibody Binding The percentages of secondary structure of recombinant HRV-B14 VP1 was also similar to published data on native HRV-B14 VP1 ([30]; pdb: 1NCQ). Quantitation of IgG1 Antibody Binding A titration of reference sera was included on every plate to construct a standard curve and act as a positive control to assess reproducibility. The correlation of variation was less than 5% between plates. The standard curve was then used to quantitate the IgG1 binding to the VP1 antigens. Human myeloma IgG1 (Sigma-Aldrich, St Louis, MO) was used as a negative control to August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 4 Antibody Binding to Human Rhinovirus Antigens Figure 3. Strong IgG1 cross-reactivity between two HRV genotypes of the same species. Each graph represents percentage of inhibition in individual sera. Subjects that had high total IgG1 titres to testing antigens were selected for competitive inhibition assays and example inhibition assays are shown. (A) Level of inhibition between HRV-A34 and HRV-A1B (HRV-A species). (B) Level of inhibition between HRV-B14 and HRV-B69 (HRV- B species). (C) Level of inhibition between HRV-C3 and HRV-C5 (HRV-C species). The inhibition of binding by an irrelevant recombinant antigen (Fel d 3), used as a negative control, are also indicated. doi:10.1371/journal.pone.0070552.g003 Figure 3. Strong IgG1 cross-reactivity between two HRV genotypes of the same species. Each graph represents percentage of inhibition in individual sera. Subjects that had high total IgG1 titres to testing antigens were selected for competitive inhibition assays and example inhibition assays are shown. (A) Level of inhibition between HRV-A34 and HRV-A1B (HRV-A species). (B) Level of inhibition between HRV-B14 and HRV-B69 (HRV- B species). (C) Level of inhibition between HRV-C3 and HRV-C5 (HRV-C species). The inhibition of binding by an irrelevant recombinant antigen (Fel d 3), used as a negative control, are also indicated. doi:10 1371/journal pone 0070552 g003 than HRV-B (83% for HRV-B14 and 89% for HRV-B69) and HRV-C (70% for HRV-C3 and 79% for HRV-C5). VP1 using the CDSSTR algorithm and reference set 4 indicated the presence of 10% a-helix and 30% b-sheet content (Table 2), comparable to the published secondary structure of HRV-A1A VP1 ([29]; pdb: 1R1A). The percentages of secondary structure of recombinant HRV-B14 VP1 was also similar to published data on native HRV-B14 VP1 ([30]; pdb: 1NCQ). VP1 using the CDSSTR algorithm and reference set 4 indicated the presence of 10% a-helix and 30% b-sheet content (Table 2), comparable to the published secondary structure of HRV-A1A VP1 ([29]; pdb: 1R1A). Competitive Inhibition Studies To examine cross-reactivity between the genotypes, the cross inhibition of the two VP1 proteins representing each species was examined. Sera from two high responder subjects were selected for each species and reciprocal competitive inhibition assays conduct- ed with antigen from the two genotypes for each species. The genotype VP1 for the HRV-A strongly cross inhibited each other (Figure 3A) as did the genotypes for HRV-B (Figure 3B). The inhibition for HRV-C was similar, although the inhibition of binding to HRV-C5 by HRV-C3 plateaued at 80% (Figure 3C). The inhibitions all reached a maximum with antigen concentra- tion of 4 mg/ml or less. Total IgG1 Antibody Binding to HRV VP1 Antigens Total IgG1 Antibody Binding to HRV VP1 Antigens Initially, total IgG1 antibody binding to each of the HRV VP1 antigens was measured. This provided a measure of IgG1 binding to each of the species, including any reactivity that may be directed to cross-reacting epitopes shared with other HRV species, or with other enteroviruses. The VP1 antigens bound IgG1 antibodies from the 63 adult sera at high titres (Figure 2). There was no significant difference between HRV-A, -B, -C or HPV Sabin VP1 titres. However, the prevalence of total IgG1 reactivity was higher for HRV-A variants (97% for HRV-A34 and 94% for HRV-A1B) Cross inhibition measured with VP1 between the species had a high degree of variation with different sera. The reciprocal August 2013 | Volume 8 | Issue 8 | e70552 August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 5 Antibody Binding to Human Rhinovirus Antigens Figure 4. Cross-reactivity between HRV-A and HRV-C. (A) An example of strong cross-inhibition in HRVA34 and HRV-C3 double-positive sera. (B) An example of poor cross-inhibition between HRV-A34 and HRV-C3 observed in a minority of subjects who had high IgG1 titres to both HRV-A and HRV-C. The inhibition of binding to HRV-A34 and HRV-C3 by an irrelevant recombinant antigen (bacterial protein, Psp-C), used as a negative control, are also indicated. doi:10.1371/journal.pone.0070552.g004 Figure 4. Cross-reactivity between HRV-A and HRV-C. (A) An example of strong cross-inhibition in HRVA34 and HRV-C3 double-positive sera. (B) An example of poor cross-inhibition between HRV-A34 and HRV-C3 observed in a minority of subjects who had high IgG1 titres to both HRV-A and HRV-C. The inhibition of binding to HRV-A34 and HRV-C3 by an irrelevant recombinant antigen (bacterial protein, Psp-C), used as a negative control, are also indicated. doi:10.1371/journal.pone.0070552.g004 The same procedure was used for the antigens from the other species. Each assay included three negative control sera, as determined by human myeloma IgG1, and reference sera. The species-specific titres determined, for each of the two genotypes of HRV-A, HRV-B and HRV-C, show the very high titres of antibody to the HRV-A species compared to usually low titres for HRV-C, and much higher prevalence of the species-specific binding antibodies to HRV-A than HRV-C (Figure 7; P,0.0001). Only 19/63 subjects had antibodies specifically attributable to HRV-C (HRV-C3) and the titres of the responders did not reach the titres found for HRV-A. Antibody Responses to Truncated HRV-C VP1 Antibody Responses to Truncated HRV-C VP1 To measure IgG1 binding to truncated HRV-C VP1, HRV-C3 VP1 was cloned and expressed from residue 14 to 275 in the pGEX-2T expression system. The purification protocol was the same as for full-length HRV-C3 VP1 and showed the same characteristics. CD analysis revealed similar structures to full- length HRV-C3 VP1, with the truncated HRV-C3 VP1 being composed of 33% b-strand and 12% a-helix (Table 2). IgG1 binding to truncated HRV-C3 VP1 was measured in 30 subjects who had high IgG1 titres to full-length HRV-C3 VP1. The IgG1 titres to truncated HRV-C3 VP1 were significantly lower than total IgG1 to full length VP1 (Figure 9A, P,0.0001) with 9/30 subjects having no reactivity to the truncated construct of HRV-C3 VP1. The sera with detectable antibody to truncated HRV-C3 VP1 were however, interestingly, the same as those with an antibody response to full-length HRV-C3 VP1 remaining after Total IgG1 Antibody Binding to HRV VP1 Antigens The HRV-B titres were similar in size and prevalence to the HRV-A. The correlation of the antibody binding between the VP1 antigens of each genotype within each species was very high (Figure 8). inhibition with the serum shown in Figure 4A shows almost 100% reciprocal inhibition between HRV-A34 genotype antigen of HRV-A and the HRV-C3 genotype of HRV-C, while the inhibition between HRV-A and HRV-C shown in Figure 4B in another subject, was low. Inhibition of IgG1 binding to HRV VP1 was negligible using the bacterial protein, Psp-C as a control (Figure 4), and HRV VP1 was similarly unable to inhibit IgG1 binding to Psp-C. Further competitive inhibitions as exemplified for the individual in Figure 5 showed there were also lesser inhibitions between all the VP1 including with HRV-B and HPV Sabin. The binding to HPV Sabin was strongly cross inhibited by the HRV-B14 genotype of HRV-B, while inhibition by HRV-A and HRV-C was minimal (Figure 5D). The pattern, however, was variable when other subjects were examined. Despite the variation, absorption assays using sera of 30 subjects who had high IgG1 titres to HRV-C (HRV-C3) revealed a much greater inhibition of antibody binding to HRV-C with HRV-A, than with HRV-B or HPV Sabin. Here, absorption with HRV-A lysate resulted in complete inhibition of binding to HRV-C3 in the majority of subjects, showing frequent high cross-reactivity (Figure 6). In comparison, absorption with a mixture of HRV-B and HPV Sabin lysates produced lower and less frequent inhibition. August 2013 | Volume 8 | Issue 8 | e70552 Species-specific IgG1 Antibody Binding to HRV VP1 Antigens In the light of the competitive inhibition studies, the species- specific titres were determined following absorption of each sera with lysates containing the VP1 of the other species. Thus antibodies to the HRV-C genotypes were measured after absorption with a mixture of HRV-A, HRV-B and HPV Sabin. August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 6 Antibody Binding to Human Rhinovirus Antigens Figure 5. An example of the level of cross-reactivity found between HRV species and HPV Sabin VP1 in an individual’s serum. (A) HRV-A VP1 antigen inhibited by HRV-B, HRV-C and HPV Sabin antigens. (B) HRV-B VP1 antigen inhibited by HRV-A, HRV-C and HPV Sabin antigens. (C) HRV-C VP1 antigen inhibited by HRV-A, HRV-B and HPV Sabin antigens. (D) HPV Sabin antigen inhibited by antigens representing the three HRV species. An irrelevant recombinant antigen (Fel d 3) was used as a negative control for each competitive inhibition assay (data not shown). doi:10.1371/journal.pone.0070552.g005 Figure 5. An example of the level of cross-reactivity found between HRV species and HPV Sabin VP1 in an individual’s serum. (A) HRV-A VP1 antigen inhibited by HRV-B, HRV-C and HPV Sabin antigens. (B) HRV-B VP1 antigen inhibited by HRV-A, HRV-C and HPV Sabin antigens. (C) HRV-C VP1 antigen inhibited by HRV-A, HRV-B and HPV Sabin antigens. (D) HPV Sabin antigen inhibited by antigens representing the three HRV species. An irrelevant recombinant antigen (Fel d 3) was used as a negative control for each competitive inhibition assay (data not shown). doi:10.1371/journal.pone.0070552.g005 HRV-A absorption. There was a strong correlation between the titres of full-length HRV-C3 VP1 which had cross-reactivity with HRV-A species removed and the truncated construct (Figure 9B). although as expected from the sequences, the competition was highest between HRV-A and -C. Given the variability between subjects observed in the pilot competitions, the assays were further developed by absorbing sera with an excess of the VP1 antigens from the heterologous HRV species and HPV Sabin and then conducting the titrations against the target VP1. The results showed a high prevalence of species-specific antibodies to HRV-A, HRV-B and HPV Sabin. It should be noted that the Sabin VP1 amino acid sequence (35% and 25% identical to HRV-A and C, respectively) can have nearly 60% identity to some Echoviruses, so this is an important consideration for future studies of HPV Sabin. Species-specific IgG1 Antibody Binding to HRV VP1 Antigens The standout result here was the low prevalence of HRV-C specific titres even though most subjects had high total IgG1 antibody binding. The two genotypes examined for each HRV species gave the same results with high correlations between the species-specific titres that concurred with the competitive inhibi- tion studies showing complete cross-competition with low doses of antigen. The pairs of genotypes that were selected to represent disparate sequences within each species (22%, 29% and 38% amino acid disparity within the HRV-A, -B and -C species, respectively) had the same species-specific binding especially for HRV-A, and where present, also to HRV-C. The response specific to an infecting genotype has not been measured in this study. Future studies could accomplish this by producing an antigen with the sequence of the known HRV genotype infecting the patient, and matching it to the patient anti-serum response. However, given the large shared IgG1 binding and the limited amino acid PLOS ONE | www.plosone.org Discussion IgG1 binding to HRV-C (HRV-C3 and HRV-C5) antigens were significantly lower than HRV-A, HRV-B and HPV Sabin titres (P,0.0001) as indicated by **. Human myeloma IgG1 was used as a negative control to determine non-specific binding and identification of negative control sera. Three negative control sera for each antigen and a titration of reference sera were included on every plate for the quantitation of IgG1 binding and to assess reproducibility of the assay. doi:10.1371/journal.pone.0070552.g007 Figure 6. Inhibition of IgG1 binding to HRV-C3 VP1 by HRV-A or HRV-B and HPV Sabin. Inhibition by lysate containing HRV-A, or a mixture of HRV-B and HPV Sabin in 30 subjects with high total IgG1 titres to HRV-C3. Percentage of cross-reactivity to the inhibiting species was calculated by (Titre after absorption/Total IgG1)6100. P,0.0001 between inhibition by HRV-A and inhibition by HRV-B and HPV Sabin. Three negative control sera and a titration of reference sera were included on every plate to assess reproducibility and quantitation of IgG1 binding. doi:10 1371/journal pone 0070552 g006 serum albumins and haemoglobins from different species can readily distinguish between antigens with 70% sequence identity [41] and the anti-Psp-A antibodies to Streptococcus pneumonia with 69% sequence identity or less, induce largely non-cross-reactive responses [32]. The house dust mite allergens Der p 5 and Blo t 5 with 38% identity show no cross-reaction despite both frequently eliciting large IgE antibody responses [42]. changes that alter neutralization [8], the antibodies against the infecting genotype might be difficult to measure, especially for HRV-A and HRV-B. The low prevalence of species-specific binding to HRV-C is unlikely to be due to the poor antigenicity of the recombinant construct since it bound strongly prior to absorption with HRV-A and the other species, and could inhibit binding to itself with the same dose-response characteristics of HRV-A. The CD measure- ments also did not indicate a deficient structure or instability. Another anomaly is that, HRV-C infection is detectable at almost the same prevalence as HRV-A and twice that of HRV-B [36] for which species-specific antibody binding was far more prevalent. The cross-reactivity and sequence identity of HRV-C with HRV- A was greater than that of HRV-B but this does not account for the high prevalence of HRV-A specific binding after absorption with HRV-C. Discussion Three negative control sera and a titration of reference sera were included on every plate to assess reproducibility and quantitation of IgG1 binding. doi:10.1371/journal.pone.0070552.g006 Figure 7. Species-specific IgG1 antibody binding (ng/ml) to HRV and HPV Sabin VP1 antigens in 63 healthy adults. Each sera was absorbed with lysates containing the VP1 of the other HRV species, as well as HPV Sabin VP1. IgG1 binding to HRV-C (HRV-C3 and HRV-C5) antigens were significantly lower than HRV-A, HRV-B and HPV Sabin titres (P,0.0001) as indicated by **. Human myeloma IgG1 was used as a negative control to determine non-specific binding and identification of negative control sera. Three negative control sera for each antigen and a titration of reference sera were included on every plate for the quantitation of IgG1 binding and to assess reproducibility of the assay. doi:10.1371/journal.pone.0070552.g007 Figure 7. Species-specific IgG1 antibody binding (ng/ml) to HRV and HPV Sabin VP1 antigens in 63 healthy adults. Each sera was absorbed with lysates containing the VP1 of the other HRV species, as well as HPV Sabin VP1. IgG1 binding to HRV-C (HRV-C3 and HRV-C5) antigens were significantly lower than HRV-A, HRV-B and HPV Sabin titres (P,0.0001) as indicated by **. Human myeloma IgG1 was used as a negative control to determine non-specific binding and identification of negative control sera. Three negative control sera for each antigen and a titration of reference sera were included on every plate for the quantitation of IgG1 binding and to assess reproducibility of the assay. doi:10.1371/journal.pone.0070552.g007 Figure 7. Species-specific IgG1 antibody binding (ng/ml) to HRV and HPV Sabin VP1 antigens in 63 healthy adults. Each sera was absorbed with lysates containing the VP1 of the other HRV species, as well as HPV Sabin VP1. IgG1 binding to HRV-C (HRV-C3 and HRV-C5) antigens were significantly lower than HRV-A, HRV-B and HPV Sabin titres (P,0.0001) as indicated by **. Human myeloma IgG1 was used as a negative control to determine non-specific binding and identification of negative control sera. Three negative control sera for each antigen and a titration of reference sera were included on every plate for the quantitation of IgG1 binding and to assess reproducibility of the assay. doi:10.1371/journal.pone.0070552.g007 Figure 7. Species-specific IgG1 antibody binding (ng/ml) to HRV and HPV Sabin VP1 antigens in 63 healthy adults. Each sera was absorbed with lysates containing the VP1 of the other HRV species, as well as HPV Sabin VP1. Discussion It is possible that for most subjects there is the phenomenon of ‘‘original antigenic sin’’ operating, where prior responses to HRV-A dictate the specificities induced by the related HRV-C and as described for influenza, can compromise the induction of protective responses [37]. HRV-C has also been reported to have greater genotypic diversity than HRV-A and HRV-B [38], although here the diverse genotypes correlated strongly. Speculatively, it might be due to the different biology perhaps with reduced growth due to its different receptor and unknown target cell and temperature requirements [39,40]. A priority would be to determine if HRV-C infection does elicit HRV-C specific-antibodies and this would, given the low prevalence, might be easy to detect serology to monitor defined infectious episodes. The HRV-C construct made with the N-terminal truncation only detected IgG1 binding in the sera of subjects that retained anti-HRV-C antibodies when the cross-reactivity to HRV-A was removed by absorption. This could accordingly make a convenient construct for screening anti-HRV-C responses as the main cross- reactivity was with HRV-A, and minimal cross-reactivity with HRV-B or HPV Sabin was found. The truncation was originally studied because it was proposed that removing the surface inaccessible alpha helical region of the VP1 might prevent intermolecular interactions of the VP1 and improve the yields in recombinant expression. The same pentameric-sized complex was, however, produced and it retained similar structure to full-length HRV-C3 VP1. The N-terminal region has separately been of interest since it becomes exposed when the VP4 protein dissociates from the capsid in the acid environment of the lysosome [43]. Furthermore, a high frequency of IgG binding to a synthetic peptide representing this region has been detected [44]. There is sequence conservation between HRV-A and HRV-C within the truncated region that might explain the phenomenon but the truncation might also influence the conformation of other parts of VP1. The study here describes recombinant constructs that can be produced in E.coli as structurally characterized, defined molecules to measure species-specific antibodies produced to HRV-A, -B and -C. The IgG binding studied to date shows evidence for a high degree of antibody production and for a high degree of cross- reactivity between the VP1 of different HRV species. August 2013 | Volume 8 | Issue 8 | e70552 Discussion The VP1 antigen constructs of human rhinovirus species A, B and C all bound high titres of IgG1 antibody with negligible binding to myeloma controls. The total IgG1 binding by the HRV-A and -C antigens varied over a 2-log range but at high titre; the highest titres probably reflect antibodies induced by multiple infections with related viruses and recent infection causing transient elevations. The range of binding of specific titres at 250 - 16107 ng/ml is similar to values obtained for other microbial and vaccine antigens [24,31–34]. All the antibody binding could be competitively inhibited with low concentrations of the homologous antigen. The recombinant glutathione-S- transferase VP1 fusion constructs purified by size exclusion chromatography were discrete, reproducible reagents that had a molecular mass equivalent to five VP1-GST dimers, which could reflect the pentameric interactions between the VP1 proteins in the apical pentameric subunits of the HRV particle [8,35]. CD analysis of the HRV species showed that they had similar secondary structures to the natural protein. The competitive inhibition performed with a sample of sera showed a sometimes high and variable degree of cross-reactivity between the different species including the HPV Sabin VP1, August 2013 | Volume 8 | Issue 8 | e70552 August 2013 | Volume 8 | Issue 8 | e70552 7 PLOS ONE | www.plosone.org Antibody Binding to Human Rhinovirus Antigens Figure 6. Inhibition of IgG1 binding to HRV-C3 VP1 by HRV-A or HRV-B and HPV Sabin. Inhibition by lysate containing HRV-A, or a mixture of HRV-B and HPV Sabin in 30 subjects with high total IgG1 titres to HRV-C3. Percentage of cross-reactivity to the inhibiting species was calculated by (Titre after absorption/Total IgG1)6100. P,0.0001 between inhibition by HRV-A and inhibition by HRV-B and HPV Sabin. Three negative control sera and a titration of reference sera were included on every plate to assess reproducibility and quantitation of IgG1 binding. doi:10.1371/journal.pone.0070552.g006 serum albumins and haemoglobins from different species can readily distinguish between antigens with 70% sequence identity Figure 6. Inhibition of IgG1 binding to HRV-C3 VP1 by HRV-A or HRV-B and HPV Sabin. Inhibition by lysate containing HRV-A, or a mixture of HRV-B and HPV Sabin in 30 subjects with high total IgG1 titres to HRV-C3. Percentage of cross-reactivity to the inhibiting species was calculated by (Titre after absorption/Total IgG1)6100. P,0.0001 between inhibition by HRV-A and inhibition by HRV-B and HPV Sabin. Discussion Most notably, while some people have high titres of species-specific antibodies to HRV-C, most do not, even though they have high The degree of cross-reactivity between the HRV-A and HRV-C is not only unusual because of its unidirectional bias but is different in magnitude to that observed for many antigens with a similar degree of sequence disparity. Hyperimmune sera produced to August 2013 | Volume 8 | Issue 8 | e70552 August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 8 Figure 9. Antibody responses to HRV-C VP1 with an N-terminal truncation in 30 adult subjects. (A) IgG1 binding (ng/ml) to truncated HRV-C3 VP1 expressed from residue 14 to 275, in comparison to full-length VP1. P,0.0001 between full-length and truncated HRV-C3 VP1. (B) Correlation of IgG1 binding (ng/ml) between full-length HRV-C3 VP1 pre-absorbed with HRV-A to remove cross-reactivity to HRV-A, and truncated HRV-C3 VP1. A reference sera and three negative sera that did not have IgG1 binding to HRV-C3, as determined by the immunoassays and immunoabsorptions, were included on every plate for quantitation of binding. doi:10.1371/journal.pone.0070552.g009 Antibody Binding to Human Rhinovirus Antigens Antibody Binding to Human Rhinovirus Antigens component of early immune response to HRV immunity and to be influenced by prior infection by all species. Supporting Information Figure S1 Example of a standard curve used for Figure 8. Intra-species correlation of IgG1 responses in 63 adult subjects. Log correlation of IgG1 binding (ng/ml) between two genotypes of (A) HRV-A (B) HRV-B and (C) HRV-C. doi:10.1371/journal.pone.0070552.g008 Figure 9. Antibody responses to HRV-C VP1 with an N-terminal truncation in 30 adult subjects. (A) IgG1 binding (ng/ml) to truncated HRV-C3 VP1 expressed from residue 14 to 275, in comparison to full-length VP1. P,0.0001 between full-length and truncated HRV-C3 VP1. (B) Correlation of IgG1 binding (ng/ml) between full-length HRV-C3 VP1 pre-absorbed with HRV-A to remove cross-reactivity to HRV-A, and truncated HRV-C3 VP1. A reference sera and three negative sera that did not have IgG1 binding to HRV-C3, as determined by the immunoassays and immunoabsorptions, were included on every plate for quantitation of binding. doi:10.1371/journal.pone.0070552.g009 y g g Figure 9. Antibody responses to HRV-C VP1 with an N-terminal truncation in 30 adult subjects. (A) IgG1 binding (ng/ml) to truncated HRV-C3 VP1 expressed from residue 14 to 275, in comparison to full-length VP1. P,0.0001 between full-length and truncated HRV-C3 VP1. Discussion (B) Correlation of IgG1 binding (ng/ml) between full-length HRV-C3 VP1 pre-absorbed with HRV-A to remove cross-reactivity to HRV-A, and truncated HRV-C3 VP1. A reference sera and three negative sera that did not have IgG1 binding to HRV-C3, as determined by the immunoassays and immunoabsorptions, were included on every plate for quantitation of binding. doi:10.1371/journal.pone.0070552.g009 component of early immune response to HRV immunity and to be influenced by prior infection by all species. component of early immune response to HRV immunity and to be influenced by prior infection by all species. Figure 8. Intra-species correlation of IgG1 responses in 63 adult subjects. Log correlation of IgG1 binding (ng/ml) between two genotypes of (A) HRV-A (B) HRV-B and (C) HRV-C. doi:10.1371/journal.pone.0070552.g008 References p 6. Hao W, Bernard K, Patel N, Ulbrandt N, Feng H, et al. (2012) Infection and Propagation of Human Rhinovirus C in Human Airway Epithelial Cells. J Virol 86: 13524–13532. 25. Hales BJ, Bosco A, Mills KL, Hazell LA, Loh R, et al. (2004) Isoforms of the major peanut allergen Ara h 2: IgE binding in children with peanut allergy. Int Arch Allergy Immunol 135: 101–107. 7. Granados A, Luinstra K, Chong S, Goodall E, Banh L, et al. (2012) Use of an improved quantitative polymerase chain reaction assay to determine differences in human rhinovirus viral loads in different populations. Diagn Microbiol Infect Dis 74: 384–387. 26. Hales BJ, Pearce LJ, Kusel MMH, Holt PG, Sly PD, et al. (2008) Differences in the antibody response to a mucosal bacterial antigen between allergic and non- allergic subjects. Thorax 63: 221–227. 8. Kistler AL, Webster DR, Rouskin S, Magrini V, Credle JJ, et al. (2007) Genome-wide diversity and selective pressure in the human rhinovirus. Virol J 4: 40. 27. Ichikawa K, Vailes L, Pomes A, Chapman M (2001) Molecular cloning, expression and modelling of cat allergen, cystatin (Fel d 3), a cysteine protease inhibitor. Clinical & Experimental Allergy 31: 1279–1286. 9. Barclay WS, Al-Nakib W, Higgins PG, Tyrrell DAJ (1989) The time course of the humoral immune response to rhinovirus infection. Epidemiol Infect 103: 659–669. 28. Brockwell D, Yu L, Cooper S, Mccleland S, Cooper A, et al. (2001) Physicochemical consequences of the perdeuteriation of glutathione S- transferase from S. japonicum. Protein Sci 10: 572–580. 10. Butler WT, Waldmann TA, Rossen RD, Douglas Jr RG, Couch RB (1970) Changes in IgA and IgG concentrations in nasal secretions prior to the appearance of antibody during viral respiratory infection in man. J Immunol 105: 584–591. 29. Kim S, Smith TJ, Chapman MS, Rossmann MG, Pevear DC, et al. (1989) Crystal structure of human rhinovirus serotype 1A (HRV1A). J Mol Biol 210: 91–111. 30. Zhang Y, Simpson AA, Ledford RM, Bator CM, Chakravarty S, et al. (2004) Structural and virological studies of the stages of virus replication that are affected by antirhinovirus compounds. J Virol 78: 11061–11069. 11. Van Kempen M, Bachert C, Van Cauwenberge P (1999) An update on the pathophysiology of rhinovirus upper respiratory tract infections. Rhinology 37: 97–103. 31. Aarntzen EH, de Vries IJM, Go¨ertz JH, Beldhuis-Valkis M, Brouwers HM, et al. References 1. Ma¨kela¨ MJ, Puhakka T, Ruuskanen O, Leinonen M, Saikku P, et al. (1998) Viruses and bacteria in the etiology of the common cold. J Clin Virol 36: 539– 542. 1. Ma¨kela¨ MJ, Puhakka T, Ruuskanen O, Leinonen M, Saikku P, et al. (1998) Viruses and bacteria in the etiology of the common cold. J Clin Virol 36: 539– 542. 19. Ledford RM, Patel NR, Demenczuk TM, Watanyar A, Herbertz T, et al. (2004) VP1 sequencing of all human rhinovirus serotypes: insights into genus phylogeny and susceptibility to antiviral capsid-binding compounds. J Virol 78: 3663–3674. 20. McErlean P, Shackelton LA, Lambert SB, Nissen MD, Sloots TP, et al. (2007) Characterisation of a newly identified human rhinovirus, HRV-QPM, discovered in infants with bronchiolitis. J Clin Virol 39: 67–75. 2. Lau SKP, Yip CCY, Tsoi H, Lee RA, So L, et al. (2007) Clinical Features and Complete Genome Characterization of a Distinct Human Rhinovirus (HRV) Genetic Cluster, Probably Representing a Previously Undetected HRV Species, HRV-C, Associated with Acute Respiratory Illness in Children. J Clin Microbiol 45: 3655–3664. 21. Hales BJ, Elliot CE, Chai LY, Pearce LJ, Tipayanon T, et al. (2013) Quantitation of IgE Binding to the Chitinase and Chitinase-Like House Dust Mite Allergens Der p 15 and Der p 18 Compared to the Major and Mid-Range Allergens. Int Arch Allergy Immunol 160: 233–240. 3. Fry AM, Lu X, Olsen SJ, Chittaganpitch M, Sawatwong P, et al. (2011) Human rhinovirus infections in rural Thailand: epidemiological evidence for rhinovirus as both pathogen and bystander. PLoS One 6: e17780. 22. Whitmore L, Wallace BA (2008) Protein secondary structure analyses from circular dichroism spectroscopy: Methods and reference databases. Biopolymers 89: 392–400. 4. Ferguson PE, Gilroy NM, Faux CE, Mackay IM, Sloots TP, et al. (2013) Human rhinovirus C in adult haematopoietic stem cell transplant recipients with respiratory illness. J Clin Virol 56: 339–343. 23. Lees JG, Miles AJ, Wien F, Wallace BA (2006) A reference database for circular dichroism spectroscopy covering fold and secondary structure space. Bioinfor- matics 22: 1955–1962. 5. Bizzintino J, Lee W, Laing I, Vang F, Pappas T, et al. (2011) Association between human rhinovirus C and severity of acute asthma in children. Eur Respir J 37: 1037–1042. 24. Hales BJ, Chai LY, Elliot CE, Pearce LJ, Zhang G, et al. (2012) Antibacterial antibody responses associated with the development of asthma in house dust mite-sensitised and non-sensitised children. Thorax 67: 321–327. References (2012) Humoral anti-KLH responses in cancer patients treated with dendritic cell-based immunotherapy are dictated by different vaccination parameters. Cancer Immunol Immunother 61: 2003–2011. 12. Stanekova´ Z, Vareckova E (2010) Conserved epitopes of influenza A virus inducing protective immunity and their prospects for universal vaccine development. Virol J 7: 351. 13. Wrammert J, Koutsonanos D, Li GM, Edupuganti S, Sui J, et al. (2011) Broadly cross-reactive antibodies dominate the human B cell response against 2009 pandemic H1N1 influenza virus infection. J Exp Med 208: 181–193. 32. Nabors GS, Braun PA, Herrmann DJ, Heise ML, Pyle DJ, et al. (2000) Immunization of healthy adults with a single recombinant pneumococcal surface protein A (PspA) variant stimulates broadly cross-reactive antibodies to heterologous PspA molecules. Vaccine 18: 1743–1754. 14. Weinfurter JT, Brunner K, Capuano SV, Li C, Broman KW, et al. (2011) Cross- reactive T cells are involved in rapid clearance of 2009 pandemic H1N1 influenza virus in nonhuman primates. PLoS Path 7: e1002381. 33. Vance E, George S, Guinan E, Wheeler C, Antin J, et al. (1998) Comparison of multiple immunization schedules for Haemophilus influenzae type b-conjugate and tetanus toxoid vaccines following bone marrow transplantation. Bone Marrow Transplant 22: 735–741. p 15. Kennedy JL, Turner RB, Braciale T, Heymann PW, Borish L (2012) Pathogenesis of rhinovirus infection. Curr Opin Virol 2: 287–293. 34. Semenova V, Steward-Clark E, Stamey K, Taylor T, Schmidt D, et al. (2004) Mass value assignment of total and subclass immunoglobulin G in a human standard anthrax reference serum. Clin Diagn Lab Immunol 11: 919–923. 16. Laza-Stanca V, Stanciu LA, Message SD, Edwards MR, Gern JE, et al. (2006) Rhinovirus replication in human macrophages induces NF-kappaB-dependent tumor necrosis factor alpha production. J Virol 80: 8248–8258. 35. McErlean P, Shackelton LA, Andrews E, Webster DR, Lambert SB, et al. (2008) Distinguishing molecular features and clinical characteristics of a putative new rhinovirus species, human rhinovirus C (HRV C). PLoS One 3: e1847. 17. Gern JE, Brooks GD, Meyer P, Chang A, Shen K, et al. (2006) Bidirectional interactions between viral respiratory illnesses and cytokine responses in the first year of life. J Allergy Clin Immunol 117: 72–78. 36. Lau SKP, Yip CCY, Lin AWC, Lee RA, So LY, et al. (2009) Clinical and molecular epidemiology of human rhinovirus C in children and adults in Hong Kong reveals a possible distinct human rhinovirus C subgroup. J Infect Dis 200: 1096–1103. 18. Acknowledgments We would like to acknowledge the participants who took part in the study, and L.A. Hazell and T.L.Y. Chai for assistance with protein antigen preparation. Figure S2 Circular dichroism (CD) analysis of recom- binant HRV-C VP1 protein. (A) CD spectrum of GST-fusion HRV-C3 VP1. (B) CD spectrum of HRV-C3 VP1 following subtractive CD analysis in which the GST control was subtracted from the fusion protein. The diagrams represent the ultraviolet spectra of the purified recombinant proteins analysed using CD spectroscopy in the range 260–190 nm. Structural analysis of the data was performed using DiChroWeb Server: CDSSTR algorithm, reference set 4. Author Contributions Conceived and designed the experiments: JI WS SRS WRT BJH. Performed the experiments: JI SRS. Analyzed the data: JI SRS WRT BJH. Conceived and designed the experiments: JI WS SRS WRT BJH. Performed the experiments: JI SRS. Analyzed the data: JI SRS WRT BJH. Contributed reagents/materials/analysis tools: WRT SRS. Wrote the paper: JI WS SRS WRT BJH. Contributed reagents/materials/analysis tools: WRT SRS. Wrote the paper: JI WS SRS WRT BJH. Supporting Information Figure S1 Example of a standard curve used for quantitation of IgG1 binding. Each assay was calibrated by interpolating the results from a titration curve constructed with recombinant GST-fusion Der p 2 and a standardised humanised anti-Der p 2 chimeric IgG1. The linear section of the Der p 2 chimeric curve used for analysis had a slope = 5.0060.07 and is indicated in red. Equivalent concentrations of recombinant VP1 antigens was used to coat wells and bound with a titration of reference sera (example of HRV-A34 shown). The 2-fold dilutions used are indicated. The linear section of the reference titration curve (indicated in red) had a slope = 4.9560.69, comparable to the Der p 2 chimeric curve. A titration curve of reference sera was included on every plate to construct a standard curve, which was then used to convert europium counts to absolute IgG1 binding titres of cross-reactive antibodies mainly to HRV-A, that bind the HRV-C. Further investigation of these responses with clinically defined subjects and paediatric populations in conjunction with virus detection is thus warranted. The antibodies might be used as a marker of CD4 T-cell responses and be extended in the examination of other antibody isotypes. They might also have critical functions in the immunopathology produced by these viruses where for example they could increase the involvement of cytokine-producing monocytes and other inflammatory cells, as well as antigen presentation. As shown for other virus infections [12–14], non-neutralising antibodies also play a key role in immunity and they would be expected to be quantitatively a large PLOS ONE | www.plosone.org August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 9 Antibody Binding to Human Rhinovirus Antigens (ng/ml). The mean europium count of the negative control, human myeloma IgG1, is indicated. For HRV-A34, the mean and standard deviation (SD) europium count for human myeloma IgG1 was 25726795 (n = 3), which was calculated to be 499 ng/ ml. Negative sera was determined using mean +3SD of human myeloma IgG1 binding for each antigen. (TIF) (TIF) Table S1 Nucleotide and protein sequences of HRV and HPV Sabin VP1 used in this study. (DOCX) 42. Weghofer M, Dall’Antonia Y, Grote M, Sto¨cklinger A, Kneidinger M, et al. (2008) Characterization of Der p 21, a new important allergen derived from the gut of house dust mites*. Allergy 63: 758–767. 41. Ro¨hm C, Zhou N, Su¨ss J, Mackenzie J, Webster R (1996) Characterization of a novel influenza hemagglutinin, H15: criteria for determination of influenza A subtypes. Virology 217: 508–516. References Rossmann MG, Arnold E, Erickson JW, Frankenberger EA, Griffith JP, et al. (1985) Structure of a human common cold virus and functional relationship to other picornaviruses. Nature 317: 145–153. August 2013 | Volume 8 | Issue 8 | e70552 August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 10 Antibody Binding to Human Rhinovirus Antigens Antibody Binding to Human Rhinovirus Antigens Antibody Binding to Human Rhinovirus Antigens 37. Kim JH, Skountzou I, Compans R, Jacob J (2009) Original antigenic sin responses to influenza viruses. J Immunol 183: 3294–3301. 38. McIntyre CL, McWilliam Leitch EC, Savolainen-Kopra C, Hovi T, Simmonds P (2010) Analysis of genetic diversity and sites of recombination in human rhinovirus species C. J Virol 84: 10297–10310. yp gy 42. Weghofer M, Dall’Antonia Y, Grote M, Sto¨cklinger A, Kneidinger M, et al. (2008) Characterization of Der p 21, a new important allergen derived from the gut of house dust mites*. Allergy 63: 758–767. p J 39. Bochkov YA, Palmenberg AC, Lee WM, Rathe JA, Amineva SP, et al. (2011) Molecular modeling, organ culture and reverse genetics for a newly identified human rhinovirus C. Nat Med 17: 627–632. g gy 43. Giranda VL, Heinz BA, Oliveira MA, Minor I, Kim KH, et al. (1992) Acid- induced structural changes in human rhinovirus 14: possible role in uncoating. Proc Natl Acad Sci U S A 89: 10213–10217. 40. Ashraf S, Brockman-Schneider R, Bochkov YA, Pasic TR, Gern JE (2012) Biological characteristics and propagation of human rhinovirus-C in differen- tiated sinus epithelial cells. Virology 436: 143–149. 44. Niespodziana K, Napora K, Cabauatan C, Focke-Tejkl M, Keller W, et al. (2012) Misdirected antibody responses against an N-terminal epitope on human rhinovirus VP1 as explanation for recurrent RV infections. FASEB J 26: 1001– 1008. PLOS ONE | www.plosone.org August 2013 | Volume 8 | Issue 8 | e70552 PLOS ONE | www.plosone.org 11 11

https://openalex.org/W4376617277

https://figshare.com/articles/journal_contribution/Figure_S3_from_Immunologic_Predictors_for_Clinical_Responses_during_Immune_Checkpoint_Blockade_in_Patients_with_Myelodysplastic_Syndromes/22353771/1/files/39778296.pdf

English

Figure S2 from Immunologic Predictors for Clinical Responses during Immune Checkpoint Blockade in Patients with Myelodysplastic Syndromes

cc-by

Supplementary Figure 3 Supplementary Figure 3 Supplementary Figure 3 pplementary Figure 3 Figure S3. Flow-cytometry gating strategy. CD, cluster of differentiation, Treg, regulatory T cell Figure S3 Flow cytometry gating strategy CD cluster of differentiation Treg regulatory Figure S3. Flow-cytometry gating strategy. CD, cluster of differentiation, Treg, regulatory T cell Figure S3. Flow-cytometry gating strategy. CD, cluster of differentiation, Treg, regulatory T cell T cell

https://openalex.org/W2466955766

https://www.eurosurveillance.org/deliver/fulltext/eurosurveillance/13/36/art18969-en.pdf?itemId=%2Fcontent%2F10.2807%2Fese.13.36.18969-en&mimeType=pdf&containerItemId=content/eurosurveillance

English

Principles and uses of HIV incidence estimation from recent infection testing - a review

Euro surveillance/Eurosurveillance

cc-by

Introduction Estimating HIV incidence, the number of new infections during a time period, is critically important for assessing the dynamics of human immunodeficiency virus (HIV) transmission and evaluating the impact of prevention policies. A conceptual improvement in surveillance methods has been made in the past ten years to make incidence estimation more feasible. By using a biomarker measurement to identify seropositive individuals who have recently been infected, incidence estimates can be obtained from a single specimen. This laboratory-based method can take advantage of the collection of specimen intended to assess prevalence (the proportion or number of persons cumulatively infected at a given time) and to obtain valid incidence data without the expensive and logistically complex requirement of following a cohort of uninfected individuals over time. However, as for other methods based on repeated prevalence data and mathematical modelling, the use of biomarkers to estimate incidence requires a substantial number of assumptions, some being difficult to assess, and an appropriate definition of the population the incidence is estimated for. Within the range of methods to identify early infection through virological markers before seroconversion, testing of pooled HIV RNA now seems to be the most appropriate approach because RNA can be detected earlier than p24 antigen, which allows characterisation of a longer time period (Figure 1). Moreover, pooling of specimens improves the predictive value of the amplification assays and substantially lowers the costs. However, in order to obtain accurate incidence estimates, this method requires the inclusion of very large sample populations, such as those provided by blood donations [2] or by the large testing programme in the United States (US) described by Pilcher et al. [3]. In this article based on the literature, we attempt to give an overview of the methods that allow estimating HIV incidence based on biomarker detection at the early stage of infection. After defining the principles, we review some typical uses of serological incidence assays and the challenges for each type of application. Principles Since the 1990s, the development of laboratory-based methods has allowed to estimate incidence of human immunodeficiency virus (HIV) infections on single samples. The tests aim to differentiate recent from established HIV infection. Incidence estimates are obtained by using the relationship between prevalence, incidence and duration of recent infection. We describe the principle of the methods and typical uses of these tests to characterise recent infection and derive incidence. We discuss the challenges in interpreting estimates and we consider the implications for surveillance systems. P r i n c i p l e s a n d u s e s o f H IV i n c i d e n c e e s t i m at i o n f r o m r e c e n t i n f e c t i o n t e s t i n g - a r e v i e w S Le Vu ([email protected])1, J Pillonel1, Caroline Semaille1, P Bernillon1, Y Le Strat1, L Meyer2, J C Desenclos1 1. Department of Infectious Diseases, HIV/AIDS-STI-HCV Unit, Institut de veille sanitaire (French Institute for Public Health Surveillance, InVS), Saint-Maurice, France 2. Department of Epidemiology, Institut national de la santé et de la recherche médicale/Institut national d’études démographiques (National Institute of Health and Medical Research/National Institute for Demographic Studies, INSERM/INED/ Paris XI U569), Le Kremlin-Bicêtre, France Principles Incidence based on detection of virological markers before seroconversion Incidence based on detection of virological markers before seroconversion Incidence based on detection of virological markers before seroconversion In 1995, Brookmeyer and Quinn introduced a simple approach for estimating HIV incidence from a cross-sectional survey [1]. They used a two-step algorithm combining diagnostic tests for the p24 antigen and HIV-1 antibodies to determine the prevalence of p24 antigenaemia among antibody-negative individuals (Figure 1). The HIV incidence rate was then calculated by using the classical epidemiologic relation between prevalence, incidence, and duration of the period between the onset of detectability of p24 and the first HIV antibodies. Overall, these methods can add remarkable value to surveillance systems based on prevalence surveys as well as HIV case reporting. The assumptions that must be fulfilled to correctly interpret the estimates are mostly similar to those required in prevalence measurement. However, further research on the specific aspect of window period estimation is needed in order to generalise these methods in various population settings. The disadvantage of this approach was that the time during which p24 antigen is detectable prior to seroconversion is short (the mean duration of this period was 22.5 days in 1995 and has become shorter since then due to the development of new diagnostic assays that allow to detect antibodies earlier [2]). The first consequence of this is that the estimation of this period comes with a considerable uncertainty which can have a large impact on the incidence estimate. The second consequence is that large samples and/or high HIV incidence are required to identify a sufficient number of individuals with detectable p24 antigen who have not seroconverted. Nevertheless, Brookmeyer and Quinn provided the conceptual framework for subsequent laboratory-based methods to estimate incidence from single cross-sectional surveys. Serologic incidence assaysi Janssen et al. were the first to describe in 1998 an approach based on a test specifically developed for the purpose of estimating incidence [4]. This approach named “Serologic testing algorithm EUROSURVEILLANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveillance.org 11 for recent HIV seroconversion (STARHS)” aimed at detecting a transient state reached after the antibody conversion. It thus offered the advantage of testing only positive individuals and defining a period sufficiently short to fulfil the requirements of stationarity of the incidence over the study period, while sufficiently long to minimise the inaccuracy in its estimation. The work of Janssen et al. can be considered as a milestone for the concept of serological methods for the estimation of HIV incidence. for recent HIV seroconversion (STARHS)” aimed at detecting a transient state reached after the antibody conversion. It thus offered the advantage of testing only positive individuals and defining a period sufficiently short to fulfil the requirements of stationarity of the incidence over the study period, while sufficiently long to minimise the inaccuracy in its estimation. The work of Janssen et al. can be considered as a milestone for the concept of serological methods for the estimation of HIV incidence. can be approximated by the prevalence [13]. And the relation is valid for a stationary population with a constant level of incidence during the study period [1]. In Figure 2, we present an example of an incidence calculation using the formula developed by Janssen et al. with a window period of 180 days [4]. Various adjustments have been made to Janssen’s formula in order to correctly express the number of people at risk and to account for misclassification of long-term infections. The first adjustment consisted in varying the assumed number of people at risk of having had a recent HIV infection during one year. As in the estimation of incidence in a cohort, HIV-negative individuals are considered at risk during the whole period, while infected individuals can be considered at risk during half a year on average [14]. Following the same principle, various applications of laboratory- based incidence estimation from cross-sectional population surveys have been described and a growing number of assays have been developed (see the article of Parry et al. in this issue).These assays measure the immunological response against the virus, based on specific HIV antibody concentration [4-6], proportion [7], isotype [8] or avidity [9]. Serologic incidence assaysi This measure should define a transient state from the onset of detectability by a standard HIV screening test to the cut-off value defining the “established” infection status of the test for recent infection (Figure 1). This period is called the window period. Because of the individual variability in antibody response, window periods may differ widely from person to person. Their mean duration is measured in advance by testing serial specimens from infected individuals with known dates of seroconversion [10]. The STARHS methods have been compared to classical incidence measurements obtained in cohorts to assess their validity [4,11,12]. Provided that the compared estimates are not affected by population sampling bias, the estimates are reported to be similar [10,12]. In addition, concerns have been expressed that the mean window period for the BED capture enzyme immunoassay (BED- CEIA) does not properly take into account people who have a very long individual window period and can be falsely classified as F i g u r e 2 Relation between HIV prevalence, recent infection rate, and the incidence estimation in a cross-sectional survey A constant incidence rate of 0.4% persons/year is observed in a population of 1,000 individuals seronegative from the beginning of year 2005. Prevalence, incidence and rate of recent infection are estimated cross-sectionally at the end of 2005, 2006 and 2007. The number of HIV-positive inviduals includes those with recent infection, tested within window period (NR), and those with established infection, tested after the window period (NE), represented respectively in light blue and dark blue in the figure. HIV-negative individuals (Nneg) are represented in grey. While incidence estimates are nearly constant over the years, the recent infection rate, being influenced by the prevalence of established infection, is decreasing. Serologic incidence assaysi HIV- HIV+ in WP HIV+ 1000 999 998 997 996 995 994 993 992 991 990 989 988 987 986 985 3 2 1 2005 2006 2007 2008 1,000 individuals Prevalence=0.40% Incidence=0.41% RI rate=50% Prevalence=0.81% Incidence=0.41% RI rate=25% Prevalence=1.21% Incidence=0.41% RI rate=17% F i g u r e 2 Relation between HIV prevalence, recent infection rate, and the incidence estimation in a cross-sectional survey HIV- HIV+ in WP HIV+ 1000 999 998 997 996 995 994 993 992 991 990 989 988 987 986 985 3 2 1 2005 2006 2007 2008 1,000 individuals Prevalence=0.40% Incidence=0.41% RI rate=50% Prevalence=0.81% Incidence=0.41% RI rate=25% Prevalence=1.21% Incidence=0.41% RI rate=17% Relation between HIV prevalence, recent infection rate, and the incidence estimation in a cross-sectional survey 1,000 individuals F i g u r e 1 Kinetics of virological markers and host immune response used to define transient states in the early phase of HIV infection Kinetics of virological markers and host immune response used to define transient states in the early phase of HIV infection 0 50 100 150 200 250 Days since infection Antigen/antibodies/RNA quantitative measure Detectability of antigen/antibody/RNA Cut-off threshold for recent infection Antibodies to HIV envelope protein p24 antigen Seroconversion RNA 3 1 2 1: RNA-to-seroconversion transient state as defined by Busch et al., 2005 [2] 2: p24-to-seroconversion transient state as defined by Brookmeyer et al., 1995 [1] 3: Antibody-based mean window period as defined by Janssen et al., 1998 [4] A constant incidence rate of 0.4% persons/year is observed in a population of 1,000 individuals seronegative from the beginning of year 2005. Prevalence, incidence and rate of recent infection are estimated cross-sectionally at the end of 2005, 2006 and 2007. The number of HIV-positive inviduals includes those with recent infection, tested within window period (NR), and those with established infection, tested after the window period (NE), represented respectively in light blue and dark blue in the figure. HIV-negative individuals (Nneg) are represented in grey. While incidence estimates are nearly constant over the years, the recent infection rate, being influenced by the prevalence of established infection, is decreasing. Antigen/antibodies/RNA quantitative measure 0 50 100 150 200 250 Days since infection Antigen/antibodies/RNA quantitative measure Detectability of antigen/antibody/RNA Cut-off threshold for recent infection Antibodies to HIV envelope protein p24 antigen Seroconversion RNA 3 1 2 Antigen/antibodies/RNA quantitative measure Antibodies to HIV envelope protein Incidence estimation Th i id i The incidence estimation is calculated as the frequency of the transient state (i.e. the prevalence of recent infection) divided by its duration (the mean window period). As stated above, this calculation is based on the relation “prevalence = incidence * mean duration”. This relation assumes that the condition, in our context “recent HIV infection”, is a rare event so that the prevalence odds Applications Whil While a comprehensive review of applications for serological incidence assays is beyond the scope of this paper, the purpose of this chapter is to point out typical settings in which they may be used. On the other hand, some assays have been developed for the specific purpose of classifying infections in individual patients as recent or established with given predictive criteria. This is the case for the enzyme immunoassay for recent HIV-1 infections (EIA-RI) developed by Barin et al. [6]. This assay uses a logistic regression classification algorithm in which the cut-off was chosen to detect individuals infected for less than 180 days with a enhanced focus on the level of specificity of detection It is to be noted that a lack of specificity, because it affects the population of established infections that is generally larger, should have a wider impact on misclassification than a lack of sensitivity, considering the low prevalence of recent infection status [30]. On-going development of the EIA-RI test aims to re-calibrate it for the purpose of incidence derivation. Identifying recent infection recent. This issue is probably a general one, affecting all the tests that have been calibrated using a disproportionate number of short term infections (for less than one year). It should have an impact on incidence estimation since the cross-sectional populations on which the method is to be applied are expected to contain a larger number of long-term infections. Two adjustments have been proposed to correct this issue about the specificity [15]. They share the principle of applying a corrective factor in the incidence formula to compensate for the false recent cases due to very long window period. Other algorithms have been proposed that, rather than correcting the formula, combine two incidence assays in order to avoid misclassification [12,16]. A particular use of incidence assays is identifying recent infection status per se, for individual patient management such as contact tracing or assessment of primary resistance. It is helpful to bear in mind that characterisation of recent infection was initially a by- product in the method described by Janssen et al. which considered incidence derivation as the main outcome. In particular, the use of the mean value of an incidence assay window period assumes that individual window periods are variable and that a certain number of individuals in a given population will have a window period shorter or longer than the mean. Consequently, some misclassifications of established infection (false positives) and of recent infection (false negatives) are to be expected. For the purpose of incidence estimation, the respective misclassifications are supposed to cancel each other out, so that the number of recent infection at a population level is correctly estimated. At the level of individual patients, however, this could lead to serious misinterpretation. Estimates are calculated as follows [4]: Estimates are calculated as follows [4]: E N R N neg N E N R N Prevalence + + + = E N R N R N rate infection Recent WP mean 365 R N neg N R N Incidence + = × + = WP mean 365 R N neg N R N Incidence × + = This illustration was inspired by the presentation of Ruigang Song “Modeling HIV Testing Behavior and Its Impact on Incidence Estimation” at the 15th International AIDS Conference, July 15, 2004, Bangkok, Thailand. 1: RNA-to-seroconversion transient state as defined by Busch et al., 2005 [2] 2: p24-to-seroconversion transient state as defined by Brookmeyer et al., 1995 [1] 3: Antibody-based mean window period as defined by Janssen et al., 1998 [4] 1: RNA-to-seroconversion transient state as defined by Busch et al., 2005 [2] 2: p24-to-seroconversion transient state as defined by Brookmeyer et al., 1995 [1] 3: Antibody-based mean window period as defined by Janssen et al., 1998 [4] WP = Window period; RI = Recent infection. 1 2 Typical applications The most common context in which incidence assays are used are prevalence sero-surveys. Some were dedicated to incidence estimation, but the majority were set up to observe the recent infection status of stored HIV-positive serum specimens. Numerous serial cross-sectional surveys have been applied in the setting of testing for HIV or other sexually transmitted diseases in countries such as the US [17-19], some European countries [20;21] or Brazil [22]. In these studies, temporal trends in incidence rate could be derived and helped to assess retrospectively epidemic phenomena among high-risk subgroups. But concerns about representativeness and selection bias can be raised about such voluntary testing sites (as reviewed below in the section ‘Issues’). Numerous serial cross-sectional surveys have been applied in the setting of testing for HIV or other sexually transmitted diseases in countries such as the US [17-19], some European countries [20;21] or Brazil [22]. In these studies, temporal trends in incidence rate could be derived and helped to assess retrospectively epidemic phenomena among high-risk subgroups. But concerns about representativeness and selection bias can be raised about such voluntary testing sites (as reviewed below in the section ‘Issues’). Similarly, already existing sentinel surveillance systems have provided insight into underlying trends in transmission in particular risk groups. Specimens gathered at enrolment in syringe exchange programmes or serial street surveys allowed the estimation of trends in HIV incidence among intravenous drug users in New York City, US [23] and San Francisco, US [24] over a long period. Similarly, already existing sentinel surveillance systems have provided insight into underlying trends in transmission in particular risk groups. Specimens gathered at enrolment in syringe exchange programmes or serial street surveys allowed the estimation of trends in HIV incidence among intravenous drug users in New York City, US [23] and San Francisco, US [24] over a long period. Expressing the proportion of recent infectioni p g p p Some applications define the proportion of recent infection in a population of positive individuals as an outcome. This is the way Puchhammer et al. analysed the results of the avidity assay among new diagnoses from case-reporting in Austria [32]. This is also the way that correlates of recent infection among new diagnoses are interpreted in France [33] (see also the article by Semaille et al. in this issue). However, this quantity that is somehow related to incidence depends also on the prevalence of non-recent infection and thus can not be considered as a good proxy for incidence. In fact, in the context of diagnostic testing, the proportion of recent infection has a lot to do with the testing framework capacity as well as the incidence rate in the population. Since the prevalence of undiagnosed infection affects the proportion of recent infection independently of any change in incidence (Figure 2), such results are difficult to interpret. For purposes of precision and as done for prevalence estimation, targeting a more general population than particular high-risk groups requires testing a very large number of people or setting the study in a country with a high incidence level. At least one of these conditions was met in studies that estimated the HIV incidence by means of recent infection testing in antenatal screening programmes in Cambodia [25], South Africa [26], the US [27] and Brazil [28], in screening programmes for blood donation in the US [2;4], France [29] and the Ivory Coast [30], and a national household survey in South Africa [31]. In all these settings, specimens are collected routinely and can be tested for recent infection retrospectively or prospectively. Some demographic and behavioural data on the targeted population are usually collected along with the specimens, both for positive and negative individuals. Taking advantage of specimens from prevalence serosurveys allows to derive incidence data for these populations with only minor expenses in terms of cost and logistics. Issues There are issues that pertain to the estimation HIV incidence by characterising recent infections. We can distinguish issues that are related to the determination of recent HIV infection from those that affect the validity of incidence estimation. Nevertheless, it can be argued that every design of an incidence study suffers from some kind of selection bias, even longitudinal studies [11]. Moreover, studying the level of the infection among the attendees of testing sites can still provide insights over time, especially in conjunction with behavioural data. Incidence estimation from HIV case-reporting data While it seems especially promising to take advantage of recent infection testing among reported HIV diagnoses at province or country level, there are several specific difficulties with regards to deriving a valid incidence measurement. Unlike cross-sectional surveys, a case-reporting system collects information only for individuals with positive test results and generally can not provide information on those who were negative. Therefore, the denominator of the formula, i. e. the number of people at risk, is not available. Another approach is needed to derive an incidence that can be generalised for the population targeted by the surveillance, and to take account of the fact that negative test results are not reported. In certain contexts, the most obvious added value of the incidence assays approach is that the incidence could not have been estimated by any other means. This is what happens when no accurate data on prior testing or exposure period can be obtained such as for the population of blood donors screened during their first donation [29]. EUROSURVEILLANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveillance.org 13 sex, mode of contamination, testing patterns, and possibly virus subtypes must be gathered along with tests results. Such an approach has been described by Lee et al. for the estimation of the national HIV incidence in the US [34]. The statistical framework considers the reported cases identified as recently infected as a sample selected from all annual new cases, with a probability of inclusion related to their testing pattern. According to this probability, each case identified as recently infected is assigned a weight, and the sum of weights provides the incidence count. This approach represents a good opportunity to improve large scale surveillance of HIV dynamics, especially where a framework of HIV case reporting already exists and can provide data on testing patterns. Limitations in determining recent infectioni The first issues are due to the limitations of the assays in detecting recent HIV infection. As the majority of assays are based on quantitative measurement of the antibody response, factors that affect the patient’s immune response lead to some misclassification. Qualitative assays such as the avidity assay may be affected to a lesser extent [36]. Even more problematic seems to be the issue of a selection bias occurring if recently infected people tended to seek testing sooner than expected because of seroconversion illness or identified recent exposure. This leads to an increase in the number of detected recent infections and an overestimation of the incidence. Remis et al. refer to this bias as the “seroconversion effect” and proposed a way to measuring it by making different incidence estimates based on varying window periods [40]. Song et al. formulated the hypothesis of independence between testing and the occurrence of infection and proposed a procedure to test this hypothesis [41]. All these biases can be found when inferring HIV incidence from case-reporting of new diagnoses which also include individuals seeking testing or health care. Firstly, people with acquired immunodeficiency syndrome (AIDS) may falsely be identified as recently infected due to declining antibody levels. The same appears to be true in some individuals in the late stage of non-AIDS HIV infection. As for the AIDS stage, clinical data or CD4+ T-cell counts would need to be collected in order to exclude these patients from the calculation and avoid overestimation. A correction for misclassification due to late-stage non-responsive patients, has been proposed by Mc Dougal et al. and Hargrove et al. [15]. Finally, as it is not always possible to test the whole positive study population for recent infection, the proportion of recent infection obtained among those tested is classically assigned to those for whom a test result is not available. This extrapolation assumes that the availability of specimens for recent infection testing is randomly determined in the population. Secondly, antiretroviral drugs affect the antibody level by decreasing the viral load [37]. Again, to correctly assess recent infection, patients with ongoing treatment need to be identified and excluded by gathering declarative information (from clinician or patient) or alternatively by detecting drugs in serum specimens by, for example, mass-spectrometry. Representativeness and selection bias A general issue of incidence estimation arises from the fact that the populations tested are not randomly selected and may not be representative of the populations at risk of infection. This is particularly the case in the context of HIV testing or sexually transmitted diseases clinics. The bias may go in either direction. People at high risk may seek testing more frequently with the consequence of raising the incidence estimation. On the other hand, people attending HIV testing settings as part of a prevention strategy might be at lower risk than people who do not do a test because they do not recognise the risk or are afraid of a positive result. Finally, another approach has been described to bypass the issue that only positive individuals are reported to the surveillance system. In Ontario, Canada, an enhanced surveillance system has been established that requires diagnostic laboratories to collect information (number and risk factor) on a random subset of individuals with a negative test result in parallel to the information on those that were positive [35]. This system then allows the use of the Janssen’s formula to derive the incidence in different risk groups. Schoenbach et al. raised this issue in 2001 and questioned the rationale of inferring HIV incidence in testing settings and in particular, whether it is possible to extrapolate these incidence estimates to a larger population [39]. With regard to generalising incidence, it may be preferable to collect specimens from surveillance settings such as blood donation facilities or antenatal clinics where people are not self-selected but tested in a systematic manner, and where large sample size can be obtained. References Development and validation of an immunoassay for identification of recent human immunodeficiency virus type 1 infections and its use on dried serum spots. J Clin Microbiol. 2005;43(9):4441-7. 28. de Freitas Oliveira CA, Ueda M, Yamashiro R, Rodrigues R, Sheppard HW, Macedo Brigido LF. Rate and incidence estimates of recent human immunodeficiency virus type 1 infections among pregnant women in Sao Paulo, Brazil, from 1991 to 2002. J Clin Microbiol. 2005;43(3):1439-42. 7. Parekh BS, Kennedy MS, Dobbs T, Pau CP, Byers R, Green T, et al. Quantitative detection of increasing HIV type 1 antibodies after seroconversion: a simple assay for detecting recent HIV infection and estimating incidence. AIDS Res Hum Retroviruses. 2002;18(4):295-307. 29. Pillonel J, Barin F, Laperche S, Bernillon P, Le Vu S, Brunet S, et al. HIV-1 Incidence among Blood Donors in France, 1992 to 2006: Use of an Immunoassay to Identify Recent Infections. Transfusion 2008;48(8):1567-75. 8. Wilson KM, Johnson EI, Croom HA, Richards KM, Doughty L, Cunningham PH, et al. Incidence immunoassay for distinguishing recent from established HIV-1 infection in therapy-naive populations. AIDS. 2004;18(17):2253-9. 30. Sakarovitch C, Rouet F, Murphy G, Minga AK, Alioum A, Dabis F, et al. Do tests devised to detect recent HIV-1 infection provide reliable estimates of incidence in Africa? JAIDS. 2007;45(1):115-22. 9. Suligoi B, Galli C, Massi M, Di Sora F, Sciandra M, Pezzotti P, et al. Precision and accuracy of a procedure for detecting recent human immunodeficiency virus infections by calculating the antibody avidity index by an automated immunoassay-based method. J Clin Microbiol. 2002;40(11):4015-20. 31. Rehle T, Shisana O, Pillay V, Zuma K, Puren A, Parker W. National HIV incidence measures--new insights into the South African epidemic. S Afr Med J. 2007;97(3):194-9. 10. McDougal JS, Pilcher CD, Parekh BS, Gershy-Damet G, Branson BM, Marsh K, et al. Surveillance for HIV-1 incidence using tests for recent infection in resource-constrained countries. AIDS. 2005;19 Suppl 2:S25-30. 32. Puchhammer-Stockl E, Schmied B, Rieger A, Sarcletti M, Geit M, Zangerle R, et al. Low proportion of recent human immunodeficiency virus (HIV) infections among newly diagnosed cases of HIV infection as shown by the presence of HIV-specific antibodies of low avidity. J Clin Microbiol. 2005;43(1):497-8. 11. Hu DJ, Vanichseni S, Mock PA, Young NL, Dobbs T, Byers RH Jr, et al. References 22. Alves K, Shafer KP, Caseiro M, Rutherford G, Falcao ME, Sucupira MC, et al. Risk factors for incident HIV infection among anonymous HIV testing site clients in Santos, Brazil: 1996-1999. J Acquir Immune Defic Syndr. 2003;32(5):551-9. 1. Brookmeyer R, Quinn TC. Estimation of current human immunodeficiency virus incidence rates from a cross-sectional survey using early diagnostic tests. Am J Epidemiol. 1995;141(2):166-72. 23. Des Jarlais DC, Perlis T, Arasteh K, Torian LV, Beatrice S, Milliken J, et al. HIV Incidence Among Injection Drug Users in New York City, 1990 to 2002: Use of Serologic Test Algorithm to Assess Expansion of HIV Prevention Services. Am J Public Health. 2005;95(8):1439-44. 2. Busch MP, Glynn SA, Stramer SL, Strong DM, Caglioti S, Wright DJ, et al. A new strategy for estimating risks of transfusion-transmitted viral infections based on rates of detection of recently infected donors. Transfusion. 2005;45(2):254-64. 24. Kral AH, Lorvick J, Gee L, Bacchetti P, Rawal B, Busch M, et al. Trends in human immunodeficiency virus seroincidence among street-recruited injection drug users in San Francisco, 1987-1998. Am J Epidemiol. 2003;157(10):915-22. 3. Pilcher CD, Fiscus SA, Nguyen TQ, Foust E, Wolf L, Williams D, et al. Detection of acute infections during HIV testing in North Carolina. N Engl J Med. 2005;352(18):1873-83. 25. Saphonn V, Parekh BS, Dobbs T, Mean C, Bun LH, Ly SP, et al. Trends of HIV-1 seroincidence among HIV-1 sentinel surveillance groups in Cambodia, 1999-2002. J Acquir Immune Defic Syndr. 2005;39(5):587-92. 4. Janssen RS, Satten GA, Stramer SL, Rawal BD, O’Brien TR, Weiblen BJ, et al. New testing strategy to detect early HIV-1 infection for use in incidence estimates and for clinical and prevention purposes. JAMA. 1998;280(1):42-8. 26. Gouws E, Williams BG, Sheppard HW, Enge B, Karim SA. High incidence of HIV-1 in South Africa using a standardized algorithm for recent HIV seroconversion. J Acquir Immune Defic Syndr. 2002;29(5):531-5. 5. Rawal BD, Degula A, Lebedeva L, Janssen RS, Hecht FM, Sheppard HW, et al. Development of a new less-sensitive enzyme immunoassay for detection of early HIV-1 infection. J Acquir Immune Defic Syndr. 2003;33(3):349-55. 27. Nesheim S, Parekh B, Sullivan K, Bulterys M, Dobbs T, Lindsay M, et al. Temporal trends in HIV Type 1 incidence among inner-city childbearing women in Atlanta: use of the IgG-capture BED-enzyme immunoassay. AIDS Res Hum Retroviruses. 2005;21(6):537-44. 6. Barin F, Meyer L, Lancar R, Deveau C, Gharib M, Laporte A, et al. Acknowledgements Acknowledgements We thank Dr Michael P. Busch for his helpful suggestions on the literature review, Dr Robert S. Remis for his valuable comments on the draft version. 20. Murphy G, Charlett A, Jordan LF, Osner N, Gill ON, Parry JV. HIV incidence appears constant in men who have sex with men despite widespread use of effective antiretroviral therapy. AIDS. 2004;18(2):265-72. 21. Dukers NH, Spaargaren J, Geskus RB, Beijnen J, Coutinho RA, Fennema HS. HIV incidence on the increase among homosexual men attending an Amsterdam sexually transmitted disease clinic: using a novel approach for detecting recent infections. AIDS. 2002;16(10):F19-24. Conclusion Thirdly, test results are affected by the virus subtype and/or the patient’s genetic background. It has been shown that all tests that have been developed mainly on specimens from patients infected with subtype B viruses give inconsistent results when used for infections with non-B subtypes. Therefore, an assessment of the test properties (cut-off and window period) in different population settings is needed before applying any method [30]. Overall, the use of laboratory-based methods to estimate HIV incidence can add remarkable value to surveillance systems based on prevalence surveys or on HIV case reporting. The estimation of HIV incidence provides a clear public health benefit in that it allows better monitoring of HIV transmission and targeting of preventive initiatives. We have seen that the application of those methods in cross-sectional settings have been well described in terms of incidence estimation and limitations, one of the most important limitations being the lack of representativeness. The assumptions that must be fulfilled to correctly interpret the estimates are to a We have seen how the correct interpretation of test results relies on the availability of clinical data that characterise the population [38]. In order to further interpret incidence estimates, data on EUROSURVEILLANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveillance.org 14 17. Schwarcz S, Kellogg T, McFarland W, Louie B, Kohn R, Busch M, et al. Differences in the temporal trends of HIV seroincidence and seroprevalence among sexually transmitted disease clinic patients, 1989-1998: application of the serologic testing algorithm for recent HIV seroconversion. Am J Epidemiol. 2001;153(10):925-34. large extent similar to those required in prevalence measurement. However, further research on the more specific aspect of window period estimation may be needed in order to generalise these methods. In particular, efforts are needed to correctly define the mean window periods for different virus subtypes and stages of infection so that the essential relation between prevalence and incidence holds true in various population settings. 18. Kellogg TA, Loeb L, Dilley J, Adler B, Louie BT, McFarland W. Comparison of Three Methods to Measure HIV Incidence Among Persons Seeking Voluntary, Anonymous Counseling and Testing. J Acquir Immune Defic Syndr. 2005;39(1):112-20. 19. Weinstock H, Dale M, Gwinn M, Satten GA, Kothe D, Mei J, et al. HIV seroincidence among patients at clinics for sexually transmitted diseases in nine cities in the United States. J Acquir Immune Defic Syndr. 2002;29(5):478-83. EUROSURVEILLANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveillance.org 1 5 EUROSURVEILLANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveillance.org 39. Schoenbach VJ, Poole C, Miller WC. Invited commentary: should we estimate incidence for undefined populations ? Am J Epidemiol. 2001;153(10):935-7. References HIV type 1 incidence estimates by detection of recent infection from a cross-sectional sampling of injection drug users in Bangkok: use of the IgG capture BED enzyme immunoassay. AIDS Res Hum Retroviruses. 2003;19(9):727-30. 33. Semaille C, Barin F, Cazein F, Pillonel J, Lot F, Brand D, et al. Monitoring the dynamics of the HIV epidemic using assays for recent infection and serotyping among new HIV diagnoses: experience after 2 years in France. J Infect Dis. 2007;196(3):377-83. 12. McDougal JS, Parekh BS, Peterson ML, Branson BM, Dobbs T, Ackers M, et al. Comparison of HIV type 1 incidence observed during longitudinal follow-up with incidence estimated by cross-sectional analysis using the BED capture enzyme immunoassay. AIDS Res Hum Retroviruses. 2006;22(10):945-52. 34. Lee LM, McKenna MT. Monitoring the incidence of HIV infection in the United States. Public health reports. 2007;122 Suppl 1:72-9. 35. Remis RS, Swantee C, Fikre-Merid M, Palmer RWH, Fearon M, Fisher M, et al. Enhancing diagnostic data for HIV surveillance: The Laboratory Enhancement Study. 15th International Conference on AIDS, Bangkok, Thailand, July 11-16, 2004. Available from: http://www.phs.utoronto.ca/ohemu/doc/intaids04LES.pdf [Accessed 22 Dec 2007] 13. Rothman KJ, Greenland S. Modern Epidemiology. Second ed. Philadelphia: Lippincott-Raven; 1998. 14. Calypte Biomedical Corporation. Calypte® HIV-1 BED Incidence EIA (IgG-Capture HIV-EIA) Package Insert. Available from: http://www.calypte.com/pdf/bed- insert.pdf [Accessed 2 Aug 2008] 36. Suligoi B, Massi M, Galli C, Sciandra M, Di Sora F, Pezzotti P, et al. Identifying recent HIV infections using the avidity index and an automated enzyme immunoassay. J Acquir Immune Defic Syndr. 2003;32(4):424-8. 15. Centers for Disease Control and Prevention. BED-CEIA Incidence and Adjustment Formula. Available from: http://www.cdc.gov/nchstp/od/gap/docs/surveillance/ BED-CEIA%20Incidence%20and%20Adjustment%20Formula.pdf [Accessed 22 Dec 2007] 37. Killian MS, Norris PJ, Rawal BD, Lebedeva M, Hecht FM, Levy JA, et al. The effects of early antiretroviral therapy and its discontinuation on the HIV- specific antibody response. AIDS Res Hum Retroviruses. 2006;22(7):640-7. 16. Constantine NT, Sill AM, Jack N, Kreisel K, Edwards J, Cafarella T, et al. Improved classification of recent HIV-1 infection by employing a two- stage sensitive/less-sensitive test strategy. J Acquir Immune Defic Syndr. 2003;32(1):94-103. 38. Fisher M, Pao D, Murphy G, Dean G, McElborough D, Homer G, et al. Serological testing algorithm shows rising HIV incidence in a UK cohort of men who have sex with men: 10 years application. AIDS. 2007;21(17):2309-14. EUROSURVEILLANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveillance.org 1 5 40. Remis RS, Palmer RWH, Raboud JM. 40. Remis RS, Palmer RWH, Raboud JM. Bias in estimates of HIV incidence based on the detuned assay : A proposed solution. 15th International Conference on AIDS, STARHS Satellite Meeting, Bangkok, Thailand. July 11, 2004. Available from: http://www.phs.utoronto.ca/ohemu/doc/Bias_Thailand.ppt [Accessed 22 Jul 2007] References Bias in estimates of HIV incidence based on the detuned assay : A proposed solution. 15th International Conference on AIDS, STARHS Satellite Meeting, Bangkok, Thailand. July 11, 2004. Available from: http://www.phs.utoronto.ca/ohemu/doc/Bias_Thailand.ppt [Accessed 22 Jul 2007] 40. Remis RS, Palmer RWH, Raboud JM. Bias in estimates of HIV incidence based on the detuned assay : A proposed solution. 15th International Conference on AIDS, STARHS Satellite Meeting, Bangkok, Thailand. July 11, 2004. Available from: http://www.phs.utoronto.ca/ohemu/doc/Bias_Thailand.ppt [Accessed 22 Jul 2007] 41. Song R, Karon JM, White E, Goldbaum G. Estimating the distribution of a renewal process from times at which events from an independent process are detected. Biometrics. 2006;62(3):838-46. 41. Song R, Karon JM, White E, Goldbaum G. Estimating the distribution of a renewal process from times at which events from an independent process are detected. Biometrics. 2006;62(3):838-46. This article was published on 4 September 2008. Citation style for this article: Le Vu S, Pillonel J, Semaille C, Bernillon P, Le Strat Y, Meyer L, Desenclos JC. Principles and uses of HIV incidence estimation from recent infection testing - a review. Euro Surveill. 2008;13(36):pii=18969. Available online: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=18969 1 6

https://openalex.org/W2137400530

https://europepmc.org/articles/pmc3291336?pdf=render

English

Influenza,<i>Campylobacter</i>and<i>Mycoplasma</i>Infections, and Hospital Admissions for Guillain-Barré Syndrome, England

Emerging infectious diseases

cc-by

RESEARCH RESEARCH Reports of Microbiologically Confirmed Infections Reports of Microbiologically Confirmed Infections Positive microbiologic diagnoses ascertained through voluntary laboratory reporting in England and Wales are recorded in the national infections database (LabBase2) (18). We obtained weekly reports of infections suspected of causing GBS, namely, Campylobacter spp., cytomegalovirus, Epstein-Barr virus, Haemophilus influenzae (B and non-B), Mycoplasma pneumoniae, and influenza (A, B, and all influenza) from 1993 through 2002. Influenza vaccination figures are available only quarterly and do not provide sufficient temporal resolution for this analysis. G uillain-Barré syndrome (GBS) is the most common cause of acute flaccid paralysis in polio-free regions. Estimated incidence in high-income countries is 0.4–4.0 cases per 100,000 population (1). Campylobacter jejuni is the most commonly identified infectious trigger for GBS. Several studies have demonstrated evidence of recent C. jejuni infection in a higher proportion of GBS case- patients than in controls (2–10). Other pathogens, includ- ing cytomegalovirus (7), Epstein-Barr virus (7), G We used the specimen date for all analyses because onset dates were rarely available. For Campylobacter, the median delay between patients’ onset date and the speci- men date was 4 days (interquartile range 3–7 days); for 90% of cases, the delay was <14 days (19). Similar data were unavailable for other pathogens. Influenza, Campylobacter and Mycoplasma Infections, and Hospital Admissions for Guillain- Barré Syndrome, England Clarence C. Tam,*† Sarah J. O’Brien,‡ and Laura C. Rodrigues* Haemophilus influenzae (11–14), and Mycoplasma pneu- moniae (7,15,16), have been suggested as possible GBS triggers, as was influenza vaccination in the United States during 1976–1977 (17). However, epidemiologic evidence that implicates these latter agents remains scarce. We con- ducted a time-series analysis to investigate temporal asso- ciations between weekly variations in reports of microbiologically confirmed infections and hospital admissions for GBS. Guillain-Barré syndrome (GBS) is the most common cause of acute flaccid paralysis in polio-free regions. Considerable evidence links Campylobacter infection with GBS, but evidence that implicates other pathogens as trig- gers remains scarce. We conducted a time-series analysis to investigate short-term correlations between weekly labo- ratory-confirmed reports of putative triggering pathogens and weekly hospitalizations for GBS in England from 1993 through 2002. We found a positive association between the numbers of reports of laboratory-confirmed influenza A in any given week and GBS hospitalizations in the same week. Different pathogens may trigger GBS in persons of different ages; among those <35 years, numbers of weekly GBS hospitalizations were associated with weekly Campylobacter and Mycoplasma pneumoniae reports, whereas among those >35 years, positive associations were with influenza. Further studies should estimate the relative contribution of different pathogens to GBS inci- dence, overall and by age group, and determine whether influenza is a real trigger for GBS or a marker for influenza vaccination. GBS Hospitalizations *London School of Hygiene and Tropical Medicine, London, United Kingdom; †Health Protection Agency, London, United Kingdom; and ‡University of Manchester, Manchester, United Kingdom Nonidentifiable GBS hospitalization data were provid- ed by Hospital Episodes Statistics (HES) (20), which Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1880 Guillain-Barré Syndrome, England records all in-patient care episodes in English National Health Service hospitals. An episode is a continuous peri- od of treatment under 1 consultant. Each episode includes patient’s age, sex, admission date, episode duration, episode number, and <14 possible International Classification of Diseases (ICD) diagnoses. Because a patient may see several consultants during a hospital stay, several episodes for the same hospitalization event may appear in HES records. Since April 1997 a unique code identifies episodes for the same patient. A series of contin- uous episodes constitutes a spell of treatment. This method of episode linkage was applied to records from January 1998 onward. Before then, only episodes classified as first episodes were used to avoid including multiple episodes for the same patient spell. Figure 1. Seasonal distribution of Guillain-Barré syndrome admis- sions during 1993–1997 and 1998–2002. Figure 1. Seasonal distribution of Guillain-Barré syndrome admis- sions during 1993–1997 and 1998–2002. A GBS admission was defined as a spell with GBS- related ICD codes (ICD-9 357.0/ICD-10 G61.0) in any of the first 3 diagnostic codes. From 1998 through 2002, 3,477 repeat spells were excluded; these are unlikely to represent independent events, e.g., the likelihood of recur- rent GBS may depend on host genetic factors. they have similar seasonal characteristics, not because one causes the other. For example, bottled water consumption increases in summer, when the incidence of salmonellosis is highest. This does not imply that bottled water is a risk factor for Salmonella infection; rather, bottled water con- sumption is influenced by ambient temperature, which itself independently influences Salmonella transmission. The exposure-outcome association could also be con- founded or obscured by other time-varying factors. For example, if influenza causes GBS, GBS admissions should increase in winter, when influenza incidence is highest. However, Campylobacter has the opposite seasonality; if Campylobacter is also associated with GBS, high numbers of GBS admissions could still occur when influenza reports are low, because these GBS cases are due to Campylobacter (or other pathogens with different season- ality). GBS Hospitalizations Time-series methods account for such temporal dependencies in data by adjusting for these long-term trends and seasonal patterns, which enables associations to be investigated over shorter periods, independent of trend and seasonal components. The 2 series of hospitalizations (1993–1997 and 1998–2002) were collapsed into weekly counts of GBS admissions. For 1993, data were only available from April 1 on. We compared the 2 periods to investigate whether inability to exclude repeat spells from 1993 through 1997 affected the seasonal pattern of GBS admissions. No major differences were seen (Figure 1), and the 2 periods were combined into a weekly time-series of 10 years. Statistical Analysis We aimed to answer the following question: Is an increase in the number of laboratory reports in any given week associated with increases in GBS hospitalizations in subsequent weeks? In choosing appropriate statistical methods, special characteristics of time-series data must be considered. Such data exhibit nonrandom patterns over time. These include long-term increasing or decreasing trends (whereby weekly GBS hospitalizations within a year are more closely related than between years) and seasonal patterns (whereby the number of GBS hospitalizations in any given week is similar to that in the same week for other years). In addition, weekly hospitalizations are count data, following a Poisson rather than a normal distribution. We used multivariable Poisson regression adapted for time-series data (21–23) to investigate the effect of week- ly variations in reports of different pathogens on the num- ber of GBS hospitalizations; we adjusted for long-term trends and seasonality. The weekly number of GBS hospi- talizations was the outcome, and the weekly number of reports for each pathogen was the exposure. We assumed a log-linear relationship between exposure and outcome, i.e., that an increase in the number of reports of a particu- lar pathogen resulted in a constant increase in the log num- ber of GBS hospitalizations throughout the range of laboratory reports. For these reasons, time-series observations cannot be considered to be independent, and statistical techniques commonly used for independent, normally distributed data (such as simple correlation) are inappropriate. Special methods that account for temporal dependence in the data are needed. Specifically, temporal dependence in time- series data can result in confounding due to long-term trends (year-on-year) and seasonal (within-year) patterns. Two variables could apparently be related in time because We adjusted for long-term (year-on-year) trends by including a variable indicating the year of hospitalization in the regression model; thus, we allowed the mean num- ber of GBS hospitalizations to vary between years. We 1881 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 RESEARCH controlled for seasonality by using Fourier terms (21,24). Fourier terms can be used to produce a smooth function of expected values for any set of periodic data (e.g., a season- al pattern). This is achieved by introducing into the regres- sion model a linear combination of pairs of sine and cosine terms (harmonics) of varying wavelengths. A harmonic is an integer fraction of 1 full wavelength (here, 1 year). Age Group Analysis Associations between GBS and infection could differ between age groups; an association between a pathogen and GBS might only become apparent in a limited age range. We performed subanalyses to investigate associa- tions in different age groups. Because the age distribution of GBS case-patients is not uniform, we categorized age into 3 broad groups: <35 years, 35–64 years, and >65 years, according to the age distribution of GBS patient admissions (Figure 2). Age group–specific models were fit similarly to those for all ages. Statistical Analysis The more harmonics used, the better the fit to the hospitaliza- tion series (i.e., the greater the level of seasonal adjust- ment). A seasonal pattern with a single peak and single trough within 1 year could be reproduced with 1 harmon- ic. In reality, seasonal patterns are more complex, and sev- eral harmonic terms are required for adequate seasonal adjustment. Given sufficient seasonal adjustment, all vari- ation in the hospitalization series explained by seasonality is removed; any remaining variation must be due to other factors or random noise. This residual variation, independ- ent of long-term trends and seasonal patterns, and its asso- ciation with weekly reports of infections was our focus of interest. We used 6 harmonics to adjust for periodic pat- terns in the data >2 months, assuming that GBS risk is increased for <2 months after infection. In addition, we introduced a variable that indicated weeks in which public holidays occurred, to adjust for artifactual variation in lab- oratory reporting and hospitalizations during these weeks. monics), and φ (holiday) is a term representing weeks with public holidays. The regression coefficient, β, is the effect of the exposure of interest (the weekly number of labora- tory reports, X). Its exponential, the relative risk (RR), reflects the ratio increase in GBS hospitalizations per unit increase in laboratory reports of pathogen (p) at lag t-l, where l ranges from 8 weeks before to 4 weeks after the GBS hospitalization. We fit separate models for each pathogen at each lag. We assessed model fit by looking at residual variation. We used the partial autocorrelation function (25) to investigate the presence of residual autocorrelation, i.e., whether residual variation in GBS hospitalizations in any given week was correlated with residual variation in other weeks. Some degree of autocorrelation at a lag of 1 week remained after adjustment for yearly and seasonal patterns. We controlled for this by adding to all models a term for the residuals lagged by 1 week (a first-order autoregressive term; AR1 in the equation). The scale parameter for stan- dard errors was set as the Pearson χ2 statistic divided by the residual degrees of freedom to allow for possible overdispersion in the data. Lag Effects Because of the time lag between infection and GBS, the number of GBS admissions is likely to be associated not with the number of laboratory reports in the same week, but with the number of reports some time before. We thus performed separate regressions with exposure variables lagged by <8 weeks. Further, because of delays in seeking healthcare, diagnosing infection, and reporting positive diagnoses to national surveillance, increases in hospitaliza- tions could precede increases in laboratory reports. To account for this possibility, we also performed regressions of GBS admissions against laboratory reports within the subsequent 4 weeks. All lags at which positive associations were found are presented. However, because of the large number of statis- tical tests performed (13 lags per pathogen per age group), a positive association at a given lag was considered poten- tially relevant only if it occurred within a cluster. A cluster The core models thus contained the logged GBS hospi- talization series as the dependent variable, indicator vari- ables for year, Fourier terms for season, and an indicator variable for weeks with public holidays. The weekly num- ber of reports of a particular pathogen, either in the same week or lagged by a certain number of weeks, was then introduced as the explanatory variable of interest. The regression equation for the models predicting the expecta- tion of the logarithm of weekly GBS hospitalizations, Y, was Figure 2. Age distribution of first hospitalization for Guillain-Barré syndrome, England, April 1993–December 2002. 1 ) ( ) ( ) ( ) ( )) ( log( AR holiday t S year X Y E y p l t + + + + + = − φ δ β α where δy represents the coefficients for each year (y), S(t) represents a smooth function of season (comprising 6 har- Figure 2. Age distribution of first hospitalization for Guillain-Barré syndrome, England, April 1993–December 2002. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1882 Guillain-Barré Syndrome, England 0.01 level of precision are presented. The RRs represent the relative increase in GBS admissions per 10% increase in the range of laboratory reports for a given pathogen at a given lag. Results Different pathogens are associated with GBS admission in different age groups. In those <35 years, the number of GBS admissions in a given week was associated with the number of Campylobacter spp. reports 5 and 4 weeks ear- lier and with the number of M. pneumoniae reports in the same week and 1 and 3 weeks later. In the 10-year study period, 11,019 primary admissions for GBS occurred: 2,929 (26.6%) patients were <35 years of age, 4,467 (40.5%) were ages 35–64 years, and 3,623 (32.9%) were >65 years. Summary statistics for the week- ly number of GBS admissions and laboratory reports for the different pathogens are found in Table 1. Table 2 gives details of the lags for each pathogen for which significant associations with GBS admissions were found. Only clusters of lags that were significant at the 0.05 significance level are presented. Within clusters, lags that were significant at the 0.01 significance level appear in bold. For example, for influenza reports in all ages, sig- nificant associations were found between the number of GBS hospitalizations in any given week and the number of influenza reports in the same week (lag 0) and the pre- vious week (lag 1); the p value for the coefficient at lag 0 was <0.01. Lags that were associated with GBS at the 0.05 significance level but did not occur in clusters are shown in Table 3. Seventeen such lags occurred, consis- tent with 1 in 20 tests giving a significant result by chance (at p<0.05) (given 312 combinations of pathogens, age groups, and lags). Among persons ages 35–64 years, a positive associa- tion was found between the number of GBS admissions in any given week and the number of all influenza reports 1 and 2 weeks earlier. In those ages >65 years, associations were found between the number of GBS admissions and the number of all influenza and influenza A reports in the current week and 1 week before hospitalization. The results were robust to varying degrees of seasonal adjustment; we repeated the analysis and adjusted for sea- sonal wavelengths of up to 4 months (3 harmonics) and 1 month (12 harmonics) and used indicator variables for month as well, all with similar results. Table 5 shows those lags that consistently appeared in clusters at all levels of seasonal adjustment. Results for influenza and M. Lag Effects For example, for influenza A, the maximum number of laboratory reports in any given week was 398, while the minimum was zero; an increase in influenza A reports of 39.8 (10% of the range) in any given week results, on average, in a 1.03-fold (or 3%) higher incidence of GBS admissions in the same week (RR = 1.032, 99% CI 1.008–1.057). Overall, a positive association was found only with influenza and influenza A at a lag of zero weeks (in the same week as GBS admission). was defined as 2 or more consecutive lags, each associat- ed with the outcome at the 0.05 significance level. This approach reduces the probability of observing chance associations due to multiple testing. Within clusters, lags significant at the 0.01 level were considered important. The adjusted coefficient from these models was used to calculate the expected increase in GBS admissions per 10% increase in the range of laboratory reports at a given lag. All statistical analyses were performed in Stata 8.0 (Stata Corp., College Station, TX, USA). Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1 Results pneumo- niae were not sensitive to the degree of seasonal adjust- ment. For Campylobacter, clusters of lags were seen with all Fourier models, but not with models that used month indicators. Table 4 presents RRs and 99% confidence intervals (CIs) for the associations shown in Table 2. Only those individual lags from Table 2 that were significant at the Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1883 RESEARCH the infections investigated. Misdiagnosis of GBS is also possible; a proportion of GBS diagnoses is likely to repre- sent false positives. We could not validate diagnoses through hospital chart review because identities were hid- den in the HES data, and we did not have access to patients’ records. However, any misclassification arising from inclusion of false-positive GBS diagnoses will be nondifferential, i.e., the likelihood of misdiagnosis with GBS is unrelated to the likelihood of diagnosis with the pathogens investigated. Inclusion of non-GBS cases could have resulted in effect dilution, but this inclusion would likely not have yielded positive associations when none truly existed. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 Discussion We found associations between the weekly number of laboratory reports of various pathogens and incidence of GBS hospitalizations. Different organisms may be respon- sible for triggering GBS in different age groups. In partic- ular, Campylobacter and M. pneumoniae appear to be associated with GBS in those <35 years, while influenza associations were seen in those >35 years. Differences in the pathogens responsible for triggering GBS in different age groups have not previously been reported. No clusters of significant lags were found for cytomegalovirus, Epstein-Barr virus, and H. influenzae infections. This could be due to low statistical power (on average, <10 reports per week were made for Epstein-Barr virus and H. influenzae), or it could indicate a very small risk or none after these infections. Laboratory reports for any condition represent only a subset of all symptomatic cases of disease in the com- munity. Our analysis assumes that, for a given condition, the seasonal pattern of laboratory reports accurately reflects the pattern of all community cases. Ascertainment of influenza is likely to be more comprehensive in winter because microbiologic investigation for this pathogen is not routinely conducted outside the influenza season (26). This could affect our ability to detect associations in different seasons, but this was not the focus of our study. Our analy- sis also assumes that the seasonal pattern of laboratory reports is accurately reflected within each age group. This may not be true if, for example, younger persons are less likely to visit the health services (and, thus, be included in Our results are subject to several limitations. HES data exclude information from private hospitals. Given the uni- versality of healthcare in England, however, the proportion of GBS cases diagnosed in private hospitals is likely to be small. Although we used only ICD codes specific for GBS, some GBS cases may be classified under nonspecific codes, namely, ICD-9 357.9 (unspecified toxic and inflam- matory neuropathy) and ICD-10 G61.9 (inflammatory polyneuropathy unspecified). However, these codes include a large proportion of cases unrelated to GBS; their inclusion in the analysis would dilute any associations with 1884 Guillain-Barré Syndrome, England captured by laboratory reports, and this will vary by pathogen (for example, severe conditions are more likely to be reported). Thus, estimates of the relative incidence of GBS due to the different pathogens cannot be obtained from these data. In addition, some evidence exists, particu- larly for C. References 1. Allos BM. Association between Campylobacter infection and Guillain-Barré syndrome. J Infect Dis. 1997;176(Suppl 2):S125–8. 2. Rees JH, Soudain SE, Gregson NA, Hughes RA. Campylobacter jejuni infection and Guillain-Barré syndrome. N Engl J Med. 1995;333:1374–9. 3. Enders U, Karch H, Toyka KV, Michels M, Zielasek J, Pette M, et al. The spectrum of immune responses to Campylobacter jejuni and gly- coconjugates in Guillain-Barré syndrome and in other neuroimmuno- logical disorders. Ann Neurol. 1993;34:136–44. 4. Guarino M, Casmiro M, D’Alessandro R. Campylobacter jejuni infection and Guillain-Barré syndrome: a case-control study. Emilia- Romagna Study Group on Clinical and Epidemiological Problems in Neurology. Neuroepidemiology. 1998;17:296–302. 5. Hao Q, Saida T, Kuroki S, Nishimura M, Nukina M, Obayashi H, et al. Antibodies to gangliosides and galactocerebroside in patients with Guillain-Barré syndrome with preceding Campylobacter jejuni and other identified infections. J Neuroimmunol. 1998;81:116–26. The short lags identified here between increases in influenza reports and subsequent GBS admissions are con- sistent with a vaccine trigger; the risk period for vaccine- related GBS is believed to be 6 weeks, and increases in vaccination coverage would be expected to precede sea- sonal rises in influenza. Vaccination could also explain the lack of an association in younger persons, because influen- za vaccination is not generally recommended in healthy persons <65 years in the United Kingdom. Influenza vac- cine coverage data indicate that for the study period, vac- cine uptake was <1% in low-risk groups ages <35 years, <10% among those ages 35–54 years, and 20%–30% among those ages >65 years (29). For elderly persons at high risk, uptake increased from 40% to 65% from 1993 through 2002 (30). These data support the hypothesis that persons in older age groups have a greater vaccine-induced risk of GBS, although a true association with the disease of influenza is still possible. Primary care–based studies investigating the influenza and influenza vaccination sta- tus of GBS patients could help resolve this issue. 6. Ho TW, Mishu B, Li CY, Gao CY, Cornblath DR, Griffin JW, et al. Guillain-Barré syndrome in northern China. Relationship to Campylobacter jejuni infection and anti-glycolipid antibodies. Brain. 1995;118:597–605. 7. Jacobs BC, Rothbarth PH, van der Meche FG, Herbrink P, Schmitz PI, de Klerk MA, et al. The spectrum of antecedent infections in Guillain-Barré syndrome: a case-control study. Neurology. 1998;51:1110–5. 8. Kaldor J, Speed BR. Guillain-Barré syndrome and Campylobacter jejuni: a serological study. Br Med J (Clin Res Ed). 1984;288:1867–70. 9. References Koga M, Yuki N, Takahashi M, Saito K, Hirata K. Close association of IgA anti-ganglioside antibodies with antecedent Campylobacter jejuni infection in Guillain-Barré and Fisher’s syndromes. J Neuroimmunol. 1998;81:138–43. 10. Kuroki S, Haruta T, Yoshioka M, Kobayashi Y, Nukina M, Nakanishi H. Guillain-Barré syndrome associated with Campylobacter infec- tion. Pediatr Infect Dis J. 1991;10:149–51. 11. Ju YY, Womersley H, Pritchard J, Gray I, Hughes RA, Gregson NA. Haemophilus influenzae as a possible cause of Guillain-Barré syn- drome. J Neuroimmunol. 2004;149:160–6. 12. Koga M, Gilbert M, Li J, Koike S, Takahashi M, Furukawa K, et al. Antecedent infections in Fisher syndrome: a common pathogenesis of molecular mimicry. Neurology. 2005;64:1605–11. Discussion jejuni, that GBS can develop after subclinical infection. Our analysis did not include asymptomatic infec- tions, so our results apply only to clinical cases of infection. These findings nevertheless raise hypotheses that merit fur- ther investigation. For example, several case reports and immunologic analyses have suggested a link between M. pneumoniae infection and GBS (7,15,16), but such a link has not been confirmed by robust epidemiologic studies. Our results suggest that studies focusing on younger GBS patients could help clarify any such association. laboratory reports) for symptoms of influenza during the influenza season. However, laboratory report data for influenza show a distinct and consistent peak during the winter months in all age groups (data not shown). Among Campylobacter spp., only C. jejuni is thought to cause GBS. As clinical isolates of Campylobacter are not routinely speciated in England and Wales, non-jejuni species could not be excluded from the analysis. However, the England and Wales Campylobacter Sentinel Surveillance Scheme indicates that 80%–90% of reports of Campylobacter infection are due to C. jejuni (19); inclu- sion of species not linked to GBS would attenuate rather than inflate any effect on GBS admissions. The regression coefficients indicate associations between the incidence of various pathogens and GBS admissions, but the coefficients themselves are not directly comparable between pathogens. This is because their mag- nitude is dependent not only on the true magnitude of the association, but also on the proportion of all cases that is Whether the associations with influenza are real or whether they reflect seasonal patterns in influenza vaccina- tion is unclear. Influenza vaccination has previously been linked to GBS. During the mass vaccination campaign Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1885 RESEARCH against swine influenza in the United States during 1976–1977, GBS incidence among vaccinees was 7-fold higher in the 6 weeks after vaccination than in nonvacci- nees (17,26). Similar analyses during subsequent influen- za seasons (with no mass vaccination) have found no increased risk (27), or a doubling of the risk (28), which suggests that differences in antigenic formulation or char- acteristics of vaccinated populations are influential factors in vaccine-related GBS risk. Here, we found associations with influenza A only; this may reflect the antigenic com- position of influenza vaccines, or differential risk resulting from antigenic differences between subtype A and B strains of influenza. References Acknowledgments 13. Mori M, Kuwabara S, Miyake M, Dezawa M, Adachi-Usami E, Kuroki H, et al. Haemophilus influenzae has a GM1 ganglioside-like structure and elicits Guillain-Barré syndrome. Neurology. 1999;52:1282–4. We gratefully acknowledge Kate Byram and Susan Alpay for providing the data on GBS hospitalizations and Sallyanne Meakins for providing the data on laboratory reports. We thank Shakoor Hajat for helpful comments on the manuscript. 14. Mori M, Kuwabara S, Miyake M, Noda M, Kuroki H, Kanno H, et al. Haemophilus influenzae infection and Guillain-Barré syndrome. Brain. 2000;123:2171–8. This study was funded by the Environmental and Enteric Diseases Department, Health Protection Agency Centre for Infections, and the Infectious Disease Epidemiology Unit, Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine. 15. Ang CW, Tio-Gillen AP, Groen J, Herbrink P, Jacobs BC, Van Koningsveld R, et al. Cross-reactive anti-galactocerebroside antibod- ies and Mycoplasma pneumoniae infections in Guillain-Barré syn- drome. J Neuroimmunol. 2002;130:179–83. 16. Ginestal RC, Plaza JF, Callejo JM, Rodriguez-Espinosa N, Fernandez-Ruiz LC, Masjuan J. Bilateral optic neuritis and Guillain- Barré syndrome following an acute Mycoplasma pneumoniae infec- tion. J Neurol. 2004;251:767–8. Dr Tam is an epidemiologist at the Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine. His main research interests include the epidemiology of infectious intestinal disease, long-term sequelae of diarrheal infections, and epidemiologic methods. 17. Safranek TJ, Lawrence DN, Kurland LT, Culver DH, Wiederholt WC, Hayner NS, et al. Reassessment of the association between Guillain- Barré syndrome and receipt of swine influenza vaccine in 1976–1977: results of a two-state study. Expert Neurology Group. Am J Epidemiol. 1991;133:940–51. 1886 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 Guillain-Barré Syndrome, England Guillain-Barré Syndrome, England 18. Wall PG, de Louvois J, Gilbert RJ, Rowe B. Food poisoning: notifi- cations, laboratory reports, and outbreaks–where do the statistics come from and what do they mean? Commun Dis Rep CDR Rev. 1996;6:R93–100. 25. Chatfield C. The analysis of time series: an introduction, 5th ed. London: Chapman and Hall; 1996. 26. Goddard NL, Joseph CA, Zambon M, Nunn M, Fleming D, Watson JM. Influenza surveillance in the United Kingdom: October 2000 to May 2001. [serial online]. Commun Dis Rep CDR. 2001;2005:S1–7. 19. Gillespie IA, O’Brien SJ, Frost JA, Adak GK, Horby P, Swan AV, et al. A case-case comparison of Campylobacter coli and Campylobacter jejuni infection: a tool for generating hypotheses. Emerg Infect Dis. 2002;8:937–42. Acknowledgments 27. Schonberger LB, Hurwitz ES, Katona P, Holman RC, Bregman DJ. Guillain-Barré syndrome: its epidemiology and associations with influenza vaccination. Ann Neurol. 1981;9(Suppl):31–8. g 20. Department of Health (England). Hospital episode statistics [cited 2005 Jan 12]. Available from http://www.hesonline.nhs.uk 28. Hurwitz ES, Schonberger LB, Nelson DB, Holman RC. Guillain- Barré syndrome and the 1978–1979 influenza vaccine. N Engl J Med. 1981;304:1557–61. 21. Schwartz J, Spix C, Touloumi G, Bacharova L, Barumamdzadeh T, le Tertre A, et al. Methodological issues in studies of air pollution and daily counts of deaths or hospital admissions. J Epidemiol Community Health. 1996;50(Suppl 1):S3–11. 29. Lasky T, Terracciano GJ, Magder L, Koski CK, Ballesteros M, Nash D, et al. The Guillain-Barré syndrome and the 1992–1993 and 1993–1994 influenza vaccines. N Engl J Med. 1998;339:1797–802. 22. Schwartz J, Levin R, Goldstein R. Drinking water turbidity and gas- trointestinal illness in the elderly of Philadelphia. J Epidemiol Community Health. 2000;54:45–51. 30. Joseph C, Goddard N, Gelb D. Influenza vaccine uptake and distribu- tion in England and Wales using data from the General Practice Research Database, 1989/90–2003/04. J Public Health (Oxf). 2005;27:371–7. 23. Schwartz J, Levin R, Hodge K. Drinking water turbidity and pediatric hospital use for gastrointestinal illness in Philadelphia. Epidemiology. 1997;8:615–20. Address for correspondence: Clarence C. Tam, Infectious Disease Epidemiology Unit, Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK; email: [email protected] 24. Vellinga A, Van Loock F. The dioxin crisis as experiment to deter- mine poultry-related campylobacter enteritis. Emerg Infect Dis. 2002;8:19–22. 24. Vellinga A, Van Loock F. The dioxin crisis as experiment to deter- mine poultry-related campylobacter enteritis. Emerg Infect Dis. 2002;8:19–22. Address for correspondence: Clarence C. Tam, Infectious Disease Epidemiology Unit, Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK; email: [email protected] Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1887 Search past issues of EID at www.cdc.gov/eid Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1887 Search past issues of EID at www.cdc.gov/eid Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1887 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 ging Infectious Diseases • www.cdc.gov/eid • Vol. 12, No. 12, December 2006 1887 1887

https://openalex.org/W2167555384

https://air.unimi.it/bitstream/2434/748112/2/1626.full.pdf

English

Platelet Isoprostane Overproduction in Diabetic Patients Treated With Aspirin

Diabetes

cc-by

Platelet Isoprostane Overproduction in Diabetic Patients Treated With Aspirin Roberto Cangemi,1 Pasquale Pignatelli,1 Roberto Carnevale,1 Carmen Nigro,1 Marco Proietti,1 Francesco Angelico,1 Davide Lauro,2 Stefania Basili,1 and Francesco Violi1 of reduced NAD phosphate (NADPH) oxidase, play a crucial role in platelet isoprostane formation (8). In contrast to TxA2, isoprostanes are chemically stable compounds that serve to propagate platelet activation, amplifying platelet re- sponse to common agonists via glycoprotein (Gp)IIb/IIIa activation (9). Accordingly, patients with hereditary de- ficiency of NOX2 showed impaired isoprostane formation and GpIIb/IIIa activation, as well as a subnormal propagation of platelet thrombus (9). Aspirin modestly influences cardiovascular events in patients with type 2 diabetes mellitus (T2DM), but the reason is unclear. The aim of the study was to determine whether in T2DM patients aspirin enhances platelet isoprostanes, which are eicosanoids with proaggregating properties derived from arachidonic acid oxidation by platelet NOX2, the catalytic subunit of reduced NAD phosphate oxidase. A cross-sectional study was performed com- paring T2DM patients, treated (n = 50) or not treated (n = 50) with 100 mg/day aspirin, with 100 nondiabetic patients, matched for age, sex, atherosclerosis risk factors, and aspirin treatment. A short-term (7 days) treatment with 100 mg/day aspirin also was performed in 36 aspirin-free diabetic and nondiabetic patients. Higher platelet recruitment, platelet isoprostane, and NOX2 acti- vation was found in diabetic versus nondiabetic patients and in aspirin-treated diabetic patients versus nontreated patients (P , 0.001). Platelet thromboxane (Tx) A2 (P , 0.001) was inhibited in all aspirin-treated patients. In the interventional study, aspirin sim- ilarly inhibited platelet TxA2 in diabetic and nondiabetic patients (P , 0.001). Platelet recruitment, isoprostane levels, and NOX2 activation showed a parallel increase in diabetic patients (P , 0.001) and no changes in nondiabetic patients. These findings suggest that in aspirin-treated diabetic patients, oxidative stress– mediated platelet isoprostane overproduction is associated with enhanced platelet recruitment, an effect that mitigates aspirin- mediated TxA2 inhibition. Diabetes 61:1626–1632, 2012 p ( ) It previously has been demonstrated that COX1 in- hibition determines a shift in arachidonic acid metabolism toward other pathways, such as the lipooxygenase system (10). We speculated that COX1 inhibition also could be associated with an increased conversion of arachidonic acid to isoprostanes in platelets. The increase of platelet isoprostanes would counterbalance the inhibition of TxA2, therefore hampering the antiplatelet effect of aspirin. From the 1I Clinica Medica, Sapienza University of Rome, Rome, Italy; and the 2Department of Internal Medicine, University of Tor Vergata, Rome, Italy. Corresponding author: Francesco Violi, [email protected]. Received 5 September 2011 and accepted 1 February 2012. DOI: 10.2337/db11-1243. Clinical trial reg. no. NCT01250340, clinicaltrials.gov. R.C. and P.P. contributed equally to this work.
2012 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by -nc-nd/3.0/ for details. ORIGINAL ARTICLE ORIGINAL ARTICLE diabetes.diabetesjournals.org Platelet Isoprostane Overproduction in Diabetic Patients Treated With Aspirin To explore this hypothesis, we performed a cross-sectional study comparing the behavior of platelet isoprostanes and TxA2 and their interplay with platelet NOX2 in diabetic and nondiabetic patients treated or not with aspirin. Platelet activation tests, including arachidonic acid–induced platelet aggregation, which is dependent upon TxA2 formation (11), and platelet recruitment, which is dependent upon ROS and isoprostane formation (9), were determined. Analysis of these variables was repeated in a prospective, short-term study of diabetic and nondiabetic patients treated for 7 days with low-dose aspirin. A di ) A ccelerated atherosclerosis is a typical feature of type 2 diabetes mellitus (T2DM). Thus, patients with T2DM have a two- to fourfold increased risk of cardiovascular diseases (coronary artery disease) and a two- to sixfold increased risk of stroke (1–3). Platelets play a major role in the etiology of athero- sclerotic disease, as shown by the significant decrease of cardiovascular events in patients treated with aspirin, an inhibitor of cyclooxygenase (COX1) that prevents platelet thromboxane (Tx) A2 formation (4,5). Platelet TxA2 over- production, combined with a significant decrease after aspirin administration, has been demonstrated in diabetic patients (6). Despite this, interventional trials with aspirin in di- abetic patients failed to show a beneficial effect in pri- mary prevention; reasons for this lack of response still are uncertain (7). RESULTS Extracellular levels of soluble NOX2-derived peptide (sNOX2-dp), a marker of NADPH oxidase activation, were detected by enzyme-linked immunosorbent assay, as previously described by Pignatelli et al. (13). The peptide was rec- ognized by the specific monoclonal antibody against the amino acidic sequence (224–268) of the extra membrane portion of NOX2, which was released in the medium upon platelet activation. Values were expressed as picograms per milliliter; intra-assay and interassay coefficients of variation were 5.2 and 6%, respectively. Cross-sectional study. Diabetic and nondiabetic patients had a similar distribution of risk factors, with the exception of BMI, which was higher in diabetic patients (Table 1). Statin therapy was more common in diabetic patients (P = 0.015). Diabetic and nondiabetic patients had similar platelet TxB2 values (Fig. 1A) and platelet aggregation percentages (95 6 3 vs. 96 6 2 LT%, respectively; P = 0.864). Diabetic patients had significantly higher platelet isoprostanes, platelet recruitment, and sNOX2-dp compared with nondiabetic patients (Fig. 1B–D). In diabetic patients, HbA1c significantly correlated with sNOX2-dp (r = 0.40, P , 0.0001) and platelet isoprostanes (r = 0.39, P , 0.0001). Diabetic patients treated or not with aspirin had similar clinical characteristics (Table 1) and received equally dis- tributed different antidiabetes treatment (data not shown). Platelet TxB2 was significantly lower in aspirin-treated di- abetic patients; conversely, aspirin-treated diabetic patients had higher platelet isoprostanes, platelet recruitment, and NOX2 activation (Fig. 1). Cross-sectional study. Diabetic and nondiabetic patients had a similar distribution of risk factors, with the exception of BMI, which was higher in diabetic patients (Table 1). Statin therapy was more common in diabetic patients (P = 0.015). Diabetic and nondiabetic patients had similar platelet TxB2 values (Fig. 1A) and platelet aggregation percentages (95 6 3 vs. 96 6 2 LT%, respectively; P = 0.864). Diabetic patients had significantly higher platelet isoprostanes, platelet recruitment, and sNOX2-dp compared with nondiabetic patients (Fig. 1B–D). In diabetic patients, HbA1c significantly correlated with sNOX2-dp (r = 0.40, P , 0.0001) and platelet isoprostanes (r = 0.39, P , 0.0001). Platelet TxB2 was measured by enzyme-linked immunosorbent assay (Amersham Pharmacia, Biotech, Little Chalfont, U.K.) and expressed as picomoles per liter. TxB2 levels in all experiments were expressed as defined in the following formula: (TxA2 in stimulated samples) 2 (TxA2 in unstimu- lated samples). Intra- and interassay coefficients of variation were 4.0 and 3.6%, respectively. Study design The minimum sample size was computed with respect to a two-tailed Student t test for independent groups, considering 1) relevant difference in isoprostane levels to be detected between groups |d| $90 pmol/L, 2) SDs homogeneous between the groups (SD = 80 pmol/L), and 3) type I error probability a = 0.05 and power 1-b = 0.90. This resulted in n = 17 per group. Statistical analysis. The minimum sample size was computed with respect to a two-tailed Student t test for independent groups, considering 1) relevant difference in isoprostane levels to be detected between groups |d| $90 pmol/L, 2) SDs homogeneous between the groups (SD = 80 pmol/L), and 3) type I error probability a = 0.05 and power 1-b = 0.90. This resulted in n = 17 per group. Categorical variables were reported as counts (percentage) and continuous variables were expressed as means 6 SD, unless otherwise indicated. Differ- ences between percentages were assessed by the x2 test or Fisher exact test. Student unpaired t test and Pearson product moment correlation analysis were used for normally distributed continuous variables. Appropriate nonparametric tests (Mann-Whitney U test and Spearman rank correlation test [Rs]) were used for all the other variables. Multiple linear regression analysis was performed to further quantify the relationship between the variables studied. Platelet sizing. To assess if there were differences between T2DM patients and control subjects in platelet size and function, we stratified in two portions (50% upper phase, 50% lower phase) the top 75% of the PRP. After the washing procedure of the two portions, we analyzed TxA2 and 8-iso-PGF2a-III production. Platelet sizing. To assess if there were differences between T2DM patients and control subjects in platelet size and function, we stratified in two portions (50% upper phase, 50% lower phase) the top 75% of the PRP. After the washing procedure of the two portions, we analyzed TxA2 and 8-iso-PGF2a-III production. Categorical variables were reported as counts (percentage) and continuous variables were expressed as means 6 SD, unless otherwise indicated. Differ- ences between percentages were assessed by the x2 test or Fisher exact test. Student unpaired t test and Pearson product moment correlation analysis were used for normally distributed continuous variables. Appropriate nonparametric tests (Mann-Whitney U test and Spearman rank correlation test [Rs]) were used for all the other variables. Multiple linear regression analysis was performed to further quantify the relationship between the variables studied. Study design Cross-sectional study. The study was performed in consecutive T2DM patients attending our metabolic outpatient clinic who were taking (n = 50) or not taking (n = 50) low-dose (100 mg/day) aspirin. T2DM was diagnosed ac- cording to the American Diabetes Association definition (12). As a control group, we selected nondiabetic outpatients taking (n = 50) or not taking (n = 50) low-dose aspirin who were matched to the diabetic group in terms of age, sex, and history of vascular disease. Low-dose aspirin treatment was defined as a self-reported daily intake of 100 mg acetylsalicylic acid at least in the previous month. Exclusion criteria were 1) recent history (,3 months) of acute vascular events, 2) clinical di- agnosis of type 1 diabetes (diagnosis of diabetes and insulin use before the age of 35 years), 3) serum creatinine level .2.5 mg/dL, 4) active infection or ma- lignancy, 5) cardiac arrhythmia or congestive heart failure, and 5) use of non- steroidal anti-inflammatory drugs, vitamin supplements, or other antiplatelet drugs, such as clopidogrel, in the previous 30 days. All participants provided written informed consent. The local ethical committee approved the study pro- tocol (approval no. Prot. 403/09-Rif. 1621/07.05.09). Diabetic patients received different antidiabetes treatments: metformin (n = 57), subcutaneous insulin (n = 25), sulfonylureas (n = 14), glinides (n = 6), glitazones (n = 3), and dipeptidyl peptidase-4 inhibitor (n = 16). Isoprostanes are a family of eicosanoids derived from arachidonic acid interaction with reactive oxidant species (ROS). Thus, ROS generated by NOX2, the catalytic subunit Interventional study. We tested the effect of short-term treatment with 100 mg/day aspirin in diabetic (n = 18) and nondiabetic (n = 18) patients, not currently under aspirin treatment and with no clinical history of vascular diseases, who were matched for sex, age, and atherosclerotic risk factors. Aspirin was given after dinner between 8:00 and 8:30 P.M., and adherence was assessed by the pill-count method. Blood samples were collected before DIABETES, VOL. 61, JUNE 2012 1626 R. CANGEMI AND ASSOCIATES aspirin ingestion and after 3 and 7 days of treatment. The study was registered in August 2010 at clinicaltrials.gov (clinical trial reg. no. NCT01250340). aspirin ingestion and after 3 and 7 days of treatment. The study was registered in August 2010 at clinicaltrials.gov (clinical trial reg. no. NCT01250340). agonist (collagen) and the second one depending on the release of mole- cules discharged by activated platelets (ADP, isoprostanes, etc.). diabetes.diabetesjournals.org Study design Cytometric analysis of platelets and leukocytes. Leukocyte content in the top 75% of the PRP was analyzed using the specific fluorescein isothiocyanate– labeled monoclonal antibody (Mab) anti-CD4 (BD International). Platelets were analyzed using the specific phycoethrin-labeled monoclonal antibodies anti- CD61 (Mab) (BD International). All assays included samples to which an irrelevant isotype-matched anti- body (fluorescein isothiocyanate–labeled IgG1 or phycoethrin-labeled IgG1) was added. Fluorescence intensity was analyzed on an Epics XL-MCL Cytom- eter (Coulter Electronics, Hialeah, FL) equipped with an argon laser at 488 nmol/L. For every histogram, 5 3 104 platelets were counted to evaluate the percentage of positive platelets. Antibody reactivity is reported as mean fluo- rescence intensity and percentage (%) of positivity. Interventional study data were analyzed performing a MANOVA with one between-subject factor (group: diabetic vs. nondiabetic patients) and one within-subject factor (time: baseline, 3 days, and 7 days after initiation of treatment). As covariates, we considered the possible random differences in age, sex, BMI, smoking habits, and hypertension between the two groups. Pairwise comparisons were performed using Bonferroni correction. Probability values ,0.05 were regarded as statistically significant. All calculations were made with Statistica 7 software for Windows (StatSoft, Tulsa, OK). Platelet NOX2, TxA2, 8-iso-PGF2a-III, and ROS. Platelet suspension was activated with arachidonic acid (1 mmol/L), and the supernatant was treated with butylhydroxytoluene and stored at 280°C. In some experiments, samples were treated with or without NOX2-blocking peptide (gp91phox ds-tat–blocking peptide, 50 mmol/L) (10 min 37°C) before activation. Study design Thus, the second phase (recruitment) is independent from the agonist (collagen) ini- tially added to the first phase of aggregation because collagen is fully taken up by platelets during the first phase (16). Aggregation of the newly added platelet portion in the presence of an existing aggregate was then measured for 5 min and expressed as a percentage (REC%) of the aggregation that had been initially reached, according to Pignatelli et al. (9). In some experi- ments, PRP samples were incubated (30 min at 37°C) with or without TxA2/ isoprostane receptor inhibitor (SQ29548, 0.1 mmol/L) or control before collagen stimulation. Laboratory analyses. All materials were from Sigma-Aldrich, unless other- wise specified. Blood analyses were performed in a blinded manner. All blood samples were taken after a 12-h fast. Between 8:00 and 9:00 A.M., subjects underwent routine biochemical evaluations, including fasting total cholesterol and glucose. Serum and platelet-poor plasma sampling. After overnight fasting (12 h) and supine rest for at least 10 min, blood samples were taken into tubes containing either 3.8% sodium citrate (ratio 9:1) or anticoagulant-free tubes and centrifuged at 300g for 10 min to obtain supernatant (plasma or serum), which was stored at 280°C until use. Flow cytometry analysis of GpIIb/IIIa activation (pituitary adenylate cyclase-activating peptide receptor 1 [PAC1] binding). Platelet GpIIb/IIIa activation was measured by PAC1 binding on platelet membranes, as previously reported, and expressed as mean fluorescence intensity (17). GpIIb/IIIa activation was analyzed after platelet suspension incubation with scalar doses of 8-iso- PGF2a (25–450 pmol/L) with or without the TxA2/isoprostane receptor antagonist SQ29548 (0.1 mmol/L), ADP (0.01–10 mmol/L) with or without 8-iso-PGF2a (300 pmol/L), and thrombin receptor agonist peptide (TRAP) (0.01–10 mmol/L). Platelet-rich plasma preparation. To obtain platelet-rich plasma (PRP), samples were centrifuged for 15 min at 180g. To avoid leukocyte contamina- tion, only the top 75% of the PRP was collected; leukocyte contamination was verified as reported below. Washed-platelet procedure. Platelet pellets were obtained by double cen- trifugation (2 3 5 min, 300g) of PRP after the addition of acid/citrate/dextrose (1:7 vol/vol) to avoid cell activation during processing. Samples were sus- pended in HEPES buffer in the presence of 0.1% albumin, pH 7.35 (2 3 105 platelets/mL, unless otherwise noted). Statistical analysis. Correlat isoprosta cruitment P , 0.001 isoprostan clinical an tes, and c formed. S which onl (b = 0.158 predicted the other Interven acteristics in Table 2 At base patients s platelet a %; P = 0.6 isoprosta cantly hi patients. aggregati ter aspiri sNOX2-dp creased a (Fig. 2). MANOV pirin trea prostanes (F = 171.6 platelet re interaction lets from blocking p of ROS (2 P = 0.003 pmol/L at P = 0.005 diabetic significant at 3 days, isoprostan at 7 days, The Tx hibited re betic pati patients ( In vitro s Effect of (0.01–10 activation vation by significant GpIIb/IIIa almost wa came dete 8-iso-PGF mmol/L) a to those a ugs (%) 81 74 0.125 88 0 0 1.000 0 ,0.001 6 SD, unless otherwise indicated. ASA, aspirin. *Diabetic vs. nondiabetic patients. †International Federation of Clinical Chemistry and Laboratory Medicine HbA1c iabetes Control and Complications Trial HbA1c (%) 2 2.15] 3 10.929. Correlation analysis. In the overall population, platelet isoprostanes significantly correlated with platelet re- cruitment (r = 0.43, P , 0.001) and sNOX2-dp (r = 0.49, P , 0.001). To further define the predictors of the platelet isoprostane level, multiple regression analysis including clinical and laboratory characteristics, presence of diabe- tes, and concomitant use of aspirin and statins was per- formed. Stepwise linear regression yielded a model in which only T2DM (b = 0.276, P , 0.001), aspirin treatment (b = 0.158, P = 0.009), and sNOX2-dp (b = 0.229, P , 0.001) predicted platelet isoprostane levels, independently from the other included variables. g ( ) Data are means 6 SD, unless otherwise indicated. ASA, aspirin. *Diabetic vs. nondiabetic patients. †International Federation of Clinical Chemistry and Laboratory Medic (mmol/mol) = [Diabetes Control and Complications Trial HbA1c (%) 2 2.15] 3 10.929. Interventional study. Clinical and anthropometric char- acteristics of interventional study participants are reported in Table 2. At baseline, platelets from diabetic and nondiabetic patients showed similar platelet TxB2 values (Fig. 2) and platelet aggregation percentages (94 6 2.2 vs. 97 6 3 LT %; P = 0.640). Conversely, platelet recruitment, platelet isoprostanes, platelet ROS, and sNOX2-dp were signifi- cantly higher in diabetic compared with nondiabetic patients. Platelet TxB2 formation (Fig. 2A) and platelet aggregation (not shown) were significantly inhibited af- ter aspirin intake in both groups. Platelet isoprostanes, sNOX2-dp, platelet ROS, and platelet recruitment in- creased after aspirin intake only in diabetic patients (Fig. 2). ( g ) MANOVA analysis showed a significant influence of as- pirin treatment in diabetic patients on platelet iso- prostanes (F = 11.9, P , 0.001), platelet ROS production (F = 171.6, P , 0.001), sNOX2-dp (F = 32.5, P , 0.001), and platelet recruitment (F = 9.6, P , 0.001). No significant interaction with covariates was found. Incubation of plate- lets from aspirin-treated diabetic patients with the NOX2- blocking peptide was associated with a significant inhibition of ROS (23.3 S.I. at 3 days, P = 0.006, and 27.9 S.I. at 7 days, P = 0.003, respectively) and platelet isoprostane (2134 pmol/L at 3 days, P , 0.001, and 2137 pmol/L at 7 days, P = 0.005) production. Conversely, in aspirin-treated non- diabetic patients, NOX2-blocking peptide had a less significant effect on platelet ROS (20.8 stimulation index at 3 days, P = 0.023, and 20.9 S.I. at 7 days, P = 0.021) and isoprostane (26 pmol/L at 3 days, P = 0.048, and 27 pmol/L at 7 days, P = 0.043) production. Data are means 6 SD, unless otherwise indicated. ASA, aspirin. *Diabetic vs. nond (mmol/mol) = [Diabetes Control and Complications Trial HbA1c (%) 2 2.15] 3 10.92 TABLE 1 Clinical and anthropometric characteristics of study participants , ) The Tx/isoprostane inhibitor SQ29548 significantly in- hibited recruitment of platelets from aspirin-treated dia- betic patients but not from aspirin-treated nondiabetic patients (Fig. 2E and F). 1628 DIABETES, VOL. 61, JUNE 2012 Correlat isoprosta cruitment P , 0.001 isoprostan clinical an tes, and c formed. S which onl (b = 0.158 predicted the other Interven acteristics in Table 2 At base patients s platelet a %; P = 0.6 isoprosta cantly hi patients. aggregati ter aspiri sNOX2-dp creased a (Fig. 2). MANOV pirin trea prostanes (F = 171.6 platelet re interaction lets from blocking p of ROS (2 P = 0.003 pmol/L at P = 0.005 diabetic significant at 3 days, isoprostan at 7 days, The Tx hibited re betic pati patients ( In vitro s Effect of (0.01–10 activation vation by significant GpIIb/IIIa almost wa came dete 8-iso-PGF mmol/L) a to those a ugs (%) 81 74 0.125 88 0 0 1.000 0 ,0.001 6 SD, unless otherwise indicated. ASA, aspirin. *Diabetic vs. nondiabetic patients. †International Federation of Clinical Chemistry and Laboratory Medicine HbA1c iabetes Control and Complications Trial HbA1c (%) 2 2.15] 3 10.929. ABLE 1 linical and anthropometric characteristics of study participants All Diabetic patients (n = 100) All Nondiabetic patients (n = 100) P* ASA treatment ASA treatment No (n = 50) P Yes (n = 50) No (n = 50) P Yes (n = 50) ge (years) 66 6 10 66 6 10 0.934 66 6 10 65 6 14 67 6 10 0.138 63 6 17 0.560 Male (%) 57 58 1.000 56 50 42 0.161 58 0.395 moking habit (%) 18 20 0.795 16 12 18 0.121 6 0.322 MI (kg/m2) 29.4 6 5.0 29.5 6 5.2 0.903 29.3 6 4.8 27.1 6 4.1 27.7 6 4.0 0.116 26.5 6 4.1 ,0.001 revious coronary artery disease (%) 31 16 0.02 46 22 12 0.028 32 0.200 revious stroke (%) 8 8 1.000 8 4 0 0.117 8 0.373 rterial hypertension (%) 81 76 0.308 86 72 82 0.044 62 0.182 otal cholesterol (mg/dL) 188.6 6 43.0 192.8 6 51.0 0.177 182.6 6 36.8 189.8 6 34.2 192.8 6 40.8 0.371 186.7 6 25.9 0.335 DL cholesterol (mg/dL) 50.2 6 19.0 50.1 6 16.6 0.942 50.4 6 21.3 51.5 6 15.7 52.5 6 20.3 0.531 50.5 6 9.0 0.606 riglycerides (mg/dL) 147.8 6 82.1 151.0 6 78.9 0.700 144.6 6 85.8 136.2 6 50.1 131.2 6 50.9 0.332 141.1 6 49.2 0.227 lood glucose (mg/dL) 142.0 6 40.7 139.8 6 32.1 0.537 144.3 6 47.0 93.5 6 5.7 93.4 6 9.5 0.952 93.6 6 13.4 ,0.001 bA1c (%)† 7.1 6 1.3 7.1 6 1.3 0.927 7.1 6 1.2 5.7 6 0.4 5.8 6 0.3 0.168 5.6 6 0.4 ,0.001 tatin therapy (%) 55 54 1.000 56 30 30 1.000 30 0.001 CE/angiotensin receptor blockers (%) 67 60 0.202 74 62 72 0.063 52 0.555 ntidiabetes drugs (%) 81 74 0.125 88 0 0 1.000 0 ,0.001 ata are means 6 SD, unless otherwise indicated. ASA, aspirin. *Diabetic vs. nondiabetic patients. †International Federation of Clinical Chemistry and Laboratory Medicine HbA1c mmol/mol) = [Diabetes Control and Complications Trial HbA1c (%) 2 2 15] 3 10 929 RESULTS The platelet isoprostane 8-iso-PGF2a-III was measured by the enzyme im- munosorbent assay method, as previously described (9), and expressed as picomoles per liter. In a subgroup of patients (n = 20), platelet isoprostane levels also were measured by gas mass chromatography (7890A GC-MS; Agi- lent Technologies) to validate the method (14). Intra- and interassay coef- ficients of variation were 5.8 and 5.0%, respectively. Diabetic patients treated or not with aspirin had similar clinical characteristics (Table 1) and received equally dis- tributed different antidiabetes treatment (data not shown). Platelet TxB2 was significantly lower in aspirin-treated di- abetic patients; conversely, aspirin-treated diabetic patients had higher platelet isoprostanes, platelet recruitment, and NOX2 activation (Fig. 1). To evaluate ROS formation, cell suspension was incubated with 29,79- dichlorofluorescin diacetate (5 mmol/L) for 15 min at 37°C. ROS production was evaluated by flow cytometric analysis and expressed as stimulation index (mean level of fluorescence in stimulated cells/mean level of fluorescence in unstimulated cells; stimulation index), as previously described (11). Intra- and interassay coefficients of variation were 4%. In nondiabetic patients, aspirin treatment significantly lowered platelet TxB2 compared with aspirin-untreated patients. However, no significant differences were found in platelet isoprostanes, platelet recruitment, and sNOX2-dp levels between nondiabetic patients treated or not with as- pirin (Fig. 1). Similar results were observed comparing control subjects with impaired fasting glucose (n = 22) taking (n = 10) or not taking (n = 12) aspirin (data not shown). Platelet aggregation was completely suppressed in aspirin- treated diabetic and nondiabetic patients (not shown). Platelet aggregation. Platelet aggregation was induced by arachidonic acid (1.0 mmol/L) and measured as light transmission (LT%) difference between PRP and platelet-poor plasma, as previously described (11). Platelet recruitment. Platelet recruitment was performed with a method modified from that described by Krötz et al. (15). Collagen (2 mg/mL)-induced platelet aggregation was measured for 10 min, then an equal portion of un- treated platelets was added to each tube, causing a reduction in light trans- mission. This method consists of two phases, the first one depending on the DIABETES, VOL. 61, JUNE 2012 1627 ISOPROSTANES IN ASPIRIN-TREATED T2DM In vitro study Effect of 8-iso-PGF2a on platelet activation. TRAP (0.01–10 mmol/L) dose-dependently increased GpIIb/IIIa activation (Fig. 3). Compared with TRAP, GpIIb/IIIa acti- vation by 8-iso-PGF2a (25–350 pmol/L) was weaker and significantly inhibited by platelet coincubation with SQ29548. GpIIb/IIIa activation by ADP also was dose dependent; it almost was undetectable with doses ,0.1 mmol/L and be- came detectable with values .1 mmol/L. The incubation of 8-iso-PGF2a (300 pmol/L) with lower ADP doses (0.01–0.1 mmol/L) amplified GpIIb/IIIa activation with values similar to those achieved with higher ADP doses. Effect of 8-iso-PGF2a on platelet activation. TRAP (0.01–10 mmol/L) dose-dependently increased GpIIb/IIIa activation (Fig. 3). Compared with TRAP, GpIIb/IIIa acti- vation by 8-iso-PGF2a (25–350 pmol/L) was weaker and significantly inhibited by platelet coincubation with SQ29548. GpIIb/IIIa activation by ADP also was dose dependent; it almost was undetectable with doses ,0.1 mmol/L and be- came detectable with values .1 mmol/L. The incubation of 8-iso-PGF2a (300 pmol/L) with lower ADP doses (0.01–0.1 mmol/L) amplified GpIIb/IIIa activation with values similar to those achieved with higher ADP doses. Influence of platelet size in eicosanoids formation. The top 75% of the PRP of diabetic and control subjects was stratified in two portions (50% upper and 50% lower phases). Arachidonic acid-induced platelet TxB2 and iso- prostanes were measured in both phases. Platelets from diabetic patients were greater in the upper 50% (Fig. 4A g Influence of platelet size in eicosanoids formation. The top 75% of the PRP of diabetic and control subjects was stratified in two portions (50% upper and 50% lower phases). Arachidonic acid-induced platelet TxB2 and iso- prostanes were measured in both phases. Platelets from diabetic patients were greater in the upper 50% (Fig. 4A diabetes.diabetesjournals.org R. CANGEMI AND ASSOCIATES d B) h diff d t t d i th l L k i i T th lit f FIG. 1. TxB2 (A) and isoprostane formation (B), platelet recruitment (REC%) (C), and sNOX2-dp (D) from arachidonic acid–activated platelets in nondiabetic and diabetic patients taking (ASA+) or not (ASA2) 100 mg acetylsalicylic acid daily in the previous month. *P < 0.001. FIG. 1. TxB2 (A) and isoprostane formation (B), platelet recruitment (REC%) (C), and sNOX2-dp (D) from arachidonic acid–activated platelets in nondiabetic and diabetic patients taking (ASA+) or not (ASA2) 100 mg acetylsalicylic acid daily in the previous month. *P < 0.001. and B), whereas no difference was detected in the lower phase. In vitro study The increase in eicosanoids formation between di- abetic patients and control subjects persisted in the two phases, suggesting that platelet size was not a determinant in the above-reported differences (Fig. 4C and D). Leukocyte contamination. To ensure the quality of platelet preparation, samples also were tested for leuko- cyte contamination. The contamination found was ,1% in all samples (an exemplificative image of flow cytometric analysis is shown in Fig. 4A and B) TABLE 2 Clinical and anthropometric characteristics of interventional study participants Diabetic patients Nondiabetic patients P n 18 18 Age (years) 62.3 6 9.4 61.6 6 10.1 0.886 Male (%) 61 56 0.735 BMI (kg/m2) 27.3 6 4.3 27.0 6 2.8 0.789 Arterial hypertension (%) 77 72 0.700 Previous coronary artery disease (%) 0 0 1.000 Previous stroke (%) 0 0 1.000 Total cholesterol (mg/dL) 188.1 6 36.4 186.8 6 37.2 0.385 HDL cholesterol (mg/dL) 51.4 6 16.6 50.8 6 15.8 0.472 Triglycerides (mg/dL) 144.5 6 63.2 134.2 6 51.8 0.239 Blood glucose (mg/dL) 140.2 6 35.4 91.5 6 9.8 ,0.001 HbA1c (%) 7.1 6 1.4 5.5 6 0.4 ,0.001 Statin therapy (%) 11 6 0.546 ACE/angiotensin receptor blockers (%) 67 61 0.729 Antidiabetes drugs (%) 89 0 ,0.001 Data are means 6 SD, unless otherwise indicated. diabetes.diabetesjournals.org DIABETES, VOL. 61, JUNE 2012 1629 DIABETES, VOL. 61, JUNE 2012 1629 ISOPROSTANES IN ASPIRIN-TREATED T2DM CUSSION s study provides evidence that, in diabetic patients ated with low-dose aspirin, platelets overproduce prostanes, an effect that is dependent upon enhanced formation of ROS generated by NOX2 activation. Pla ROS plays a central role in the propagation of pla aggregation by inactivating platelet nitric oxide, relea ADP, and producing isoprostanes (9,18). Previous stu 2. TxB2 (A) and isoprostane (B) formation, sNOX2-dp release (C), and ROS production (D) from arachidonic acid-activated platel diabetic and diabetic patients at baseline and after 3 and 7 days of aspirin (100 mg) treatment. E and F: Platelet recruitment (REC%) in di nondiabetic patients at baseline and after 3 and 7 days of low-dose aspirin treatment in the presence or not of SQ29548 (0.1 mmol/L). *P < FIG. 2. TxB2 (A) and isoprostane (B) formation, sNOX2-dp release (C), and ROS production (D) from arachidonic acid-activated platelets in nondiabetic and diabetic patients at baseline and after 3 and 7 days of aspirin (100 mg) treatment. In vitro study E and F: Platelet recruitment (REC%) in diabetic and nondiabetic patients at baseline and after 3 and 7 days of low-dose aspirin treatment in the presence or not of SQ29548 (0.1 mmol/L). *P < 0.001. formation of ROS generated by NOX2 activation. Platelet ROS plays a central role in the propagation of platelet aggregation by inactivating platelet nitric oxide, releasing ADP, and producing isoprostanes (9,18). Previous studies DISCUSSION The NOX2 inhibitor significantly inhibited platelet ROS and isoprostanes in both diabetic and nondiabetic patients with a reduction of the two variables that was, however, much higher in T2DM patients. Comparing diabetic and nondiabetic patients treated or not with aspirin, a great behavioral difference between the two platelet eicosanoids was detected. Thus, in nondiabetic patients, aspirin inhibited platelet TxB2 without affecting platelet isoprostanes. Conversely, diabetic patients on as- pirin treatment showed platelet TxB2 inhibition but a higher isoprostanes production compared with aspirin-untreated diabetic patients. The increase in isoprostane levels was associated with enhanced platelet NOX2 activation. This reinforces the hypothesis that NOX2 is crucial for the ROS overproduction observed in T2DM (21) and suggests a role for NOX2 in platelet isoprostane overproduction in T2DM. Furthermore, our data reinforce previous reports showing that NOX2 upregulation occurs in diabetes not only in the endothelial cells but also in platelets (22). Hyperglycemia is likely to play a major role as also indicated by the direct correlation between HbA1c and sNOX2-dp, but the underlying mechanism still is controversial (23). A crucial issue of the study was to establish if platelet isoprostane increase observed after aspirin treatment was functionally relevant in diabetic patients (i.e., it was as- sociated with platelet activation). We found no changes in platelet recruitment in nondiabetic but an increase in di- abetic patients. Of note, platelet incubation with the TxA2/ isoprostane receptor antagonist prevented platelet recruit- ment increase in diabetic but not in nondiabetic patients. We speculated that this different behavior could depend upon the 8-iso-PGF2a concentration expressed by platelets from the two groups. To evaluate this, we conducted in vitro experiments to analyze the interplay between isoprostane concentration and GpIIb/IIIa activation (9). A dose-response curve with scalar concentration of 8-iso-PGF2a demon- strated that GpIIb/IIIa was activated in a range of 250–350 pmol/L, which is close to the level of isoprostanes produced by platelets from aspirin-treated diabetic patients (~300 pmol/L) but higher than that expressed by platelets from aspirin-treated nondiabetic patients (~200 pmol/L). To- gether, these data indicate that in diabetic patients aspirin enhances the platelet production of isoprostanes up to functionally relevant concentrations, so enhancing platelet recruitment via GpIIb/IIIa activation. ( ) The different effect of aspirin on TxB2 and isoprostanes likely accounts for the divergent behavior of platelet function tests in aspirin-treated diabetic patients. Although arachidonic acid–induced platelet aggregation was inhibi- ted, platelet recruitment was enhanced. DISCUSSION This study provides evidence that, in diabetic patients treated with low-dose aspirin, platelets overproduce isoprostanes, an effect that is dependent upon enhanced 1630 DIABETES, VOL. 61, JUNE 2012 diabetes.diabetesjournals.org R. CANGEMI AND ASSOCIATES FIG. 3. GpIIb/IIIa activation in platelets treated with a scalar dose of 8-iso-PGF2a (25–450 pmol/L) treated or not with SQ29548 (0.1 mmol/L), ADP (0.01–10 mmol/L) treated or not with 8-iso-PGF2a (300 pmol/L), and TRAP (0.01–10 mmol/L) (values are mean of experiments performed in five healthy subjects). *P < 0.001. FIG. 3. GpIIb/IIIa activation in platelets treated with a scalar dose of 8-iso-PGF2a (25–450 pmol/L) treated or not with SQ29548 (0.1 mmol/L), ADP (0.01–10 mmol/L) treated or not with 8-iso-PGF2a (300 pmol/L), and TRAP (0.01–10 mmol/L) (values are mean of experiments performed in five healthy subjects). *P < 0.001. FIG. 4. Platelet sizing and leukocyte contamination in the top 75% of the FIG. 4. Platelet sizing and leukocyte contamination in the top 75% of the PRP in nondiabetic patients (A) and in diabetic patients (B). Evaluation of TxB2 (C) and isoprostanes (D) production in the stratified two portions of the top 75% of the PRP (50% upper phase, 50% bottom phase; up and down, respectively). Values are means of experiments performed in five nondiabetic patients and five diabetic patients. *P < 0.001. FIG. 3. GpIIb/IIIa activation in platelets treated with a scalar dose of 8-iso-PGF2a (25–450 pmol/L) treated or not with SQ29548 (0.1 mmol/L), ADP (0.01–10 mmol/L) treated or not with 8-iso-PGF2a (300 pmol/L), and TRAP (0.01–10 mmol/L) (values are mean of experiments performed in five healthy subjects). *P < 0.001. FIG. 4. Platelet sizing and leukocyte contamination in the top 75% of the PRP in nondiabetic patients (A) and in diabetic patients (B). Evaluation of TxB2 (C) and isoprostanes (D) production in the stratified two portions of the top 75% of the PRP (50% upper phase, 50% bottom phase; up and down, respectively). Values are means of experiments performed in five nondiabetic patients and five diabetic patients. *P < 0.001. showed that diabetic patients have a systemic isoprostane overproduction (19,20), but the role of platelets has never been examined. The cross-sectional study showed enhanced platelet isoprostane formation and platelet recruitment along with NOX2 activation in diabetic versus nondiabetic patients without aspirin treatment. isoprostane overproduction. This hypothesis was supported by ex vivo experiments where platelets were incubated with a specific NOX2 inhibitor. diabetes.diabetesjournals.org DIABETES, VOL. 61, JUNE 2012 1631 DISCUSSION To further support these findings, we measured platelet TxB2 and isoprostanes in diabetic and nondiabetic patients before and after short- term aspirin treatment. Although both groups disclosed a similar platelet TxB2 inhibition, platelet isoprostanes showed a different behavior. In nondiabetic patients, platelet isoprostane formation was not influenced by as- pirin, and in diabetic patients isoprostanes increased 3 and 7 days after aspirin treatment. This change in isoprostane levels was coincidental with enhanced platelet ROS for- mation and NOX2 activation, suggesting a cause-and-effect relationship between NOX2-generated ROS and platelet The study has pathophysiologic and clinical implica- tions. The upregulation of NOX2 in T2DM has important functional consequences, as it is associated with enhanced ROS and isoprostane production and ultimately with platelet activation. This finding supports and extends previous data indicating that platelet ROS formation are implicated in the DIABETES, VOL. 61, JUNE 2012 1631 ISOPROSTANES IN ASPIRIN-TREATED T2DM thrombus formation process (18) and provides further in- sight into the association between T2DM and thrombosis. subjects with and without prior myocardial infarction. N Engl J Med 1998; 339:229–234 subjects with and without prior myocardial infarction. N Engl J Med 1998; 339:229–234 3. Sander D, Kearney MT. Reducing the risk of stroke in type 2 diabetes: pathophysiological and therapeutic perspectives. J Neurol 2009;256:1603– 1619 The upregulation of NOX2-generated ROS is likely to have a negative impact on the antiplatelet effect of aspirin in diabetic patients. In T2DM, the inhibition of COX1 by aspirin could favor a shift of arachidonic acid toward other metabolic pathways with ensuing enhancement of ROS- mediated nonenzymatic oxidation of arachidonic acid and subsequent isoprostane overproduction. Accordingly, pre- vious studies showed that arachidonic acid activates NADPH oxidase via interaction with p47phox, a subunit of NADPH oxidase (24,25). 4. FitzGerald GA. Mechanisms of platelet activation: thromboxane A2 as an amplifying signal for other agonists. Am J Cardiol 1991;68:11B– 15B 5. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002;324: 71–86 6. Ferroni P, Basili S, Falco A, Davì G. Platelet activation in type 2 diabetes mellitus. J Thromb Haemost 2004;2:1282–1291 7. De Berardis G, Sacco M, Strippoli GF, et al. Aspirin for primary prevention of cardiovascular events in people with diabetes: meta-analysis of random- ised controlled trials. DISCUSSION BMJ 2009;339:b4531 Such effect represents a plausible explanation for the lower efficacy of aspirin in preventing cardiovascular events in diabetic patients, as the divergent effect on platelet TxA2 and isoprostanes has a different impact on platelet activa- tion. The suppression of the early phase of platelet aggre- gation by TxA2 may be counterbalanced by the increased propagation of platelet aggregation elicited by isoprostane overproduction. The inhibition of this last effect by in vitro addition of TxA2/isoprostane receptor antagonist suggests a potential benefit of drugs that antagonize TxA2/isoprostane receptors in patients with T2DM. Prevention of isoprostane overproduction via inhibition of NOX2-generated ROS is another attractive option to be considered. Statins have been reported to downregulate systemic isoprostanes with a mechanism that may involve inhibition of NADPH oxi- dase (13,26). Therefore, it could be interesting to examine if statins improve the antiplatelet effect of aspirin via in- hibition of platelet isoprostanes. 8. Violi F, Sanguigni V, Carnevale R, et al. Hereditary deficiency of gp91 (phox) is associated with enhanced arterial dilatation: results of a multi- center study. Circulation 2009;120:1616–1622 9. Pignatelli P, Carnevale R, Di Santo S, et al. Inherited human gp91phox deficiency is associated with impaired isoprostane formation and platelet dysfunction. Arterioscler Thromb Vasc Biol 2011;31:423–434 10. Sánchez-Borges M, Capriles-Hulett A, Caballero-Fonseca F. NSAID- induced urticaria and angioedema: a reappraisal of its clinical manage- ment. Am J Clin Dermatol 2002;3:599–607 11. Pignatelli P, Lenti L, Sanguigni V, et al. Carnitine inhibits arachidonic acid turnover, platelet function, and oxidative stress. Am J Physiol Heart Circ Physiol 2003;284:H41–H48 12. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2011;34(Suppl. 1):S62–S69 13. Pignatelli P, Carnevale R, Cangemi R, et al. Atorvastatin inhibits gp91phox circulating levels in patients with hypercholesterolemia. Arterioscler Thromb Vasc Biol 2010;30:360–367 14. Lee CY, Huang SH, Jenner AM, Halliwell B. Measurement of F2-isoprostanes, hydroxyeicosatetraenoic products, and oxysterols from a single plasma sample. Free Radic Biol Med 2008;44:1314–1322 p p A limitation of the study is the lack of analysis of aspirin compliance in the cross-sectional study. However, both diabetic and control groups had similar values of platelet TxB2, indicating that a scarce adherence to aspirin treatment, if any, was well balanced between the two. Furthermore, platelet function analysis was performed in vitro and ex vivo, which may not reflect platelet activation in vivo. DISCUSSION Pharmaco- logic study with an inhibitor of isoprostane receptors could be useful to explore our study hypothesis in vivo. 15. Krötz F, Sohn HY, Gloe T, et al. NAD(P)H oxidase-dependent platelet su- peroxide anion release increases platelet recruitment. Blood 2002;100:917– 924 16. Santos MT, Valles J, Marcus AJ, et al. Enhancement of platelet reactivity and modulation of eicosanoid production by intact erythrocytes: a new approach to platelet activation and recruitment. J Clin Invest 1991;87:571– 580 17. Pignatelli P, Pulcinelli FM, Celestini A, et al. The flavonoids quercetin and catechin synergistically inhibit platelet function by antagonizing the in- tracellular production of hydrogen peroxide. Am J Clin Nutr 2000;72:1150– 1155 In conclusion, we provide evidence that, in T2DM patients, low-dose aspirin enhances platelet isoprostanes as a con- sequence of NOX2-generated ROS upregulation. This effect mitigates the antiplatelet effect of aspirin and may account for its lower clinical efficacy in T2DM compared with other atherosclerotic settings. 18. Krötz F, Sohn HY, Pohl U. Reactive oxygen species: players in the platelet game. Arterioscler Thromb Vasc Biol 2004;24:1988–1996 19. Davì G, Ciabattoni G, Consoli A, et al. In vivo formation of 8-iso-prostaglandin f2alpha and platelet activation in diabetes mellitus: effects of improved metabolic control and vitamin E supplementation. Circulation 1999;99: 224–229 20. Sampson MJ, Gopaul N, Davies IR, Hughes DA, Carrier MJ. Plasma F2 isoprostanes: direct evidence of increased free radical damage during acute hyperglycemia in type 2 diabetes. Diabetes Care 2002;25: 537–541 ACKNOWLEDGMENTS This study was supported by a grant from “Fondazione Roma” 2009. 21. Guzik TJ, Mussa S, Gastaldi D, et al. Mechanisms of increased vascular superoxide production in human diabetes mellitus: role of NAD(P)H oxi- dase and endothelial nitric oxide synthase. Circulation 2002;105:1656–1662 No potential conflicts of interest relevant to this article were reported. R.Car., C.N., and M.P. researched data. R.Can. and P.P. wrote the manuscript and researched data. F.A. and D.L. reviewed and edited the manuscript. S.B. researched data. F.V. designed the study and wrote the manuscript. F.V. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. R.Car., C.N., and M.P. researched data. R.Can. and P.P. wrote the manuscript and researched data. F.A. and D.L. reviewed and edited the manuscript. S.B. researched data. F.V. designed the study and wrote the manuscript. F.V. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. 22. Ferreiro JL, Angiolillo DJ. Diabetes and antiplatelet therapy in acute cor- onary syndrome. Circulation 2011;123:798–813 23. Giacco F, Brownlee M. Oxidative stress and diabetic complications. Circ Res 2010;107:1058–1070 24. Ayilavarapu S, Kantarci A, Fredman G, et al. Diabetes-induced oxidative stress is mediated by Ca2+-independent phospholipase A2 in neutrophils. J Immunol 2010;184:1507–1515 25. Shiose A, Sumimoto H. Arachidonic acid and phosphorylation synergisti- cally induce a conformational change of p47phox to activate the phagocyte NADPH oxidase. J Biol Chem 2000;275:13793–13801 1. Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epide- miology, pathophysiology, and management. JAMA 2002;287:2570–2581 g g g 2. Haffner SM, Lehto S, Rönnemaa T, Pyörälä K, Laakso M. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic 1. Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epide- miology, pathophysiology, and management. JAMA 2002;287:2570–2581 2. Haffner SM, Lehto S, Rönnemaa T, Pyörälä K, Laakso M. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic diabetes.diabetesjournals.org REFERENCES 26. De Caterina R, Cipollone F, Filardo FP, et al. Low-density lipoprotein level reduction by the 3-hydroxy-3-methylglutaryl coenzyme-A inhibitor sim- vastatin is accompanied by a related reduction of F2-isoprostane forma- tion in hypercholesterolemic subjects: no further effect of vitamin E. Circulation 2002;106:2543–2549 1. Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epide- miology, pathophysiology, and management. JAMA 2002;287:2570–2581 2. Haffner SM, Lehto S, Rönnemaa T, Pyörälä K, Laakso M. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic DIABETES, VOL. 61, JUNE 2012 1632 diabetes.diabetesjournals.org

https://openalex.org/W4315607362

https://zenodo.org/records/7524704/files/Numerical%20solutions%20for%20crack%20problems%20during%20elastomer%20forming.pdf

English

Numerical solutions for crack problems during elastomer forming

Eastern-European journal of enterprise technologies

cc-by

Engineering technological systems: Reference for Chief Technology Specialist at an industrial enterprise Engineering technological systems: Reference for Chief Technology Specialist at an industrial enterprise UDC 621 DOI: 10.15587/1729-4061.2022.268285 UDC 621 UDC 621 Generically, thermoset elastomers are often referred to as rubber. It is characterized by the chemical bonding between polymer chains. One of the important problems that plague elas­ tomer manufacturing and rubber parts under service is cracks. Predicting the main factors affecting crack propagation tra­ jectories during forming and after curing time is an important challenge. For this purpose, numerical analysis was implemen­ ted by using the commercial ABAQUS/CAE software package. A three-dimensional model was established to predict the impor­ tant factors that affect this process. During the analysis, the effect of forming velocity and the amount of kinetic energy on the distortion of rubber material and crack propagation is explored in detail by using different forming punch velocities. The drop velocity of the upper insert (punch) on the rubber pad was taken as 10 m/s, 7 m/s, and 5 m/s, respectively. Consequently, while each forming velocity will generate different kinetic energy between the interaction surfaces, the change in crack beha­vior and the normal stress can be monitored in different positions. As a result, among these velocities, it was found that the low forming velocity of the upper insert (punch) is better than the others in forming rubber where cracks and distortions were at minimum values. Also, the amount of kinetic energy is low enough in the case of low speeds and can affect the results significantly. In addition, it was found that the generated stresses have a sig­ nificant impact on the crack development in a specific area, espe­ cially near the fillets and sharp edges. It was concluded that calculating the parameters affecting the crack growth and pre­ dicting the crack propagation trajectories using the finite element method is a significant method for predicting and solving crack problems before tool fabrication NUMERICAL SOLUTIONS FOR CRACK PROBLEMS DURING ELASTOMER FORMING B a d r K a m o o n Corresponding author Department of Machinery and Equipment Engineering Technical College Al-Mussaib* E-mail: [email protected] S a l a m O . D a h i Department of Machinery and Equipment Engineering Technical College Al-Mussaib* H a m z a h K a d h i m Department of Mechanical Karbala Technical Institute* *Al-Furat Al-Awsat Technical University Imam Ali Bridge, Najaf, Iraq, 54003 Keywords: rubber crack, rubber pad, numerical analysis, elastomer forming, crack propagation, ABAQUS How to Cite: Kamoon, B., O. Dahi, S., Kadhim, H. (2022). Numerical solutions for crack problems during elasto­ mer forming. Eastern-European Journal of Enterprise Technologies, 6 (1 (120)), 83–90. doi: https://doi.org/10.15587/ 1729-4061.2022.268285 1. Introduction Therefore, research on the development of this direction is concentrated on the problem under the current study to determine better visualization and recommendations. The behavior of elastomer material during forming varies according to many factors such as shape geometry, physical properties, and loading conditions. This variation is difficult to predict due to the nonlinearity behavior of this material. Consequently, building new knowledge and a clear vision by adopting the numerical technique will lead to minimizing this problem and prevent product failure. Cracking in parts under service is a great challenge in the industrial field due to the main impact of crack beha­ vior on part performance. Cracks start normally with mi­ cro-cracks and develop with elevated working temperature, load change, and other factors. Many researchers spot the light on the mechanism of crack initiation and development following different methods and analyses. Some studies used a traditional fatigue method to deal with this type of problem, and others adopted fracture mechanism theories to estimate and forecast the trajectories of crack growth rate. According to much literature and engi­ neering database in this field, the parameters related to crack behavior at different steps such as preparation, manufactur­ ing, and working conditions were determined based on the infield engineering data. Copyright © 2022, Authors. This is an open access article under the Creative Commons CC BY license 2. Literature review and problem statement Rubber parts are used in many industrial applications such as automobiles, hydrants, gaskets valves, and fittings. The major service problems of elastomers are that they are quickly degraded more than expected creating many diffi­ culties for utilities to cost and programs to plan preventive maintenance. Elastomers are normally characterized by their low tensile strength (1–10 MPa), high extensibility, and electrical insulation. Thermoset elastomers can be highly af­ fected by weak stress and return to their original shape after the stress is removed. The problem of estimating the imposed Consequently, these studies almost depend on the expe­ rience that performs poorly and only provides rough solu­ tions, while estimation according to accurate analysis and up-to-date calculation by numerical investigation are better. Since the elastomer industry is developing rapidly, numerical analysis is becoming the best solution to predict and deter­ mine crack problems before the component failed. 83 Copyright © 2022, Authors. This is an open access article under the Creative Commons CC BY license 6/1 ( 120 ) 2022 Eastern-European Journal of Enterprise Technologies ISSN 1729-3774 that rubber materials have a larger crack propagation energy when the viscoelastic peak is low. that rubber materials have a larger crack propagation energy when the viscoelastic peak is low. strain that leads to cracks propagating under different tem­ peratures was investigated. However, the effects of strain rate on the value of the critical length during crack propagation still need to be solved [1, 2]. The method in [14] attempts to provide a suitable pre­ diction of the tearing energy by calculating the dissipated energy due to different inelastic processes. The results show that the classical method overestimates the critical tearing energy by approximately 15 %. In [3, 4], the elastomer membrane under tension and the propagation of dynamic cracks in elastomer parts, and the crack speed measured along the crack zone were investigated analytically. The methods involve stretching an elastomer part and then initiating a small crack. It was found that the crack propagation zone follows the stretch direction of the crack. This method is limited to the analysis of cracks with constant crack speed and direction. It was found that the crack propagation energy in front of viscoelastic deformations of rubber and the role of this pro­ pagation for sliding friction with rolling resistance and rubber typically depend on the nature of the surface itself [15, 16]. 2. Literature review and problem statement The above sources deal with many cases of crack propa­ gation trajectories, and each one has dealt with different aspects. The majority of these sources are involved with crack development under different loading and working con­ ditions. It will be more valuable to explore other important effects such as the effects of forming velocity and kinetic ener­gy on the crack trajectories. The current research aimed to describe the cracks’ behaviors under different forming velocities. The effects of speed change between the contact surfaces and layer thickness on crack development were in­ vestigated. The ABAQUS/CAE finite element software was adopted to simulate the growth of these cracks at different punch-forming velocities. Some sidewall cracking may be linked to the generation of stress during rubber shrinkage and early thermal contraction. These problems may cause a rapid increase in crack propagation and need to be solved. The intensity of stresses and strains was investigated to de­ scribe the crack extension while working. The works [5, 6] show that the most popular approach to simulate the cracks’ behavior during their life cycle is the fi­ nite element method. Many finite element codes can describe the hyperplastic behavior of rubber as a nonlinear elasticity behavior. Large deformations and large displacements occur during the forming process. By using finite element software, the growth and behavior of cracks can be estimated with less time and minimum cost. Graphical demonstrations and numerical computations are always used successfully to ob­ serve the effect of the geometry conditions, elastic constants ratio, and stresses on the stress intensity factors at the crack tips. The approach here is used for the two-dimensional crack problems and cannot be extended to three-dimensional ones due to geometry constraint problems. In [7], LS-DYNA finite element software is used for mo­ deling and simulating a rubber pad in the stamping process. The forming parameters such as rubber pad hardness, model, and friction coefficient of the blanks were successfully deter­ mined, and the result was 50 % CPU time-saving. However, some problems in this method are linked to the appearance of negative volume due to the formation of small-size elements. 3. The aim and objectives of the study The aim of the current study is to identify regularities of the crack development trajectories during elastomers form­ ing by using a numerical analysis method. In [8], simulations and experiments were carried out by using a family group of a matrix of the polydimethylsilo­ xane (PDMS) embedded elastomer with two spherical glass beads. The aim is to generate high triaxial stress. The result shows that the increase in loading will lead to an increase in stress on the deformed surface and cracks may spread out to critical points. Also, the micro-crack may grow and transit at some critical points. The objectives below have been adopted to achieve the aim: d h k d d ff f The objectives below have been adopted to achieve the aim: – to predict the crack’s trajectories under different form­ ing velocities, and explore the kinetic energy required for this process by developing a finite element model; – to estimate the cause of stresses affecting the mechani­ cal behavior of the rubber bushing subject during the forming process. In [9], ABAQUS/Standard finite-element code is used to analyze and simulate semi-elliptical surface cracks for compo­ site aluminum plates. The uniaxial tensile load was applied on the repaired cracked plate to evaluate the failure strength and stress intensity factors (SIF). The results show that the stress concentration depends on both the patch thickness and crack depth. However, increasing layer thickness will increase the possibility of surface cracks with depths greater than expected. 4. 2. Finite Element Process In this work, the numerical analyses were performed by us­ ing commercial FEM software (ABAQUS/CAE) to establish analysis and simulation of this model. The parts are modeled by using CATIA V5, and then exported to ABAQUS/CAE for the simulation process. The completed procedure and steps of analysis are shown by the flowchart in Fig. 5. Fig. 2. Product under analysis in three dimensions Table 1 Mechanical properties of rubber No. Type A B Density (kg/m3) Poisson’s Ratio 1 Shore 90 2.8 0.71 1.1×10–9 0.5 The forming process of the rubber part starts with filling the upper insert cavity with the rubber blanks material. This rubber consists of some components such as base rubber, cur­ ing agent, and filler like carbon black. Also, some components may be added like adhesion agents, chemical additives, and antioxidants. The steel type used for forming parts (punch and die) is tool steel with a Young’s modulus of elasticity 206 GPa and 0.3 Poisson’s ratio. Making a contact between rubbers and insert surfaces is the initial step in the simulation process. The gap between the upper and lower inserts represents the product thickness. The contact step between the surfaces is illustrated in Fig. 6. Making a contact between rubbers and insert surfaces is the initial step in the simulation process. The gap between the upper and lower inserts represents the product thickness. The contact step between the surfaces is illustrated in Fig. 6. During forming, the rubber folds will be distributed in the gap between the upper and lower inserts to form the final product shape The tooling for this manufacturing process encompas­ ses many important parts. The main parts are the upper core (punch) and lower insert (die). During forming, the rubber folds will be distributed in the gap between the upper and lower inserts to form the final product shape. To avoid and prevent any interference between the parts during forming, and to decrease the error and time during the simulation, all these mold parts and assembly will be modeled according to their actual dimensions and tolerances. This modeling will support a better understanding of building the simulation process according to actual conditions and enhance the final results. The CATIA V5 R16 software is adopted for modeling this process. 4. 1. Part geometry and material properties h h bb b h Mold assembly parts during forming Fig. 4. Cross-section of the assembly tools during the part ejection Fig. 1. The rubber product including all dimensions Fig. 1. The rubber product including all dimensions Fig. 2. Product under analysis in three dimensions Fig. 2. Product under analysis in three dimensions Fig. 4. Cross-section of the assembly tools during the part ejection For accurate results, only the interaction parts (upper insert, lower insert, and rubber material) will be simulated in the assembly. 4. 1. Part geometry and material properties h h bb b h In this research, rubber bushing geometry is built by CAD (3-D) using the CATIA V5 design program in order to process and develop the finite element model. Fig. 1, 2 illustrate the geometry of a rubber pad used for the analysis process in 2-D and 3-D, respectively. In [10], a nonlinear finite element analysis is used to model and analyzed dynamic crack propagation in rubber. As a result, the influence of some parameters such as surface energy, viscoelastic dissipation, and inertia effects has been investigated. The fracture energy in the vicinity and sur­ rounding of the crack tip is a problem to be solved. The rubber type used in this research is MAT27, and the main mechanical properties of this type of rubber are listed in Table 1. Shore units are normally used in measuring rubber hardness. The papers [11, 12] show that by identifying the problems of potential forming, it is possible to reduce the time, cost, and cycle of product development by using numerical simu­ lation (ABAQUS). Also, the part quality can be improved during manufacturing to ensure appropriate processes. Mathematically, the strain energy density ( ) during rubber forming can be defined by the following equations: ¬ , = − ( )+ − ( ) − ( )+ − ( ) + A K B K C K D K 1 2 3 2 3 2 3 3 1 1 (1) (1) The paper [13] approved that the viscoelastic modulus highly affects crack propagation in rubber. The nucleation of cracks is stress-aided, so the crack nucleation rate will increase if the viscoelastic loss peak is shifted. It was found where C = 0.5A+B, K1, K2, K3 is an invariant of right Cauchy Tensor C; 84 Engineering technological systems: Reference for Chief Technology Specialist at an industrial enterprise D = A(5v–2)+B(K2v–5)/2(1–2v), (2) where v is Poisson’s ratio. Fig. 1. The rubber product including all dimensions Fig. 3. Mold assembly parts during forming Fig. 4. Cross-section of the assembly tools during the part ejection Fig. 3. Mold assembly parts during forming Fig. 4. Cross-section of the assembly tools during the part ejection 2) Fig. 3. Mold assembly parts during forming Fig. 4. Cross-section of the assembly tools during the part ejection D = A(5v–2)+B(K2v–5)/2(1–2v), here v is Poisson’s ratio. Fig. 1. The rubber product including all dimensions D = A(5v–2)+B(K2v–5)/2(1–2v), (2) where v is Poisson’s ratio. Fig. 3. 4. 2. Finite Element Process Based on elastic layered system theory, the boundary conditions involve a full constraint of the lower insert to pre­ vent it from any motion in any direction. The boundary con­ ditions were set as no rotation and no displacement at some other supporting parts. Also, a uniform pressure of 1000 MPa will be applied to the upper insert. Fig. 7 shows the boundary conditions and applied pressure. In terms of parts deformation, the rubber sheet is classi­ fied as a deformable part, while the upper and lower inserts are classified as a rigid body. The relationship between the rigid body and the deformable part is represented as master and slave. The procedure of forming the rubber sheet includes three sequence steps. These steps involve aligning the rubber material between the punch and die, moving the punch down the blank to press the material inside the die cavity, and finally ejecting the product. These steps are illustrated in Fig. 3, 4 below. 85 Eastern-European Journal of Enterprise Technologies ISSN 1729-3774 6/1 ( 120 ) 2022 Start Parts creation Materials definition Parts assembly and instances creation Parts meshing Interaction, Boundary condition, Apply load Step definition Create job and job submission to analysis End Output results Fig. 5. Flowchart of the completed procedure and steps of analysis Fig. 6. Contact behavior between surfaces before forming Start Parts creation Materials definition Parts assembly and instances creation Parts meshing Interaction, Boundary condition, Apply load Step definition Create job and job submission to analysis End Output results Fig. 5. Flowchart of the completed procedure and steps of analysis Start Parts creation Materials definition Parts assembly and instances creation Parts meshing Interaction, Boundary condition, Apply load Step definition Create job and job submission to analysis End Output results Fig. 5. Flowchart of the completed procedure and steps of analysis it is modeled as a hyperelastic model, which is defined as a strain energy function. As a large strain occurs, the tetrahe­ dral mesh type is used for meshing the rubber parts. Other parts are meshed by hexahedral mesh type to yield reasonable results. The finite element mesh of the parts is shown in Fig. 8. Calculating the number and position of elements is an important step that allows estimating the values of strain and stress at each element and leads to accurate results. Labeling all the important elements will enable us to extract accurate results. 4. 2. Finite Element Process In this model, the total elements of the whole model are 17,676 elements. Fig. 8. Parts mesh and elements labeling representation Fig. 5. Flowchart of the completed procedure and steps of analysis Fig. 5. Flowchart of the completed procedure and steps of analysis Fig. 8. Parts mesh and elements labeling representation In Fig. 8 E-model, the hyperelastic rubber component is modeled using the CAX4H element used in the analysis of this model, and the mesh size of this highly non-linear rubber component model is between 1.4 mm to 1.6 mm. These types of elements are suitable for simplifying the boundary condi­ tions and loading. Fig. 6. Contact behavior between surfaces before forming Determination of the number and position of elements is an important step that allows estimating the values of strain and stress at each element and leads to accurate results. In this mo­del, the total elements of the whole model are 17,676 elements. The FRICTION, ROUGH option is normally used in ABAQUS to represent the frictional behavior as sticking friction. During contact and interference between rigid bo­ dies (masters) and deformable parts (slaves), the assumption was that the rigid body has a lubricated surface and the fric­ tion coefficient between them is a small value (0.05). Fig. 6. Contact behavior between surfaces before forming Fig. 7. The boundary conditions and applied pressure representation Rubber behavior during forming is classified as non­ linear elastic. Rubber folds will be distributed under the applied pressure to the vicinity cavity, which equals product thickness. The cavity gap between the upper and lower in­ serts will be filled with rubber to form the final product. The nonlinearity behavior of rubber is a function of strain energy and is referred to the properties of this material as an elastic material. 5. Results of numerical analysis during elastomer forming Fig. 7. The boundary conditions and applied pressure representation 5. 1. Results of rubber formation under different form­ ing velocities The upper part will move downward to be in contact with the rubber and stop after the part forming process is completed. The upper part will move up toward the cavity and press the material to the desired shapes before stopping. A feasible model was established, and the ABAQUS finite element software was used for numerical analysis. In the cur­ rent work, the aim is to analyze the impact of deformations on the initiation of crack development in rubber material. During fracture initiation, critical tension occurs driven by the strain softening. Failure and crack description in simu­ Rubber is a very soft material and considered a deform­ able material (fully hyperelastic) with a high strain level so 86 Engineering technological systems: Reference for Chief Technology Specialist at an industrial enterprise lations that were plugged by ABAQUS was performed using fine meshes to investigate the influences. speed will allow facilitating the process and obtaining a high­ ly accurate product. With ABAQUS/CAE, the forming process is simulated in two steps. The high values of strain generated due to the applied pressure will decrease or be prevented by increasing the contact zones between the insert surfaces and the rubber pad. At the start of the forming process, the strain will reach its peak values due to the movement of the highly energized particles and highly dislocation forces, espe­ cially between the contact forming surfaces. The high strain values generated due to the high kinetic energy will increase the contact zones between the insert sur­ faces and the rubber pad. At the start of the forming process, the strain will reach its peak values due to the movement of the highly energized particles and highly dislocation forces, especially between the contact forming surfaces. Fig. 9. The contour plot of forming the rubber pad at high punch drop Fig. 10. Energy-time plot during the first step Fig. 11. The contour plot of the forming at high punch drop Fig. 12. Energy-time plot at 7 m/s punch speed Fig. 9. The contour plot of forming the rubber pad at high punch drop Fig. 10. Energy-time plot during the first step y g The shear stress, normal stress, and stress intensity of the crack tip under different load­ ing conditions were estimated. Also, at the crack tip, the shear stress and principal stress intensity can reflect the cracking situation. 5. 1. Results of rubber formation under different form­ ing velocities In the first step of the analysis, the upper insert force is applied. The punch (rigid body) is moved down 70 mm in the second step of the analysis, and the peak velocity occurs in the middle of the time period. Because the fundamental mode of vibration is low, and the rubber blank is very flexible, the simu­ lation would take more time to obtain the analysis results. Different forming velocities were used as 10 m/s, 7 m/s, and 5 m/s, re­ spectively. When the punch drops from the hydraulic press with a velocity of 10 m/s, it was found that the distortion zone of rubber is too large, with many visible cracks in the adjacent zones (red zones). This can be at­ tributed to the high amount of kinetic energy applied to the model. This case is illustrated in Fig. 9 below. The kinetic energy in rubber elements during folding for the above step is too high as shown by the energy-time plot in Fig. 10. Fig. 10. Energy-time plot during the first step Based on the above results, the punch drop velocity should be less valued and taken to be 7 m/s. However, the distortion occurred but the folding of rubber layers is less than the previous one. In addition, it can be seen that there are fewer cracks in adjacent zones than expected (fewer red color zones). The contour plot in Fig. 11 shows that the rubber product started to relax from distortion and changed to green color, and only a small zone is affected by distortion with minimum cracks. Fig. 11. The contour plot of the forming at high punch drop In this forming step, the amount of kinetic energy in rubber elements will be less than in the previous step as shown by the ener­ gy-time plot in Fig. 12. Fig. 11. The contour plot of the forming at high punch drop In the third and last forming step using a punch drop velocity of 5 m/s, the rubber layers are folded smoothly, and the amount of kinetic energy is less than in the previous steps. The product is expected completed uniformly without distortion and cracks. This forming step is illustrated by the contour plot in Fig. 13. Fig. 14 shows that a minimum of kinetic energy is consumed at this forming speed. 5. 1. Results of rubber formation under different form­ ing velocities The contours of axial stresses of the deformed mesh show that the maximum stresses will occur in the sharp and the fillet zones (radius) and along the parting line between the upper and lower inserts. The generation of stresses during the forming process is illustrated by the contour graph in Fig. 15. Fig. 16. Elements and nodes topology along the fillet and surface curvature After upsetting the product, the curing starts and the part begins to shrink to the actual dimensions. However, the shrinkage in the small value dimension is not visible, especially at the wall thickness. Sometimes, and due to many reasons such as non-homogenous pressure distribution, the thickness in fillets and corners will be greater than in other zones. This will lead to an increase in the thermal stresses due to high-tem­ perature concentration in these zones. The dimension stability is one of the major problems in forming a hyperelastic material. Fig. 15. Сontour graph of generated stress during the forming process One of the most possible solutions for this problem is to add some additives such as fillers or fiber to increase the binding forces between material crystals. 5. 1. Results of rubber formation under different form­ ing velocities The results above reveal the important effect of the forming speed on cracks and non-uniform distortion generated in rubber products. Choosing the optimum forming Fig. 12. Energy-time plot at 7 m/s punch speed 87 6/1 ( 120 ) 2022 Eastern-European Journal of Enterprise Technologies ISSN 1729-3774 Fig. 13. The contour plot of the last forming at 5 m/s punch drop Fig. 14. Energy-time plot at 5 m/s punch speed The pressure adjustment and calibration during action onto the rubbers will prevent high strains by increasing the contact area between the blanks and inserts. Large defor­ mations can be undergone by softer materials and lead to generating high strain-stress concentrations. Material orientation and dis­ tribution during the final step of simulation in the forming process under some variables like pressure and temperature are like differ­ ent surfaces meeting together to contact with each other and generating a single surface. The imposed constraints on the part thick­ ness will build connecting elements and nodes along the fillet radius and surface curvature according to the shape topology to form the final thickness, as shown in Fig. 16. Fig 14 Energy-time plot at 5 m/s punch speed In the numerical model, the position of the crack threshold was obtained and evaluated. Crack propagating near the fillet zones (R) and the sharp edge is developing steadily due to non-uniform stretch forming in this area and this causes high stress concentrations. Larger strains will occur during the crack. This strain concentration is important because the material structure will evolve with strain development. It is possible to model the non­ linearities of the physical model as boundary nonlinearities like friction and contact, material nonlinearity, and large displacement effects. Fig. 14. Energy-time plot at 5 m/s punch speed 5. 2. Results of stress generation and mechanical be­ havior of rubber bushing during forming Fig. 16. Elements and nodes topology along the fillet and surface curvature The distortion and contact instabilities will take place during the rubber pad forming due to large strain occurring in the simulation process. The shear stress, normal stress, and stress intensity of the crack tip under different loading conditions were estimated. Also, at the crack tip, the shear stress and prin­ cipal stress intensity can reflect the cracking situation. 6. Discussion of the analysis results of forming the elastomer pad The low-speed forming will lead to smooth folding of the rubber layers with less generated energy, and the possibility of crack generation will decrease. 2. The mechanical behavior of rubber material during forming is a complex process. During punch drop, the ele­ ments near the fillets and sharp zones may be suffering from the lack of enough compression and filling according to the forming speed. This leads to making these areas weak and highly stressed zone, and the possibility of crack initiation is 25 % higher than in the case of uniform pressure distribution. These non-uniform dislocations are considered the main reason for forming some weak points that develop and spread out with time to initiate a crack. Also, it is a high impact on the mechanical properties of the final product. The load should be distributed uniformly to prevent cracks due to the high impact of non-uniform pressure distribution on crack initiation and development. In the other simulated tests, there is considerable defor­ mation in many layers. It is possible to explore that the rub­ ber material is deformed very fast in the pressure direction. The forming rubber appears to be stuck to the inner surface of the cavity due to the friction force. In the last simulation step, the deformed rubber will contact all cavity surfaces to fill the 5 mm thickness between the upper and lower inserts. During forming, the relative movement between the rubber and upper insert leads to the distribution of the molten uni­ formly inside the cavity. During impact, the rubber elements distort too much and penetrate the elements. To minimize these influences, which affect the system stability, the pres­ sure should not exceed 0.95 MPa. Finally, developing a finite element model is a suitable way of understanding the behavior and mechanism of elas­ tomer forming. This will help the manufacturer to evaluate problem conditions more comprehensively and make true decisions in design before tooling investment. As a comparison with the finding of other researchers in the same field, the finding of the current work is nearly close to the work [1]. It was found that the crack propagation behavior in rubber materials due to high stresses causes the rubber to lose some of its elasticity and allows surface cracks to appear eventually. Some limitations were recorded in this study, and this is normally related to the hyperelasticity be­ havior of this material. 6. Discussion of the analysis results of forming the elastomer pad The finite element can mimic the general behavior of the rubber pad very well during different punch velocity drop and loading steps. It is considered a better choice to model hyperelastic materials like rubber. The behavior of rubber Fig. 15. Сontour graph of generated stress during the forming process 88 Engineering technological systems: Reference for Chief Technology Specialist at an industrial enterprise During simulation, the crack’s trajectories under different punch-forming velocities are investigated. material is always time-dependent, and the strain rate has the main impact on the stiffness increase in rubber forming. During these sequence steps, the interaction between the parts according to the rule of master (non-deformable part) and slave (deformable part) will make the rubber deform to fill the cavity space according to the forming speed. The high forming speed will result in a nonhomogeneous layer distribution, and the random movements of nodes and ele­ ments will cause a successive dislocation between the layer’s boundaries. In this case, and due to high particle movements, the crystals will collide with each other and gain high energy leading to high distortion and non-uniform forming. The effects of a random increase in the forming speed and load on crack propagation are largely due to the increase in kinetic energy, and this will increase the possibility of crack initia­ tion by more than 15 %. In some particular areas, the mechanical response of the stress intensity is normally a high-order polynomial, and due to large deformations, high stress strains at the crack zone can be absorbed by soft materials as explained in Fig. 15. The rate of crack propagation has a comprehensive effect and characterization on the local morphology of the part. Cracks will follow different paths and develop in many di­ rections such as vertical or horizontal. The value of stress at a circular crack strongly depends on crack sharpness, which affects the material failure as mentioned in Fig. 16. Distortion and contact instabilities will take place during the rubber pad forming due to large strain occurring in the simulation process. The movement and dislocation of rub­ ber elements will generate kinetic energy according to the impact between the interaction surfaces (Fig. 10, 12, 14). Numerical mass damping is used to prevent some oscillations that happened during the first interaction between the rub­ ber and other parts, and the hourglass mode is always used to prevent this problem. Conflict of interest The authors declare that they have no conflict of inte­ rest in relation to this research, whether financial, personal, authorship or otherwise, that could affect the research and its results presented in this paper. 1. Scetta, G. (2020). Fatigue cracking of thermoplastic elastomers. Universit Paris sciences et lettres. Available at: https://pastel. archives-ouvertes.fr/tel-03149063 2. Samarth, N. B., Mahanwar, P. A. (2021). Degradation of Polymer & Elastomer Exposed to Chlorinated Water – A Review. Open Journal of Organic Polymer Materials, 11 (01), 1–50. doi: https://doi.org/10.4236/ojopm.2021.111001 6. Discussion of the analysis results of forming the elastomer pad The nature of element type randomly deforms under some boundary conditions. In this case, the thickness may vary in some zones. The disadvantage in dealing with this analysis method normally requires a long time of simulation running to find accurate results. Conse­ quently, determining the suitable boundary conditions and interaction types between the contact parts will significantly enhance the results. 7. Conclusions 2nd International Conference g y y, ( ), p // g/ / y , , Öztürk, E., Türköz, M., Dilmeç, M. (2017). 2D Finite Element Analysis of Rubber Pad Forming Process. 2nd International Conference Öztürk, E., Türköz, M., Dilmeç, M. (2017). 2D Finite Element Analysis of Rubber Pad Forming Process. 2nd International Conference 7. Oscar, J., Centeno, G. (2017). Finite Element Modeling Of Rubber Bushing For Crash Simulation Experimental Tests and Validation. Division of Structural Mechanics, Lund University. Available at: https://www.byggmek.lth.se/fileadmin/byggnadsmekanik/publi­ cations/tvsm5000/web5163.pdf 7. Oscar, J., Centeno, G. (2017). Finite Element Modeling Of Rubber Bushing For Crash Simulation Experimental Tests and Validation. Division of Structural Mechanics, Lund University. Available at: https://www.byggmek.lth.se/fileadmin/byggnadsmekanik/publi­ cations/tvsm5000/web5163.pdf 8. Iv ez, I., Braun, M. (2017). Numerical analysis of surface cracks repaired with single and double patches of composites. Journal of Composite Materials, 52 (8), 1113–1120. doi: https://doi.org/10.1177/0021998317722044 8. Iv ez, I., Braun, M. (2017). Numerical analysis of surface cracks repaired with single and double patches of composites. Journal of Composite Materials, 52 (8), 1113–1120. doi: https://doi.org/10.1177/0021998317722044 9. Elmukashfi, E., Kroon, M. (2020). Numerical Modeling and Analysis of Dynamic Crack Propagation in Rubber. 13th International Conference on Fracture. Beijing. Available at: https://www.researchgate.net/publication/346898728 9. Elmukashfi, E., Kroon, M. (2020). Numerical Modeling and Analysis of Dynamic Crack Propagation in Rubber. 13th International Conference on Fracture. Beijing. Available at: https://www.researchgate.net/publication/346898728 10. Magid, H. M., Dabis, B. K., Abed alabas Siba, M. (2021). Analysis of the main factors affecting mass production in the plastic molding process by using the finite element method. Eastern-European Journal of Enterprise Technologies, 6 (1 (114)), 65–71. doi: https://doi.org/10.15587/1729-4061.2021.248375 10. Magid, H. M., Dabis, B. K., Abed alabas Siba, M. (2021). Analysis of the main factors affecting mass production in the plastic molding process by using the finite element method. Eastern-European Journal of Enterprise Technologies, 6 (1 (114)), 65–71. doi: https://doi.org/10.15587/1729-4061.2021.248375 11. Kadhim, K. J., Jaber, J. A., Ibrihim, H. R. (2022). Implementation of finite element analysis for solving the constraints in form­ ing process of large steel parts. Eastern-European Journal of Enterprise Technologies, 4 (1 (118)), 64–71. doi: https://doi.org/ 10.15587/1729-4061.2022.263452 11. Kadhim, K. J., Jaber, J. A., Ibrihim, H. R. (2022). Implementation of finite element analysis for solving the constraints in form­ ing process of large steel parts. Eastern-European Journal of Enterprise Technologies, 4 (1 (118)), 64–71. doi: https://doi.org/ 10.15587/1729-4061.2022.263452 12. Persson, B. N. J., Albohr, O., Heinrich, G., Ueba, H. (2005). Crack propagation in rubber-like materials. 7. Conclusions 7. Conclusions 1. The analysis and evaluation of the rubber-forming pro­ cess through different steps were carried out according to the relevant geometrical properties of the specific component. References 1. Scetta, G. (2020). Fatigue cracking of thermoplastic elastomers. Universit Paris sciences et lettres. Available at: https://pastel. archives-ouvertes.fr/tel-03149063 h h d f l l d hl d 2. Samarth, N. B., Mahanwar, P. A. (2021). Degradation of Polymer & Elastomer Exposed to Chlorinated Water – A Review. Open Journal of Organic Polymer Materials, 11 (01), 1–50. doi: https://doi.org/10.4236/ojopm.2021.111001 89 Eastern-European Journal of Enterprise Technologies ISSN 1729-3774 6/1 ( 120 ) 2022 6/1 ( 120 ) 2022 3. Corre, T., Coret, M., Verron, E., Lebl , B. (2021). Non steady-state intersonic cracks in elastomer membranes under large static strain. Journal of Theoretical, Computational and Applied Mechanics. doi: https://doi.org/10.46298/jtcam.6906 3. Corre, T., Coret, M., Verron, E., Lebl , B. (2021). Non steady-state intersonic cracks in elastomer membranes under large static strain. Journal of Theoretical, Computational and Applied Mechanics. doi: https://doi.org/10.46298/jtcam.6906 4. Poulain, X., Lefèvre, V., Lopez-Pamies, O., Ravi-Chandar, K. (2017). Damage in elastomers: nucleation and growth of cavities, micro-cracks, and macro-cracks. International Journal of Fracture, 205 (1), 1–21. doi: https://doi.org/10.1007/s10704-016-0176-9 4. Poulain, X., Lefèvre, V., Lopez-Pamies, O., Ravi-Chandar, K. (2017). Damage in elastomers: nucleation and growth of cavities, micro-cracks, and macro-cracks. International Journal of Fracture, 205 (1), 1–21. doi: https://doi.org/10.1007/s10704-016-0176-9 4. Poulain, X., Lefèvre, V., Lopez-Pamies, O., Ravi-Chandar, K. (2017). Damage in elastomers: nucleation and growth of cavities, micro-cracks, and macro-cracks. International Journal of Fracture, 205 (1), 1–21. doi: https://doi.org/10.1007/s10704-016-0176-9 5. Wang, H., Wu, Y., Yang, J., Wang, H. (2021). Numerical Simulation on Reflective Cracking Behavior of Asphalt Pave Sciences, 11 (17), 7990. doi: https://doi.org/10.3390/app11177990 6. Hamzah, K. B., Nik Long, N. M. A., Senu, N., Eshkuvatov, Z. K. (2021). Numerical Solution for Crack Phenomenon in Dissimilar Materials under Various Mechanical Loadings. Symmetry, 13 (2), 235. doi: https://doi.org/10.3390/sym13020235 Halkacı, H. S., Öztürk, E., Türköz, M., Dilmeç, M. (2017). 2D Finite Element Analysis of Rubber Pad Forming Process. 2nd International Conference on Science, Ecology and Technology-2016 (ICONSETE’2016). Available at: https://www.researchgate.net/publication/312198030 6. Hamzah, K. B., Nik Long, N. M. A., Senu, N., Eshkuvatov, Z. K. (2021). Numerical Solution for Crack Phenomenon in Dissimilar Materials under Various Mechanical Loadings. Symmetry, 13 (2), 235. doi: https://doi.org/10.3390/sym13020235 Halkacı, H. S., Öztürk, E., Türköz, M., Dilmeç, M. (2017). 2D Finite Element Analysis of Rubber Pad Forming Process. 7. Conclusions Journal of Physics: Con­ densed Matter, 17 (44), R1071–R1142. doi: https://doi.org/10.1088/0953-8984/17/44/r01 12. Persson, B. N. J., Albohr, O., Heinrich, G., Ueba, H. (2005). Crack propagation in rubber-like materials. Journal of Physics: Con densed Matter, 17 (44), R1071–R1142. doi: https://doi.org/10.1088/0953-8984/17/44/r01 13. Elmukashfi, E. (2015). Modeling of fracture and damage in rubber under dynamic and quasi-static conditions. School of Engineer­ ing Sciences, Department of Solid Mechanics, Royal Institute of Technology. Stockholm. Available at: http://kth.diva-portal.org/ smash/get/diva2:876354/FULLTEXT01.pdf 13. Elmukashfi, E. (2015). Modeling of fracture and damage in rubber under dynamic and quasi-static conditions. School of Engineer­ ing Sciences, Department of Solid Mechanics, Royal Institute of Technology. Stockholm. Available at: http://kth.diva-portal.org/ smash/get/diva2:876354/FULLTEXT01.pdf 14. Maiorova, K., Vorobiov, I., Andrieiev, O., Lupkin, B., Sikulskiy, V. (2022). Forming the geometric accuracy and roughness of holes when drilling aircraft structures made from polymeric composite materials. Eastern-European Journal of Enterprise Technologies, 2 (1 (116)), 71–80. doi: https://doi.org/10.15587/1729-4061.2022.254555 14. Maiorova, K., Vorobiov, I., Andrieiev, O., Lupkin, B., Sikulskiy, V. (2022). Forming the geometric accuracy and roughness of holes when drilling aircraft structures made from polymeric composite materials. Eastern-European Journal of Enterprise Technologies, 2 (1 (116)), 71–80. doi: https://doi.org/10.15587/1729-4061.2022.254555 15. Korzhyk, V., Khaskin, V., Grynyuk, A., Peleshenko, S., Kvasnytskyi, V., Fialko, N. et al. (2022). Comparison of the features of the for­ mation of joints of aluminum alloy 7075 (Al-Zn-Mg-Cu) by laser, microplasma, and laser-microplasma welding. Eastern-European Journal of Enterprise Technologies, 1 (12 (115)), 38–47. doi: https://doi.org/10.15587/1729-4061.2022.253378 90 90

https://openalex.org/W4220871485

https://www.mdpi.com/1422-0067/23/6/3053/pdf?version=1647247806

English

Prediction of the Neurotoxic Potential of Chemicals Based on Modelling of Molecular Initiating Events Upstream of the Adverse Outcome Pathways of (Developmental) Neurotoxicity

International journal of molecular sciences

cc-by





 Citation: Gadaleta, D.; Spînu, N.; Roncaglioni, A.; Cronin, M.T.D.; Benfenati, E. Prediction of the Neurotoxic Potential of Chemicals Based on Modelling of Molecular Initiating Events Upstream of the Adverse Outcome Pathways of (Developmental) Neurotoxicity. Int. J. Mol. Sci. 2022, 23, 3053. https:// doi.org/10.3390/ijms23063053 Academic Editor: Stephen C. Bondy Received: 8 February 2022 Accepted: 8 March 2022 Published: 11 March 2022  Citation: Gadaleta, D.; Spînu, N.; Roncaglioni, A.; Cronin, M.T.D.; Benfenati, E. Prediction of the Neurotoxic Potential of Chemicals Based on Modelling of Molecular Initiating Events Upstream of the Adverse Outcome Pathways of (Developmental) Neurotoxicity. Int. J. Mol. Sci. 2022, 23, 3053. https:// doi.org/10.3390/ijms23063053 Academic Editor: Stephen C. Bondy Received: 8 February 2022 Accepted: 8 March 2022 Published: 11 March 2022 Citation: Gadaleta, D.; Spînu, N.; Roncaglioni, A.; Cronin, M.T.D.; Benfenati, E. Prediction of the Neurotoxic Potential of Chemicals Based on Modelling of Molecular Initiating Events Upstream of the Adverse Outcome Pathways of (Developmental) Neurotoxicity. Int. J. Mol. Sci. 2022, 23, 3053. https:// doi.org/10.3390/ijms23063053 Keywords: molecular initiating events; neurotoxicity; adverse outcome pathways; QSAR Domenico Gadaleta 1,* , Nicoleta Spînu 2, Alessandra Roncaglioni 1, Mark T. D. Cronin 2 and 1 Laboratory of Environmental Chemistry and Toxicology, Department of Environmental Health Sciences, Istituto di Ricerche Farmacologiche Mario Negri IRCCS, Via Mario Negri 2, 20156 Milan, Italy; [email protected] (A.R.); [email protected] (E.B.) 2 School of Pharmacy and Biomolecular Sciences, Liverpool John Moores University, Byrom Street, Liverpool L3 3AF, UK; [email protected] (N.S.); [email protected] (M.T.D.C.) * Correspondence: [email protected] 1 Laboratory of Environmental Chemistry and Toxicology, Department of Environmental Health Sciences, Istituto di Ricerche Farmacologiche Mario Negri IRCCS, Via Mario Negri 2, 20156 Milan, Italy; [email protected] (A.R.); [email protected] (E.B.) 2 p p j * Correspondence: [email protected] Abstract: Developmental and adult/ageing neurotoxicity is an area needing alternative methods for chemical risk assessment. The formulation of a strategy to screen large numbers of chemicals is highly relevant due to potential exposure to compounds that may have long-term adverse health consequences on the nervous system, leading to neurodegeneration. Adverse Outcome Pathways (AOPs) provide information on relevant molecular initiating events (MIEs) and key events (KEs) that could inform the development of computational alternatives for these complex effects. We propose a screening method integrating multiple Quantitative Structure–Activity Relationship (QSAR) models. The MIEs of existing AOP networks of developmental and adult/ageing neurotoxicity were modelled to predict neurotoxicity. Random Forests were used to model each MIE. Predictions returned by single models were integrated and evaluated for their capability to predict neurotoxicity. Specifically, MIE predictions were used within various types of classifiers and compared with other reference standards (chemical descriptors and structural fingerprints) to benchmark their predictive capability. Overall, classifiers based on MIE predictions returned predictive performances comparable to those based on chemical descriptors and structural fingerprints. The integrated computational approach described here will be beneficial for large-scale screening and prioritisation of chemicals as a function of their potential to cause long-term neurotoxic effects.



 International Journal of Molecular Sciences International Journal of Molecular Sciences International Journal of Molecular Sciences International Journal of Molecular Sciences 1. Introduction The human brain is exceptionally sensitive to injury, and several neurodevelopmental processes have been shown to be highly vulnerable to external factors [1,2]. Such pro- cesses include neural progenitor cell proliferation, apoptosis, cell migration, neuronal and glial differentiation, neurite outgrowth and branching, myelination, synaptogenesis and neuronal network formation, the ontogeny of neurotransmitters and receptors, the development of the blood–brain barrier, and the developmental changes in the adolescent brain [3–5]. Disruption of any of these processes may lead to potentially adverse alterations in neuroanatomy, neurophysiology, and neurochemistry. It has been estimated that devel- opmental neurotoxicity (DNT) disorders affect 10–15% of all births [6], and the prevalence of autism and attention-deficit hyperactivity disorders is increasing worldwide [1]. In addi- tion, DNT disorders documented in children and adolescents could be a precursor of the development of neurodegenerative diseases (NDs) later in life [7]. NDs (e.g., Alzheimer’s and Parkinson’s) are widely investigated pathologies due to the low efficacy of current therapies [8,9], the severe functional impairments they impose on daily life activities, and the resulting high familial, social, and financial costs of patient care [10]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2022, 23, 3053. https://doi.org/10.3390/ijms23063053 Int. J. Mol. Sci. 2022, 23, 3053 2 of 18 Overall, genetic factors seem to account for about 30–40% of all cases of DNT disor- ders [11]. Evidence has been reported that exposure to chemical stressors, e.g., industrial chemicals in the environment, is a key determinant in the occurrence of neurological disorders [11,12]. Thousands of chemicals have been reported to have adverse effect on neurodevelopment or to be toxic the nervous system in adults [11,12]. However, the total number of known neurotoxic substances is likely to be an underestimation of the true num- ber released into the environment [11]. There is, therefore, a need to develop strategies able to screen the large number of chemicals to which the population is exposed daily and that may have possible long-term adverse health consequences to the brain. 1. Introduction Guideline-based DNT studies involve the use of a large number of animals for an extended period of time, making this kind of study significantly resource-intensive and not suitable for large-scale screening [13,14]. Computational toxicology has been shown to be a cost- and time-efficient alternative to traditional toxicity testing methods [15]. Quantitative Structure–Activity Relationship (QSAR) models are computational methods that have been primarily applied for their capability to identify the toxicity of chemicals as a function of their structural attributes [16]. The increased availability of data obtained from in vitro bioactivity testing has made the development of QSARs easier. Moreover, QSARs require relatively few resources and are rapid, which have been key factors in their increased use assisting in filling toxicological data gaps for chemicals with high production volumes. Toxicology has recently undergone a paradigm shift towards the use of alternative testing methods based on knowledge of the biological modes of action and pathways that are responsible for adverse effects, defined as adverse outcome pathways (AOPs). An AOP is a logical construct that connects an upstream molecular initiating event (MIE) (e.g., the interaction of a chemical with a molecular target) to a downstream adverse outcome (AO), progressing through a series of key events (KEs) [17,18]. According to this concept, compounds of unknown hazards can be assigned to various levels of concern based on the number of activated MIEs and the extent of their activation. Moreover, chemicals activating similar MIEs/KEs with respect to known toxic chemicals will be more likely to be toxic themselves [19]. Several authors have recently highlighted a possible synergism between the AOP concept and QSAR modelling in toxicology [19–26]. Thus, it is possible to utilise QSAR models to predict the potential of chemicals to modulate MIEs and to prioritise chemicals as a function of their toxicological profile. In the present manuscript, we present an integrated computational system to predict neurotoxic potential that relies on the identification of the MIEs activated by chemicals. MIEs upstream of neurotoxicity were identified from recently published AOP networks, then QSAR models were developed for the prediction of each MIE. Predictions from QSARs for individual MIEs were evaluated for their capability to discriminate neurotoxic and non-neurotoxic compounds as part of an integrated computational prediction system. 1. Introduction MIEs were used as independent variables in various machine learning approaches and compared for their predictive power with other widely used methods for chemical description, i.e., fingerprints and molecular descriptors. The predictions returned by QSARs presented here may represent an effective first tier of an Integrated Approaches to Testing and Assess- ment (IATA) to rapidly screen many chemicals, providing information regarding potential MIEs and associated mechanisms of toxicity and thus, helping to prioritise chemicals for additional and better-targeted screening/testing, e.g., in vitro testing [27,28]. 2. Results Performance is the average of metrics obtained over 100 different training-test splits. MIE TP FP TN FN SEN SPE BA MCC AUC AChE 555.6 63.0 887.2 68.0 0.89 0.93 0.91 0.83 0.96 AMPAR 13.4 20.0 651.0 1.2 0.92 0.97 0.94 0.60 0.99 CAR 9.0 6.8 668.6 1.2 0.88 0.99 0.94 0.72 0.95 CYP2E1 4.0 35.2 645.2 1.0 0.80 0.95 0.87 0.29 0.90 GABAR 20.0 11.0 649.0 6.0 0.77 0.98 0.88 0.69 0.96 KAR 4.4 17.4 663.0 0.6 0.88 0.97 0.93 0.42 0.97 NADHOX 15.2 4.4 665.4 0.4 0.97 0.99 0.98 0.87 1.00 NIS 11.0 0.6 673.4 0.4 0.97 1.00 0.98 0.96 1.00 NMDAR 50.0 27.8 604.4 3.4 0.94 0.96 0.95 0.75 0.98 PXR 35.8 35.8 601.8 13.0 0.73 0.94 0.84 0.57 0.92 RYR 11.0 0.6 673.6 0.2 0.98 1.00 0.99 0.96 0.99 THRα 60.0 23.4 599.2 2.8 0.96 0.96 0.96 0.81 0.99 THRβ 110.2 37.8 500.0 38.4 0.74 0.93 0.84 0.67 0.93 TTR 14.8 44.0 624.0 3.8 0.80 0.93 0.87 0.42 0.94 VGSC 28.4 12.8 639.2 5.0 0.85 0.98 0.92 0.76 0.97 The statistics of the balanced random forest (BRF) models for MIEs were extremely good and confirmed the high quality of the information included in ChEMBL database and its suitability as a source of data for modelling the interactions of ligands with molecular targets upstream of biological pathways (e.g., nuclear receptors and enzymes) [29,30]. One possible reason for this high performance is the high structural homogeneity of the active samples. This is not surprising, as many records included in ChEMBL are congeneric sets of candidate drugs, while negative samples in the MIE datasets are more structurally heterogeneous. This aspect is reasonable, as ligands for enzymes and receptors are often required to possess specific pharmacophoric features to interact with binding sites, leading to preferred structural moieties being shared among the different ligands. External validation returned BA values in the range of 0.84–0.99. The statistics were balanced between sensitivity (SEN) and specificity (SPE); although SEN was higher in most cases, slightly lower values for the inactive class were most common in all datasets. As for Matthews Correlation Coefficient (MCC) values, several of the models showed values below the average, such as the BRFs for CYP2E1, KAR, and TTR. 2. Results This aspect is reasonable, as ligands for enzymes and receptors are often required to possess specific pharmacophoric features to interact with binding sites, leading to preferred structural moieties being shared among the different ligands. External validation returned BA values in the range of 0.84–0.99. The statistics were balanced between sensitivity (SEN) and specificity (SPE); although SEN was higher in most cases, slightly lower values for the inactive class were most common in all datasets. As for Matthews Correlation Coefficient (MCC) values, several of the models showed values below the average, such as the BRFs for CYP2E1, KAR, and TTR. MCC has beenproposed for the evaluation of classification between two very unbalanced categories; however, it was observed that this parameter can sometimes be biased by high unbalance datasets on terms of the categories (i.e., fewer than 20% of chemicals included in the smallest category) [24]. Indeed, the datasets mentioned above are among those with a lower number of actives. Table 2 shows statistics for the neurotoxicity-predicting models. The performance of each classifier was calculated as the average of 100 iterations of five-fold cross validation. Figure 1 (Figure 1a: k-nearest neighbours (k-NN); Figure 1b: random forest (RF); Figure 1c: neural network (NNET)) shows the distribution of the BAs of models based on MIE predictions compared with DRAGON descriptors and extended fingerprints. In the case of k-NN classifiers, the MIE predictions show higher performance (BA avg = 0.72) with respect to chemical descriptors (BA avg = 0.70), and fingerprints do not perform as well. MIEs and descriptors have a similar peak in the distribution of BAs between 0.70–0.75. RFs are the top-performing classifiers overall with respect to kNN and NNET, always reaching average BAs higher than 0.65 and having maximum values closer to 0.90. In this case, DRAGON descriptors were the top-performing variables (BA avg = 0.83), followed by predictions of QSARs based on fingerprints (BA avg = 0.74) and MIEs (BA avg = 0.73). In the case of NNET, Table 1. External validation of QSAR models for MIEs based on ChEMBL data. For each MIE predicting QSAR the average number of true positives (TP), false positives (FP), true negatives (TN), and false negatives (FN) were reported. The metrics for evaluating the predictivity of the models were sensitivity (SEN), specificity (SPE), balanced accuracy (BA), Matthew’s correlation coefficient (MCC) and area under the ROC curve (AUC). 2. Results Performance statistics from external validation for the QSARs for MIEs are reported in Table 1. A complete description of the MIEs listed in Table 1 and of their abbreviations is reported in Section 4.1. Int. J. Mol. Sci. 2022, 23, 3053 3 of 18 Table 1. External validation of QSAR models for MIEs based on ChEMBL data. For each MIE predicting QSAR the average number of true positives (TP), false positives (FP), true negatives (TN), and false negatives (FN) were reported. The metrics for evaluating the predictivity of the models were sensitivity (SEN), specificity (SPE), balanced accuracy (BA), Matthew’s correlation coefficient (MCC) and area under the ROC curve (AUC). Performance is the average of metrics obtained over 100 different training-test splits. MIE TP FP TN FN SEN SPE BA MCC AUC AChE 555.6 63.0 887.2 68.0 0.89 0.93 0.91 0.83 0.96 AMPAR 13.4 20.0 651.0 1.2 0.92 0.97 0.94 0.60 0.99 CAR 9.0 6.8 668.6 1.2 0.88 0.99 0.94 0.72 0.95 CYP2E1 4.0 35.2 645.2 1.0 0.80 0.95 0.87 0.29 0.90 GABAR 20.0 11.0 649.0 6.0 0.77 0.98 0.88 0.69 0.96 KAR 4.4 17.4 663.0 0.6 0.88 0.97 0.93 0.42 0.97 NADHOX 15.2 4.4 665.4 0.4 0.97 0.99 0.98 0.87 1.00 NIS 11.0 0.6 673.4 0.4 0.97 1.00 0.98 0.96 1.00 NMDAR 50.0 27.8 604.4 3.4 0.94 0.96 0.95 0.75 0.98 PXR 35.8 35.8 601.8 13.0 0.73 0.94 0.84 0.57 0.92 RYR 11.0 0.6 673.6 0.2 0.98 1.00 0.99 0.96 0.99 THRα 60.0 23.4 599.2 2.8 0.96 0.96 0.96 0.81 0.99 THRβ 110.2 37.8 500.0 38.4 0.74 0.93 0.84 0.67 0.93 TTR 14.8 44.0 624.0 3.8 0.80 0.93 0.87 0.42 0.94 VGSC 28.4 12.8 639.2 5.0 0.85 0.98 0.92 0.76 0.97 The statistics of the balanced random forest (BRF) models for MIEs were extremely good and confirmed the high quality of the information included in ChEMBL database and its suitability as a source of data for modelling the interactions of ligands with molecular targets upstream of biological pathways (e.g., nuclear receptors and enzymes) [29,30]. One possible reason for this high performance is the high structural homogeneity of the active samples. This is not surprising, as many records included in ChEMBL are congeneric sets of candidate drugs, while negative samples in the MIE datasets are more structurally heterogeneous. 2. Results MCC has beenproposed for the evaluation of classification between two very unbalanced categories; however, it was observed that this parameter can sometimes be biased by high unbalance datasets on terms of the categories (i.e., fewer than 20% of chemicals included in the smallest category) [24]. Indeed, the datasets mentioned above are among those with a lower number of actives. Table 2 shows statistics for the neurotoxicity-predicting models. The performance of each classifier was calculated as the average of 100 iterations of five-fold cross validation. Figure 1 (Figure 1a: k-nearest neighbours (k-NN); Figure 1b: random forest (RF); Figure 1c: neural network (NNET)) shows the distribution of the BAs of models based on MIE predictions compared with DRAGON descriptors and extended fingerprints. In the case of k-NN classifiers, the MIE predictions show higher performance (BA avg = 0.72) with respect to chemical descriptors (BA avg = 0.70), and fingerprints do not perform as well. MIEs and descriptors have a similar peak in the distribution of BAs between 0.70–0.75. RFs are the top-performing classifiers overall with respect to kNN and NNET, always reaching average BAs higher than 0.65 and having maximum values closer to 0.90. In this case, DRAGON descriptors were the top-performing variables (BA avg = 0.83), followed by predictions of QSARs based on fingerprints (BA avg = 0.74) and MIEs (BA avg = 0.73). In the case of NNET, Int. J. Mol. Sci. 2022, 23, 3053 4 of 18 MIE predictions and DRAGON descriptors (BA avg = 0.74) were characterised by an almost equal distribution profile, while fingerprints had lower performance (BA avg = 0.67). MIE predictions and DRAGON descriptors (BA avg = 0.74) were characterised by an almost equal distribution profile, while fingerprints had lower performance (BA avg = 0.67). MIE predictions and DRAGON descriptors (BA avg = 0.74) were characterised by an almost equal distribution profile, while fingerprints had lower performance (BA avg = 0.67). Table 2. Performance of the three classifiers (kNN, RF, NNET) using MIE predictions, chemical descriptors, and extended fingerprints as independent variables. For each method, the average number of true positives (TP), false positives (FP), true negatives (TN) false negatives (FN) and not classified (NC) were reported. The metrics to evaluate the predictivity of the models were sensitivity (SEN), specificity (SPE), balanced accuracy (BA), Matthew’s correlation coefficient (MCC), and area under the ROC curve (AUC). 2. Results Performance is the average of five-fold cross-validation results obtained over 500 iterations (100 fold-splitting procedures and five parameter combinations). Classifier Variable TP FP TN FN NC SEN SPE BA MCC AUC K-NN MIE predictions 30.5 11.4 19.6 7.5 0.0 0.80 0.63 0.72 0.44 0.76 Descriptors 29.5 11.5 19.5 8.5 0.0 0.78 0.63 0.70 0.41 0.76 Fingerprints 14.6 2.3 28.5 22.2 1.4 0.40 0.92 0.66 0.37 0.75 MLP- NNET MIE predictions 29.4 9.4 21.6 8.6 0.0 0.77 0.70 0.74 0.47 0.78 Descriptors 30.2 9.8 21.2 7.8 0.0 0.79 0.68 0.74 0.48 0.79 Fingerprints 28.1 12.4 18.6 9.9 0.0 0.74 0.60 0.67 0.34 0.69 RF MIE predictions 31.1 11.6 19.2 6.4 0.7 0.83 0.62 0.73 0.47 0.77 Descriptors 32.9 6.4 24.4 4.9 0.4 0.87 0.79 0.83 0.66 0.91 Fingerprints 32.9 11.9 18.8 4.8 0.5 0.87 0.61 0.74 0.51 0.80 Figure 1. Distribution of balanced accuracies calculated among the various QSARs developed to predict neurotoxic potential. Balanced accuracies are grouped based on the algorithm used: (a) k-Nearest Neighbours; (b) Random Forest; (c) Neural Network. Blue bars refer to models developed based on MIE predictions, red bars refer to models based on DRAGON descriptors, and yellow bars refer to models based on Extended Fingerprints. Dashed lines indicate the mean accuracy value achieved by each group of models. Classifier Variable TP FP TN FN NC SEN SPE BA MCC AUC K-NN MIE predictions 30.5 11.4 19.6 7.5 0.0 0.80 0.63 0.72 0.44 0.76 Descriptors 29.5 11.5 19.5 8.5 0.0 0.78 0.63 0.70 0.41 0.76 Fingerprints 14.6 2.3 28.5 22.2 1.4 0.40 0.92 0.66 0.37 0.75 MLP- NNET MIE predictions 29.4 9.4 21.6 8.6 0.0 0.77 0.70 0.74 0.47 0.78 Descriptors 30.2 9.8 21.2 7.8 0.0 0.79 0.68 0.74 0.48 0.79 Fingerprints 28.1 12.4 18.6 9.9 0.0 0.74 0.60 0.67 0.34 0.69 RF MIE predictions 31.1 11.6 19.2 6.4 0.7 0.83 0.62 0.73 0.47 0.77 Descriptors 32.9 6.4 24.4 4.9 0.4 0.87 0.79 0.83 0.66 0.91 Fingerprints 32.9 11.9 18.8 4.8 0.5 0.87 0.61 0.74 0.51 0.80 Figure 1. Distribution of balanced accuracies calculated among the various QSARs developed to predict neurotoxic potential. Balanced accuracies are grouped based on the algorithm used: (a) k-Nearest Neighbours; (b) Random Forest; (c) Neural Network. Blue bars refer to models developed based on MIE predictions, red bars refer to models based on DRAGON descriptors, and yellow bars refer to models based on Extended Fingerprints. 3. Discussion In the present work, a new integrated computational system was proposed for predict- ing the neurotoxic potential of chemicals as a function of their capability to trigger MIE (i.e., interaction and modulation of relevant receptors and enzymes) upstream of neurotoxicity. QSARs were developed to predict MIE induction while BRFs were applied to handle the unbalanced training data. In the last part of the manuscript, MIE predictions were used to classify neurotoxic and non-neurotoxic compounds and compared for their predictivity with other approaches to describe the structure of chemicals, namely, chemical descriptors and extended fingerprints. g p Overall, in two out of three cases MIEs perform comparably with chemical descriptors and better than fingerprints, which are considered the gold standard for describing chemical structures within QSARs [33]. The only exception is given by neurotoxicity models based on RF, where DRAGON descriptors (BA avg = 0.83) performed better than their counterparts based on MIEs (BA avg = 0.73). This is likely to be due to the fact that RFs perform better if trained on a larger pool of variables, such as the pool of descriptors provided by DRAGON (i.e., several thousand) [34,35]. Despite this, one of the key advantages afforded by the use of MIE responses in place of the classical structural representation of molecules is the interpretability of the predictions. Indeed, a complete profile of the neuronal receptors and enzymes that are activated is given together with the overall neurotoxicity outcome, providing insights into the possible mode of action of a predicted neurotoxic chemical. This aspect was further verified by predicting a series of known neurotoxicants with the models predicting MIEs. The mode of action for these neurotoxicants was reported in a review by Masjosthusmann and coworkers [36] who gathered information from the literature about the targets upstream of the neurotoxicological pathway of these chemicals [11,12,37]. Interestingly, QSARs for the prediction of MIEs were able to identify the correct mode of action of several neurotoxicants. For example, dichlofenthion (97-17-6), Parathion (56-38-2), paraoxon (311-45-5), diazinon (333-41-5), physostigmin (57-47-6), ibogaine (83-74-9), and dichlovoros (62-73-7) were correctly predicted to stimulate cholinergic neurotransmission through AChE inhibition, while 3-Nitropropionic acid (504-88-1), glyphosate (1071-83-6), and argiopine (105029-41-2) were predicted to interact with NAMDR receptors. 2. Results Dashed lines indicate the mean accuracy value achieved by each group of models. Classifier Variable TP FP TN FN NC SEN SPE BA MCC AUC K-NN MIE predictions 30.5 11.4 19.6 7.5 0.0 0.80 0.63 0.72 0.44 0.76 Descriptors 29.5 11.5 19.5 8.5 0.0 0.78 0.63 0.70 0.41 0.76 Fingerprints 14.6 2.3 28.5 22.2 1.4 0.40 0.92 0.66 0.37 0.75 MLP- NNET MIE predictions 29.4 9.4 21.6 8.6 0.0 0.77 0.70 0.74 0.47 0.78 Descriptors 30.2 9.8 21.2 7.8 0.0 0.79 0.68 0.74 0.48 0.79 Fingerprints 28.1 12.4 18.6 9.9 0.0 0.74 0.60 0.67 0.34 0.69 RF MIE predictions 31.1 11.6 19.2 6.4 0.7 0.83 0.62 0.73 0.47 0.77 Descriptors 32.9 6.4 24.4 4.9 0.4 0.87 0.79 0.83 0.66 0.91 Fingerprints 32.9 11.9 18.8 4.8 0.5 0.87 0.61 0.74 0.51 0.80 Figure 1. Distribution of balanced accuracies calculated among the various QSARs developed to predict neurotoxic potential. Balanced accuracies are grouped based on the algorithm used: (a) k-Nearest Neighbours; (b) Random Forest; (c) Neural Network. Blue bars refer to models developed based on MIE predictions, red bars refer to models based on DRAGON descriptors, and yellow bars refer to models based on Extended Fingerprints. Dashed lines indicate the mean accuracy value achieved by each group of models. The relative importance of MIEs for neurotoxicity prediction was evaluated using the methods described in Section 4.6. Table S1 in the Supplementary Materials reports the impact of the removal of specific MIEs on the performance (BAs) of models for neurotoxicity, while Table S2 sorts the various MIEs included in the RF models for neurotoxicity by their variable importance. Both BAs and variable importance are an average of the values calculated over the various modelling iterations; see Section 4.5. Thyroid elements seem to be relevant (THRs, TPO and TTR) to neurotoxicity; in particular, the exclusion of TTR always leads to a reduction in BA average. TTR is the fourth MIE in terms of variable importance. THRs, TPO and TTR are involved in the biosynthesis, metabolism, and transportation of the thyroid hormone, respectively. Among the two isoforms of thyroid receptors, THRβ is consistently the most important (first descriptor for variable importance), while the removal of THRα did not negatively affect performance. Among ionotropic glutamate receptors, AMPAR and KAR seem to be more linked to neurotoxicity than NMDAR. AMPA/kainite receptor-mediated neurotoxicity was reported Int. J. Mol. Sci. 2. Results 2022, 23, 3053 5 of 18 to possibly play a role in neuronal neurodegeneration in amyotrophic lateral sclerosis [31] and in the injury of basal forebrain cholinergic neurons in diseases such as Alzheimer’s [32]. VGSC and, to a lesser extent, GABAR are consistently relevant within models developed for neurotoxicity (Table S1). The relevance of VGSC and GABAR is confirmed in the RF variable importance analyses (Table S4). The role of SMARTS for protein adduct formation is unclear: the descriptor is the least relevant from the variable importance analyses, and on the whole, its exclusion does not affect the average performance of RFs. On the other hand, its removal has a detrimental effect on the performance of both the K-NN and NNET models. 3. Discussion Indeed, the two former chemicals are reported to stimulate glutamatergic neurotransmission and cause excitotoxicity after activation of NMDA, leading to oxidative stress and cell death, while the latter was reported to inhibit glutamatergic neurotransmission after blockage of the post-synaptic receptors. Rotenone (83-79-4) and dieldrin (60-57-1) were correctly predicted to inhibit complex I (NADH dehydrogenase) and to cause reactive oxygen species (ROS)-induced degeneration of dopaminergic neurons and locomotor deficit. In addition to increased biological relevance, MIE predictions simplify models to a reduced number of variables (i.e., fewer than 20), while in the case of descriptors and fingerprints several hundred variables may be included in the models. The RFs developed here for the prediction of single MIEs returned satisfactory predic- tive performance and were confirmed to be a valuable method in the field of computational toxicology. The statistical performance of the models presented here confirmed our previ- Int. J. Mol. Sci. 2022, 23, 3053 6 of 18 ous findings that using RFs with internal balancing of categories (BRF) represents one of the best methods for handling the unbalanced distribution typical of biological data [24,38]. g yp g In the present manuscript, the use of biological information (i.e., MIEs) instead of the classical structural description of molecules was proposed for the development of QSARs. The use of biological data (e.g., biological assays) utilised as input variables to develop QSARs has been increasingly explored in the recent literature [39]. This strategy is justified by the fact that QSARs historically had difficulty predicting complex systemic endpoints encompassing several mechanisms, which are difficult to model together. In the case of neurotoxicity, the brain is an extremely complex organ comprising a variety of highly specialised neuronal cell types that differ in function, expression of brain regions, and stages of development [40]. These different cells are all potential targets that can be disrupted by neurotoxicants with different possible mechanisms of toxicity [4]. Another limitation of QSARs is that they rely on the principle that analogies in chemical structure always result in analogies in toxicity. However, the existence of activity cliffs, i.e., compounds with high structural similarity together with unexpectedly high activity differences, were reported for high-tier endpoints characterised by multiple mechanisms of toxicity [41]. On the other hand, the use of information from AOPs and biological assays allows for the fragmentation of complex endpoints into simpler ones based on mechanistic knowledge. 3. Discussion These “sub-endpoints” are easier to address with a single computational model, as they describe the interaction of a chemical with a single molecular target that triggers a specific response. Overall, this strategy allows for a reduction in the complexity of the challenge of capturing the complex relationships existing between the structure of a chemical and its high-level systemic toxicity [4]. The development of new machine learning and artificial intelligence-based approaches is highly desirable, as it allows for the detection of chemicals with potential neurotoxicity and DNT effects in a more time- and resource-efficient way compared to traditional in vivo testing. In addition, data from in silico screenings based on AOPs can provide a scientifically sound rationale to make decisions relating to assessment of the safety of chemicals. The mechanistic nature of AOPs provides knowledge to guide the design of new IATAs to meet regulatory needs [28,42]. In particular, these in silico predictions can be used to provide information regarding the potential MIEs of chemicals, to help prioritise or deprioritise certain chemicals for further testing, and to provide indications for better-targeted follow- up in vitro evaluations [43]. In the specific case of neurotoxicity, a wide range of in vitro tests has been proposed, each evaluating a different MIE/KE of the complex network upstream of the adverse outcome [27]. Predictions of MIE provided by QSARs may give indications of which assays to prioritise among the wide battery of tests available. In this regard, in silico models represent an ideal first tier of a multi-step IATA for the prediction of the neurotoxicity of chemicals which involves multiple alternative testing methods. 4.1. Data Selection for Molecular Initiating Events (MIEs) MIEs linked to neurotoxicity were identified from the AOP networks [44] published by Spînu et al. [45] and Li et al. [27]. The MIEs selected along with their associated molecular targets (i.e., receptors, enzymes) are reported in Table 3. In certain cases, multiple molecular targets are associated with a single MIE (e.g., three different glutamate receptors were considered for MIE A), while a single target may be repeated in multiple MIEs (e.g., NADHOX was common to MIEs C and N). The molecular targets involved in the MIEs and their relevance in neurotoxicity are briefly described below. • Glutamate ionotropic receptors, i.e., N-methyl-D-aspartate (NMDAR), alpha-amino-3- hydroxy-5-methyl-4-isoxazolepropionate (AMPAR) and kainate (KAR) are responsible for excitatory synaptic transmission and synaptic plasticity, which are fundamental for learning and memory [46]. Sustained over-activation of these receptors (MIE A) can induce excitotoxicity due to increased Ca2+ influx, with consequent cell death, memory problems, and convulsions [4]. Analogously, the chronic blockage of NMDAR Int. J. Mol. Sci. 2022, 23, 3053 7 of 18 7 of 18 by chemicals during synaptogenesis (MIE B) disrupts neuronal network formation, resulting in the impairment of learning and memory processes [47] and increasing the risk of developing Alzheimer’s-type NDs in later life [2]. p g yp Protein adduct formation is the covalent interaction between an electrophilic chemical and the nucleophilic part of a protein, and may lead to damage of the protein and the potential loss of its function. This may affect thiol- and seleno-containing proteins, which offer antioxidant protection [48]. The binding of xenobiotics (e.g., heavy metals and mercury) to these or other proteins during brain development (MIE D and H) may lead to several functional impairments, such as in learning and memory. Cytochrome P450 2E1 (CYP2E1) is relevant to this mechanism as well, as it is one of the enzymes responsible for the metabolism of small compounds. The induction of CYP2E1 (MIE E) leads to an increase in reactive metabolites, which can form protein adducts. For example, a high concentration of ethanol leads to an increased expression of CYP2E1 and consequent increased production of acetaldehyde metabolite, which can form protein adducts [49]. The consequences include oxidative stress, lipid peroxidation, unfolded protein responses and, ultimately the apoptosis of neuronal cells [50]. 4.1. Data Selection for Molecular Initiating Events (MIEs) • The function of the Na+/I−symporter (NIS) is critical for the physiological production and maintenance of thyroid hormone levels in the serum, as it mediates the transport of iodide into thyroid cells. Its inhibition (MIE F) results in decreased thyroid hormone synthesis, with effects on neurocognitive function in children [51,52]. • Acetylcholinesterase (AChE) is an enzyme present in both central and peripheral nervous systems and in muscular motor plaques. It is responsible for the enzymatic cleavage of the neurotransmitter acetylcholine [53]. Inhibition of AChE (MIE I), e.g., by organophosphates and carbamates, leads to an increase in levels of acetylcholine and overstimulation of both muscarinic and nicotinic receptors, resulting in multiple adverse outcomes affecting a wide variety of functions [54]. g y • Ryanodine-sensitive Ca2+ channels (RyR) contribute to neurotransmission and synap- tic plasticity. Polychlorinated biphenyl (PCB) exposure has been reported to alter intracellular Ca2+ levels and to interfere with normal neuronal dendritic growth and plasticity in a RyR-dependent manner (MIE L) [55]. p y y p • Thyroid hormone receptors α and β (THRα and THRβ) mediate the effects of thy- roid hormones, while thyroperoxidase (TPO) and deionidase are involved in the biosynthesis/catabolism of thyroid hormones. Transtyretrin serum binding protein (TTR), monocarboxylate transporters 8 and 10, and the solute carrier organic anion transporter family member 1C14 (OATP1C1) are involved in the transportation of thyroid hormones at various levels [56]. Interference at any of these levels (MIEs G and Q-T) may lead to decreased thyroxine (T4) and thyroid hormones in the brain, and ultimately alter neurodevelopmental processes such as neuronal proliferation, apoptosis, migration, neurite outgrowth, and neuronal network connectivity [57,58], culminating in irreversible mental retardation and motor deficits [59]. It has been reported that PCBs induce activation of xenobiotic nuclear receptors, e.g., the constitu- tive androstane receptor (CAR) and the pregnane X receptor (PXR), which represent MIE P, leading to thyroid hormone disruption during cochlear development and potentially resulting in permanent auditory loss [60]. p y g p y The complexes of the respiratory chain play a pivotal role in neuronal and glial cell sur- vival and cell death, as they regulate both energy metabolism and apoptotic/necrotic pathways. The interaction of xenobiotics with these enzymes can interfere in various ways with their normal functionality, e.g., inhibiting the production of ATP (MIE M) or interfering with the redox cycle (MIE N and O), with consequent increased production of ROS and oxidative stress. 4.1. Data Selection for Molecular Initiating Events (MIEs) p ( ) q • Ionotropic GABA receptors (GABAR) are ligand-gated ion channels which play im- portant roles in inhibitory neurotransmission [64]. Interference with GABA signalling (MIE V) during development and after brain maturation is likely to cause such varied adverse outcomes as autism, mental retardation, epilepsy, and schizophrenia [59]. Chemically-induced epileptic seizures can be caused by the binding of neurotoxicants (e.g., barbiturates, benzodiazepines, and picrotoxin) to the active sites of the GABA receptor [65]. Data relative to each of the molecular targets identified in Table 3 were extracted from the ChEMBL database [66] using a protocol adapted from Bosc et al. [30]. ChEMBL Target IDs for each molecular target linked to an MIE are listed in Table 4. When available, only data relative to Homo sapiens were considered. For each target, only bioactivities with pChEMBL values were chosen. This term refers to all the comparable measures of half-maximal responses (molar IC50, XC50, EC50, AC50, Ki, Kd, potency and ED50) on a negative logarithmic scale [67]. Different pChEMBL thresholds to classify bioactivity values were evaluated. Ultimately, selected pChEMBL data were flagged as active or inactive based on a pChEMBL threshold of 5.0 (10 µM), providing datasets with a reasonable number of active samples for modelling. pChEMBL-like activities with standard relation “>“ or “≥“ (i.e., not associated with a precise activity value) were included as inactive. Only activities that were not flagged as potential duplicates, with no data_validity_comment and with an activity_comment that was not ‘inconclusive’, ‘undetermined’, or ‘not determined’ were considered. Endpoints characterised by few or no active compounds (i.e., thyroid hormone transporters, monocarboxylate transporters 8 and 10 and OATP1C1, NADH-cytochrome b5 reductase, deiodinase, and thyroperoxidase) were excluded from the modelling. For the modelling of NADH oxide reductase activity Bos taurus data were used, as human data were not available. pCheMBL data distributions were heavily skewed in the majority of cases towards positive values. Table 3. Molecular Initiating Events associated with Developmental Neurotoxicity, adapted from Spînu et al. [45] and Li et al. [27]. Table 3. Molecular Initiating Events associated with Developmental Neurotoxicity, adapted from Spînu et al. [45] and Li et al. [27]. 4.1. Data Selection for Molecular Initiating Events (MIEs) When available only data relative to Homo sapiens were considered. For each target, only bioactivities with pChEMBL values were chosen. This term refers to all the comparable measures of half-maximal responses (molar IC50, XC50, EC50, AC50, Ki, Kd, potency and ED50) on a negative logarithmic scale [67]. Different pChEMBL thresholds to classify bioactivity values were evaluated. Ultimately, selected pChEMBL data were flagged as active or inactive based on a pChEMBL threshold of 5.0 (10 µM), providing datasets with a reasonable number of active samples for modelling. pChEMBL-like activities with standard relation “>“ or “≥“ (i.e., not associated with a precise activity value) were included as inactive. Only activities that were not flagged as potential duplicates, with no data_validity_comment and with an activity_comment that was not ‘inconclusive’, ‘undetermined’, or ‘not determined’ were considered. Endpoints characterised by few or no active compounds (i.e., thyroid hormone transporters, monocarboxylate transporters 8 and 10 and OATP1C1, NADH-cytochrome b5 reductase, deiodinase, and thyroperoxidase) were excluded from the modelling. For the modelling of NADH oxide reductase activity Bos taurus data were used, as human data were not available. pCheMBL data distributions were heavily skewed in the majority of cases towards positive values. Table 3. Molecular Initiating Events associated with Developmental Neurotoxicity, adapted from Spînu et al. [45] and Li et al. [27]. ID MIE Target Reference A Binding of agonist, Ionotropic glutamate receptors Glutamate [NMDA] receptor [45] A Binding of agonist, Ionotropic glutamate receptors Glutamate receptor ionotropic kainate [45] A Binding of agonist, Ionotropic glutamate receptors Glutamate receptor ionotropic AMPA [45] B Binding of antagonist, NMDA receptors Glutamate [NMDA] receptor [45] C Binding of inhibitor, NADH-ubiquinone oxidoreductase (complex I) Mitochondrial complex I (NADH dehydrogenase) [45] D Binding, SH/SeH proteins involved in protection against oxidative stress Aspecific1 [45] E CYP2E1 Activation Cytochrome P450 2E1 [45] F Inhibition, Na+/I−symporter (NIS) Sodium/iodide cotransporter [45] G Thyroperoxidase, Inhibition Thyroid peroxidase 1 [45] y • Voltage-gated sodium channels (VGSC) are the primary molecules responsible for the control of the electrophysiological potentials of electrically excitable cells. Various isoforms exist, with isoforms 1, 2, 3, and 6 reported to be mainly expressed in the central nervous system [62]. Neurotoxic effects in mammals have been associated with the ability of some neurotoxicants (e.g., p,p’-DDT and pyrethroids) to bind to and disrupt VGSC (MIE U), with consequent behavioural effects [4,63]. 4.1. Data Selection for Molecular Initiating Events (MIEs) Oxidative stress contributes to a loss of func- tion of hippocampal neural progenitor cells and a decline in learning and memory performance [4]. Moreover, the inhibition of NADH-quinone oxidoreductase (NAD- HOX) (MIE C) by pesticides or toxins (e.g., neurotoxin 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine, MPTP) has been reported to cause mitochondrial dysfunction The complexes of the respiratory chain play a pivotal role in neuronal and glial cell sur- vival and cell death, as they regulate both energy metabolism and apoptotic/necrotic pathways. The interaction of xenobiotics with these enzymes can interfere in various ways with their normal functionality, e.g., inhibiting the production of ATP (MIE M) or interfering with the redox cycle (MIE N and O), with consequent increased production of ROS and oxidative stress. Oxidative stress contributes to a loss of func- tion of hippocampal neural progenitor cells and a decline in learning and memory performance [4]. Moreover, the inhibition of NADH-quinone oxidoreductase (NAD- HOX) (MIE C) by pesticides or toxins (e.g., neurotoxin 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine, MPTP) has been reported to cause mitochondrial dysfunction Int. J. Mol. Sci. 2022, 23, 3053 8 of 18 8 of 18 and degeneration of dopaminergic neurons of the nigro-striatal area, with consequent motor deficits typical of Parkinson’s disease [61]. motor deficits typical of Parkinson’s disease [61]. • Voltage-gated sodium channels (VGSC) are the primary molecules responsible for the control of the electrophysiological potentials of electrically excitable cells. Various isoforms exist, with isoforms 1, 2, 3, and 6 reported to be mainly expressed in the central nervous system [62]. Neurotoxic effects in mammals have been associated with the ability of some neurotoxicants (e.g., p,p’-DDT and pyrethroids) to bind to and disrupt VGSC (MIE U), with consequent behavioural effects [4,63]. • Ionotropic GABA receptors (GABAR) are ligand-gated ion channels which play im portant roles in inhibitory neurotransmission [64]. Interference with GABA signalling (MIE V) during development and after brain maturation is likely to cause such varied adverse outcomes as autism, mental retardation, epilepsy, and schizophrenia [59] Chemically-induced epileptic seizures can be caused by the binding of neurotoxicants (e.g., barbiturates, benzodiazepines, and picrotoxin) to the active sites of the GABA receptor [65]. Data relative to each of the molecular targets identified in Table 3 were extracted from the ChEMBL database [66] using a protocol adapted from Bosc et al. [30]. ChEMBL Target IDs for each molecular target linked to an MIE are listed in Table 4. 4.1. Data Selection for Molecular Initiating Events (MIEs) Target Code CheMBL ID Species MIE ACT INA Acetylcholinesterase AChE CHEMBL220 Human I 3076 4793 Glutamate receptor ionotropic AMPA AMPAR CHEMBL2096670 Human A 73 3355 Nuclear receptor subfamily 1 group I member 3 (Constitutive Androstane Receptor) CAR CHEMBL5503 Human P 51 3377 Cytochrome P450 2E1 CYP2E1 CHEMBL5281 Human E 25 3402 GABA-A receptor; alpha-1/beta-2/gamma-2 GABAR CHEMBL2095172 Human V 129 3298 Glutamate receptor ionotropic kainate KAR CHEMBL2109241 Human A 25 3402 Mitochondrial complex I (NADH dehydrogenase) NADHOX CHEMBL614865 Bos taurus C, N 78 3349 Sodium/iodide cotransporter NIS CHEMBL2331047 Human F 56 3371 Glutamate [NMDA] receptor NMDAR CHEMBL2094124 Human A, B 267 3161 Pregnane X receptor PXR CHEMBL3401 Human P 244 3188 Ryanodine receptors 1 RYR CHEMBL2062 CHEMBL4403 CHEMBL1846 Human L 56 3371 Thyroid hormone receptor alpha THRα CHEMBL1860 Human S 311 3116 Thyroid hormone receptor beta THRβ CHEMBL1947 Human S 728 2704 Table 3. Cont. Table 3. Cont. ID MIE Target Reference H Protein Adduct Formation Aspecific 2 [45] I Binding of inhibitors to acetylcholinesterase (AChE) Acetylcholinesterase [27] L Binding of non-dioxin-like polychlorinated biphenyls with ryanodine receptor (RyR) Ryanodine receptors 1, 2 and 3 [27] M Interaction uncouplers with oxidative phosphorylation Aspecific 3 [27] N Binding of redox cycling chemicals with NADH-quinone oxidoreductase Mitochondrial complex I (NADH dehydrogenase) [27] O Binding of redox cycling chemicals with NADH cytochrome b5 reductase NADH-cytochrome b5 reductase [27] P Xenobiotic nuclear receptor activation Pregnane X receptor [27] P Xenobiotic nuclear receptor activation Nuclear receptor subfamily 1 group I member 3 (Constitutive Androstane Receptor) [27] Q Interference with thyroid serum binding protein Transthyretin [27] R Deiodinase inhibition Deiodinase 4 [27] S Thyroid receptor binding Thyroid hormone receptor beta [27] S Thyroid receptor binding Thyroid hormone receptor alpha [27] T Thyroid hormone transporter interference Monocarboxylate transporter 8 4 [27] T Thyroid hormone transporter interference Monocarboxylate transporter 10 4 [27] T Thyroid hormone transporter interference Solute carrier organic anion transporter family member 1C1 4 [27] U Binding of pyrethroids to voltage-gated sodium channels (VGSC) Sodium channel protein type N alpha subunit [27] V Binding of antagonist to γ-aminobutyric acid receptor GABAAR GABA-A receptor; alpha-1/beta-2/gamma-2 [27] 1 No data found in ChEMBL, QSAR from Gadaleta et al. [68] was used. 2 Replaced with the use of reactivity SMARTS [69]. 3 No specific targets, not considered for modelling. 4 No data found in ChEMBL, not considered for modelling. 4.1. Data Selection for Molecular Initiating Events (MIEs) ID MIE Target Reference A Binding of agonist, Ionotropic glutamate receptors Glutamate [NMDA] receptor [45] A Binding of agonist, Ionotropic glutamate receptors Glutamate receptor ionotropic kainate [45] A Binding of agonist, Ionotropic glutamate receptors Glutamate receptor ionotropic AMPA [45] B Binding of antagonist, NMDA receptors Glutamate [NMDA] receptor [45] C Binding of inhibitor, NADH-ubiquinone oxidoreductase (complex I) Mitochondrial complex I (NADH dehydrogenase) [45] D Binding, SH/SeH proteins involved in protection against oxidative stress Aspecific1 [45] E CYP2E1 Activation Cytochrome P450 2E1 [45] F Inhibition, Na+/I−symporter (NIS) Sodium/iodide cotransporter [45] G Thyroperoxidase, Inhibition Thyroid peroxidase 1 [45] Int. J. Mol. Sci. 2022, 23, 3053 9 of 18 Table 3. Cont. ID MIE Target Reference H Protein Adduct Formation Aspecific 2 [45] I Binding of inhibitors to acetylcholinesterase (AChE) Acetylcholinesterase [27] L Binding of non-dioxin-like polychlorinated biphenyls with ryanodine receptor (RyR) Ryanodine receptors 1, 2 and 3 [27] M Interaction uncouplers with oxidative phosphorylation Aspecific 3 [27] N Binding of redox cycling chemicals with NADH-quinone oxidoreductase Mitochondrial complex I (NADH dehydrogenase) [27] O Binding of redox cycling chemicals with NADH cytochrome b5 reductase NADH-cytochrome b5 reductase [27] P Xenobiotic nuclear receptor activation Pregnane X receptor [27] P Xenobiotic nuclear receptor activation Nuclear receptor subfamily 1 group I member 3 (Constitutive Androstane Receptor) [27] Q Interference with thyroid serum binding protein Transthyretin [27] R Deiodinase inhibition Deiodinase 4 [27] S Thyroid receptor binding Thyroid hormone receptor beta [27] S Thyroid receptor binding Thyroid hormone receptor alpha [27] T Thyroid hormone transporter interference Monocarboxylate transporter 8 4 [27] T Thyroid hormone transporter interference Monocarboxylate transporter 10 4 [27] T Thyroid hormone transporter interference Solute carrier organic anion transporter family member 1C1 4 [27] U Binding of pyrethroids to voltage-gated sodium channels (VGSC) Sodium channel protein type N alpha subunit [27] V Binding of antagonist to γ-aminobutyric acid receptor GABAAR GABA-A receptor; alpha-1/beta-2/gamma-2 [27] 1 No data found in ChEMBL, QSAR from Gadaleta et al. [68] was used. 2 Replaced with the use of reactivity SMARTS [69]. 3 No specific targets, not considered for modelling. 4 No data found in ChEMBL, not considered for modelling. Table 4. List of endpoints modelled using ChEMBL data. For each endpoint, the reference MIE, ChEMBL ID relative to the molecular target, species, and composition of the Training and Test sets are reported; ACT is the number of active compounds, while INA is the number of inactive compounds. 4.1. Data Selection for Molecular Initiating Events (MIEs) The entry with the maximum pChEMBL activity was selected in the case of duplicate structures in order to maximise the number of active samples. Table 4 reports the final distribution of active and inactive chemicals for each dataset. The training sets for each of the modelled MIEs are available in the Supplementary Materials (Table S3). 4.2. QSARs for Molecular Initiating Events ChEMBL datasets from Table 4 were used to develop 15 QSARs for molecular targets involved in the MIEs. Extended fingerprints (Daylight Chemical Information Systems, Inc., 2019) were calculated for each compound with a KNIME implementation [71] of the CDK toolkit (https://cdk.github.io/ (accessed on 7 March 2022)) and used as input for QSAR modelling. The BRF [72] implemented in KNIME was used for QSAR development. This technique artificially alters the class distribution in each tree. A sampling without repetition was made to select compounds, allowing all of the active compounds to always be selected together with an equal number of randomly selected inactive compounds from the training set in order to assure balancing between categories [73]. The number of trees in each BRF was set to 100. Models were validated by splitting each data set into a training (80%) and a test set (20%) by applying a stratification sampling to the activity classes. The splitting procedure was repeated 100 times using different random splits, ensuring that each chemical in the datasets was included in the test set the same number of times in order to avoid bias due to the molecules present in the different sets. The performance using the test set was calculated for each iteration, then the final performance of each model was calculated by averaging the statistical parameters obtained using the test sets relative to each of the 100 iterations. 4.3. Thyroperoxidase (TPO) Modelling As no data were found for TPO inhibition from ChEMBL, the QSAR model for predict- ing TPO inhibition proposed by Gadaleta et al. [68] was used. The model was developed from data related to the Amplex UltraRed-thyroperoxidase (AUR-TPO) assay. For posi- tive hit-calls, only high selective inhibitors were used for the development of the model. These data were characterised by a demarcated separation of the AUR TPO assay log IC20 Table 4. Cont. 4.2. QSARs for Molecular Initiating Events ChEMBL datasets from Table 4 were used to develop 15 QSARs for molecular targets involved in the MIEs. Extended fingerprints (Daylight Chemical Information Systems, Inc., 2019) were calculated for each compound with a KNIME implementation [71] of the CDK toolkit (https://cdk.github.io/ (accessed on 7 March 2022)) and used as input for QSAR modelling. The BRF [72] implemented in KNIME was used for QSAR development. This technique artificially alters the class distribution in each tree. A sampling without repetition was made to select compounds, allowing all of the active compounds to always be selected together with an equal number of randomly selected inactive compounds from the training set in order to assure balancing between categories [73]. The number of trees in each BRF was set to 100. Models were validated by splitting each data set into a training (80%) and a test set (20%) by applying a stratification sampling to the activity classes. The splitting procedure was repeated 100 times using different random splits, ensuring that each chemical in the datasets was included in the test set the same number of times in order to avoid bias due to the molecules present in the different sets. The performance using the test set was calculated for each iteration, then the final performance of each model was calculated by averaging the statistical parameters obtained using the test sets relative to each of the 100 iterations. 4.1. Data Selection for Molecular Initiating Events (MIEs) Target Code CheMBL ID Species MIE ACT INA Transthyretin TTR CHEMBL3194 Human Q 93 3340 Sodium channel protein type N alpha subunit 2 VGSC CHEMBL1845 CHEMBL4187 CHEMBL5163 CHEMBL5202 Human U 167 3260 1 All the three isoforms of RYR were considered. 2 Isoforms 1, 2, 3 and 6 were considered. Table 4. Cont. 1 All the three isoforms of RYR were considered. 2 Isoforms 1, 2, 3 and 6 were considered. In order to prevent skew towards positive values, which is different from the natural distribution of biological data (i.e., few active, many inactive compounds), each dataset was further enriched with the chemicals included in the datasets of the remaining endpoints. These chemicals were treated as ‘decoys’ and assumed to be inactive. Due to the very large number of data available, AChE data were not used to enrich inactive samples of other endpoints to avoid the creation of datasets excessively unbalanced towards inactives. A semi-automated curation procedure [70] was applied to SMILES strings retrieved from ChEMBL in order to neutralise ionised chemical structures, remove counterions, and discard inorganics, organometallics, and mixtures. Removal of duplicate structures was carried out automatically at the InChI level. The entry with the maximum pChEMBL activity was selected in the case of duplicate structures in order to maximise the number of active samples. p Table 4 reports the final distribution of active and inactive chemicals for each dataset. The training sets for each of the modelled MIEs are available in the Supplementary Materials (Table S3). 4.1. Data Selection for Molecular Initiating Events (MIEs) bl f d d ll d h d h d h f Table 4. List of endpoints modelled using ChEMBL data. For each endpoint, the reference MIE, ChEMBL ID relative to the molecular target, species, and composition of the Training and Test sets are reported; ACT is the number of active compounds, while INA is the number of inactive compounds. Target Code CheMBL ID Species MIE ACT INA Acetylcholinesterase AChE CHEMBL220 Human I 3076 4793 Glutamate receptor ionotropic AMPA AMPAR CHEMBL2096670 Human A 73 3355 Nuclear receptor subfamily 1 group I member 3 (Constitutive Androstane Receptor) CAR CHEMBL5503 Human P 51 3377 Cytochrome P450 2E1 CYP2E1 CHEMBL5281 Human E 25 3402 GABA-A receptor; alpha-1/beta-2/gamma-2 GABAR CHEMBL2095172 Human V 129 3298 Glutamate receptor ionotropic kainate KAR CHEMBL2109241 Human A 25 3402 Mitochondrial complex I (NADH dehydrogenase) NADHOX CHEMBL614865 Bos taurus C, N 78 3349 Sodium/iodide cotransporter NIS CHEMBL2331047 Human F 56 3371 Glutamate [NMDA] receptor NMDAR CHEMBL2094124 Human A, B 267 3161 Pregnane X receptor PXR CHEMBL3401 Human P 244 3188 Ryanodine receptors 1 RYR CHEMBL2062 CHEMBL4403 CHEMBL1846 Human L 56 3371 Thyroid hormone receptor alpha THRα CHEMBL1860 Human S 311 3116 Thyroid hormone receptor beta THRβ CHEMBL1947 Human S 728 2704 Int. J. Mol. Sci. 2022, 23, 3053 10 of 18 10 of 18 Table 4. Cont. Target Code CheMBL ID Species MIE ACT INA Transthyretin TTR CHEMBL3194 Human Q 93 3340 Sodium channel protein type N alpha subunit 2 VGSC CHEMBL1845 CHEMBL4187 CHEMBL5163 CHEMBL5202 Human U 167 3260 1 All the three isoforms of RYR were considered. 2 Isoforms 1, 2, 3 and 6 were considered. In order to prevent skew towards positive values, which is different from the natural distribution of biological data (i.e., few active, many inactive compounds), each dataset was further enriched with the chemicals included in the datasets of the remaining endpoints. These chemicals were treated as ‘decoys’ and assumed to be inactive. Due to the very large number of data available, AChE data were not used to enrich inactive samples of other endpoints to avoid the creation of datasets excessively unbalanced towards inactives. A semi-automated curation procedure [70] was applied to SMILES strings retrieved from ChEMBL in order to neutralise ionised chemical structures, remove counterions, and discard inorganics, organometallics, and mixtures. Removal of duplicate structures was carried out automatically at the InChI level. 4.5. Neurotoxicity Data Predictions using single QSARs for MIEs of neurotoxicity were evaluated for their capability to predict the neurotoxic potential of chemicals. Neurotoxicity data were re- trieved from Kosnik et al. [76], who listed data for a total of 73 compounds (41 neuroactive and 32 non-neuroactive). This is a sub-selection of a list of the EPA’s ToxCast chemicals, previously tested by Strickland et al. [77] for their neural network function in vitro as mea- sured on primary cortical cultures grown on microelectrode arrays and then subsequently retested to confirm the measured activities. SMILES were retrieved from the chemical name and CAS number using the semi-automated data retrieval and curation procedure from Gadaleta et al. [70]. Four compounds (three neuroactive and one non-neuroactive) were removed because they were mixtures, inorganics, and/or organometallics, leading to a final dataset of 69 chemicals. The final list of 69 compounds along with their neurotoxic classification is reported in the Supplementary Materials (Table S4). 4.4. Reactivity SMARTS MIEs D (Binding and SH/SeH proteins involved in protection against oxidative stress) and H (Protein Adduct Formation) do not refer to an interaction with a specific receptor/enzyme; rather, they describe non-specific covalent binding to biological macro- molecules (i.e., proteins.). Because this type of binding refers to the intrinsic reactivity of molecules, SMARTS compiled by Enoch et al. [69] describing electrophilic protein binding reactions (71 SMARTS) were used to account for the two MIEs. Chemicals matching at least one of the SMARTS were flagged as positive (1); otherwise, they were negative (0). 4.3. Thyroperoxidase (TPO) Modelling As no data were found for TPO inhibition from ChEMBL, the QSAR model for predict- ing TPO inhibition proposed by Gadaleta et al. [68] was used. The model was developed from data related to the Amplex UltraRed-thyroperoxidase (AUR-TPO) assay. For posi- tive hit-calls, only high selective inhibitors were used for the development of the model. These data were characterised by a demarcated separation of the AUR-TPO assay log IC20 value from confounding activities reported by a luciferase inhibition assay (flagging for non-specific enzyme inhibition) and a cytotoxicity assay. The QSAR was based on a BRF developed with the imbalance-learner and scikit-learn Python libraries [74] and based on Int. J. Mol. Sci. 2022, 23, 3053 11 of 18 11 of 18 160 DRAGON descriptors [75] with a training set of 723 chemicals. Additional details on the predictive performance of the model can be found in [68]. 160 DRAGON descriptors [75] with a training set of 723 chemicals. Additional details on the predictive performance of the model can be found in [68]. 4.6. Neurotoxicity Modelling The MIEs for the 69 chemicals with data for neuroactivity were predicted with the 15 BRFs developed from the entire datasets in Table 4 using the BRF model to predict AUR-TPO from Gadaleta et al. [68], and were profiled with the SMARTS for electrophilic activity compiled by Enoch et al. [69]. The predictions for the 69 chemicals from Kosnik et al. [76] were reported in the form of probabilities associated with predictions, and are shown in Table S4 of the Supplementary Materials. Probabilities ranged from 0 to 1, and in the case of BRF are the percentage of trees within the BRF returning a ‘positive’ prediction. As a consequence, probabilities higher than 0.50 flag for positive predictions, while probabilities lower than 0.50 flag for negative predictions. Predictions equal to 0.50 were considered “not classified”. The predictions generated by the MIE models for the 69 chemicals were used as inde- pendent variables to develop new integrated QSAR models for predicting the neurotoxicity of chemicals. Three different classifiers able to naturally handle a high number of independent variables were trained based on the neurotoxicity data. Five different settings were applied for the various classifiers, as implemented in KNIME [71]. 1. K-Nearest Neighbours (k-NN) [78]: Euclidean distance was used to calculate the similarity between the target and the neighbours. K was varied from 1 and 9, with a step of 2. p 2. Random Forest (RF) [79]: the number of trees was varied from 50 and 250, with a step of 50. 3. Multi-linear Perceptron–Artificial Neural Networks (NNET) [80,81]: one hidden layer was used, with the number of hidden neurons varied from 2 to 12 with a step of 2. 3. Multi-linear Perceptron–Artificial Neural Networks (NNET) [80,81]: one hidden layer was used, with the number of hidden neurons varied from 2 to 12 with a step of 2. In order to verify the capability of MIE prediction to discriminate neuroactive and non-neuroactive compounds, QSARs based on MIEs were compared with other models developed with the same algorithms (kNN, RF and NNET) and different independent Int. J. Mol. Sci. 2022, 23, 3053 12 of 18 12 of 18 variables (i.e., extended structural fingerprints [82] and chemical descriptors). Chemical descriptors were calculated by means of DRAGON software [75,83]. The initial pool of descriptors calculated by DRAGON was pruned by constant and semi-constant values (standard deviation < 0.001). 4.6. Neurotoxicity Modelling Descriptors having at least one missing value were also discarded. In the case of highly correlated descriptors (absolute pair correlation > 0.90), only the one with the highest number of correlated descriptors was retained, while the others were discarded. This procedure led to a final pool of 747 descriptors. p p p Model performance was evaluated by five-fold cross validation. Fold-splitting was performed by applying a stratified sampling of the neurotoxicity categories. For each classifier and each selection of parameters, the seed applied when performing the split was maintained; thus, the various folds were always the same. The splitting procedure was repeated 100 times using the same list of 100 random seeds for each combination of classifiers and parameters; then, statistics were collected for each iteration. Considering the variation of splits and parameters, a total of 500 iterations were performed for each of the three classifiers. The same procedure was repeated in turns using MIE predictions, fingerprints, and DRAGON descriptors as independent variables, for a total of 1500 models developed. In the case of NNETs, DRAGON descriptors were preliminarily normalised in a range of 0–1, as NNETs are sensitive to the normalisation of independent variables. p Figure 2 summarises the entire procedure described above, including data extraction and curation, MIE modelling, and neurotoxicity modeling. p Figure 2 summarises the entire procedure described above, including data extraction and curation, MIE modelling, and neurotoxicity modeling. 4.7. Evaluation of MIE Relative Importance 4.7. Evaluation of MIE Relative Importance The relative importance of the various MIEs on the neurotoxicity predictions was evaluated in two ways. The relative importance of the various MIEs on the neurotoxicity predictions was evaluated in two ways. (1) MIEs were iteratively removed, then QSARs for neurotoxicity were developed with the remaining features, as described in Section 4.6. BAs were averaged among the various iterations and compared to the reference values of models developed using all of the variables. A reduction in performance after the removal of a specific MIE flags a strong relationship between the excluded MIE and neurotoxicity. On the contrary, MIEs are considered less relevant if their exclusion does not vary or improve baseline performance. (2) Variable importance was calculated for each MIE within RF models. A score was calculated based on the attribute usage statistics in the RF for each descriptor by counting how many times it was selected for a split (#split) and at which rank (“level”; the first two levels were considered) among all available attributes (#candidates) in the trees of the ensemble: Variable importance = #splits(level 0)/#candidates(level 0) + #splits(level 1)/#candidates(level 1) (1) (1) Variable importance = #splits(level 0)/#candidates(level 0) + #splits(level 1)/#candidates(level Variable importance calculated in this way was averaged among the various mod- elling iterations. Variable importance calculated in this way was averaged among the various mod- elling iterations. 13 of 18 13 of 18 Int. J. Mol. Sci. 2022, 23, 3053 Figure 2. Modeling workflow. The colours of the various blocks refer to the paragraph in Materials and Methods that describes the specific steps of the workflow. Data from ChEMBL for 15 targets relevant for the MIEs of neurotoxicity (red) were classified based on the threshold pChEMBL = 5; negative samples were enriched with data using “>” and “≥” qualifiers and with chemicals from other MIE data that were treated as decoys. QSARs for MIEs (blue) were developed from these datasets using the BRF method. Datasets were iteratively partitioned into training and test sets and their external performance was calculated as the average of the various iterations; then, the models were retrained on the whole datasets. Neurotoxicity data (green) were retrieved from [76] and curated at the SMILES level. Predictions from the thyroperoxidase model (violet) by [72] and reactivity SMARTS (cyan) by [75] were combined with the predictions from the 15 MIE modes and used as independent variables to develop neurotoxicity QSAR models (orange). 5. Conclusions In the present manuscript, a new take on the traditional QSAR methodology was proposed to predict neurotoxicity by employing the biological information associated with chemicals (in the form of ligand-based predictions of MIE activation data) in place of the traditional structural data. The main advantage of this approach is that it can both return a prediction of the adverse outcome and provide insights into the specific mechanisms and molecular events that trigger toxicity. Emphasising the specific mechanisms of ac- tion behind neurotoxicity will increase the confidence of scientists and regulators in the predictions returned by these models. Having information about the activated molecular targets that are responsible for an apical effect may, in some cases, provide indications to chemists of possible modifications to the structure of hazardous chemicals, allowing for the designing of safe alternatives. Despite their increasing usage, the application of the AOP framework in computational toxicology remains hampered by numerous and serious challenges. In general, an AOP is always a simplification of more complex and articulated biological pathways. Indeed, for certain biological processes, there are gaps in definitive knowledge of all responsible molecular determinants and mechanisms. In the case of neurotoxicity, there is a lack of understanding of all the MIEs involved in the alteration of downstream KEs as well as the occurrence of the AOs [4]. Several of the MIEs initially identified from AOPs were not included in the modelling presented here due to the shortage of data. Gaps in knowledge regarding chemical concentrations and time of exposure to trigger MIE/KEs prevent the development of quantitative approaches and limit the development of AOPs for adult and developmental neurotoxicity mainly to qualitative ones [4,27]. Considering the fact that the AOPs studied in this work are likely to be incomplete, the results described here are even more encouraging. Indeed, the future availability of more high-quality data is likely to improve the predictive capability of single QSARs for MIEs, while the future availability of more detailed AOPs and the inclusion of additional MIEs will complete the overall infrastructure, possibly leading to a more accurate and reliable prediction of apical endpoints. The incorporation of exposure and toxicokinetics, i.e., absorption (e.g., blood–brain barrier penetration), distribution, metabolism, and excretion data represent a possible additional improvement of the results presented herein [84]. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijms23063053/s1. Author Contributions: Conceptualization, D.G. 4.7. Evaluation of MIE Relative Importance kNN, RF, and NNET were used to develop models. The use of MIE predictions as independent variables was benchmarked with fingerprints and DRAGON descriptors; then, the performance of the obtained models was compared with five-fold cross validation. Figure 2. Modeling workflow. The colours of the various blocks refer to the paragraph in Materials and Methods that describes the specific steps of the workflow. Data from ChEMBL for 15 targets relevant for the MIEs of neurotoxicity (red) were classified based on the threshold pChEMBL = 5; negative samples were enriched with data using “>” and “≥” qualifiers and with chemicals from other MIE data that were treated as decoys. QSARs for MIEs (blue) were developed from these datasets using the BRF method. Datasets were iteratively partitioned into training and test sets and their external performance was calculated as the average of the various iterations; then, the models were retrained on the whole datasets. Neurotoxicity data (green) were retrieved from [76] and curated at the SMILES level. Predictions from the thyroperoxidase model (violet) by [72] and reactivity SMARTS (cyan) by [75] were combined with the predictions from the 15 MIE modes and used as independent variables to develop neurotoxicity QSAR models (orange). kNN, RF, and NNET were used to develop models. The use of MIE predictions as independent variables was benchmarked with fingerprints and DRAGON descriptors; then, the performance of the obtained models was compared with five-fold cross validation. Int. J. Mol. Sci. 2022, 23, 3053 14 of 18 14 of 18 5. Conclusions and N.S.; methodology, D.G.; validation, D.G., data curation, D.G. and N.S.; writing—original draft preparation, D.G.; writing—review and editing, N.S., A.R., M.T.D.C. and E.B.; supervision, A.R., M.T.D.C. and E.B.; project administration, A.R., M.T.D.C. and E.B.; funding acquisition, M.T.D.C. and E.B. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the European Union’s Horizon 2020 research and innova- tion programme under Grant Agreement No. 963845 (ONTOX) and the European Union’s Marie Skłodowska-Curie Action ‘in3’: MSCA-ITN-2016, Grant No. 721975. ta Availability Statement: ThedatapresentedinthisstudyareavailableintheSupplementary Materials. Data Availability Statement: ThedatapresentedinthisstudyareavailableintheSupplementary Materials. Acknowledgments: D.G. acknowledges the grant from the LUSH Prize 2020 (Category: Young Researchers) from LUSH and Ethical Consumer. Acknowledgments: D.G. acknowledges the grant from the LUSH Prize 2020 (Category: Young Researchers) from LUSH and Ethical Consumer. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. Abbreviations AMPAR alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate receptor AChE acetylcholinesterase AO adverse outcome Int. J. Mol. Sci. 2022, 23, 3053 15 of 18 15 of 18 AOP adverse outcome pathway AUC area under the ROC curve AUR-TPO Amplex UltraRed-thyroperoxidase BA balanced accuracy BRF balanced random forest CAR constitutive androstane receptor CYP2E1 cytochrome P450 2E1 DNT developmental neurotoxicity FN false negative FP false positive GABAR GABA receptor IATA integrated approaches to testing and assessment KA kainite receptor KE key event k-NN k-nearest neighbors MCC Matthew’s correlation coefficient MIE molecular initiating event NADHOX NADH-quinone oxidoreductase NIS Na+/I−symporter ND neurodegenerative disease NNET neural networks NMDAR N-methyl-D-aspartate receptor OATP1C1 solute carrier organic anion transporter family member 1C14 PCB polychlorinated biphenyls PXR pregnane X receptor QSAR quantitative structure-activity relationship RF random forest ROS reactive oxygen species RyR ryanodine-sensitive Ca2+ channel SEN sensitivity SPE specificity THRα thyroid hormone receptor α THRβ thyroid hormone receptor β TN true negative TP true positive TPO thyroperoxidase TTR transtyretrin serum binding protein VGSC voltage-gated sodium channel 9. Trippier, P.C.; Jansen Labby, K.; Hawker, D.D.; Mataka, J.J.; Silverman, R.B. Target-and Mechanism-Based Therapeutics for Neurodegenerative Diseases: Strength in Numbers. J. Med. Chem. 2013, 56, 3121–3147. [CrossRef] 7. Bandeen-Roche, K.; Glass, T.A.; Bolla, K.I. Cumulative Lead Dose and Cognitive Function in Older Adults. Altern. Med. Rev. 2010, 15, 112–113. [CrossRef] [PubMed] References Ankley, G.T.; Bennett, R.S.; Erickson, R.J.; Hoff, D.J.; Hornung, M.W.; Johnson, R.D.; Mount, D.R.; Nichols, J.W.; Russom, C.L.; Schmieder, P.K. Adverse Outcome Pathways: A Conceptual Framework to Support Ecotoxicology Research and Risk Assessment. Environ. Toxicol. Chem. Int. J. 2010, 29, 730–741. [CrossRef] 18. Vinken, M. The Adverse Outcome Pathway Concept: A Pragmatic Tool in Toxicology. Toxicology 2013, 312, 158–165. [CrossRef] 18. Vinken, M. The Adverse Outcome Pathway Concept: A Pragmatic Tool in Toxicology. Toxicology 2013, 312, 158–165. [CrossRef] 19. Leist, M.; Ghallab, A.; Graepel, R.; Marchan, R.; Hassan, R.; Bennekou, S.H.; Limonciel, A.; Vinken, M.; Schildknecht, S.; Waldmann, T.; et al. Adverse Outcome Pathways: Opportunities, Limitations and Open Questions. Arch. Toxicol. 2017, 91, 3477–3505. [CrossRef] 20. Allen, T.E.; Goodman, J.M.; Gutsell, S.; Russell, P.J. A History of the Molecular Initiating Event. Chem. Res. Toxicol. 2016, 29, 2060–2070. [CrossRef] 21. Allen, T.E.; Goodman, J.M.; Gutsell, S.; Russell, P.J. Quantitative Predictions for Molecular Initiat Dimensional Quantitative Structure–Activity Relationships. Chem. Res. Toxicol. 2019, 33, 324–332. [Cro 22. Benigni, R. Building Predictive Adverse Outcome Pathway Models: Role of Molecular Initiating Events and Structure–Activity Relationships. Appl. Vitr. Toxicol. 2017, 3, 265–270. [CrossRef] p pp 23. Cronin, M.T.; Richarz, A.-N. Relationship between Adverse Outcome Pathways and Chemistry-Based in Silico Models to Predict Toxicity. Appl. Vitr. Toxicol. 2017, 3, 286–297. [CrossRef] y pp 24. Gadaleta, D.; Manganelli, S.; Roncaglioni, A.; Toma, C.; Benfenati, E.; Mombelli, E. QSAR Modeling of ToxCast Assays Relevant to the Molecular Initiating Events of AOPs Leading to Hepatic Steatosis. J. Chem. Inf. Model. 2018, 58, 1501–1517. [CrossRef] [PubMed] 25. Patlewicz, G.; Simon, T.W.; Rowlands, J.C.; Budinsky, R.A.; Becker, R.A. Proposing a Scientific Confidence Framework to Help Support the Application of Adverse Outcome Pathways for Regulatory Purposes. Regul. Toxicol. Pharmacol. 2015, 71, 463–477. [CrossRef] [PubMed] 26. Tollefsen, K.E.; Scholz, S.; Cronin, M.T.; Edwards, S.W.; de Knecht, J.; Crofton, K.; Garcia-Reyero, N.; Hartung, T.; Worth, A.; Patlewicz, G. Applying Adverse Outcome Pathways (AOPs) to Support Integrated Approaches to Testing and Assessment (IATA). Regul. Toxicol. Pharmacol. 2014, 70, 629–640. [CrossRef] 27. Li, J.; Settivari, R.; LeBaron, M.J.; Marty, M.S. An Industry Perspective: A Streamlined Screening Strategy Using Alternative Models for Chemical Assessment of Developmental Neurotoxicity. Neurotoxicology 2019, 73, 17–30. [CrossRef] [PubMed] 28. Marx-Stoelting, P.; de LM Solano, M.; Aoyama, H.; Adams, R.H.; Bal-Price, A.; Buschmann, J.; Chahoud, I.; Clark, R.; Fang, T.; Fujiwara, M.; et al. References The Macroeconomics of Dementia—Will the World Economy Get Alzheimer’s Disease? Arch. Med. Res. 2012, 43, 705–709. [CrossRef] [ ] 11. Grandjean, P.; Landrigan, P.J. Neurobehavioural Effects of Developmental Toxicity. Lancet Neurol. 2014, 13, 330–338. [CrossRef] 12. Grandjean, P.; Landrigan, P.J. Developmental Neurotoxicity of Industrial Chemicals. Lancet 2006, 368, 2167–2178. [CrossRef] [ ] 11. Grandjean, P.; Landrigan, P.J. Neurobehavioural Effects of Developmental Toxicity. Lancet Neurol. 2014, 13, 330–338. [CrossRef] djean, P.; Landrigan, P.J. Neurobehavioural Effects of Developmental Toxicity. Lancet Neurol. 2014, 13, 330–3 djean, P.; Landrigan, P.J. Developmental Neurotoxicity of Industrial Chemicals. Lancet 2006, 368, 2167–2178. j g p y 12. Grandjean, P.; Landrigan, P.J. Developmental Neurotoxicity of Industrial Chemicals. Lancet 2006, 368, 2 , A.; Pistollato, F.; Sachana, M.; Bopp, S.K.; Munn, S.; Worth, A. Strategies to Improve the Regulatory Ass 13. Bal-Price, A.; Pistollato, F.; Sachana, M.; Bopp, S.K.; Munn, S.; Worth, A. Strategies to Improve the Regulatory Assessment of Developmental Neurotoxicity (DNT) Using in Vitro Methods. Toxicol. Appl. Pharmacol. 2018, 354, 7–18. [CrossRef] [PubMed] , A.; Pistollato, F.; Sachana, M.; Bopp, S.K.; Munn, S.; Worth, A. Strategies to Improve the Regulatory Ass mental Neurotoxicity (DNT) Using in Vitro Methods. Toxicol. Appl. Pharmacol. 2018, 354, 7–18. [CrossRef] [P 13. Bal-Price, A.; Pistollato, F.; Sachana, M.; Bopp, S.K.; Munn, S.; Worth, A. Strategies to Improve the Regulatory Assessment of Developmental Neurotoxicity (DNT) Using in Vitro Methods. Toxicol. Appl. Pharmacol. 2018, 354, 7–18. [CrossRef] [PubMed] ; , ; , ; pp, ; , ; , g p g y tal Neurotoxicity (DNT) Using in Vitro Methods. Toxicol. Appl. Pharmacol. 2018, 354, 7–18. [CrossRef] [PubM 14. Tsuji, R.; Crofton, K.M. Developmental Neurotoxicity Guideline Study: Issues with Methodology, Evaluation and Regulation. Congenit. Anom. 2012, 52, 122–128. [CrossRef] [PubMed] g 15. Collins, F.S.; Gray, G.M.; Bucher, J.R. Transforming Environmental Health Protection. Science 2008, 319, 9 15. Collins, F.S.; Gray, G.M.; Bucher, J.R. Transforming Environmental Health Protection. Science 2008, 319, 906. [CrossRef] [PubMed] 16. Dearden, J.C. The History and Development of Quantitative Structure-Activity Relationships (QSARs). In Oncology: Breakthroughs in Research and Practice; IGI Global: Hershey, PA, USA, 2017; pp. 67–117. , ; y, ; , J g , , 16. Dearden, J.C. The History and Development of Quantitative Structure-Activity Relationships (QSARs). I in Research and Practice; IGI Global: Hershey, PA, USA, 2017; pp. 67–117. y pp 17. References 1. Landrigan, P.J.; Lambertini, L.; Birnbaum, L.S. A Research Strategy to Discover the Environmental Causes of Autism and Neurodevelopmental Disabilities. Environ. Health Perspect. 2012, 120, a258–a260. [CrossRef] [PubMed] 1. Landrigan, P.J.; Lambertini, L.; Birnbaum, L.S. A Research Strategy to Discover the Environme Neurodevelopmental Disabilities. Environ. Health Perspect. 2012, 120, a258–a260. [CrossRef] [PubM p p [ ] [ ] 2. Landrigan, P.J.; Sonawane, B.; Butler, R.N.; Trasande, L.; Callan, R.; Droller, D. Early Environmental Orig Di i L t Lif E i H lth P t 2005 113 1230 1233 [C R f] [P bM d] p p n, P.J.; Sonawane, B.; Butler, R.N.; Trasande, L.; Callan, R.; Droller, D. Early Environmental Origins of Neurode p p 2. Landrigan, P.J.; Sonawane, B.; Butler, R.N.; Trasande, L.; Callan, R.; Droller, D. Early Environmental Disease in Later Life. Environ. Health Perspect. 2005, 113, 1230–1233. [CrossRef] [PubMed] 2. Landrigan, P.J.; Sonawane, B.; Butler, R.N.; Trasande, L.; Callan, R.; Droller, D. Early Environmental Origins of Neurodegenerative Disease in Later Life. Environ. Health Perspect. 2005, 113, 1230–1233. [CrossRef] [PubMed] 2. Landrigan, P.J.; Sonawane, B.; Butler, R.N.; Trasande, L.; Callan, R.; Droller, D. Early Environmental Origins of Neurodegenerative Disease in Later Life. Environ. Health Perspect. 2005, 113, 1230–1233. [CrossRef] [PubMed] g , J ; , ; , ; , ; , ; , y Disease in Later Life. Environ. Health Perspect. 2005, 113, 1230–1233. [CrossRef] [PubMed] 3. Lein, P.; Silbergeld, E.; Locke, P.; Goldberg, A.M. In Vitro and Other Alternative Approaches to Developmental Neurotoxicity Testing (DNT). Environ. Toxicol. Pharmacol. 2005, 19, 735–744. [CrossRef] [PubMed] , ; g , ; , ; g, pp p y Testing (DNT). Environ. Toxicol. Pharmacol. 2005, 19, 735–744. [CrossRef] [PubMed] 6. Bloom, B.; Cohen, R.A.; Freeman, G. Summary Health Statistics for US Children: National Health Interview Survey, 2009; National Center for Health Statistics: Hyattsville, MD, USA, 2010; Volume 247, pp. 1–82. 7. Bandeen-Roche, K.; Glass, T.A.; Bolla, K.I. Cumulative Lead Dose and Cognitive Function in Older Adults. Altern. Med. Rev. 2010, 15, 112–113. [CrossRef] [PubMed] 8. Narayan, P.; Ehsani, S.; Lindquist, S. Combating Neurodegenerative Disease with Chemical Prob Chem. Biol. 2014, 10, 911–920. [CrossRef] 9. Trippier, P.C.; Jansen Labby, K.; Hawker, D.D.; Mataka, J.J.; Silverman, R.B. Target-and Mechanism-Based Therapeutics for Neurodegenerative Diseases: Strength in Numbers. J. Med. Chem. 2013, 56, 3121–3147. [CrossRef] 16 of 18 Int. J. Mol. Sci. 2022, 23, 3053 16 of 18 10. Banerjee, S. References 25th Anniversary of the Berlin Workshop on Developmental Toxicology: DevTox Database Update, Challenges in Risk Assessment of Developmental Neurotoxicity and Alternative Methodologies in Bone Development and Growth. Reprod. Toxicol. 2021, 100, 155–162. [CrossRef] [PubMed] 29. Lenselink, E.B.; Ten Dijke, N.; Bongers, B.; Papadatos, G.; Van Vlijmen, H.W.; Kowalczyk, W.; Ijzerman, A.P.; Van Westen, G.J. Beyond the Hype: Deep Neural Networks Outperform Established Methods Using a ChEMBL Bioactivity Benchmark Set. J. Cheminform. 2017, 9, 1–14. [CrossRef] f 30. Bosc, N.; Atkinson, F.; Felix, E.; Gaulton, A.; Hersey, A.; Leach, A.R. Large Scale Comparison of QSAR and Conformal Prediction Methods and Their Applications in Drug Discovery. J. Cheminform. 2019, 11, 4. [CrossRef] 31. Couratier, P.; Sindou, P.; Hugon, J.; Vallat, J.-M.; Dumas, M. Cell Culture Evidence for Neuronal Degeneration in Amyotrophic Lateral Sclerosis Being Linked to Glutamate AMPA/Kainate Receptors. Lancet 1993, 341, 265–268. [CrossRef] 31. Couratier, P.; Sindou, P.; Hugon, J.; Vallat, J.-M.; Dumas, M. Cell Culture Evidence for Neuronal Degeneration in Amyotrophic Lateral Sclerosis Being Linked to Glutamate AMPA/Kainate Receptors. Lancet 1993, 341, 265–268. [CrossRef] 32. Weiss, J.; Yin, H.-Z.; Choi, D. Basal Forebrain Cholinergic Neurons Are Selectively Vulnerable to AMPA/Kainate Receptor- Mediated Neurotoxicity. Neuroscience 1994, 60, 659–664. [CrossRef] g p 32. Weiss, J.; Yin, H.-Z.; Choi, D. Basal Forebrain Cholinergic Neurons Are Selectively Vulnerable to AMPA/Kainate Receptor- Mediated Neurotoxicity. Neuroscience 1994, 60, 659–664. [CrossRef] 33. Muratov, E.N.; Bajorath, J.; Sheridan, R.P.; Tetko, I.V.; Filimonov, D.; Poroikov, V.; Oprea, T.I.; Baskin, I.I.; Varnek, A.; Roitberg, A.; et al. QSAR without Borders. Chem. Soc. Rev. 2020, 49, 3525–3564. [CrossRef] [PubMed] 34. Polishchuk, P.G.; Muratov, E.N.; Artemenko, A.G.; Kolumbin, O.G.; Muratov, N.N.; Kuz’min, V.E. Application of Random Forest Approach to QSAR Prediction of Aquatic Toxicity. J. Chem. Inf. Model. 2009, 49, 2481–2488. [CrossRef] [PubMed] 35 S t ik V Li A T C C lb J C Sh id R P F t B P R d F t A Cl ifi ti d R i T l 34. Polishchuk, P.G.; Muratov, E.N.; Artemenko, A.G.; Kolumbin, O.G.; Muratov, N.N.; Kuz’min, V.E. Appl Approach to QSAR Prediction of Aquatic Toxicity. J. Chem. Inf. Model. 2009, 49, 2481–2488. [CrossRef] 34. Polishchuk, P.G.; Muratov, E.N.; Artemenko, A.G.; Kolumbin, O.G.; Muratov, N.N.; Kuz’min, V.E. Application of Random Forest Approach to QSAR Prediction of Aquatic Toxicity. J. Chem. Inf. Model. 2009, 49, 2481–2488. [CrossRef] [PubMed] 35. Svetnik, V.; Liaw, A.; Tong, C.; Culberson, J.C.; Sheridan, R.P.; Feuston, B.P. References Random Forest: A Classification and Regression Tool 35. Svetnik, V.; Liaw, A.; Tong, C.; Culberson, J.C.; Sheridan, R.P.; Feuston, B.P. Random Forest: A Classifica for Compound Classification and QSAR Modeling. J. Chem. Inf. Comput. Sci. 2003, 43, 1947–1958. [Cro 36. Masjosthusmann, S.; Barenys, M.; El-Gamal, M.; Geerts, L.; Gerosa, L.; Gorreja, A.; Kühne, B.; Marchetti, N.; Tigges, J.; Viviani, B.; et al. Literature Review and Appraisal on Alternative Neurotoxicity Testing Methods. EFSA Support. Publ. 2018, 15, 1410E. [CrossRef] 17 of 18 17 of 18 Int. J. Mol. Sci. 2022, 23, 3053 37. Choi, J.; Polcher, A.; Joas, A. Systematic Literature Review on Parkinson’s Disease and Childhood Leukaemia and Mode of Actions for Pesticides. EFSA Support. Publ. 2016, 13, 955E. [CrossRef] pp 38. Gadaleta, D.; Vukovi´c, K.; Toma, C.; Lavado, G.J.; Karmaus, A.L.; Mansouri, K.; Kleinstreuer, N.C.; Benfenati, E.; Roncaglioni, A. SAR and QSAR Modeling of a Large Collection of LD 50 Rat Acute Oral Toxicity Data. J. Cheminform. 2019, 11, 58. [CrossRef] g g y f ortés-Ciriano, I.; Dehaen, W.; Kˇríž, P.; van Westen, G.J.; Tetko, I.V.; Bender, A.; Svozil, D. QSAR-Derived A 39. Škuta, C.; Cortés-Ciriano, I.; Dehaen, W.; Kˇríž, P.; van Westen, G.J.; Tetko, I.V.; Bender, A.; Svozil, D. QSAR-Derived Affinity Fingerprints (Part 1): Fingerprint Construction and Modeling Performance for Similarity Searching, Bioactivity Classification and Scaffold Hopping. J. Cheminform. 2020, 12, 39. [CrossRef] pp g f 40. Rice, D.; Barone, S., Jr. Critical Periods of Vulnerability for the Developing Nervous System: Evidence from Humans and Animal Models. Environ. Health Perspect. 2000, 108 (Suppl. 3), 511–533. pp g 40. Rice, D.; Barone, S., Jr. Critical Periods of Vulnerability for the Developing N Models. Environ. Health Perspect. 2000, 108 (Suppl. 3), 511–533. p , ( pp ), 41. Cruz-Monteagudo, M.; Medina-Franco, J.L.; Perez-Castillo, Y.; Nicolotti, O.; Cordeiro, M.N.D.; Borges, F. Activity Cliffs in Drug Discovery: Dr Jekyll or Mr Hyde? Drug Discov. Today 2014, 19, 1069–1080. [CrossRef] p ( pp ) gudo, M.; Medina-Franco, J.L.; Perez-Castillo, Y.; Nicolotti, O.; Cordeiro, M.N.D.; Borges, F. Activity Cliffs i uz-Monteagudo, M.; Medina-Franco, J.L.; Perez-Castillo, Y.; Nicolotti, O.; Cordeiro, M.N.D.; Borges, F. Activ scovery: Dr Jekyll or Mr Hyde? Drug Discov. Today 2014, 19, 1069–1080. [CrossRef] g J g y r Jekyll or Mr Hyde? Drug Discov. Today 2014, 19, 1069–1080. [CrossRef] y g y mpagne, F.A.; Cory-Slechta, D.A.; Dishaw, L.; Faustman, E.; Mundy, W.; Segal, D.; Sobin, C.; Starkey, C. 42. References Carlson, L.M.; Champagne, F.A.; Cory-Slechta, D.A.; Dishaw, L.; Faustman, E.; Mundy, W.; Segal, D.; Sobin, C.; Starkey, C.; Taylor, M. Potential Frameworks to Support Evaluation of Mechanistic Data for Developmental Neurotoxicity Outcomes: A Symposium Report. Neurotoxicol. Teratol. 2020, 78, 106865. [CrossRef] p g y y g y Taylor, M. Potential Frameworks to Support Evaluation of Mechanistic Data for Developmental Neurotoxicity Outcomes: A Symposium Report. Neurotoxicol. Teratol. 2020, 78, 106865. [CrossRef] y p p 43. Fritsche, E.; Grandjean, P.; Crofton, K.M.; Aschner, M.; Goldberg, A.; Heinonen, T.; Hessel, E.V.; Hogberg, H.T.; Bennekou, S.H.; Lein, P.J. Consensus Statement on the Need for Innovation, Transition and Implementation of Developmental Neurotoxicity (DNT) Testing for Regulatory Purposes. Toxicol. Appl. Pharmacol. 2018, 354, 3–6. [CrossRef] [PubMed] 44. Villeneuve, D.L.; Crump, D.; Garcia-Reyero, N.; Hecker, M.; Hutchinson, T.H.; LaLone, C.A.; Landesmann, B.; Lettieri, T.; Munn, S.; Nepelska, M.; et al. Adverse Outcome Pathway (AOP) Development I: Strategies and Principles. Toxicol. Sci. 2014, 142, 312–320. [CrossRef] [PubMed] 45. Spînu, N.; Bal-Price, A.; Cronin, M.T.; Enoch, S.J.; Madden, J.C.; Worth, A.P. Development and Ana Pathway Network for Human Neurotoxicity. Arch. Toxicol. 2019, 93, 2759–2772. [CrossRef] 46. Schrattenholz, A.; Soskic, V. NMDA Receptors Are Not Alone: Dynamic Regulation of NMDA Receptor Structure and Function by Neuregulins and Transient Cholesterol-Rich Membrane Domains Leads to Disease-Specific Nuances of Glutamate-Signalling. Curr. Top. Med. Chem. 2006, 6, 663–686. [CrossRef] [PubMed] p 47. Toscano, C.D.; Guilarte, T.R. Lead Neurotoxicity: From Exposure to Molecular Effects. Brain Res. Rev. 2005, 49, 529–554. [CrossRef] 48. Farina, M.; Rocha, J.B.; Aschner, M. Mechanisms of Methylmercury-Induced Neurotoxicity: Evidence from Experimental Studies. Life Sci. 2011, 89, 555–563. [CrossRef] 47. Toscano, C.D.; Guilarte, T.R. Lead Neurotoxicity: From Exposure to Molecular Effects. Brain Res. Rev. 48. Farina, M.; Rocha, J.B.; Aschner, M. Mechanisms of Methylmercury-Induced Neurotoxicity: Evidenc Life Sci. 2011, 89, 555–563. [CrossRef] 49. Haorah, J.; Ramirez, S.H.; Floreani, N.; Gorantla, S.; Morsey, B.; Persidsky, Y. Mechanism of Alcohol-Induced Oxidative Stress and Neuronal Injury. Free. Radic. Biol. Med. 2008, 45, 1542–1550. [CrossRef] j y 50. Valencia-Olvera, A.C.; Morán, J.; Camacho-Carranza, R.; Prospéro-García, O.; Espinosa-Aguirre, J.J. CYP2E1 Induction Leads to Oxidative Stress and Cytotoxicity in Glutathione-Depleted Cerebellar Granule Neurons. Toxicol. Vitr. 2014, 28, 1206–1214. [CrossRef] 51. De la Vieja, A.; Dohan, O.; Levy, O.; Carrasco, N. Molecular Analysis of the Sodium/Iodide Symporter: Impact on Thyroid and Extrathyroid Pathophysiology. Physiol. Rev. 2000, 80, 1083–1105. [CrossRef] 52. References Dohan, O.; De la Vieja, A.; Paroder, V.; Riedel, C.; Artani, M.; Reed, M.; Ginter, C.S.; Carrasco, N. The Sodium/Iodide Symporter (NIS): Characterization, Regulation, and Medical Significance. Endocr. Rev. 2003, 24, 48–77. [CrossRef] 53. Darvesh, S.; Hopkins, D.A.; Geula, C. Neurobiology of Butyrylcholinesterase. Nat. Rev. Neurosci. 2003, 4, 131–138. [CrossRef] 54. US Environmental Protection Agency. The Use of Data on Cholinesterase Inhibition for Risk Assessments of Organophosphorous and Carbamate Pesticides; US Environmental Protection Agency: Washington, DC, USA, 2000. 54. US Environmental Protection Agency. The Use of Data on Cholinesterase Inhibition for Risk Assessments of Organophosphorous and Carbamate Pesticides; US Environmental Protection Agency: Washington, DC, USA, 2000. g y g 55. Holland, E.B.; Feng, W.; Zheng, J.; Dong, Y.; Li, X.; Lehmler, H.-J.; Pessah, I.N. An Extended Structure–Activity Relationship of Nondioxin-like PCBs Evaluates and Supports Modeling Predictions and Identifies Picomolar Potency of PCB 202 towards Ryanodine Receptors. Toxicol. Sci. 2017, 155, 170–181. [CrossRef] [PubMed] y p 56. Paul Friedman, K.; Watt, E.D.; Hornung, M.W.; Hedge, J.M.; Judson, R.S.; Crofton, K.M.; Houck, K.A.; Simmons, S.O. Tiered High-Throughput Screening Approach to Identify Thyroperoxidase Inhibitors within the ToxCast Phase I and II Chemical Libraries. Toxicol. Sci. 2016, 151, 160–180. [CrossRef] 57. Zoeller, R.; Rovet, J. Timing of Thyroid Hormone Action in the Developing Brain: Clinical Observations and Experimental Findings. J. Neuroendocrinol. 2004, 16, 809–818. [CrossRef] [PubMed] 57. Zoeller, R.; Rovet, J. Timing of Thyroid Hormone Action in the Developing Brain: Cli Findings. J. Neuroendocrinol. 2004, 16, 809–818. [CrossRef] [PubMed] g 58. Bernal, J. Thyroid Hormone Receptors in Brain Development and Function. Nat. Clin. Pract. Endocrinol. Metab. 2007, 3, 249–259. [CrossRef] [PubMed] 59. Westerholz, S.; De Lima, A.; Voigt, T. Regulation of Early Spontaneous Network Activity and GABAergic Neurons Development by Thyroid Hormone. Neuroscience 2010, 168, 573–589. [CrossRef] 60. Crofton, K.M.; Zoeller, R.T. Mode of Action: Neurotoxicity Induced by Thyroid Hormone Disruption during Development— Hearing Loss Resulting from Exposure to PHAHs. Crit. Rev. Toxicol. 2005, 35, 757–769. [CrossRef] [PubMed] 61. Van Maele-Fabry, G.; Hoet, P.; Vilain, F.; Lison, D. Occupational Exposure to Pesticides and Parkinson Review and Meta-Analysis of Cohort Studies. Environ. Int. 2012, 46, 30–43. [CrossRef] [PubMed] 62. Goldin, A.L. Resurgence of Sodium Channel Research. Annu. Rev. Physiol. 2001, 63, 871–894. [CrossRef] 63. Soderlund, D.M. Molecular Mechanisms of Pyrethroid Insecticide Neurotoxicity: Recent Advances. Arch. Toxicol. 2012, 86, 165–181. [CrossRef] 18 of 18 Int. J. Mol. Sci. 2022, 23, 3053 64. McGonigle, I.; Lummis, S.C. References y pp uaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V. Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. g y 75. Kode: DRAGON 7.0.8. 2017. Available online: https://chm.kode-solutions.net/products_dragon.php (accessed on 7 March 2022). 75. Kode: DRAGON 7.0.8. 2017. Available online: https://chm.kode-solutions.net/products_dragon.php 75. Kode: DRAGON 7.0.8. 2017. Available online: https://chm.kode-solutions.net/products_dragon.php (accessed on 7 March 2022). 76. Kosnik, M.B.; Strickland, J.D.; Marvel, S.W.; Wallis, D.J.; Wallace, K.; Richard, A.M.; Reif, D.M.; Shafer, T.J. Concentration– Response Evaluation of ToxCast Compounds for Multivariate Activity Patterns of Neural Network Function. Arch. Toxicol. 2020, 94, 469–484. [CrossRef] [PubMed] 77. Strickland, J.D.; Martin, M.T.; Richard, A.M.; Houck, K.A.; Shafer, T.J. Screening the ToxCast Phase II Libraries for Alterations in Network Function Using Cortical Neurons Grown on Multi-Well Microelectrode Array (MwMEA) Plates. Arch. Toxicol. 2018, 92, 487–500. [CrossRef] [PubMed] 78. Altman, N.S. An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression. Am. Stat. 1992, 46, 175–185. 78. Altman, N.S. An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression. Am. Stat. 1992, 46, 175–185. 79. Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [CrossRef] Random Forests. Mach. Learn. 2001, 45, 5–32. [CrossR 80. Dudek, A.Z.; Arodz, T.; Gálvez, J. Computational Methods in Developing Quantitative Structure-Activity Relationships (QSAR): A Review. Comb. Chem. High Throughput Screen. 2006, 9, 213–228. [CrossRef] [PubMed] 80. Dudek, A.Z.; Arodz, T.; Gálvez, J. Computational Methods in Developing Quantitative Structur A Review. Comb. Chem. High Throughput Screen. 2006, 9, 213–228. [CrossRef] [PubMed] 81. Jain, A.K.; Mao, J.; Mohiuddin, K.M. Artificial Neural Networks: A Tutorial. Computer 1996, 29, 31–44. [CrossRef] 82. Daylight Chemical Information Systems, Inc. 6. Fingerprints—Screening and Similarity. 2019. Available online: https://www. d l h /d h l/d / h / h fi h l ( d ) p 82. Daylight Chemical Information Systems, Inc. 6. Fingerprints—Screening and Similarity. 2019. Available online: https://www. daylight.com/dayhtml/doc/theory/theory.finger.html (accessed on 26 January 2022). y g y y y g ( J y ) 83. Todeschini, R.; Consonni, V. Handbook of Molecular Descriptors; John Wiley & Sons: Hoboken, NJ, USA, 2008; Volume 11. 84. Blaauboer, B.J. The Integration of Data on Physico-Chemical Properties, in Vitro-Derived Toxicity Data and Physiologically Based Kinetic and Dynamic as Modelling a Tool in Hazard and Risk Assessment. A Commentary. Toxicol. Lett. 2003, 138, 161–171. y g y y y g ( J y ) 83. Todeschini, R.; Consonni, V. 64. McGonigle, I.; Lummis, S.C. Molecular Characterization of Agonists That Bind to an Insect GABA Receptor. Biochemistry 2010, 49, 2897–2902. [CrossRef] [PubMed] References Molecular Characterization of Agonists That Bind to an Insect GABA Receptor. Biochemistry 2010, 49, 2897–2902. [CrossRef] [PubMed] [ ] [ ] 65. Gong, P.; Hong, H.; Perkins, E.J. Ionotropic GABA Receptor Antagonism-Induced Adverse Outcome Pathways for Potential Neurotoxicity Biomarkers. Biomark. Med. 2015, 9, 1225–1239. [CrossRef] [PubMed] y 66. Gaulton, A.; Hersey, A.; Nowotka, M.; Bento, A.P.; Chambers, J.; Mendez, D.; Mutowo, P.; Atkinson, F.; Bellis, L.J.; Cibrián-Uhalte, E.; et al. The ChEMBL Database in 2017. Nucleic Acids Res. 2017, 45, D945–D954. [CrossRef] [PubMed] 67. Bento, A.P.; Gaulton, A.; Hersey, A.; Bellis, L.J.; Chambers, J.; Davies, M.; Krüger, F.A.; Light, Y.; Mak, L.; McGlinchey, S. The ChEMBL Bioactivity Database: An Update. Nucleic Acids Res. 2014, 42, D1083–D1090. [CrossRef] [PubMed] 68. Gadaleta, D.; d’Alessandro, L.; Marzo, M.; Benfenati, E.; Roncaglioni, A. Quantitative Structure-Ac of the Amplex Ultrared Assay to Predict Thyroperoxidase Inhibitory Activity. Front. Pharmacol. 202 68. Gadaleta, D.; d’Alessandro, L.; Marzo, M.; Benfenati, E.; Roncaglioni, A. Quantitative Structure-Activity Relationship Modeling of the Amplex Ultrared Assay to Predict Thyroperoxidase Inhibitory Activity. Front. Pharmacol. 2021, 12, 713037. [CrossRef] 69. Enoch, S.J.; Madden, J.C.; Cronin, M.T.D. Identification of Mechanisms of Toxic Action for Skin Sensit Pattern Based Approach. SAR QSAR Environ. Res. 2008, 19, 555–578. [CrossRef] pp 70. Gadaleta, D.; Lombardo, A.; Toma, C.; Benfenati, E. A New Semi-Automated Workflow for Chemical Data Retrieval and Quality Checking for Modeling Applications. J. Cheminform. 2018, 10, 60. [CrossRef] g g pp f 71. Berthold, M.R.; Cebron, N.; Dill, F.; Gabriel, T.R.; Kötter, T.; Meinl, T.; Ohl, P.; Thiel, K.; Wiswedel, B. KNIME-the Konstanz Information Miner: Version 2.0 and Beyond. AcM SIGKDD Explor. Newsl. 2009, 11, 26–31. [CrossRef] y p 72. Chen, C.; Liaw, A.; Breiman, L. Using Random Forest to Learn Imbalanced Data. Univ. Calif. Berkel y Liaw, A.; Breiman, L. Using Random Forest to Learn Imbalanced Data. Univ. Calif. Berkeley 2004, 110, 24. 73. Dal Pozzolo, A.; Boracchi, G.; Caelen, O.; Alippi, C.; Bontempi, G. Credit Card Fraud Detection and Concept-Drift Adaptation with Delayed Supervised Information. In Proceedings of the 2015 International Joint Conference on Neural Networks (IJCNN), Killarney, Ireland, 12–16 July 2015; pp. 1–8. y y pp 74. Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, et al. Scikit-Learn: Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. References Handbook of Molecular Descriptors; John Wiley & Sons: Hoboken, NJ, USA, 2008; Volume 11. 84 Blaauboer B J The Integration of Data on Physico Chemical Properties in Vitro Derived Toxicity Data and Physiologically Based y g y y y g ( y ) 83. Todeschini, R.; Consonni, V. Handbook of Molecular Descriptors; John Wiley & Sons: Hoboken, NJ, USA, 2008; Volume 11. 83. Todeschini, R.; Consonni, V. Handbook of Molecular Descriptors; John Wiley & Sons: Hoboken, NJ, USA, 84. Blaauboer, B.J. The Integration of Data on Physico-Chemical Properties, in Vitro-Derived Toxicity Data and Physiologically Based Kinetic and Dynamic as Modelling a Tool in Hazard and Risk Assessment. A Commentary. Toxicol. Lett. 2003, 138, 161–171. [CrossRef] [CrossRef]

https://openalex.org/W4286486466

http://oro.open.ac.uk/86238/1/86238.pdf

English

Orbital and dynamical analysis of the system around HR 8799

Astronomy & astrophysics

cc-by

Policy This document has been downloaded from Open Research Online, The Open University's repository of research publications. This version is being made available in accordance with Open Research Online policies available from Open Research Online (ORO) Policies Open Research Online Citation Zurlo, A.; Goździewski, K.; Lazzoni, C.; Mesa, D.; Nogueira, P.; Desidera, S.; Gratton, R.; Marzari, F.; Langlois, M.; Pinna, E.; Chauvin, G.; Delorme, P.; Girard, J. H.; Hagelberg, J.; Henning, Th.; Janson, M.; Rickman, E.; Kervella, P.; Avenhaus, H.; Bhowmik, T.; Biller, B.; Boccaletti, A.; Bonaglia, M.; Bonavita, M.; Bonnefoy, M.; Cantalloube, F.; Cheetham, A.; Claudi, R.; D’Orazi, V.; Feldt, M.; Galicher, R.; Ghose, E.; Lagrange, A.-M.; le Coroller, H.; Ligi, R.; Kasper, M.; Maire, A.-L.; Medard, F.; Meyer, M.; Peretti, S.; Perrot, C.; Puglisi, A. T.; Rossi, F.; Rothberg, B.; Schmidt, T.; Sissa, E.; Vigan, A. and Wahhaj, Z. (2022). Orbital and dynamical analysis of the system around HR 8799. Astronomy & Astrophysics, 666 A133. ABSTRACT In addition, we observed HR 8799 with the instrument LUCI at the Large Binocular Telescope (LBT). Methods. All the observations were reduced with state-of-the-art algorithms implemented to apply the spectral and angular differential imaging method. We re-reduced the SPHERE data obtained during the commissioning of the instrument and in three open-time programs to have homogeneous astrometry The precise position of the four planets with respect to the host star was calculated by system discovered with the direct imaging technique, it has been observed extensively since 1998. This wide baseline of astrometric measurements, counting over 50 observations in 20 years, permits a detailed orbital and dynamical analysis of the system. Aims. To explore the orbital parameters of the planets, their dynamical history, and the planet-to-disk interaction, we made follow-up observations of the system during the VLT/SPHERE guaranteed time observation program. We obtained 21 observations, most of them in favorable conditions. In addition, we observed HR 8799 with the instrument LUCI at the Large Binocular Telescope (LBT). g y p y y y Aims. To explore the orbital parameters of the planets, their dynamical history, and the planet-to-disk interaction, we made follow-up observations of the system during the VLT/SPHERE guaranteed time observation program. We obtained 21 observations, most of them in favorable conditions. In addition, we observed HR 8799 with the instrument LUCI at the Large Binocular Telescope (LBT). Methods. All the observations were reduced with state-of-the-art algorithms implemented to apply the spectral and angular differential imaging method. We re-reduced the SPHERE data obtained during the commissioning of the instrument and in three open-time programs to have homogeneous astrometry. The precise position of the four planets with respect to the host star was calculated by exploiting the fake negative companions method. We obtained an astrometric precision of the order of 6 mas in the worst case and 1 mas in the best case. To improve the orbital fitting, we also took into account all of the astrometric data available in the literature. From the photometric measurements obtained in different wavelengths, we estimated the masses of the planets following the evolutionary models. Results. We obtained updated parameters for the orbits with the assumption of coplanarity, relatively small eccentricities, and periods very close to the 2:1 resonance. We also refined the dynamical mass of each planet and the parallax of the system (24.49 ± 0.07 mas), which overlap with the recent Gaia eDR3/DR3 estimate. Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This article is published in open access under the Subscribe-to-Open model. Subscribe to A&A to support open access publication. ABSTRACT Hydrodynamical simulations suggest that inward migration of the planets caused by the interaction with the disk might be responsible for the planets being locked in resonance. We also conducted detailed N-body simulations indicating possible positions of a putative fifth planet with a mass below the present detection limits of ≃3 MJup. Key words. planets and satellites: dynamical evolution and stability – planet-disk interactions – stars: individual: HR8799 – instrumentation: adaptive optics – astrometry – techniques: image processing Hinz et al. 2010; Zuckerman et al. 2011; Bell et al. 2015), γ Dor-type variable star (Gray & Kaye 1999) with λ Boo-like abun- dance patterns. The mass of the star is 1.47+0.12 −0.17 M⊙(Sepulveda & Bowler 2022) and its distance is 40.88 ± 0.08 pc from Gaia measurements (Gaia Collaboration 2020). Orbital and dynamical analysis of the system around HR 8799 New astrometric epochs from VLT/SPHERE and LBT/LUCI A Z l 1 2 3 K G ´d i ki4 C L i5 6 D M 6 P N i 1 S D id 6 R G 6 F M i7 (Affiliations can be found after the references) (Affiliations can be found after the references) Received 25 April 2022 / Accepted 15 June 2022 Received 25 April 2022 / Accepted 15 June 2022 A133, page 1 of 25 Open Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This article is published in open access under the Subscribe-to-Open model. Subscribe to A&A to support open access publication. Versions If this document is identified as the Author Accepted Manuscript it is the version after peer review but before type setting, copy editing or publisher branding Astronomy & Astrophysics Astronomy & Astrophysics A&A 666, A133 (2022) https://doi.org/10.1051/0004-6361/202243862 © A. Zurlo et al. 2022 Orbital and dynamical analysis of the system around HR 8799 New astrometric epochs from VLT/SPHERE and LBT/LUCI A. Zurlo1,2,3 , K. Go´zdziewski4, C. Lazzoni5,6, D. Mesa6, P. Nogueira1, S. Desidera6, R. Gratton6, F. Marzari7, M. Langlois8,3, E. Pinna9, G. Chauvin10, P. Delorme10, J. H. Girard11, J. Hagelberg12, Th. Henning13, M. Janson14, E. Rickman15, P. Kervella16, H. Avenhaus13, T. Bhowmik1, B. Biller17, A. Boccaletti16, M. Bonaglia9, M. Bonavita18,6 M. Bonnefoy10, F. Cantalloube3, A. Cheetham12, R. Claudi6, V. D’Orazi6, M. Feldt13, R. Galicher16, E. Ghose9, A.-M. Lagrange10, H. le Coroller3, R. Ligi20, M. Kasper21, A.-L. Maire10, F. Medard10, M. Meyer19, S. Peretti12, C. Perrot16, A. T. Puglisi9, F. Rossi9, B. Rothberg22,23, T. Schmidt16, E. Sissa6, A. Vigan3, and Z. Wahhaj24 ABSTRACT Context. HR 8799 is a young planetary system composed of four planets and a double debris belt. Being the first multi-planetary system discovered with the direct imaging technique, it has been observed extensively since 1998. This wide baseline of astrometric measurements, counting over 50 observations in 20 years, permits a detailed orbital and dynamical analysis of the system. Context. HR 8799 is a young planetary system composed of four planets and a double debris belt. Being the first multi-planetary system discovered with the direct imaging technique, it has been observed extensively since 1998. This wide baseline of astrometric measurements, counting over 50 observations in 20 years, permits a detailed orbital and dynamical analysis of the system. Aims. To explore the orbital parameters of the planets, their dynamical history, and the planet-to-disk interaction, we made follow-up observations of the system during the VLT/SPHERE guaranteed time observation program. We obtained 21 observations, most of them in favorable conditions. In addition, we observed HR 8799 with the instrument LUCI at the Large Binocular Telescope (LBT). Methods. All the observations were reduced with state-of-the-art algorithms implemented to apply the spectral and angular differential imaging method. We re-reduced the SPHERE data obtained during the commissioning of the instrument and in three open-time programs to have homogeneous astrometry. The precise position of the four planets with respect to the host star was calculated by exploiting the fake negative companions method. We obtained an astrometric precision of the order of 6 mas in the worst case and 1 mas in the best case. To improve the orbital fitting, we also took into account all of the astrometric data available in the literature. From the photometric measurements obtained in different wavelengths, we estimated the masses of the planets following the evolutionary models. Context. HR 8799 is a young planetary system composed of four planets and a double debris belt. Being the first multi-planetary system discovered with the direct imaging technique, it has been observed extensively since 1998. This wide baseline of astrometric measurements, counting over 50 observations in 20 years, permits a detailed orbital and dynamical analysis of the system. Aims. To explore the orbital parameters of the planets, their dynamical history, and the planet-to-disk interaction, we made follow-up observations of the system during the VLT/SPHERE guaranteed time observation program. We obtained 21 observations, most of them in favorable conditions. 1. Introduction Among the exoplanets discovered with the high-contrast imag- ing technique (e.g., Chauvin et al. 2004; Lagrange et al. 2010; Rameau et al. 2013; Keppler et al. 2018; Bohn et al. 2021), the system around HR 8799 is undoubtedly one of the most interesting. This is mainly because HR 8799 hosts a greater num- ber of planets detected with high-contrast imaging than any other system, only three systems that host two planets were detected: PDS 70 (Keppler et al. 2018; Haffert et al. 2019), TYC 8998-760-1 (Bohn et al. 2020), and β Pic (Lagrange et al. 2010; Nowak et al. 2020). HR 8799 is a perfect laboratory with which to study dynamical interaction in young planetary sys- tems, with its four planets (HR 8799 bcde; Marois et al. 2008, 2010) and a double debris belt (see, e.g., Su et al. 2009; Hughes et al. 2011; Matthews et al. 2014; Booth et al. 2016; Faramaz et al. 2021). The host star HR 8799 is a young (∼42 Myr, This system is an optimal target for high-contrast imaging observations, as its four planets are easily detectable with state- of-the-art imagers; their contrasts are about 2–8 × 10−6; and the separation of the closest planet HR 8799 e is ∼390 mas, which is further than the inner working angle of most current instru- ments designed for direct imaging. For these reasons, the system has been observed dozens of times starting from 1998, with dif- ferent instruments, wavelengths, and configurations. Thanks to this rich pool of archival data, HR 8799 can be studied in detail from an astrometric point of view, being the only multi-planetary system for which there are tens of astrometric data points on a baseline of more than 20 yr. These measurements began with A&A 666, A133 (2022) Table 1. Summary of the observations of HR 8799 with IRDIS and IFS during SHINE. Table 1. Summary of the observations of HR 8799 with IRDIS and IFS during SHINE. uncertainties of ∼20 mas, but have since reached unprecedented precision of below a milliarcsecond (0.1–0.2 mas) with optical interferometry (GRAVITY Collaboration 2019). Works regarding the analysis of the dynamical interaction of the four planets have been presented; most of these agree on the near coplanarity of the planets, and that they are locked in a 1:2:4:8 resonance, (see, e.g., Fabrycky & Murray-Clay 2010; Esposito et al. 2013; Konopacky et al. 2016; Zurlo et al. 1. Introduction 2016; Wang et al. 2018; GRAVITY Collaboration 2019; Go´zdziewski & Migaszewski 2020). From the age of the system and the luminosity of the planets, evolutionary hot-start models derive masses of around 5–7 MJup (Marois et al. 2010; Currie et al. 2011; Sudol & Haghighipour 2012). These masses are compatible with the dynamical studies because small masses help the orbits to remain stable (e.g., Wang et al. 2018). On the other hand, Brandt et al. (2021) found a dynamical mass for the innermost planet HR 8799 e of 9.6+1.9 −1.8 MJup assuming that planets c, d, and e share the same mass within ∼20%. This result is 2 MJup higher than the prediction from the evolutionary models. UT date Filter ∆FoV (◦) S/NMin S/NMax 2014-07-12 DB_H23 17 9 44 2014-08-13 (a) BB_J 23 0 36 2014-12-04 (b) BB_H 9 14 24 2014-12-05 (b) BB_H 8 11 25 2014-12-06 (b) BB_H 8 15 25 2014-12-08 (a,b) BB_H 8 6 25 2015-07-03 DB_K12 18 12 75 2015-07-29 (a,b) DB_J23 82 4 51 2015-07-30 (b) DB_K12 81 21 123 2015-09-27 DB_K12 24 10 71 2016-11-17 DB_H23 17 15 52 2017-06-14 DB_H23 19 14 67 2017-10-07 (a) BB_H 76 0 0 2017-10-11 BB_H 74 45 90 2017-10-12 BB_H 78 24 100 2018-06-18 DB_H23 34 20 78 2018-08-17 BB_H 73 29 57 2018-08-18 BB_H 73 42 106 2019-10-31 DB_K12 24 22 62 2019-11-01 DB_K12 54 38 114 2021-08-20 DB_H23 20 21 59 Notes. (a)Discarded for bad weather. (b)IRDIS alone. UT date Filter ∆FoV (◦) S/NMin S/NMax In this paper, we present all the astrometric measurements obtained during the whole guaranteed time observation (GTO) of VLT/SPHERE for HR 8799 bcde. Ad hoc observations were designed for the orbital follow-up of the four planets; HR 8799 was observed 21 times in total with SPHERE, and only four observations were rejected based on quality criteria. To comple- ment the SPHERE measurement, we obtained one astrometric epoch with the instrument LUCI, installed at the LBT. This instrument observed HR 8799 as part of the commissioning phase of the new adaptive optics (AO) system. With a total of 18 epochs and the addition of all the astrometric measure- ments available in the literature (69 astrometric epochs in total), we performed the dynamical analysis of the system and the planet-to-disk interaction. Notes. (a)Discarded for bad weather. (b)IRDIS alone. The outline of the paper is as follows: in Sect. 1. Introduction 2, we present SPHERE and LUCI observations; in Sect. 3, we describe the reduction methods applied and the astrometric results that we obtained. In Sect. 4, we present the astrometric fitting for the four planets of HR 8799 and in Sect. 5 a possible interpretation of the history of the system. We also explored the possibility of the presence of a fifth planet, studying the planets–disk interaction (Sect. 6). We provide our conclusions in Sect. 7. the IWA is ∼0.′′12. For a detailed description of the observing sequence, we refer the reader to Zurlo et al. (2014, 2016). In gen- eral, the working sequence includes an image of the off-axis star point spread function (PSF) for the flux calibration, a long coro- nagraphic sequence with the satellite spots (Langlois et al. 2013) mode, a second image of the stellar PSF, and the sky images. The waffle mode was used on purpose to assure maximum astro- metric precision. The long waffle sequence was taken in pupil stabilized mode in order to apply the angular differential imag- ing (ADI; Marois et al. 2006) method. While IRDIS has a field of view of ∼11 × 11′′, IFS is smaller (1.7′′ × 1.7′′), and only the two interior planets are visible. The SHINE observations are sum- marized in Table 1. For almost all the epochs, the observation was with IRDIS and IFS working in parallel. On a few occa- sions, IRDIS was used alone; these observations are marked in the table. We discarded data for time periods where the condi- tions did not permit a clear detection of the planets or there was a very poor signal-to-noise ratio (S/N). In particular, we rejected data from 2014 August 13 ( only IFS data were considered), 2014 December 08, 2015 July 29, and 2017 October 07. All the other observations are taken into account in this analysis. 2.1. VLT/SPHERE HR 8799 was observed several times during the SpHere INfrared survey for Exoplanets (SHINE, papers I, II, III; Desidera et al. 2021; Vigan et al. 2021; Langlois et al. 2021) during the GTO of VLT/SPHERE. The instrument SPHERE (Beuzit et al. 2019) is a planet finder equipped with an extreme AO system (SAXO; Fusco et al. 2006; Petit et al. 2014) to characterize substel- lar companions with high-contrast imaging. The near-infrared (NIR) arm includes the IR dual-band imager and spectrograph (IRDIS; Dohlen et al. 2008) and an integral field spectrograph (IFS; Claudi et al. 2008). During the observations, these two sub- systems observed the target in parallel. HR 8799 was periodically observed for astrometric monitoring, the setup of the observa- tions included three different filter pairs: IRDIS in H2H3 bands (λH2 = 1.593 µm, λH3 = 1.667 µm), in BB_H (λH = 1.625 µm), and K1K2 bands (λK1 = 2.102 µm, λK2 = 2.255 µm). The coro- nagraph used for the shortest wavelengths was an apodized Lyot with a mask diameter of 185 mas and an inner working angle (IWA) of 0.′′09 (see Boccaletti et al. 2008), while for K1K2 band 2.2. Large Binocular Telescope/SOUL-LUCI HR 8799 was observed with LBT/SOUL-LUCI1 during its com- missioning on night 2020 September 29. LUCI1 (Seifert et al. 2010) is a NIR spectro-imager that can work in a diffraction- limited regime with a sampling of 15 mas pix−1. The AO A133, page 2 of 25 A133, page 2 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Table 2. List of astrometric points from IRDIS observations. 2.2. Large Binocular Telescope/SOUL-LUCI No coronagraphic masks are available on LUCI, and therefore to avoid any saturation, HR 8799 was observed adopt- ing a sub-windowing of 256 × 256 pixels (corresponding to a field of view (FoV) of 3.84 × 3.84′′). In this way, we were able to set a minimum exposure time of 0.34 s. We observed this target in pupil stabilization mode with two narrow-band filters: FeII (λFeII = 1.646 µm; ∆λ = 0.018 µm) and H2 (λH2 = 2.124 µm; ∆λ = 0.023 µm). The observations with the two filters were alternated during the whole sequence. correction is provided by SOUL (Pinna et al. 2019), the upgrade of the FLAO system (Esposito et al. 2010). During this obser- vation, the SOUL system was correcting 500 modes with a rate of 1.7 kHz. No coronagraphic masks are available on LUCI, and therefore to avoid any saturation, HR 8799 was observed adopt- ing a sub-windowing of 256 × 256 pixels (corresponding to a field of view (FoV) of 3.84 × 3.84′′). In this way, we were able to set a minimum exposure time of 0.34 s. We observed this target in pupil stabilization mode with two narrow-band filters: FeII (λFeII = 1.646 µm; ∆λ = 0.018 µm) and H2 (λH2 = 2.124 µm; ∆λ = 0.023 µm). The observations with the two filters were alternated during the whole sequence. 2.2. Large Binocular Telescope/SOUL-LUCI Date Planet b Planet c Planet d Planet e ∆RA ; ∆Dec (mas) ∆RA ; ∆Dec (mas) ∆RA ; ∆Dec (mas) ∆RA ; ∆Dec (mas) 2014-07-12 1570 ± 3; 704 ± 3 −521 ± 3; 790 ± 9 −391 ± 2; −530 ± 2 −387 ± 2; −10 ± 3 2014-12-04 1574 ± 3; 701 ± 2 −514 ± 3; 798 ± 4 −399 ± 4; −525 ± 4 −389 ± 8; 11 ± 4 2014-12-05 1574 ± 4; 701 ± 3 −512 ± 3; 798 ± 4 −400 ± 4; −523 ± 4 −390 ± 7; 12 ± 4 2014-12-06 1573 ± 3; 701 ± 3 −512 ± 3; 797 ± 4 −403 ± 4; −524 ± 4 −383 ± 8; 11 ± 4 2015-07-03 1579 ± 1; 694 ± 1 −498 ± 1; 806 ± 1 −417 ± 1; −517 ± 1 −383 ± 9; 33 ± 5 2015-07-30 1580 ± 5; 689 ± 3 −495 ± 2; 806 ± 2 −419 ± 2; −516 ± 1 −386 ± 1; 36 ± 1 2015-09-27 1580 ± 1; 688 ± 1 −494 ± 1; 811 ± 1 –426 ± 1; −512 ± 1 −382 ± 9; 50 ± 5 2016-11-17 1589 ± 2; 666 ± 1 −464 ± 2; 824 ± 2 –454 ± 2; −490 ± 2 −378 ± 4; 90 ± 2 2017-06-14 1591 ± 1; 653 ± 1 −449 ± 1; 835 ± 1 −472 ± 2; −482 ± 2 −373 ± 3; 118 ± 2 2017-10-11 1595 ± 1; 647 ± 1 −441 ± 1; 839 ± 1 −480 ± 1; −477 ± 1 −369 ± 1; 128 ± 1 2017-10-12 1595 ± 1; 647 ± 1 −441 ± 1; 839 ± 1 −480 ± 1; −477 ± 1 −371 ± 2; 128 ± 2 2018-06-18 1601 ± 1; 635 ± 1 −424 ± 1; 848 ± 1 −497 ± 2; −463 ± 2 −358 ± 2; 156 ± 2 2018-08-17 1601 ± 2; 632 ± 3 −421 ± 1; 850 ± 1 −502 ± 1; −461 ± 1 −357 ± 1; 162 ± 2 2018-08-18 1600 ± 1; 632 ± 1 −421 ± 1; 851 ± 1 −502 ± 2; −458 ± 1 −358 ± 2; 163 ± 1 2019-10-31 1606 ± 2; 615 ± 2 −392 ± 2; 875 ± 2 −532 ± 3; −425 ± 2 −338 ± 2; 215 ± 2 2019-11-01 1611 ± 1; 611 ± 1 −388 ± 2; 870 ± 2 −530 ± 2; −430 ± 1 −335 ± 1; 210 ± 1 2021-08-20 1626 ± 1; 578 ± 2 −339 ± 2; 890 ± 2 −563 ± 2; −391 ± 2 −287 ± 4; 272 ± 2 Table 2. 3. Data reduction 3.1. SPHERE p From the IRDIS photometry, we also estimated the mass of each planet using evolutionary models. In particular, we applied evolutionary models from Baraffe et al. (2003, 2015) to the IRDIS filters photometry. The age of the system used in this estimation is 42 Myr and the parallax is 24.46 ± 0.05 mas, as in Gaia EDR3. The error reported is the standard deviation between different epochs taken with the same filter. Results are listed in Table 4. The reduction of the IRDIS data was carried out entirely with the Data Center (DC; Delorme et al. 2017) which uses the stan- dard SPHERE pipeline SpeCal (Galicher et al. 2018). For the astrometric calibration, we used a true north orientation of – 1.75 deg and a pixel scale value of 12.25 mas as reported in Maire et al. (2016). The reduction algorithm used by the DC for HR 8799 is the T-LOCI (Marois et al. 2014). To provide a homogeneous reduction of all the IRDIS data, we processed the observations presented in Zurlo et al. (2016) again, which were previously reduced with custom routines, which included seven epochs in 2014 (we excluded the sequence of 2014 August 13 for poor quality). The four epochs taken for variability monitor- ing – once per night (2014 December 4, 2014 December 5, 2014 December 6, 2014 December 8) – and presented in Apai et al. (2016) were reprocessed with the DC. As in Apai et al. (2016), we excluded the sequence of December 8 for poor conditions. In the same way, we used the DC to reduce the data from the vari- ability monitoring of Biller et al. (2021), with observation dates: 2.2. Large Binocular Telescope/SOUL-LUCI List of astrometric points from IRDIS observations. 2015 July 29, 2015 July 30, 2017 October 07, 2017 October 11, and 2017 October 12 (two epochs were excluded for poor qual- ity), and 2018 August, 17–18. Finally, we re-processed the IRDIS data presented by Wahhaj et al. (2021) and taken on October 31, 2019, and November 1, 2019. In the DC, a “fake negative planets” (e.g., Zurlo et al. 2014, and references therein) algo- rithm is implemented to calculate the precise position of each planet with respect to the central star. A summary of all the SPHERE astrometric points, both from the SHINE survey and published observations presented in Zurlo et al. (2016), Apai et al. (2016), Biller et al. (2021), and Wahhaj et al. (2021) is listed in Tables 2 (IRDIS) and 3 (IFS). We refer the reader to these latter publications for further information about the observ- ing conditions. The results of the updated astrometry are also listed in Table A.1, together with all the other data from different instruments presented in the literature. 2015 July 29, 2015 July 30, 2017 October 07, 2017 October 11, and 2017 October 12 (two epochs were excluded for poor qual- ity), and 2018 August, 17–18. Finally, we re-processed the IRDIS data presented by Wahhaj et al. (2021) and taken on October 31, 2019, and November 1, 2019. In the DC, a “fake negative planets” (e.g., Zurlo et al. 2014, and references therein) algo- rithm is implemented to calculate the precise position of each planet with respect to the central star. A summary of all the SPHERE astrometric points, both from the SHINE survey and published observations presented in Zurlo et al. (2016), Apai et al. (2016), Biller et al. (2021), and Wahhaj et al. (2021) is listed in Tables 2 (IRDIS) and 3 (IFS). We refer the reader to these latter publications for further information about the observ- ing conditions. The results of the updated astrometry are also listed in Table A.1, together with all the other data from different instruments presented in the literature. correction is provided by SOUL (Pinna et al. 2019), the upgrade of the FLAO system (Esposito et al. 2010). During this obser- vation, the SOUL system was correcting 500 modes with a rate of 1.7 kHz. Table 3. List of astrometric points from IFS observations. Table 3. List of astrometric points from IFS observations. Table 3. List of astrometric points from IFS observations. Table 3. List of astrometric points from IFS observations. Date Planet d Planet e ∆RA ; ∆Dec (mas) ∆RA ; ∆Dec (mas) 2014-07-12 −400 ± 4; −512 ± 4 −389 ± 1; −22 ± 2 2014-08-13 −396 ± 1; −524 ± 1 −389 ± 1; −17 ± 2 2015-07-03 −424 ± 4; −509 ± 3 −391 ± 1; 33 ± 2 2015-09-27 −420 ± 4; −513 ± 4 −392 ± 1; 39 ± 3 2016-11-17 −464 ± 1; −486 ± 2 −382 ± 2; 94 ± 6 2017-06-14 −473 ± 2; −476 ± 2 −377 ± 1; 115 ± 3 2017-10-11 −480 ± 2; −478 ± 2 −371 ± 1; 129 ± 2 2017-10-12 −492 ± 5; −463 ± 6 −370 ± 2; 135 ± 3 2018-06-18 −495 ± 1; −460 ± 2 −360 ± 2; 162 ± 2 2018-08-17 −509 ± 2; −452 ± 3 −361 ± 2; 166 ± 1 2018-08-18 −503 ± 1; −456 ± 2 −359 ± 1; 167 ± 2 2019-10-31 −527 ± 3; −432 ± 3 −337 ± 2; 210 ± 3 2019-11-01 −528 ± 1; −435 ± 2 −337 ± 2; 208 ± 1 2021-08-20 −569 ± 1; −390 ± 2 −292 ± 2; 276 ± 2 in the literature. Recently, Go´zdziewski & Migaszewski (2020) (GM2020 hereafter) reported an exact-resonance configuration (or “periodic orbits model” (PO) hereafter). This model is designed to explain the astrometric measurements through con- straints on geometry, planetary masses, and astrometric parallax of the system. Direct determination of planetary masses is crucial for calibrating astrophysical cooling models and for deter- mining the origin and long-term orbital evolution of the entire system, including its debris disk components. Here, we consider a less stringent condition on the stability of the system, assuming that it may be close to exact resonance but not necessarily fully periodic. This assumption may be nat- ural in the sense that migration may result in a system that is not exactly in a configuration of PO (e.g. Ramos et al. 2017). We want to determine whether such near-resonance models can yield statistically different or perhaps even better best-fitting solutions than those that appear in the PO scenario. 4.2. Mass and parallax priors We conducted MCMC sampling from the posterior defined through the general merit function in Eq. (1) and astrophysical and geometrical priors. The most important astrophysical priors are the masses of the star and the planets, and the geometrical prior is the parallax. Table 3. List of astrometric points from IFS observations. We define the fol- lowing merit functions as in Go´zdziewski & Migaszewski (2018, 2020): ln L(x) = −1 2χ2(x) −1 2 Nobs X i=1 h ln θ2 i,α + ln θ2 i,δ i −Nobs ln(2π), χ2(x) = Nobs X i=1  [αi −α(ti, x)]2 θ2 i,α + [δi −δ(ti, x)]2 θ2 i,δ , (1) (1) was switched off to allow the rotation of the FOV and be able to use the ADI method. Before performing the high-contrast imaging method, we obtained the position of the stellar PSF for each frame using the FIND procedure as implemented for IDL. Exploiting these positions, we then precisely registered the whole dataset, positioning the stellar PSFs at the center of each frame. The ADI was then implemented by exploiting the princi- pal component analysis (PCA; Soummer et al. 2012) technique. For the reduction, we tested a different number of principal com- ponents, but found that the best solution was to use ten of them. We were able to recover all the known companions with S/N ranging between 7 and 10. where (αi, δi) are the right ascension (RA) and declination (Dec) measurements at time ti; α(ti, x), δ(ti, x) are for their ephemeris (model) values at the same moments as implied by the adopted x parameters; and θ2 i,α and θ2 i,δ are the nominal measurement uncer- tainties in RA and Dec scaled in quadrature with the so-called error floor, θ2 i,α = (σ2 i,α + σ2 α,δ) and θ2 i,δ = (σ2 i,δ + σ2 α,δ), for each datum, respectively. Also, if Nobs is the number of observations then N = 2Nobs, because RA and Dec are measured in a single detection. The error floor can be introduced for unmodeled mea- surement uncertainties, such that the resulting best-fitting model should yield the reduced χ2 ν ≃1. However, in this work, as we rely on MCMC sampling and the error floor introduces little qualitative variability into the solutions, we omit this parameter, thereby also simplifying the astrometric model. We obtained the precise position of each planet, introducing fake negative companions and changing its position to minimize the standard deviation in a small region around the planet itself. The results of this procedure are listed in Table 5. 3.2. SOUL+LUCI The data were taken with a randomized offset of the position of the PSF on the detector. This was used to create a back- ground obtained by calculating the median of all the frames. This background was then subtracted from each science frame. The final dataset was composed of 60 and 61 files for the FeII and H2 observations, respectively. In both cases, each file was com- posed of 112 frames for a total exposure time of 2284.80 s and 2322.88 s, respectively. During the observations, the derotator A133, page 3 of 25 A133, page 3 of 25 A&A 666, A133 (2022) 4.1. Dynamical constraints on the astrometry Filter Planet e (MJup) Planet d (MJup) Planet c (MJup) Planet b (MJup) BB_H 8.9 ± 1.1 8.4 ± 0.9 8.5 ± 0.9 6.4 ± 0.6 D_H2 7.0 ± 0.3 6.6 ± 0.3 6.7 ± 0.3 4.6 ± 0.3 D_H3 8.5 ± 0.3 8.1 ± 0.2 8.3 ± 0.3 6.8 ± 0.3 D_K1 9.6 ± 0.3 9.3 ± 0.3 9.3 ± 0.2 6.7 ± 0.3 D_K2 10.5 ± 0.1 10.5 ± 0.1 10.4 ± 0.1 9.0 ± 0.3 Notes. The age of the system is assumed to be 42 Myr. The error is the standard deviation between different observations with the same filter. Table 5. List of astrometric points from the LBT/LUCI observation. Date Planet b Planet c Planet d Planet e ∆RA; ∆Dec (mas) ∆RA; ∆Dec (mas) ∆RA; ∆Dec (mas) ∆RA; ∆Dec (mas) 2020-09-29 1620 ± 1; 591 ± 3 −364 ± 2; 883 ± 1 −551 ± 1; −415 ± 4 −315 ± 3 ; 242 ± 5 Table 4. Values for the mass of each planet calculated using the evolutionary models in the different IRDIS filters. Filter Planet e (MJup) Planet d (MJup) Planet c (MJup) Planet b (MJup) m is assumed to be 42 Myr. The error is the standard deviation between different observations with the same filter. Table 5. List of astrometric points from the LBT/LUCI observation. Date Planet b Planet c Planet d Planet e ∆RA; ∆Dec (mas) ∆RA; ∆Dec (mas) ∆RA; ∆Dec (mas) ∆RA; ∆Dec (mas) 2020-09-29 1620 ± 1; 591 ± 3 −364 ± 2; 883 ± 1 −551 ± 1; −415 ± 4 −315 ± 3 ; 242 ± 5 significantly shifted from Π ≃24.22±0.08 mas in the Gaia DR2 and 24.8 ± 0.7 mas in Gaia DR1 catalog, respectively. The par- allax determinations in the Gaia catalogs appear subtly biased, depending on the luminosity and spectral type (Lindegren et al. 2021). However, compared to the uncertainties of the dynam- ical estimates here, the predicted parallax correction for Gaia eDR3/DR3 of <0.1 mas (a few tens of µas) would be insignifi- cant. Indeed, the parallax correction predicted by Lindegren et al. (2021) yields 24.50±0.05 mas (Kervella et al. 2022). This appar- ently very small shift of the order of 1σ uncertainty may still be meaningful when compared to the dynamical estimates, as is found below. 4.3. The exact Laplace resonance revisited It is common in the literature to assume smaller planet masses, that is, of ≃7 MJup, given the results of dynamical N-body simulations. In particular, Wang et al. (2018) report difficulty in finding dynamically stable solutions for planet masses larger than ≃7 MJup. Simultaneously, Go´zdziewski & Migaszewski (2018, 2020) found rigorously stable systems locked in the Laplace resonance for higher mass ranges as well that is, of, ≃8–9 MJup. Given systematic observationally constrained shifts to higher masses, as predicted by the exact Laplace MMR model, we decided to apply the mass priors in the ≃9 MJup range. Brandt et al. (2021) estimated the mass of HR 8799e as 9.6+1.9 −1.8 MJup based on the secular variation of the proper motion of the parent star in Gaia and HIPPARCOS catalogs. Moreover, in a very recent work, Kervella et al. (2022) similarly determined the dynamical mass of 12 ± 3.5 MJup for HR 8799e. This latter is less precise than in the prior work of Brandt et al. (2021), because Kervella et al. (2022) did not account for the actual orbital geometry. These high mass ranges for HR 8799e are consistent with the larger mass priors. Therefore, in some experiments we also used the mass priors Nm with the innermost mass changed to the value computed in Brandt et al. (2021).i In the first step, we verified whether or not the resonant model in GM2020 based on measurements in Konopacky et al. (2016) fits the recent measurements listed in Tables 2, 3, and 5. We found that this model visually and significantly deviates from the updated data set, especially for planet HR 8799b. Therefore, we refined the PO model using the same parametrization as for the merit function; see Eq. (1) in GM2020. In the formulation presented by these latter authors, the primary parameters of the astrometric model consist of all planet masses mi, i = 1, . . . , 4, the osculating period ratio κ for the two innermost planets – which selects the particular four-body Laplace resonance chain– , as well as the reference epoch τ, three Euler angles (I, ωrot, Ω) rotating the coplanar resonant configuration to the sky (observer) plane, and the system parallax Π. This means 11 free parameters. We note that all remaining orbital parameters are constrained through the N-body dynamics confined to the manifold of strictly periodic solutions. 4.1. Dynamical constraints on the astrometry We also implied Gaussian priors Nm = [8.7 ± 1.7, 8.7 ± 1.0, 8.7 ± 1.0, 6 ± 0.7] MJup for the masses of the innermost to the outermost planet, respectively, based on the hot-start cooling models by Baraffe et al. (2003), following discussion in Wang et al. (2018) and confirmed by our estimates collected in Table 4. The mass ranges for BB_H are consistent with the priors in Wang et al. (2018), and D_H3, D_K1 overlap with the determinations from the earlier astrometric model in GM2020. As also demonstrated by GM2020 and confirmed below, there are many discrepancies regarding the mass hierarchy; according to the resonant model, HR 8799d is the most massive one, and the masses of HR 8799e and HR 8799c appear strongly anti-correlated. Masses for D_K2 seem to be too far into the high-end range regarding the dynamical stability, especially that of planet HR 8799b. 4.1. Dynamical constraints on the astrometry Given the literature regarding astrometric fits of the HR 8799 system (e.g., Wang et al. 2018; Go´zdziewski & Migaszewski 2020, and references therein), it is now widely recognized that the present observation time-window does not make it possible to determine long-term stable orbital solutions of HR HR 8799 without invoking particular dynamical constraints. Such con- straints may arise from the coplanarity of the planets’ orbits and their relatively small eccentricities, which implies a ratio of orbital periods close to 2:1 for successive pairs of planets, that is, 2:1 mean-motion resonances (MMRs). This assumption can be further supported by the likely origin of the system from planetary migration (e.g., Wang et al. 2018); see also Sect. 5 in this paper. Also, the recent analysis of high-contrast images and resulting orbital solutions in Wahhaj et al. (2021) indicate that planets HR 8799e and HR 8799c are most consistent with co-planar and resonant orbits. The dynamical multiple MMR scenario therefore seems to be well justified and documented Regarding the stellar mass, we considered two fixed values of m⋆= 1.52 M⊙determined by Baines et al. (2012) as m⋆= 1.516+0.038 −0.024 M⊙for the age of ≃33 Myr, and m⋆= 1.47+0.12 −0.17 M⊙, following the most recent dynamical estimate by Sepulveda & Bowler (2022). The same value was used by Wang et al. (2018) in their earlier work. Kervella et al. (2022) determined m⋆= 1.50 ± 0.08 M⊙, which overlaps with the two estimates. We note here that the stellar mass is fixed in our astrometric N-body model because it also constrains the parallax Π of the whole system. These two parameters are strongly correlated through the near-III Keplerian law. Moreover, the parallax tends to be systematically tightly bounded in subsequent Gaia catalogs. The most recent Gaia eDR3 estimate of Π = 24.460 ± 0.045 mas is accurate to 0.1%, and it is a meaningful prior for the MCMC sampling. We note that the parallax did not change in the final Gaia DR3 catalog; now it is listed as Π = 24.462 ± 0.046 mas. A133, page 4 of 25 A133, page 4 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Table 4. Values for the mass of each planet calculated using the evolutionary models in the different IRDIS filters. 4.3. The exact Laplace resonance revisited We performed the MCMC sampling from the posterior defined with the merit function in Eq. (1). This merit function is based on a numerical procedure for computing periodic solu- tions in a co-planar N-planet system, as described in GM2020 and implemented by Cezary Migaszewski (priv. comm.). To con- duct the MCMC sampling, we used the affine sampler in emcee As the parallax prior, we defined the most recent Gaia eDR3/DR3 estimate of Π = 24.46 ± 0.05 mas. This value is A133, page 5 of 25 A133, page 5 of 25 A&A 666, A133 (2022) Table 6. Osculating, heliocentric elements of the best-fitting solutions at the epoch of 1998.830. 4.3. The exact Laplace resonance revisited Model IVPO: m⋆= 1.52 M⊙, Π = 24.2585686 mas (24.26 ± 0.05) mas, χ2 ν = 4.32, ν = 441, dim p = 11 m (MJup) a (au) e I (deg) Ω(deg) ϖ (deg) M (deg) HR 8799e 8.1 ± 0.7 16.28 ± 0.03 0.147 ± 0.001 110.8 ± 0.5 336.8 ± 0.5 8.24465258 16.2817072 0.147508399 110.765660 336.793683 HR 8799d 9.5 ± 0.1 26.78 ± 0.08 0.115 ± 0.001 29.4 ± 0.9 59.8 ± 0.8 9.48438158 26.7888442 0.01154150 29.4676779 59.6448288 HR 8799c 7.7 ± 0.5 41.40 ± 0.10 0.055 ± 0.002 26.9 ± 0.2 62.2 ± 0.4 92.1 ± 0.5 145.6 ± 0.8 7.64047381 41.3945527 0.05485344 26.8923597 62.2416412 92.1126122 145.7203489 HR 8799b 6.0 ± 0.4 71.95 ± 0.13 0.017 ± 0.001 44.4 ± 3.0 309.2 ± 2.6 6.01476193 71.9477733 0.017597398 44.6384383 309.291519 Model IVPO: m⋆= 1.47 M⊙, Π = 24.5256617 mas (24.53 ± 0.04) mas, χ2 ν = 4.32, ν = 441, dim p = 11 m (MJup) a (au) e I (deg) Ω(deg) ϖ (deg) M (deg) HR 8799e 7.6 ± 0.9 16.10 ± 0.03 0.148 ± 0.001 110.8 ± 0.5 336.8 ± 0.3 7.4060918 16.1038912 0.147675953 110.816707 336.828050 HR 8799d 9.2 ± 0.1 26.46 ± 0.09 0.115 ± 0.002 29.1 ± 1.0 60.0 ± 1.0 9.18999484 26.45172667 0.114600334 29.0594738 60.1752106 HR 8799c 7.7 ± 0.7 40.97 ± 0.10 0.054 ± 0.002 26.9 ± 0.2 62.2 ± 0.4 92.3 ± 0.6 145.5 ± 0.7 7.79611845 40.9861801 0.05380902 26.8734271 62.1852189 92.3879462 145.5347082 HR 8799b 5.8 ± 0.4 71.14 ± 0.15 0.017 ± 0.002 43.50 ± 3.0 310.35 ± 3.0 5.80235941 71.1427851 0.01667073 42.9704380 310.895065 Model IV∆n: m⋆= 1.47 M⊙, Π = 24.5256617 mas (24.49 ± 0.07) mas, χ2 ν = 3.9, ν = 429, dim p = 23 m (MJup) a (au) e I (deg) Ω(deg) ϖ (deg) M (deg) HR 8799e 7.5 ± 0.6 16.00 ± 0.10 0.150 ± 0.004 108.8 ± 1.3 336.00 ± 1.5 7.40609179 16.1038912 0.147675954 110.816707 336.828050 HR 8799d 9.3 ± 0.5 26.55 ± 0.12 0.114 ± 0.004 28.66 ± 1.2 60.0 ± 1.0 9.18999484 26.45172667 0.11460033 29.05947383 60.1752106 HR 8799c 7.8 ± 0.6 41.00 ± 0.15 0.050 ± 0.003 26.5 ± 0.5 63.1 ± 0.5 91.8 ± 1.0 145.0 ± 1.0 7.79611845 40.9861801 0.05380902 26.8734271 62.1852189 92.3879462 145.5347082 HR 8799b 5.8 ± 0.8 71.3 ± 0.2 0.017 ± 0.002 42.3 ± 1.6 310.8 ± 2.0 5.80235941 71.1427851 0.016670734 42.9704380 310.895065 es. 4.3. The exact Laplace resonance revisited The stellar mass is m⋆= 1.52 M⊙or m⋆= 1.47 M⊙. Uncertainties for parameters in models IVPO and IV∆n are determined as the 16th h percentiles of the samples. Parameter values with 8–9 significant digits are provided in order to reproduce the results of particular integrati ameter ν = Nobs −dim p is for the degrees of freedom, where Nobs denotes the number of (α, δ) measurements, and dim p is the number of ameters. Notes. The stellar mass is m⋆= 1.52 M⊙or m⋆= 1.47 M⊙. Uncertainties for parameters in models IVPO and IV∆n are determined as the 16th and 86th percentiles of the samples. Parameter values with 8–9 significant digits are provided in order to reproduce the results of particular integrations. Parameter ν = Nobs −dim p is for the degrees of freedom, where Nobs denotes the number of (α, δ) measurements, and dim p is the number of free parameters. package (Foreman-Mackey et al. 2013). We initiated 1536 walk- ers in a small hyper-ball in the parameter space around the initial condition in Table 6 determined with a search with the Powell non-gradient minimization algorithm. Since the auto- correlation time appears relatively short, of ∼100 iterations only, we ended up with 1000 iterations that make it possible to derive the posterior distribution, given the large number of walkers. appears strongly correlated with geometric parameters of the system, i.e., the Euler angles and the parallax. These parameters show mutually weaker but still significant correlations.i i Parameters of the best-fitting solutions with their formal uncertainties and particular solutions from the posterior that may be useful for future numerical studies are collected in Table 6, referred to as Model IVPO for the two stellar masses of m⋆= 1.52 M⊙and m⋆= 1.47 M⊙, respectively. The derived posterior is illustrated in 2-dim projections and 1-dim histograms of the MCMC samples for the primary param- eters (Fig. 1). This experiment reveals significant correlations between different parameters already reported in GM2020 that still cannot be reduced with the new data. Besides a strong me−mc anti-correlation, there is also me−κ correlation and md −κ anti-correlation. The remaining two masses are free from sig- nificant mutual correlations, however, the mass of HR 8799b For a stellar mass of m⋆= 1.47M⊙, the HR 8799d mass seems to be constrained to ≃9.2 MJup within a relatively very small uncertainty of ≃0.1 MJup. 4.3. The exact Laplace resonance revisited This mass estimate is slightly larger than that found by GM2020. The mass of the outer- most planet HR 8799b may be determined as ≃5.8 MJup which is similar to ≃5.5 MJup found by these latter authors, yet with also smaller uncertainty of ≃0.3 MJup. Moreover, for the larger stellar A133, page 6 of 25 A133, page 6 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. me = 7.64+0.89 0.91 9.00 9.15 9.30 9.45 md[MJ] md = 9.18+0.08 0.08 6.4 7.2 8.0 8.8 mc[MJ] mc = 7.63+0.64 0.63 5.2 5.6 6.0 6.4 6.8 mb[MJ] mb = 5.84+0.30 0.30 1.968 1.976 1.984 1.992 = 1.98+0.01 0.00 24.45 24.50 24.55 24.60 24.65 [mas] = 24.53+0.03 0.03 26.50 26.75 27.00 27.25 27.50 Inc[deg] Inc = 26.90+0.16 0.17 6 7 8 9 10 me[MJ] 61.6 62.0 62.4 62.8 [deg] 9.00 9.15 9.30 9.45 md[MJ] 6.4 7.2 8.0 8.8 mc[MJ] 5.2 5.6 6.0 6.4 6.8 mb[MJ] 1.968 1.976 1.984 1.992 24.45 24.50 24.55 24.60 24.65 [mas] 26.50 26.75 27.00 27.25 27.50 Inc[deg] 61.6 62.0 62.4 62.8 [deg] = 62.25+0.30 0.32 Fig. 1. MCMC posterior for the best-fitting, strictly resonant model derived for m⋆= 1.47 MSun. Parameters included in the diagram are for th planet masses and orbital period ratio κ = Pd/Pe = κ, and parallax Π, inclination I, and the nodal angle Ω(the orbital scale ρ, PO epoch shift t and the common rotation angle ωrot in the orbital planet are skipped). The crossed lines mark the best-fitting PO solution in terms of the smalle χ2 that was used as a starting point to initiate the emcee walkers; see Tables 6 and 7 (Model 1) for parameters of this solution. Uncertainties fo the parameters are determined as the 16th and 86th percentile of the samples around the median values. A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. mc[MJ] mb[MJ] mc[MJ] Fig. 1. MCMC posterior for the best-fitting, strictly resonant model derived for m⋆= 1.47 MSun. Parameters included in the diagram are for the planet masses and orbital period ratio κ = Pd/Pe = κ, and parallax Π, inclination I, and the nodal angle Ω(the orbital scale ρ, PO epoch shift t0 and the common rotation angle ωrot in the orbital planet are skipped). 4.3. The exact Laplace resonance revisited The crossed lines mark the best-fitting PO solution in terms of the smallest χ2 that was used as a starting point to initiate the emcee walkers; see Tables 6 and 7 (Model 1) for parameters of this solution. Uncertainties for the parameters are determined as the 16th and 86th percentile of the samples around the median values. Fig. 1. MCMC posterior for the best-fitting, strictly resonant model derived for m⋆= 1.47 MSun. Parameters included in the diagram are for the planet masses and orbital period ratio κ = Pd/Pe = κ, and parallax Π, inclination I, and the nodal angle Ω(the orbital scale ρ, PO epoch shift t0 and the common rotation angle ωrot in the orbital planet are skipped). The crossed lines mark the best-fitting PO solution in terms of the smallest χ2 that was used as a starting point to initiate the emcee walkers; see Tables 6 and 7 (Model 1) for parameters of this solution. Uncertainties for the parameters are determined as the 16th and 86th percentile of the samples around the median values. mass of 1.52 M⊙, we derived slightly larger masses of HR 8799d ≃9.6 ± 0.1 MJup and HR 8799b ≃6.0 ± 0.4 MJup. This relation confirms the predicted strong stellar mass–parallax correlation and indirectly favors m⋆= 1.47 M⊙. The dynamical parallax determinations appear to be very accurate and mean- ingful, overlapping closely with the independent geometric Gaia measurements. The astrometric model for the updated data window yields Π ≃24.26 ± 0.05 mas, which is consistent with the earlier esti- mate by GM2020. The parallax estimate of Π ≃24.53±0.04 mas derived for 1.47 M⊙overlaps to 1σ with Π ≃24.460 ± 0.045 mas in Gaia eDR3/DR3. Moreover, as noted above, when applying the parallax correction of the Lindegren et al. (2021) eDR3 catalog, Kervella et al. (2022) obtained Π = 24.50 ± 0.05 mas which is even closer to the value derived from the PO model. Regarding the HR 8799e–HR 8799d mass anti-correlation, we conducted additional MCMC sampling for the primary parameters in the PO model for fake data series. We prepared two or three synthetic observations per year, extending the real observations window by ≃20 yr around the best-fitting periodic A133, page 7 of 25 A133, page 7 of 25 A&A 666, A133 (2022) Fig. 2. 4.4. Near 8:4:2:1 MMR model (3) To characterize the system near the 2:1 MMR chain, we rely on the notion of the proper mean motions as the fundamental fre- quencies fi in the framework of conservative N-body dynamics. We resolve these frequencies with the refined Fourier frequency analysis (e.g., Laskar 1993; Nesvorny & Ferraz-Mello 1997, Numerical Analysis of Fundamental Frequencies, NAFF). To perform the NAFF, we consider the time series if ∆f(k) accounts only for the proper mean motions according to the d’Alembert rule. The resonant configuration may therefore be called the generalized Laplace resonance (Papaloizou 2015). This type of MMR, which possibly drives the orbital evolu- tion of HR 8799, is characterized with the proper mean motions fulfilling the following relation (e.g., GM2020): Sλ,i = { ai(t j) exp iλi(t j) }, SM,i = { ai(t j) exp iMi(t j) }, ∆f(k) = n1 −2n2 + n3 −2n4. (4) (4) where j = 0, 1, . . ., t j = j∆t, and ai(t), λi(t) and Mi are the canonical osculating semi-major axis, the mean longitude, and the mean anomaly for planet i, i = 1, 2, 3, 4, respectively, and ∆t is the sampling interval (i is the imaginary unit). These canon- ical elements inferred in the Jacobi or Poincaré reference frame make it possible to account for mutual interactions between the planets to the first order in the masses. Similarly to the classic Laplace resonance in the Galilean moons system, we consider the generalized MMR as a chain of two- body MMRs, Similarly to the classic Laplace resonance in the Galilean moons system, we consider the generalized MMR as a chain of two- body MMRs, 2ni+1 −ni + gj ≃2ni+1 −ni ≃0, i = 1, 2, 3, j = i, i + 1, for subsequent pairs of planets. Here, gj are for the proper fre- quencies (mode) associated with the pericenter rotation. In the exact resonance, which we consider as the PO in the reference frame rotating with a selected planet (Hadjidemetriou 1977), the apsidal lines of all planets rotate synchronously, with the same angular velocity relative to the inertial frame. The condition in Eq. (4) for the exact resonance (periodic motion in the rotating frame) may be expressed through the proper mean motions of the planets (e.g., Delisle 2017): for subsequent pairs of planets. Here, gj are for the proper fre- quencies (mode) associated with the pericenter rotation. 4.3. The exact Laplace resonance revisited Temporal evolution of the osculating orbital elements in the astrocentric (blue curves) and Jacobian (yellow curves) frames, respectively, for the same N-body initial condition implying a stable, near 8:4:2:1 MMR system. We note that the elements overlap for the first planet according to the construction of the Jacobian reference frame. Fig. 2. Temporal evolution of the osculating orbital elements in the astrocentric (blue curves) and Jacobian (yellow curves) frames, respectively, for the same N-body initial condition implying a stable, near 8:4:2:1 MMR system. We note that the elements overlap for the first planet according to the construction of the Jacobian reference frame. zeroth-order mean-motion resonance implies one of the possible linear combinations of the fundamental frequencies (the proper mean-motions), model with deviations and uncertainties of ∼1 mas. It appears that all correlations except those between the masses and κ would be almost eliminated. A probable cause of that effect may be the almost perfectly aligned gravitational tugs of these planets in the present particular time window of the observa- tions. If the PO model is correct, then likely only highly precise data similar to the most accurate GRAVITY measurement in GRAVITY Collaboration (2019) could break this degeneracy. This was indicated by GM2020. We note that the GRAVITY measurement is depicted with a star in all figures illustrating orbital solutions. ∆f(k) ≡ N X i=1 ki fi, (2) (2) where k ≡[k1, . . . , kN] is a vector of non-zero integers, and PN i=1 |ki| , 0, which yields small |∆f|. Simultaneously, we may determine the critical angle corresponding to η = arg min |∆f(n)|: where k ≡[k1, . . . , kN] is a vector of non-zero integers, and PN i=1 |ki| , 0, which yields small |∆f|. Simultaneously, we may determine the critical angle corresponding to η = arg min |∆f(n)|: θη(t) ≡ N X i=1 ηiλi, (3) 4.5. Astrometric fits to near-resonance configurations To sample the posterior defined with Eq. (1), we adopted priors described in Sect. 4.2. We did not imply any other particu- lar limits on the anticipated near-resonant N-body solutions, besides uniform priors for orbital parameters in sufficiently wide ranges. In this sense, the assumed prior set is minimal. Moreover, the N-body model is parameterized with (mi, ai, xi ≡ ei cos ϖi, yi ≡ei sin ϖi, Mi), i = 1, . . . , 4, that is, mass, semi- major axis, Poincaré elements composed of eccentricity and argument of pericenter, and the mean anomaly at the osculat- ing epoch for each orbit, as well as two angles (I, Ω) determining the orbital plane of the system, and the astrometric parallax Π. This means 23 free parameters.i In the top-right panel Fig. 4, for HR 8799c, blue circles repre- senting measurements in Konopacky et al. (2016) and red circles and yellow hexagons for SPHERE measurements are slightly shifted with respect to each other, and relative to the orbital model arcs. This is further visible in Figs. 5 and 6, which illustrate residuals of the best-fitting model in Table 6 in the ∆RA–∆Dec-plane as well as individual ∆RA(t) and ∆Dec(t) panels. While the SPHERE data are roughly uniformly clustered around the origin (0, 0), the blue circles exhibit systematic pat- terns, especially for planets HR 8799b and HR 8799c, besides a much larger spread. This effect is curious once we recall the two data sets consisting of uniformly reduced, essentially homoge- neous measurements in Konopacky et al. (2016) and this work, respectively. It is likely that the instruments and/or reduction pipelines are not fully compatible. The effect is quite subtle but still noticeable. It should be noted that the implicitly defined frequency prior (Eq. (5)) makes it difficult to derive the posterior distribution. In many experiments performed with a different number of MCMC walkers of up to 256 and up to 256 000 iterations, we could obtain an acceptance ratio as low as 0.1. Therefore, we understand the MCMC sampling with the ∆n prior as dynami- cally constrained optimization (rejection sampling) which helps to determine parameter ranges for the best-fitting, stable solu- tions. Also, due to implicit frequency prior definition, which has to be calibrated consistently with the expected dynamical stability of the system, the method has a heuristic character. 4.4. Near 8:4:2:1 MMR model In the exact resonance, which we consider as the PO in the reference frame rotating with a selected planet (Hadjidemetriou 1977), the apsidal lines of all planets rotate synchronously, with the same angular velocity relative to the inertial frame. The condition in Eq. (4) for the exact resonance (periodic motion in the rotating frame) may be expressed through the proper mean motions of the planets (e.g., Delisle 2017): i We note that the canonical orbital elements evolve differently from the common astrocentric elements. This is illustrated in Fig. 2. Subsequent panels are for the semi-major axis and eccen- tricity of HR 8799d,c,b, respectively, computed for the interval of a few tens of outermost orbits, and marked with different col- ors. It turns out that these elements span much wider ranges in the common astrocentric Keplerian frame. This effect is particu- larly noticeable for the two outermost planets. A small change in the N-body initial condition may result in substantial displace- ment of the orbit in the Keplerian (a, e)–plane. Therefore, the geometric elements should be understood as a formal representa- tion of the Cartesian N-body initial conditions, and the canonical elements are more suitable for the qualitative characterization of the orbits. ∆n ≡ 3 X i=1 ni ni+1 −2 ! = 0, (5) (5) given the proper mean-motions resolved from the time series S M,i. Numerically, we find that the exact resonance yields ∆n ≃ 10−13 rad d−1 and ∆f ≃10−15 rad d−1 for 3 × 8192 samples of ∆t = 2048 days. Considering a particular four-body MMR chain that can explain astrometric observations of the HR 8799 system, the A133, page 8 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. We used the resonance constraint in Eq. (5) as one of the pri- ors for the MCMC sampling of the posterior defined through the maximum likelihood function in Eq. (1). This indirect frequency prior is Gaussian with a small variance σ∆n to be determined later. Laplace resonance center, which is selected as the starting PO model and that may explain missing mass correlations found for the PO model. 4.4. Near 8:4:2:1 MMR model While the fit quality in terms of ln L or χ2 may be somewhat improved with the quasi-periodic model, the RMS of the best-fitting models remains almost the same in the two approaches. It may be concluded that the quasi-periodic model does not lead to any qualitative change in the dynamical status of the system, apart from making it possible to systematically explore long-term stable solutions, which are not necessarily regular. We discuss this further in Sect. 4.6.i We determined the frequency variance σ∆n based on exten- sive numerical experiments that relied on calibrating the ∆n range with the Lagrange (geometric) stability interval, which should not be shorter than the stellar age of 40–50 Myr, as pre- dicted in the most recent work (Brandt et al. 2021). These authors also rule out ages for HR 8799 of greater than ≈300 Myr. The proper choice of the ∆n prior is crucial to bound (resolve) a long- term stable system in a very short integration time that spans merely a few hundred of the outermost orbits, ≃2 × 105 yr, which is also the instability timescale outside the Laplace resonance. We find, as described below in Sect. 4.6, that the astrometric HR 8799 solutions are long-term stable if |∆n| is roughly less than 10−5–10−6 rad d−1, as described above. Figure 4 illustrates the best-fitting orbital solutions (red curves) over-plotted for models within 1σ (grey curves) ran- domly selected from the MCMC samples such that their RMS varies between 7 and 9 mas. The best-fitting models with RMS ≃7.6 mas are marked as red curves. All of those solutions are depicted for roughly one osculating period for each planet. Clearly, the orbits do not close for all companions. This effect is best visible for the outermost ( top-middle panel) and HR 8799d ( bottom-right panel) planet, respectively. In the latter case, the range of the orbit splitting can be compared to a relatively large uncertainty on the HST measurements; for more accurate mea- surements, the splitting would be even more significant. This effect illustrates strong, short-term mutual interactions. Here, we follow Go´zdziewski & Migaszewski (2018), who indicated that Keplerian orbits are already inadequate for representing the proper astrometric solutions despite limited observational orbital arcs. 4.5. Astrometric fits to near-resonance configurations Nevertheless, we may gain much insight into the parameter ranges of stable near-resonance models consistent with the astrometric observations. Another trend of residuals is revealed in Fig. 6 for planet HR 8799e. The SPHERE data seem to be clearly arranged along a curve in the RA coordinate, revealing a systematic deviation in time with the ephemeris. This effect is especially clear for IRDIS data and may be noted in the lower right ∆RA–∆Dec panel and the ∆RA(t) panel, as the yellow hexagons spread along the ∆RA = 0 axis. This may indicate a systematic effect in IRDIS data reduction because it does not seem to appear for IFS data. However, the time–∆RA(t) panel is suggestive of this trend for both sets of measurements. If it has no instrumental origin, then it may indicate an unmodeled component of the astromet- ric model, such as the presence of a yet unknown object (e.g., a distant moon) or a different curvature of the orbit due to dif- ferent parameters and geometry (e.g., noncoplanarity). We note here the perfect position of the GRAVITY measurement, but this one point is insufficient to dismiss the trend. We performed many MCMC experiments varying σ∆n within the range of 10−7–10−9 rad d−1 and parameters of the emcee samplers. The example posterior for σ∆n = 10−8 rad d−1 is illustrated in Fig. 3 for parameters selected consistently with the posterior for PO shown in Fig. 1. During the sampling, we recorded all elements of the tested models, and the semi- amplitude of the critical angles of the Laplace resonance for the integration time spanning ≃200 outermost periods. Fig. 3 shows a collection of samples with |∆n| < 10−5 rad d−1, and the semi-amplitude of Laplace resonance argument |∆θ| < 60◦. This choice is explained in the following Sect. 4.6, in which we analyze the orbital evolution of particular selected solutions. Overall, when comparing the posterior distributions in Fig. 1 derived for the PO model, and in Fig. 4 for the quasi-periodic model, we may notice similar ranges for the parameters. A basic distinction relies on the different character of the solutions: while the PO posterior covers strictly stable, resonant solutions, the ∆n posterior also involves weakly chaotic solutions around the 4.6. Frequency priors versus the dynamical stability 4.6. Frequency priors versus the dynamical stability Regarding the dynamical character of solutions derived in Sect. 4.5, we analyzed representative examples of the best-fitting models marked in Fig. 4. We selected MCMC samples in Fig. 3 A133, page 9 of 25 A&A 666, A133 (2022) − − Fig. 3. Posterior samples for stable solutions characterized with |∆n| < 10−5 rad d−1, and the semi-amplitude of the libration of the Laplac esonance argument |∆θ| < 60◦. The starting solution that initiated nearby MCMC walkers in the emcee samplers is the periodic configuratio n Tables 6 and 7 (Model 1). Uncertainties for the parameters are determined as the 16th and 86th percentiles of the samples around the media values. A&A 666, A133 (2022) − − Fig. 3. Posterior samples for stable solutions characterized with |∆n| < 10−5 rad d−1, and the semi-amplitude of the libration of the Laplace resonance argument |∆θ| < 60◦. The starting solution that initiated nearby MCMC walkers in the emcee samplers is the periodic configuration in Tables 6 and 7 (Model 1). Uncertainties for the parameters are determined as the 16th and 86th percentiles of the samples around the median values for m⋆= 1.47 M⊙and from other MCMC sampling for m⋆= 1.52 M⊙. These initial conditions were integrated with the IAS15 integrator from the REBOUND package (Rein & Spiegel 2015) for 1 Gyr or up to disruption of the system when a collision or ejection of a planet occurs. The results are shown in a sequence of plots in Figs. 7 and 8. argument is for the outermost pair of planets and librates around the center at ≃180◦. As it comes to the critical argument of the Laplace resonance, it librates with a very small amplitude of just ≃5◦around the resonance center at ≃15◦. The periodic, perfectly regular character of this solution manifests as time evolution of the eccentricities. The middle column (Model 2) in Fig. 7 is for a model slightly displaced from the exact resonance, as measured by |∆n| ≃ 10−7 rad d−1, yet it yields slightly better fit quality. The critical arguments evolve in the same manner, as for the exact resonance, but their amplitudes and eccentricities become noticeably larger. This solution also yields a fully stable configuration of the plan- ets. It is regular in the sense of the Lyapunov exponent, as it Systems illustrated in Fig. 7 are for long-term stable solutions that survived for at least 1 Gyr. 4.6. Frequency priors versus the dynamical stability The left column is for exactly res- onant Model 1 and its elements displays Table 7 and also Table 6. This solution is stable forever “by design” in the framework of the N body dynamics. All but one critical angle of the two- body MMRs between subsequent pairs of planets librate around centers in the [−90◦, 90◦]–range. The exceptional critical A133, page 10 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Fig. 4. Best-fitting solutions to the four-planet model in Table 6 for m⋆= 1.47 M⊙, illustrated on the sky plane as randomly selected MCMC samples from the ∆n posterior. The y-axis corresponds to the north (N), and the x-axis corresponds to the east (E) direction, respectively (we note that the numerical values of ∆α are opposite in sign with respect to the formal left-hand direction of the RA α). Red-filled circles are for the new or re-reduced IRDIS measurements in this paper, yellow hexagons are for the IFS measurements, green hexagons are for the LUCI points, light-blue filled circles are for measurements in Konopacky et al. (2016), and dark-blue circles are for data in other references collected in Tables 2, 3, and 5 and summarised in Table A.1. Diamonds are for the GPI data, and a star symbol is for the most accurate GRAVITY point. Red curves mark stable solutions with the lowest RMS ≃7.6–8 mas, randomly selected from the MCMC samples, and darker grey curves are for other orbital arcs derived for stable models up to RMS ≃9–10 mas. All model orbits have been derived for all measurements available to date, and cover roughly one osculating orbital period for every planet, respectively. The osculating epoch is 1998.830. We also labeled the observational epochs encompassing the orbital arcs with data. The upper left panel is for the global view of the system on the sky plane, and subsequent panels are for its close-ups and interesting regions. Fig. 4. Best-fitting solutions to the four-planet model in Table 6 for m⋆= 1.47 M⊙, illustrated on the sky plane as randomly selected MCMC samples from the ∆n posterior. The y-axis corresponds to the north (N), and the x-axis corresponds to the east (E) direction, respectively (we note that the numerical values of ∆α are opposite in sign with respect to the formal left-hand direction of the RA α). 4.6. Frequency priors versus the dynamical stability Red-filled circles are for the new or re-reduced IRDIS measurements in this paper, yellow hexagons are for the IFS measurements, green hexagons are for the LUCI points, light-blue filled circles are for measurements in Konopacky et al. (2016), and dark-blue circles are for data in other references collected in Tables 2, 3, and 5 and summarised in Table A.1. Diamonds are for the GPI data, and a star symbol is for the most accurate GRAVITY point. Red curves mark stable solutions with the lowest RMS ≃7.6–8 mas, randomly selected from the MCMC samples, and darker grey curves are for other orbital arcs derived for stable models up to RMS ≃9–10 mas. All model orbits have been derived for all measurements available to date, and cover roughly one osculating orbital period for every planet, respectively. The osculating epoch is 1998.830. We also labeled the observational epochs encompassing the orbital arcs with data. The upper left panel is for the global view of the system on the sky plane, and subsequent panels are for its close-ups and interesting regions. left column in Fig. 8), 600 Myr (Model 5, the middle col- umn), and just 200 Myr (Model 6, the right column). Despite this, for the initial few hundred million years, extending safely beyond most approximations of the stellar age, which range between 30–160 Myr (Baines et al. 2012; Sepulveda & Bowler 2022), the systems are bounded and locked in the resonance. Simultaneously, Models 3 and 4 yield masses of the inner planet in the ≃10 MJup range, and Model 4 is especially inter- esting given its low χ2 compared to the initial starting PO configuration. These models yield a declining mass hierarchy that resembles that of the outer Solar System, and a mass of HR 8799e is consistent with the dynamical estimate in Brandt et al. (2021). can be seen in Fig. 9 showing dynamical maps in terms of the MEGNO fast indicator (Cincotta et al. 2003). The mean expo- nential growth factor of nearby orbits (MEGNO; also known as ⟨Y⟩) is a numerical technique designed to efficiently character- ize the stability of the N-body solutions in terms of the maximal Lyapunov exponent (MLE). We implemented MEGNO for the planetary problem in Go´zdziewski et al. (2001) and in our CPU- parallelized µFARM numerical package. 4.6. Frequency priors versus the dynamical stability It is also clear that this solution lies close to the edge of the resonance island (dark blue color). Finally, the right column (Model 3) in Fig. 7 is for a marginally stable solution, in the sense of nonzero MLE, with |∆n| ≃10−5 rad d−1. Although it survived for the 1 Gyr inte- gration interval, one of the critical angles in the outermost 2:1 MMR rotates. This solution is peculiar in the sense that, despite rotations of one of the 2:1 MMR critical arguments, the semi-amplitudes of other critical angles are similar to those for quasi-periodic Model 2. These examples are to justify that the system may be dynami- cally long-term stable in the planet mass range of ≃10 MJup, even if detuned from the exact resonance and mildly chaotic. In all unstable cases, one of the critical arguments of the 2:1 MMR of the two outermost planets progressively increases its libration amplitude and eventually begins to rotate. In this sense, the outer- most pair HR 8799b–c is the weakest link in the resonance chain, provoking instability of the whole system displaced from the res- onance. The time for the onset of instability can be relatively Model 3 in Fig. 7 is also remarkable when we compare it with a sequence of solutions in Fig. 8 illustrating chaotic, yet still long-term stable models characterised by |∆n| ≃10−6 rad d−1. All these models self-destruct between 800 Myr (Model 4, the A133, page 11 of 25 A133, page 11 of 25 A&A 666, A133 (2022) −40 −20 0 20 40 ΔRA [mas] −40 −30 −20 −10 0 10 20 30 40 ΔDEC [mas] HR8799b 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔRA [mas] 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔDEC [mas] −40 −20 0 20 40 ΔRA [mas] −40 −30 −20 −10 0 10 20 30 40 ΔDEC [mas] HR8799c 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔRA [mas] 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔDEC [mas] Fig. 5. Residuals to the selected best-fitting, near-resonant four-planet Model 2 in Table 7 for planets HR 8799b and HR 8799c; stellar mass is m⋆= 1.47 M⊙. 4.6. Frequency priors versus the dynamical stability The y-axis corresponds to the north (N), and the x-axis corresponds to the east (E) direction, respectively (we note that the numerical values of ∆RA are sign-opposite to regarding the formal left-hand direction of the RA axis). Red filled circles, and yellow and green hexagons are for the IRDIS, IFS, and LUCI measurements reported here, respectively, and dark-blue and light-blue (light-grey) filled circles are for measurements in previous papers and in Konopacky et al. (2016), respectively. Yellow diamonds are for GPI, the grey diamonds are for the early HST data. 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔRA [mas] 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔDEC [mas] −40 −20 0 20 40 ΔRA [mas] −40 −30 −20 −10 0 10 20 30 40 ΔDEC [mas] HR8799b [ ] −40 −20 0 20 40 ΔRA [mas] −40 −30 −20 −10 0 10 20 30 40 ΔDEC [mas] HR8799c 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔRA [mas] 1995 2000 2005 2010 2015 2020 2025 time [yr] −40 −20 0 20 40 ΔDEC [mas] Fig. 5. Residuals to the selected best-fitting, near-resonant four-planet Model 2 in Table 7 for planets HR 8799b and HR 8799c; stellar mass is m⋆= 1.47 M⊙. The y-axis corresponds to the north (N), and the x-axis corresponds to the east (E) direction, respectively (we note that the numerical values of ∆RA are sign-opposite to regarding the formal left-hand direction of the RA axis). Red filled circles, and yellow and green hexagons are for the IRDIS, IFS, and LUCI measurements reported here, respectively, and dark-blue and light-blue (light-grey) filled circles are for measurements in previous papers and in Konopacky et al. (2016), respectively. Yellow diamonds are for GPI, the grey diamonds are for the early HST data. very long, as shown by the evolution of Models 4 and 5. In any case, the libration amplitudes of the critical angles seem to be a weak indicator of instability, because there is no clear rela- tionship between these amplitudes and the time of instability. Also, Model 5 illustrates the difficulty in predicting the system behavior based on the variation of critical angles, especially if the system stability is tested for a limited period of time. 4.6. Frequency priors versus the dynamical stability The apparently regular, quasi-periodic, and bounded evolution of the critical angles for ≃200 Myr does not prevent the system from eventually becoming unstable after ≃400–500 Myr. A similar effect may be observed for Model 6, although in this case, the critical argument of the Laplace resonance varies irregularly at the beginning of the integration. 4.7. Resonant structure of the inner debris disk Based on the updated orbital solutions in Table 6, we revised and extended simulations of the dynamical structure of the inner debris disk in Go´zdziewski & Migaszewski (2018). These experi- ments rely on the concept of the so-called ⟨Y⟩-model. We assume that the planets form a system of primaries in safely stable orbits robust to small perturbations. We then inject bodies with masses significantly smaller than those of the primaries and on orbits with different semi-major axes and eccentricities span- ning the interesting region, and randomly selected orbital angles. Next, we integrate the synthetic configuration and determine its stability with the MEGNO (⟨Y⟩) fast indicator. Calibra- tion experiments spanning orbital evolution of debris disks in HR 8799 for up to 70 Myr are described in detail in Go´zdziewski & Migaszewski (2018). A comparison of the results of direct numerical integrations with the outcomes of the ⟨Y⟩-model con- firms that these two approaches are consistent with one another, yet the ⟨Y⟩-based method is much more CPU efficient and there- fore makes it possible to obtain a clear representation of the structure of stable solutions. This algorithm is especially effec- tive for strongly interacting systems. Also, these simulations It is worth noting that all of the example solutions yield astro- metric fits that differ little in quality in terms of χ2 and RMS. The models are difficult to distinguish statistically and visually from each other. This may mean that for a coplanar, near-resonance, or resonance model we can hardly differentiate between per- fectly regular and chaotic evolution of the system, as long as the instability time is sufficiently long relative to the age of the star. However, these two types of configurations occur near the exact Laplace resonance. A133, page 12 of 25 A133, page 12 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. = Fig. 6. Residuals to the best-fitting, near-resonant four-planet Model 2 in Table 7 for m⋆= 1.47 M⊙, a continuation of Fig. 8 for planets HR 8799d and HR 8799e. See the caption of Fig. 5 for labels. The yellow star marks the GRAVITY measurement for HR 8799e. = Fig. 6. Residuals to the best-fitting, near-resonant four-planet Model 2 in Table 7 for m⋆= 1.47 M⊙, a continuation of Fig. 8 for planets HR 8799d and HR 8799e. See the caption of Fig. 5 for labels. 4.7. Resonant structure of the inner debris disk The yellow star marks the GRAVITY measurement for HR 8799e. = ing, near-resonant four-planet Model 2 in Table 7 for m⋆= 1.47 M⊙, a continuation of Fig. 8 for planets HR 8799d of Fig. 5 for labels. The yellow star marks the GRAVITY measurement for HR 8799e. the epoch 1998.830 up to the given final epoch t. For instance, t = 2009.575 (middle-right panel) is for the first detection of the innermost planet in Marois et al. (2010). revealed that the close proximity of the four primaries to the exact Laplace resonance makes the system stability robust even to apparently significant perturbations caused by probe masses as large as ≃2 MJup. We mark the model orbits of the two inner planets with red curves, and the astrometric measurements are over-plotted on them. Instant positions of the planets are marked with shaded large filled circles (for the first HST epoch), and large filled cir- cles for positions of the planets at the particular epoch labeled in the lower right corner of each panel. The time range of about 30 yr is more than half the orbital period of HR 8799e and roughly one orbital period of an asteroid involved in the 2:1e MMR with this planet. We considered two types of probe objects: Ceres-like aster- oids with a mass of 10−6 MJup and Jupiter-like planets with a mass of 1 MJup. To calculate the ⟨Y⟩values for the synthetic sys- tems, we integrated the N-body equations of motion and their variational equations with the GraggBulirsch–Stoer integrator (Hairer et al. 2000) for 3 Myr, which covers 6 × 104 orbital periods of HR 8799e and 7 × 103 orbital periods of HR 8799b, respectively. That integration time is consistent with the typical 104 outermost orbits required to achieve ⟨Y⟩convergence. We also note that the most significant interactions are exerted by the inner planets. We chose the integration time that is optimal from the CPU overhead point of view. Positions of the test particles are marked with different col- ors depending on their dynamical status: yellow and orange dots are for objects involved in 1:1e MMR with the innermost planet HR 8799e; green dots are for the 3:2e MMR, blue dots are for the 2:1e MMR, and gray dots are for the other stable asteroids. As can be seen in Fig. 4.7. Resonant structure of the inner debris disk The left column is for the exact Laplace resonance, the middle column is for a model displaced from the PO but rigorously stable, and he right column is for a >1 Gyr stable solution characterized by rotation of one critical angle of the 2:1 MMR in the outermost pair. The rows from he top to bottom are for all critical angles of the 2:1 MMR, the critical angle of the generalized Laplace MMR, and eccentricities for subsequent pairs of planets. See Table 7 for the initial conditions labeled from 1 to 3 at the top-left corner of each column. A&A 666, A133 (2022) Fig. 7. Orbital evolution of selected long-term-stable solutions with similar astrometric fit quality, but exhibiting qualitatively different stability signatures. The left column is for the exact Laplace resonance, the middle column is for a model displaced from the PO but rigorously stable, and the right column is for a >1 Gyr stable solution characterized by rotation of one critical angle of the 2:1 MMR in the outermost pair. The rows from the top to bottom are for all critical angles of the 2:1 MMR, the critical angle of the generalized Laplace MMR, and eccentricities for subsequent pairs of planets. See Table 7 for the initial conditions labeled from 1 to 3 at the top-left corner of each column. orbital phases, (a0, λe −λ0) (bottom panel). In both graphs, we mark the test particles with the same color scheme as in the previous plot, and the low-order resonances with HR 8799e are labeled. We also plot the geometrical collision curve (in gray) with planet HR 8799e constrained through the apocenter dis- tance of the inner orbit equal to the pericenter distance of the planet, a0(1 + e0) = ae(1 −ee). Also, the curve depicted in red is an image of the collision curve shifted by ∆a0 = 4 au towards the star and clearly marks a boundary of stable particles. the (a0, λe −λ0)-plane. Both graphs mark two families of aster- oids in 1:1e MMR – depicted in gold colour (e0 < 0.3) and in orange (e0 > 0.3). The low-eccentricity objects resemble clas- sic Trojan asteroids in the Solar System, relative by ±60◦to the planet. The second unusual family of highly eccentric particles in 1:1e MMR can be found on the sky plane far beyond the orbit of HR 8799e. 4.7. Resonant structure of the inner debris disk 10, the outer edge of the disk is highly nonsymmetric and quickly evolves in time. Its temporal struc- ture strongly changes even for a relatively very short interval of 25 years spanned by the astrometric observations of the system. The structure of the disk has a clear resonant structure that was also noted by Go´zdziewski & Migaszewski (2018). This is demonstrated in Fig. 11, which shows the distribution of canoni- cal osculating elements inferred in the Jacobi reference frame, in the (a0, e0)-plane (top panel), and on the plane of relative Positions of the test particles are marked with different col- ors depending on their dynamical status: yellow and orange dots are for objects involved in 1:1e MMR with the innermost planet HR 8799e; green dots are for the 3:2e MMR, blue dots are for the 2:1e MMR, and gray dots are for the other stable asteroids. As can be seen in Fig. 10, the outer edge of the disk is highly nonsymmetric and quickly evolves in time. Its temporal struc- ture strongly changes even for a relatively very short interval of 25 years spanned by the astrometric observations of the system. The results for the less massive Ceres-like asteroids are illus- trated in Fig. 10 and Fig. 11. In this experiment, we sampled the semi-major axis a0 ∈[4, 18] au and eccentricity e0 ∈[0, 0.8] of these objects. We collected 3 × 105 N-body initial conditions with |⟨Y⟩−2| < 0.05 that represent the structure of long-term- stable orbits in the inner debris disk. Subsequent panels in Fig. 10 illustrate snapshots of the disk at other epochs following the first observation (t0 =1998.830) as seen on the sky plane. To obtain the snapshots, we numerically integrated the whole set of ini- tial conditions for planets and the Ceres-like objects defined at The structure of the disk has a clear resonant structure that was also noted by Go´zdziewski & Migaszewski (2018). This is demonstrated in Fig. 11, which shows the distribution of canoni- cal osculating elements inferred in the Jacobi reference frame, in the (a0, e0)-plane (top panel), and on the plane of relative A133, page 13 of 25 A133, page 13 of 25 A&A 666, A133 (2022) Fig. 7. Orbital evolution of selected long-term-stable solutions with similar astrometric fit quality, but exhibiting qualitatively different stability signatures. A133, page 14 of 25 4.7. Resonant structure of the inner debris disk At some epochs (e.g., 2011.83), these appear closer to the orbit of HR 8799d, which may be counter-intuitive. Given the 1:1e MMR dynamical classification, they would be expected to share the same orbit as the planet. y y p The bottom graph shows objects involved in low-order MMRs with HR 8799e that are apparently widely spread on the sky plane, but they appear in narrow and well-localized islands on the planes of orbital elements; this is particularly clear on p This simulation reveals the dynamical structure composed of long-term-stable but mutually noninteracting asteroids in the system, on the same orbit as the innermost planet, whose orbital A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Fig. 8. Orbital evolution of selected long-term-stable chaotic solutions with similar astrometric fit quality. The rows from the top to bottom are or all critical angles of the 2:1 MMR, the critical angle of the generalized Laplace MMR, and eccentricities for subsequent pairs of planets. See Table 7 for the initial conditions labeled from 5 to 8 at the top of each column. y y Fig. 8. Orbital evolution of selected long-term-stable chaotic solutions with similar astrometric fit quality. The rows from the top to bottom are for all critical angles of the 2:1 MMR, the critical angle of the generalized Laplace MMR, and eccentricities for subsequent pairs of planets. See Table 7 for the initial conditions labeled from 5 to 8 at the top of each column. the warm disk is estimated down to 610 au (e.g., Su et al. 2009; Matthews et al. 2014; Chen et al. 2014), which coincides with the 2:1e MMR (Fig.11); see also Sect 5. Given the topic is complex and somewhat out of the scope of the present work, we postpone in-depth analysis of the dynamical structure of the warm dust disk to a future paper. evolution is governed by the gravitational interactions with the massive Jovian companions. Such objects should not be signif- icantly affected by nongravitational forces, such as Poynting- Robertson drag or the Yarkovsky effect. However, such forces may be crucial for the investigation and modeling of the actual emission profile of dust produced by collisional dynamics. The results indicate that the asteroid breakup events may be frequent and violent. 4.7. Resonant structure of the inner debris disk Given the large eccentricities and the instant mix- ing of different resonant fractions of these objects in the wide inner region between 4 au (and below) and 18 au, including wide Lagrange clumps of planet HR 8799e, their orbital velocity dis- persion may be significant. The interplay of all these factors – and especially the clearing rate of the dust due to radiation of the young and bright host star – determines the emission profile. The present observational evidence is limited, and the position of 4.8. Putative fifth Jupiter-like planet We performed a very similar experiment for test particles of mass 1 MJup, a hypothetical planet below the current detection level that may exist in the inner part of the HR 8799 system. The results are illustrated in Fig. 12 in a similar way to in Figs. 10–11. We collected 105 stable solutions. Again, the positions of the hypothetical planet are marked at characteristic epochs; for A133, page 15 of 25 A133, page 15 of 25 A&A 666, A133 (2022) Fig. 9. Dynamical maps in the Keplerian, osculating astrocentric semi-major-axis–eccentricity plane for the quasi-periodic, near-resonant config- uration described by Model 2. All masses and elements but the map coordinates are fixed at their best-fitting values in Table 7. The osculating epoch is 1998.830. Stable solutions are determined with |⟨Y⟩−2| ≃0 and marked with blue color. Fig. 9. Dynamical maps in the Keplerian, osculating astrocentric semi-major-axis–eccentricity plane for the quasi-periodic, near-resonant config- uration described by Model 2. All masses and elements but the map coordinates are fixed at their best-fitting values in Table 7. The osculating epoch is 1998.830. Stable solutions are determined with |⟨Y⟩−2| ≃0 and marked with blue color. 7 and 16 au lies above roughly three Jupiter masses, as also con- cluded here; see Fig. 13 and discussion in Sect. 6. Therefore, we believe that steadily improved imaging techniques, gaining better contrast and lower detection limits, combined with dynam- ical simulations similar to those conducted in this section may eventually reveal the fifth planet. However, if a relatively massive planet of mass ≃1 MJup exists in the inner part of the system, the outer edge of the debris disk carved out by this planet would have a much smaller radius than predicted at ≃10 au under the current observational configuration of four planets (Fig. 11), and related to the above-mentioned detections of warm dust at 6–10 au (e.g., Su et al. 2009; Hughes et al. 2011; Matthews et al. 2014; Chen et al. 2014). We defer the analysis of this situation to a future work as well, because of its complexity arising from different MMR scenarios. example, near the first epoch of the HST detection, up to the last epoch of measurements in this work, and one epoch ahead of that time. Compared to the previous case, the distribution of stable objects on the plane of the sky is much narrower and more restricted. 4.8. Putative fifth Jupiter-like planet As in the case of low-mass asteroids, the positions of the putative planets can be seen to change rapidly relative to the background model orbits and astrometric measurements. 7 and 16 au lies above roughly three Jupiter masses, as also con- cluded here; see Fig. 13 and discussion in Sect. 6. Therefore, we believe that steadily improved imaging techniques, gaining better contrast and lower detection limits, combined with dynam- ical simulations similar to those conducted in this section may eventually reveal the fifth planet. However, if a relatively massive planet of mass ≃1 MJup exists in the inner part of the system, the outer edge of the debris disk carved out by this planet would have a much smaller radius than predicted at ≃10 au under the current observational configuration of four planets (Fig. 11), and related to the above-mentioned detections of warm dust at 6–10 au (e.g., Su et al. 2009; Hughes et al. 2011; Matthews et al. 2014; Chen et al. 2014). We defer the analysis of this situation to a future work as well, because of its complexity arising from different MMR scenarios. We note that the stability limit (red curve) in the bottom plot for the (a0, e0)-plane is offset by ∆a0 = 6 au with respect to the collision curve with planet HR 8799e shown in gray. In addition to the clear dependence of the structure of stable solutions on the probe mass, this simulation indicates that the hypothetical planet may be located only on very narrow islands in the orbital param- eter space, limited to low-order resonances with the innermost planet. The temporal evolution of these islands in the sky can be useful for analyzing AO images and provides clues as to where an additional, Jovian planet might be expected. If such objects exist beyond ≃7 au, they should be involved in a 2:1e, 3:1e, or 5:2e MMR with the inner planet, respectively. These low-order reso- nances may be preferred if the system has undergone a migration in the past. We discuss this further in Sect. 5.ii 5. Possible history of the system Convergent migration due to tidal interaction with the nesting circumstellar disk is thought to be the most reliable mechanism for producing resonance trapping in multiple-planet systems (Masset & Snellgrove 2001; Lee & Peale 2002; Moorhead & Adams 2005; Thommes 2005; Beaugé et al. 2006; Crida et al. 2008; D’Angelo & Marzari 2012). The different masses of the planets and progressive inside-out depletion of gas in the disk due to the presence of the planet itself and photo-evaporation lead to different migration speeds for the planets, which may end up in resonance. Numerical modeling by Hands et al. (2014) and Szuszkiewicz & Podlewska-Gaca (2012) shows that during inward migration, planets are often trapped in resonances like the 2:1 and 3:2 (the most frequent). Attempts to find the hypothetical innermost fifth planet have so far been unsuccessful. Very recently, Wahhaj et al. (2021), using SPHERE measurements also reported here, did not detect planet HR 8799f at the most plausible locations, namely 7.5 and 9.7 au, down to mass limits of 3.6 and 2.8 MJup, respectively. Neither did these authors detect any new candidate companions at the smallest observable separation, of namely 0.1′′ or ≃4.1 au, overlapping with the semi-major axes range in Figs. 10–12. Wahhaj et al. (2021) conclude that the planet may still exist with a mass of 2–3.6 MJup at 7.5 au (3:1e MMR) or 1.5–2.8 MJup at 10 au, which is in the region we closely investigated in this sec- tion. The contrast curve in the interesting zone between roughly While in most resonant cases, such as Gliese 876, HD 82943, and HD 37124 (Wright et al. 2011), the planets are close to the A133, page 16 of 25 A133, page 16 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. 7. Osculating, heliocentric elements of the best-fitting solutions at the epoch of 1998.83 for models illustrated in Fig. 4 and in Figs. 5. Possible history of the system 7– Model 1: m⋆=1.47 M⊙, Π = 24.525662 mas, χ2 = 1593.22, RMS = 7.41 mas, log |∆n[rad d−1]| = –12.0 m (MJup) a (au) e I (deg) Ω(deg) ω (deg) M (deg) HR8799e 7.4060918 16.1038912 0.1476760 26.8734271 62.1852189 110.8167067 -23.1719500 HR8799d 9.1899948 26.4517267 0.1146003 26.8734271 62.1852189 29.0594738 60.1752106 HR8799c 7.7961185 40.9861801 0.0538090 26.8734271 62.1852189 92.3879462 145.5347082 HR8799b 5.8023594 71.1427851 0.0166707 26.8734271 62.1852189 42.9704380 -49.1049354 Model 2: m⋆=1.47 M⊙, Π = 24.451807 mas, χ2 = 1568.19, RMS = 7.66 mas, log |∆n[rad d−1]| = –6.2 m (MJup) a (au) e I (deg) Ω(deg) ω (deg) M (deg) HR8799e 7.8680000 15.9753973 0.1524708 26.2394739 63.7770393 108.3445443 335.6210695 HR8799d 9.6490000 26.6396632 0.1145654 26.2394739 63.7770393 28.3126706 59.7394429 HR8799c 7.3081300 40.9406369 0.0517422 26.2394739 63.7770393 91.3551907 145.0637721 HR8799b 5.6590700 71.3289844 0.0173411 26.2394739 63.7770393 41.6627885 310.5199311 Model 3: m⋆=1.52 M⊙, Π = 24.347839 mas, χ2 = 1624.73, RMS = 7.66 mas, log |∆n[rad d−1]| = –4.5 m (MJup) a (au) e I (deg) Ω(deg) ω (deg) M (deg) HR8799e 8.9817200 16.1163303 0.1548634 27.3470617 62.9005413 108.0955330 336.2997163 HR8799d 9.0680400 26.8768634 0.1068892 27.3470617 62.9005413 29.2629241 60.0277429 HR8799c 7.2212900 41.0832702 0.0627005 27.3470617 62.9005413 91.9708727 144.7587008 HR8799b 6.1950900 71.7822044 0.0192741 27.3470617 62.9005413 44.4483815 308.9107410 Model 4: m⋆=1.52 M⊙, Π = 24.305464 mas, χ2 = 1564.70, RMS = 7.64 mas, log |∆n[rad d−1]| = –6.2 m [MJup] a [au] e I (deg) Ω(deg) ω (deg) M (deg) HR8799e 10.6577500 16.1274692 0.1496852 26.9747476 63.6149718 107.6133926 336.6560551 HR8799d 9.5739000 26.8852716 0.1169394 26.9747476 63.6149718 27.5720349 60.0200173 HR8799c 6.8413200 41.2618982 0.0555455 26.9747476 63.6149718 93.1349542 143.0235158 HR8799b 5.7085500 71.9446080 0.0217920 26.9747476 63.6149718 45.1149621 307.6876430 Model 5: m⋆=1.47 M⊙, Π = 24.466345 mas, χ2 = 1591.33, RMS = 7.63 mas, log |∆n[rad d−1]| = –6.9 m [MJup] a [au] e I (deg) Ω(deg) ω (deg) M (deg) HR8799e 7.1130900 15.9976012 0.1545297 26.5969851 62.8989533 108.8687608 336.1919760 HR8799d 9.8864900 26.5290674 0.1122973 26.5969851 62.8989533 27.8297522 61.2151462 HR8799c 7.7520500 41.1188833 0.0497536 26.5969851 62.8989533 92.4070918 144.9763772 HR8799b 5.5380200 71.3013150 0.0172797 26.5969851 62.8989533 44.6821435 308.5634968 Model 6: m⋆=1.47 M⊙, Π = 24.539234 mas, χ2 = 1568.59, RMS = 7.63 mas, log |∆n[rad d−1]| = –6.2 m [MJup] a [au] e I (deg) Ω(deg) ω (deg) M (deg) HR8799e 7.3387300 15.9246991 0.1553777 26.5990408 62.7197878 108.3820177 336.0139713 HR8799d 10.0516400 26.5092120 0.1117405 26.5990408 62.7197878 29.2671129 59.8339479 HR8799c 8.0973100 40.9565104 0.0509428 26.5990408 62.7197878 91.8212168 145.6257159 HR8799b 5.8932800 71.1073028 0.0167537 26.5990408 62.7197878 41.9648749 311.2692413 Table 7. 5. Possible history of the system Osculating, heliocentric elements of the best-fitting solutions at the epoch of 1998.83 for models illustrate ntric elements of the best-fitting solutions at the epoch of 1998.83 for models illustrated in Fig. 4 and in Figs. 7–8. The resonance trapping in this system at that distance from the star is confirmed by hydrodynamical simulations performed with the FARGO code (Masset 2000). Fully radiative models have been adopted where the energy equation contains viscous heating and radiative cooling through the disk surface. A polar grid with 682 × 512 elements is used to cover the disk, extend- ing from 1 to 120 au. The initial surface gas density is given by Σ = Σ0r−1/2 with Σ0 = 50 g cm−2. This low density is moti- vated by the evolved state of the disk when the planets are fully formed. To test the sequential resonance trapping, we start the star, in the case of HR 8799 the planets are significantly far away. This implies that the primordial disk from which they formed should have extended beyond 100 au which is compatible with the mass of the star being ∼1.5 M⊙. The most robust scenario is that in which the planets were fully formed further out in the disk and then migrated inwards until they became progressively trapped in the multiple resonances. This scenario raises the ques- tion of whether these planets were formed by core accretion (Mizuno 1980; Pollack et al. 1996) or by gravitational instability (Cameron 1978; Boss 1997). star, in the case of HR 8799 the planets are significantly far away. This implies that the primordial disk from which they formed should have extended beyond 100 au which is compatible with the mass of the star being ∼1.5 M⊙. The most robust scenario is that in which the planets were fully formed further out in the disk and then migrated inwards until they became progressively trapped in the multiple resonances. This scenario raises the ques- tion of whether these planets were formed by core accretion (Mizuno 1980; Pollack et al. 1996) or by gravitational instability (Cameron 1978; Boss 1997). A133, page 17 of 25 A133, page 17 of 25 A&A 666, A133 (2022) Fig. 10. Evolution of the inner debris disk as seen on the plane of the sky. 5. Possible history of the system The subsequent panels show the instantaneous positions of two inner planets and small asteroids in stable orbits (small colored dots) with a mass of 10−6 MJup at the epoch labeled in the lower-left corner of each graph. We collected 3.3 × 105 particles. Their distribution in the (a0, e0)-plane of osculating, canonical elements at the initial epoch 1998.830 is shown in Fig. 11. The largest filled circles correspond to the positions of the planets HR 8799e and HR 8799d at the snapshot epoch (large filled circles) and he initial osculating epoch (shaded filled circles), respectively. The asteroid colors are for the lowest order MMR types, as indicated in the bottom wo plots, or those belonging to the innermost quasi-homogeneous disk (gray points). The 2024.83 epoch covers roughly one orbital period in a 2:1e MMR resonance (dark blue circles) with the innermost planet. A&A 666, A133 (2022) A&A 666, A133 (2022) Fig. 10. Evolution of the inner debris disk as seen on the plane of the sky. The subsequent panels show the instantaneous positions of two inner planets and small asteroids in stable orbits (small colored dots) with a mass of 10−6 MJup at the epoch labeled in the lower-left corner of each graph. We collected 3.3 × 105 particles. Their distribution in the (a0, e0)-plane of osculating, canonical elements at the initial epoch 1998.830 is shown in Fig. 11. The largest filled circles correspond to the positions of the planets HR 8799e and HR 8799d at the snapshot epoch (large filled circles) and the initial osculating epoch (shaded filled circles), respectively. The asteroid colors are for the lowest order MMR types, as indicated in the bottom two plots, or those belonging to the innermost quasi-homogeneous disk (gray points). The 2024.83 epoch covers roughly one orbital period in a 2:1e MMR resonance (dark blue circles) with the innermost planet. inner planets on already resonant orbits; they are not affected by the disk, and so do not migrate. The outer planet instead feels the disk perturbations and migrates inward until it is trapped in the 2:1 resonance and stops migrating. This behavior is illustrated in Fig. 14 where in the top panel the semi–major axis of the fourth planet is shown while in the middle and bottom panels the crit- ical argument of the 2:1 resonance between the third and fourth planet and the Laplace resonance argument is illustrated. 6. Planets–disk interaction Along with the four giant planets detected so far, the architecture of HR 8799 is enriched by the presence of an extended debris disk with two components. The cold Kuiper-like component was resolved in the FIR (Matthews et al. 2014) and at millimeter wavelengths (Booth et al. 2016; Wilner et al. 2018) and was detected up to 450 au from the star, with a halo extending to thousands of astronomical units (au). The warm component was instead never fully resolved and its inferred position of 9.3 au is given by modeling the IR excess at 155 K (Chen et al. 2014) in the SED of the star under the assumption of dust particles behaving like black bodies. The two components are separated by a huge dust-free gap that extends from 109 au (Wilner et al. 2018) to regions interior to the orbit of HR 8799e. p Figure 13 shows a summary of the results discussed above. The four detected planets are represented as pink circles together with the extension of their chaotic zone (pink shaded area). As it emerges from the image, the inner extension of the clearing zone gets very close to the inner belt (blue vertical line) whereas it stops roughly 20 au from the outer belt. Even a very low mass planet could carve the remaining part of the gap and, consistently with our observations, it would be too small to be detected with an instrument such as SPHERE, as proven by the detection lim- its (red curve; IRDIS data taken on 2017 October 12) shown in the figure. Simulations of the inner debris disk in the framework of the N-body resonant model of the system in Sect. 4.7 indicate possible localizations of such a planet. We want to stress that this is only a preliminary analysis of the planet–disk interaction. More detailed studies will have to consider the dynamical inter- action of the fifth planet with all four of the detected ones as well as the possibility of having a larger object locked in MMR with at least one of the other known planets, as discussed in the following. 5. Possible history of the system simulation, we increased Σ0 to 100 g/cm−2 in order to speed up the migration. Figure 15 shows the gas density distribution after 8 Kyr from the beginning of the simulation. The planets create a common gap where some overdensities are still present in the corotation regions of each planet. The semi-major axes of the planets are illustrated in the bottom panel, rescaled to be shown in a single plot. They migrate inwards while trapped in resonance, suggesting that the present position of the planets is not necessarily the one where they became trapped in resonance. The capture in the mutual resonance may have occurred farther out, followed by an inward migration until the disk dissipates. This may push the location where the planet initially grew even It is noteworthy that, once all the planets are trapped in res- onance, they share a common gap and migrate inwards. We performed an additional simulation where all four planets are affected by the disk perturbations and can freely migrate. In this A133, page 18 of 25 A133, page 18 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Fig. 11. Structure of the inner debris disk composed of 3 × 105 Ceres-like asteroids of mass 10−6 MJup on stable orbits, seen in the (a0, e0)-plane of Jacobian (canonical) osculating elements at the initial 1998.830 epoch. The colors of the asteroids indicate the lowest order MMR in which they are involved with HR 8799e, as indicated in the diagrams, or belong to the innermost, quasi-homogeneous disk (gray points) according to Fig. 10. The light gray curve indicates the collision of the orbits with HR 8799e and the red curve is an image of the collision curve shifted by ∆a0 ≃4 au toward the star. We note that loose points above the collision curve in the top panel represent immediately scattered asteroids on regular but hyperbolic (open) orbits. Fig. 11. Structure of the inner debris disk composed of 3 × 105 Ceres-like asteroids of mass 10−6 MJup on stable orbits, seen in the (a0, e0)-plane of Jacobian (canonical) osculating elements at the initial 1998.830 epoch. The colors of the asteroids indicate the lowest order MMR in which they are involved with HR 8799e, as indicated in the diagrams, or belong to the innermost, quasi-homogeneous disk (gray points) according to Fig. 10. 5. Possible history of the system The light gray curve indicates the collision of the orbits with HR 8799e and the red curve is an image of the collision curve shifted by ∆a0 ≃4 au toward the star. We note that loose points above the collision curve in the top panel represent immediately scattered asteroids on regular but hyperbolic (open) orbits. farther out. It is noteworthy that the wider oscillations observed in the critical arguments of the resonances and the semi-major axis of the planets compared to a pure N-body problem are related to the presence of the perturbations of the disk on the planets. the inner edge of the gap. On the other hand, the chaotic zone for HR 8799b can clear the orbit of the planet up to 91.4 au. Given the position of the outer edge of the gap at 109 au, we can try to infer the characteristics of a fifth planet able to extend the chaotic zone up to the detected position of the edge. Following a very simple approach, we can use the same analytical tools to estimate the mass and semi-major axis of a planet able to carve a gap extending from 91.4 au to 109 au, that is, the free dynamical region left beyond the orbit of HR 8799b. As a result, we obtain a 0.12 MJup planet orbiting at 100 au. 6. Planets–disk interaction ) g As the presence of the gap is tightly correlated with the plan- ets in it, we can assess the planets–disk interactions and infer whether or not the planets detected are responsible for carving the entire gap or whether or not there is further dynamical space for hosting other undetected planets. For this analytical study, the only planets taken into account are the two closest to the edges of the gap, HR 8799 b and e. The orbital parameters and masses adopted are the ones listed in Tables 6 and 7 for Model 1. To calculate the extension of the region from which dust particles are scattered away from the orbit of the planet (chaotic zone), we used Eqs. (9) and (10) of Lazzoni et al. (2018) for HR 8799b and HR 8799e, respectively. As a result, we obtained that the chaotic zone of the inner planet extends down to 9.8 au, which is con- sistent with the direct N-body simulations in Sect. 4.7. Given the uncertainties on the position of the inner belt and on its width, we can safely state that HR 8799e is likely responsible for shaping Indeed, the results of numerical N-body simulations in Go´zdziewski & Migaszewski (2018) based on the ⟨Y⟩-model described in Sect. 4.7 partially confirm the analytical predictions and address the likely real resonant configuration of the planets. These latter authors simulated the inner edge of the outer debris disk, also accounting for the presence of a hypothetical fifth planet with a mass between 0.1 and 1.66 MJup. These simula- tions are based on a quasi-periodic, near-resonant orbital model A133, page 19 of 25 A133, page 19 of 25 A&A 666, A133 (2022) A&A 666, A133 (2022) A&A 666, A133 (2022) Fig. 12. Possible instant positions (small gray and colored dots) of a hypothetical and still undetected planet of mass 1 MJup on the sky plane, at the epoch labeled in the lower-left corner of each graph. We collected 105 initial conditions. The bottom plot shows their distribution in the (a0, e0)– plane of canonical osculating elements at the initial epoch of 1998.830, similarly to Fig. 11. As in Fig. 10, the largest filled circles correspond to the positions of the planets HR 8799e and HR 8799d at the snapshot epoch (large filled circles) and the initial osculating epoch (shaded filled circles), respectively. 6. Planets–disk interaction Colors mark the lowest order MMR types, as labeled in the bottom plot. The light-gray curve marks the collision zone with HR 8799e, the thinner curve marks the stability zone for m0 = 10−6 MJup (see Fig. 10), and the red curve is the image of the geometric collision curve shifted by ∆a0 ≃6 au toward the star. A133, page 20 of 25 Fig. 12. Possible instant positions (small gray and colored dots) of a hypothetical and still undetected planet of mass 1 MJup on the sky plane, at the epoch labeled in the lower-left corner of each graph. We collected 105 initial conditions. The bottom plot shows their distribution in the (a0, e0)– plane of canonical osculating elements at the initial epoch of 1998.830, similarly to Fig. 11. As in Fig. 10, the largest filled circles correspond to the positions of the planets HR 8799e and HR 8799d at the snapshot epoch (large filled circles) and the initial osculating epoch (shaded filled circles), respectively. Colors mark the lowest order MMR types, as labeled in the bottom plot. The light-gray curve marks the collision zone with HR 8799e, the thinner curve marks the stability zone for m0 = 10−6 MJup (see Fig. 10), and the red curve is the image of the geometric collision curve shifted by ∆a0 ≃6 au toward the star. A133, page 20 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Fig. 14. Resonance trapping of the HR 8799 system during the migra tion phase. The upper panel shows the evolution of the semi-major axi of the outer planet at the moment of capture in resonance with the inne three bodies. After the resonance trapping, the semi-major axis remain constant. The middle panel shows the critical argument of the resonance between the two outer planets. After the capture in resonance, the criti cal argument librates. The same occurs for the Laplace critical argumen shown in the bottom panel. Fig. 13. HR 8799 architecture: two belts (blue lines), four detected plan- ets (pink circles), and a putative fifth planet (light-blue circle). The pink shaded area represents the extension of the chaotic zone whereas the gray shaded zone represents the extension of the outer disk. The con- trast curve for the system is shown in red. Fig. 13. 7. Conclusions Fig. 14. Resonance trapping of the HR 8799 system during the migra- tion phase. The upper panel shows the evolution of the semi-major axis of the outer planet at the moment of capture in resonance with the inner three bodies. After the resonance trapping, the semi-major axis remains constant. The middle panel shows the critical argument of the resonance between the two outer planets. After the capture in resonance, the criti- cal argument librates. The same occurs for the Laplace critical argument shown in the bottom panel. The system around HR 8799 is a unique laboratory with which to study the mutual gravitational interactions between the planets and their relation with the circumstellar disk in the early evolu- tionary stages of the system. To better understand the dynamics of this system, we followed up HR 8799 with SPHERE at the Very Large Telescope and with LUCI at the Large Binocu- lar Telescope to refine the orbital parameters of the system. We reduced the new data with state-of-the-art algorithms that apply the ADI technique, and for consistency, we also repeated the reduction of published SPHERE data from open-time pro- grams. Precise astrometry for the four planets was obtained for 21 epochs from SPHERE and one epoch from LUCI. million years), even if they are ∼2 MJup bigger than the masses proposed in the previous dynamical studies. We find masses of 8–9 MJup for planets HR 8799 e, d, and c, while for the exte- rior planet HR 8799b we estimate a smaller mass of ≃6 MJup. Moreover, the dynamical parallax is consistent with 1σ uncer- tainty with the corrected, independently determined Gaia eDR3 value of 24.50 ± 0.05 mas, reinforcing both the adopted mass of the host star m⋆= 1.47 M⊙and the assumed resonant or close-to-resonant configuration of the system. We performed a detailed exploration of the orbital parame- ters for the four planets with dynamical constraints imposed by the 8:4:2:1 Laplace resonance. This model was updated using 68 epochs from our reduction and the literature. As a result, we re- derived the dynamical masses of the planets and the parallax of the system with minimal prior information. 6. Planets–disk interaction HR 8799 architecture: two belts (blue lines), four detected plan- ets (pink circles), and a putative fifth planet (light-blue circle). The pink shaded area represents the extension of the chaotic zone whereas the gray shaded zone represents the extension of the outer disk. The con- trast curve for the system is shown in red. of the four known planets derived through migration simula- tions. However, given that Go´zdziewski & Migaszewski (2018) adopted the larger mass for the star of m⋆= 1.52M⊙and a larger parallax Π = 25.4 mas, the whole system appears more compact than predicted at present – these authors found the osculating astrocentric semi-major axis of HR 8799b ≃67.1 au, compared to ≃71 au in the present models. The updated values of the par- allax and the stellar mass translate to orbits expanded by a few au. According to the simulations, the inner edge of the outer disk is globally highly nonsymmetric, and planets with a mass of between 0.1 and 1.66 MJup may be present beyond the clear- ing zone, with semi-major axis >90 au. These planets could be involved in low-order resonances, such as 3:2b, 5:3b, 2:1b, or 5:2b with the outermost planet HR 8799b and in low and moder- ate eccentricity orbits; see Fig. 9 in Go´zdziewski & Migaszewski (2018). The fifth planet, even if it had a small mass, would be strongly affecting the shape of the inner parts of the outer debris disk. 7. Conclusions The derived masses are consistent with the prediction from the evolutionary models and allow long-term stability of the orbits (over a few hundred Regarding the quality of the N-body astrometric models con- sidered in this work, we did not find any significant or qualitative improvement of multi-parameter near-resonant configurations A133, page 21 of 25 A133, page 21 of 25 A&A 666, A133 (2022) Fig. 15. Orbital evolution of the HR 8799 system during the migration phase. The upper panel shows the gas density of the disk. The bottom panel shows the semi-major axes of the planets rescaled with constant values for ease of visualization in the same plot. planet HR 8799e, and there is no need for an additional inner planet to explain it. However, we also determined the dynami- cal structure of this region with direct N-body simulations. The results may be useful for predicting the positions of small-mass objects below the present detection limits of ≃3 MJup. Acknowledgements. We are thankful to the referee for the constructive report. A.Z. acknowledges support from the FONDECYT Iniciación en investigación project number 11190837 and ANID – Millennium Science Initiative Program – Center Code NCN2021_080. K.G. is very grateful to Dr Cezary Migaszewski for sharing a template C code for computing periodic orbits in a few planet sys- tems and explanations regarding the PO approach. K.G. thanks the staff of the Pozna´n Supercomputer and Network Centre (PCSS, Poland) for the generous long-term support and computing resources (grant No. 529). K.G. also thanks the staff of the Tricity Supercomputer Centre, Gda´nsk (Poland) for comput- ing resources on the Tryton cluster that were used to conduct some numerical experiments and for their great and professional support. A.V. acknowledges funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme, grant agreements no. 757561 (HiRISE). The LBTO AO group would like to acknowledge the assistance of R.T. Gatto with night observations. We would also like to thank J. Power for assisting in preparation of the observations using ADI mode. For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any author accepted manuscript version arising from this submis- sion. 7. Conclusions SPHERE is an instrument designed and built by a consortium consisting of IPAG (Grenoble, France), MPIA (Heidelberg, Germany), LAM (Marseille, France), LESIA (Paris, France), Laboratoire Lagrange (Nice, France), INAF – Osservatorio di Padova (Italy), Observatoire de Genève (Switzerland), ETH Zurich (Switzerland), NOVA (Netherlands), ONERA (France), and ASTRON (Netherlands), in collaboration with ESO. SPHERE was funded by ESO, with additional contributions from CNRS (France), MPIA (Germany), INAF (Italy), FINES (Switzerland), and NOVA (Netherlands). SPHERE also received funding from the European Commission Sixth and Seventh Framework Programmes as part of the Optical Infrared Coordination Network for Astronomy (OPTICON) under grant number RII3-Ct-2004-001566 for FP6 (2004–2008), grant number 226604 for FP7 (2009–2012) and grant number 312430 for FP7 (2013–2016). References Fig. 15. Orbital evolution of the HR 8799 system during the migration phase. The upper panel shows the gas density of the disk. The bottom panel shows the semi-major axes of the planets rescaled with constant values for ease of visualization in the same plot. Apai, D., Kasper, M., Skemer, A., et al. 2016, ApJ, 820, 40 Baines, E. K., White, R. J., Huber, D., et al. 2012, ApJ, 761, 57 Baraffe, I., Chabrier, G., Barman, T. S., Allard, F., & Hauschildt, P. H. 2003, A&A, 402, 701 Baraffe, I., Homeier, D., Allard, F., & Chabrier, G. 2015, A&A, 577, A42 Beaugé, C., Michtchenko, T. A., & Ferraz-Mello, S. 2006, MNRAS, 365, 1160 Bell, C. P. M., Mamajek, E. E., & Naylor, T. 2015, MNRAS, 454, 593 parametrized with 23 free orbital elements and masses compared to the previously proposed, strictly resonant periodic configu- ration described by 11 free parameters by GM2020. However, given that near-resonant, chaotic configurations may survive for hundreds of millions of years, it is still very difficult to dif- ferentiate such solutions from a strictly resonant configuration of the system. On the other hand, the well-defined PO N-body model with a minimal number of free parameters may be used as a template solution in numerical experiments and simula- tions requiring sufficiently long stable orbital evolution of the planets. These may include simulations of debris disks and yet undetected objects. Bergfors, C., Brandner, W., Janson, M., Köhler, R., & Henning, T. 2011, A&A, 528, A134 Beuzit, J. L., Vigan, A., Mouillet, D., et al. 2019, A&A, 631, A155 Biller, B. A., Apai, D., Bonnefoy, M., et al. 2021, MNRAS, 503, 743 Boccaletti, A., Abe, L., Baudrand, J., et al. 2008, SPIE Conf. Ser., 7015, 70151B Bohn, A. J., Kenworthy, M. A., Ginski, C., et al. 2020, ApJ, 898, L16 Bohn, A. J., Ginski, C., Kenworthy, M. A., et al. 2021, A&A, 648, A73 Booth, M., Jordan, A., Casassus, S., et al. 2016, MNRAS, 476, 2845 Boss A P 1997 Science 276 1836 Beuzit, J. L., Vigan, A., Mouillet, D., et al. 2019, A&A, 631, A155 Biller, B. A., Apai, D., Bonnefoy, M., et al. 2021, MNRAS, 503, 743 Bohn, A. J., Kenworthy, M. A., Ginski, C., et al. 2020, ApJ, 898, L16 Bohn, A. J., Ginski, C., Kenworthy, M. A., et al. References 2021, A&A, 648, A73 B h M J d A C S l 2016 MNRAS 476 2845 Boss, A. P. 1997, Science, 276, 1836 Brandt, G. M., Brandt, T. D., Dupuy, T. J., Michalik, D., & Marleau, G.-D. 2021, ApJ, 915, L16 Cameron, A. G. W. 1978, Moon and Planets, 18, 5 Chauvin, G., Lagrange, A.-M., Dumas, C., et al. 2004, A&A, 425, L29 As the system shows planets locked in MMRs, we ran numer- ical simulations to explore the possible migration histories which could have led to the present architecture of the system. Hydro- dynamical modeling suggests that the planets migrated inwards and that a different migration speed or different formation times drove the system into the present quasi-exact resonance.i Chen, C. H., Mittal, T., Kuchner, M., et al. 2014, ApJ, 211, 25 Cincotta, P. M., Giordano, C. M., & Simó, C. 2003, Physica D Nonlinear Phenomena, 182, 151 Claudi, R. U., Turatto, M., Gratton, R. G., et al. 2008, SPIE Conf. Ser., 7014, 70143E Crida, A., Sándor, Z., & Kley, W. 2008, A&A, 483, 325 Currie, T., Burrows, A., Itoh, Y., et al. 2011, ApJ, 729, 128 Currie, T., Fukagawa, M., Thalmann, C., Matsumura, S., & Plavchan, P. 2012, ApJ, 755, L34 We looked for the dynamical space where a putative fifth planet could sit using analytical formulae and estimates, and suggest that only a low-mass planet in the outer part of the system would agree with the currently accepted scenario. Its location would be in between planet HR 8799b and the exter- nal debris belt, and its mass would be too small (<0.1 MJup) to be detected by current instrumentation for high-contrast imag- ing. On the other hand, the interior belt is heavily shaped by p Currie, T., Burrows, A., Girard, J. H., et al. 2014, ApJ, 795, 133 Currie, T., Burrows, A., Girard, J. H., et al. 2014, D’Angelo, G., & Marzari, F. 2012, ApJ, 757, 50 De Rosa, R. J., Esposito, T. M., Gibbs, A., et al. 2020, SPIE Conf. Ser., 11447, 114475A Delisle, J.-B. 2017, A&A, 605, A96 Delorme, P., Meunier, N., Albert, D., et al. 2017, in SF2A-2017: Proceedings of the Annual meeting of the French Society of Astronomy and Astrophysics, eds. C. Reylé, P. Di Matteo, F. Herpin, et al. A133, page 22 of 25 y p Masset, F. 2000, A&AS, 141, 165 16 LESIA, Observatoire de Paris, Université PSL, CNRS, Sorbonne Université, Université de Paris, 5 place Jules Janssen, 92195 Meudon, France Masset, F., & Snellgrove, M. 2001, MNRAS, 320, L55 Matthews, B., Kennedy, G., Sibthorpe, B., et al. 2014, ApJ, 780, 97 Mizuno, H. 1980, Progr. Theor. Phys., 64, 544 Moorhead, A. V., & Adams, F. C. 2005, Icarus, 178, 517 17 SUPA, Institute for Astronomy, University of Edinburgh, Blackford Hill, Edinburgh EH9 3HJ, UK Nesvorny, D., & Ferraz-Mello, S. 1997, A&A, 320, 672 Nowak, M., Lacour, S., Lagrange, A. M., et al. 2020, A 18 School of Physical Sciences, Faculty of Science, Technology, Engi- neering and Mathematics, The Open University, Walton Hall, Milton Keynes MK7 6AA, UK Papaloizou, J. C. B. 2015, Int. J. Astrobiol., 14, 291 p Petit, C., Sauvage, J.-F., Fusco, T., et al. 2014, SPIE Conf. Ser., 9148, 0 Petit, C., Sauvage, J.-F., Fusco, T., et al. 2014, SPIE , , g , , , , , , , Pinna, E., Rossi, F., Puglisi, A., et al. 2019, in Adaptive Optics for Extremely g Pinna, E., Rossi, F., Puglisi, A., et al. 2019, in A y 19 Institute for Particle Physics and Astrophysics, ETH Zurich, Wolfgang-Pauli-Strasse 27, 8093 Zurich, Switzerland g Large Telescopes 6 – Conference Proceedings g p p y Large Telescopes 6 – Conference Proceedings Pollack, J. B., Hubickyj, O., Bodenheimer, P., et al. 1996, Icar Pollack, J. B., Hubickyj, O., Bodenheimer, P., et al. 1996, Icarus, 124, 62 Rameau, J., Chauvin, G., Lagrange, A.-M., et al. 2013, ApJ, 772, L15 20 Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Lagrange, 06304 Nice, France 21 Ramos, X. S., Charalambous, C., Benítez-Llambay, P., & Beaugé, C. 2017, A&A, 602, A101 21 European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748 Garching, Germany Rein, H., & Spiegel, D. S. 2015, MNRAS, 446, 1424 p g Seifert, W., Ageorges, N., Lehmitz, M., et al. 2010, in Proc. SPIE, Vol. 7735, Seifert, W., Ageorges, N., Lehmitz, M., et al. 2010, in Proc. SPIE, Vol. 7735, Ground-based and Airborne Instrumentation for Astronomy III, 77357W 22 LBT Observatory, University of Arizona, 933 N.Cherry Ave, Tucson AZ 85721, USA Seifert, W., Ageorges, N., Lehmitz, M., et al. 2010, in Proc. SPIE, Vol. 77 Ground-based and Airborne Instrumentation for Astronomy III, 77357W g g Ground-based and Airborne Instrumentation for Astronomy III, 77357W Sepulveda, A. G., & Bowler, B. P. , , , , 2010, Nature, 468, 1080 Marois, C., Correia, C., Véran, J.-P., & Currie, T. 2014, in IAU Symposium, 299, IAU Symposium, eds. M. Booth, B. C. Matthews, & J. R. Graham, 48 s, C., Correia, C., Véran, J.-P., & Currie, T. 2014, in IAU Symposiu 15 European Space Agency (ESA), ESA Office, Space Telescope Sci- ence Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA 16 Marois, C., Correia, C., Véran, J. P., & Currie, T. 2014, in IAU Symposium IAU Symposium, eds. M. Booth, B. C. Matthews, & J. R. Graham, 48 y p U Symposium, eds. M. Booth, B. C. Matthews, & J. R. Graham, 48 y p Masset, F. 2000, A&AS, 141, 165 Szuszkiewicz, E., & Podlewska-Gaca, E. 2012, Origins Life Evol. Biosphere, 42, 113 Thommes, E. W. 2005, ApJ, 626, 1033 Vigan, A., Fontanive, C., Meyer, M., et al. 2021, A&A, 651, A72 Wahhaj, Z., Milli, J., Romero, C., et al. 2021, A&A, 648, A26 Wang, J. J., Graham, J. R., Dawson, R., et al. 2018, AJ, 156, 192 Wertz, O., Absil, O., Gómez González, C. A., et al. 2017, A&A, 598, A83 Wilner, D. J., MacGregor, M. A., Andrews, S. M., et al. 2018, ApJ, 855, 56 Wright, J. T., Veras, D., Ford, E. B., et al. 2011, ApJ, 730, 93 Zuckerman, B., Rhee, J. H., Song, I., & Bessell, M. S. 2011, ApJ, 732, 61 Zurlo, A., Vigan, A., Mesa, D., et al. 2014, A&A, 572, A85 Zurlo, A., Vigan, A., Galicher, R., et al. 2016, A&A, 587, A57 113 Dohlen, K., Langlois, M., Saisse, M., et al. 2008, SPIE Conf. Ser., 7014, 70143L Thommes, E. W. 2005, ApJ, 626, 1033 osito, S., Riccardi, A., Fini, L., et al. 2010, SPIE Conf. Ser., 7736, 1 p Esposito, S., Mesa, D., Skemer, A., et al. 2013, A&A, 549, A52 Fabrycky, D. C., & Murray-Clay, R. A. 2010, ApJ, 710, 1408 Faramaz, V., Marino, S., Booth, M., et al. 2021, AJ, 161, 271 Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman, J. 2013, PASP, 125, 306 Fukagawa, M., Itoh, Y., Tamura, M., et al. 2009, ApJ, 696, L1 Fusco, T., Rousset, G., Sauvage, J.-F., et al. 2006, Opt. Express, 14, 7515 Fusco, T., Rousset, G., Sauvage, J.-F., et al. 2006, Opt. Express, 14, 7515 , , , , g , , , p p , , Gaia Collaboration 2020, VizieR Online Data Catalog: I/350 Gaia Collaboration 2020, VizieR Online Data Catalog: I/350 Galicher, R., Marois, C., Macintosh, B., Barman, T., & Konopacky, Q. 2011, ApJ, 739, L41 Galicher, R., Marois, C., Macintosh, B., Barman, T., & Konopacky, Q. 2011, ApJ, 739, L41 p Galicher, R., Boccaletti, A., Mesa, D., et al. 2018, A&A, 615, A92 1 Núcleo de Astronomía, Facultad de Ingeniería, Universidad Diego Portales, Av. Ejercito 441, Santiago, Chile e-mail: [email protected] 1 Núcleo de Astronomía, Facultad de Ingeniería, Universidad Diego Portales, Av. Ejercito 441, Santiago, Chile e-mail: [email protected] Go´zdziewski, K., & Migaszewski, C. 2018, ApJS, 238, 6 Go´zdziewski, K., & Migaszewski, C. 2020, ApJ, 902, L40 Go´zdziewski, K., Bois, E., Maciejewski, A. J., & Kiseleva-Eggleton, L. 2001, A&A, 378, 569 Go´zdziewski, K., Bois, E., Maciejewski, A. J., & Kiseleva-Eggleton, L. 2001, A&A, 378, 569 p 2 Millennium Nucleus on Young Exoplanets and their Moons (YEMS), Universidad Diego Portales, Av. Ejercito 441, Santiago, Chile GRAVITY Collaboration (Lacour, S., et al.) 2019, A&A, 623, A11 ( ) Gray, R. O., & Kaye, A. B. 1999, AJ, 118, 2993 Gray, R. O., & Kaye, A. B. 1999, AJ, 118, 2993 Hadjidemetriou, J. D. 1977, Celest. Mech., 16, 61 3 Aix-Marseille Université, CNRS, LAM (Laboratoire d’Astrophysique de Marseille) UMR 7326, 13388 Marseille, France Haffert, S. Y., Bohn, A. J., de Boer, J., et al. 2019, Haffert, S. Y., Bohn, A. J., de Boer, J., et al. 2019, Nat. Astron., 3, 749 Hairer, E., Nørsett, S., & Wanner, G. 2000, Solving Ordinary Differential Hairer, E., Nørsett, S., & Wanner, G. J. 1993, Celest. Mech. Dyn. Astron., 56, 191 Laskar, J. 1993, Celest. Mech. Dyn. Astron., 56, 191 9 INAF – Osservatorio Astrofisico di Arcetri, L.go E. Fermi 5, 50125 Firenze, Italy y Lazzoni, C., Desidera, S., Marzari, F., et al. 2018, A&A, 611, A43 Lazzoni, C., Desidera, S., Marzari, F., et al. 2018, Lee, M. H., & Peale, S. J. 2002, ApJ, 567, 596 10 IPAG, Univ. Grenoble Alpes, CNRS, IPAG, 38000 Grenoble, Franc Lindegren, L., Bastian, U., Biermann, M., et al. 2021, A&A, 649, A4 , p , , , , 11 Space Telescope Science Institute (STScI), 3700 San Martin Dr, Baltimore, MD 21218, USA Maire, A. L., Skemer, A. J., Hinz, P. M., et al. 2015, A&A, 579, A2 Maire, A.-L., Langlois, M., Dohlen, K., et al. 2016, SPIE Conf. Ser., 9908, 990834 12 Geneva Observatory, University of Geneva, 51 ch. Pegasi, 1290 Versoix, Switzerland Marois, C., Lafrenière, D., Doyon, R., Macintosh, B., & Nadeau, D. 2006, ApJ, 641, 556 13 Max Planck Institute for Astronomy, Königstuhl 17, 69117, Heidel- berg, Germany Marois, C., Macintosh, B., Barman, T., et al. 2008, Science, 322, 1348 Marois, C., Zuckerman, B., Konopacky, Q. M., Macintosh, B., & Barman, T. 2010, Nature, 468, 1080 14 Department of Astronomy, Stockholm University, 106 91 Stockholm, Sweden 15i 113 2000, Solving Ordinary Differential Equations I Nonstiff Problems, 2nd edn. (Berlin: Springer) Equations I Nonstiff Problems, 2nd edn. (Berlin: Springer) 4 Institute of Astronomy, Faculty of Physics, Astronomy and Infor- matics, Nicolaus Copernicus University, Grudziadzka 5, 87-100 Toru´n, Poland 5 ands, T. O., Alexander, R. D., & Dehnen, W. 2014, MNRAS, 445, Hinz, P. M., Rodigas, T. J., Kenworthy, M. A., et al. 2010, ApJ, 716, 417 g y p Hughes, A. M., Wilner, D. J., Andrews, S. M., et al. 2011, ApJ, 740, 38 5 University of Exeter, Astrophysics Group, Physics Building, Stocker Road, Exeter EX4 4QL, UK Keppler, M., Benisty, M., Müller, A., et al. 2018, A&A, 617, A4 Kervella, P., Arenou, F., & Thévenin, F. 2022, A&A, 657, A7 6 INAF – Osservatorio Astronomico di Padova, Vicolo dell’Osservatorio 5, 35122 Padova, Italy Konopacky, Q. M., Marois, C., Macintosh, B. A., et al. 2016, AJ, 152, 28 Lafrenière, D., Marois, C., Doyon, R., & Barman, T. 2009, ApJ, Lagrange, A. M., Bonnefoy, M., Chauvin, G., et al. 2010, Science, 329, 57 y 7 Department of Physics and Astronomy, University of Padova, Via Marzolo 8, 35131 Padova, Italy g g y Langlois, M., Vigan, A., Moutou, C., et al. 2013, in Proceedings of the Third AO4ELT Conference ed S Esposito & L Fini 63 Langlois, M., Vigan, A., Moutou, C., et al. 2013, in Proceedings of the Third Langlois, M., Vigan, A., Moutou, C., et al. 2013, in Proceedings of the Third AO4ELT Conference, ed. S. Esposito & L. Fini, 63 g g AO4ELT Conference, ed. S. Esposito & L. Fini, 63 , , y 8 CRAL, UMR 5574, CNRS, Université de Lyon, ENS, 9 avenue Charles André, 69561 Saint-Genis-Laval Cedex, Francei Langlois, M., Gratton, R., Lagrange, A. M., et al. 2021 Langlois, M., Gratton, R., Lagrange, A. M., et al. 2021, A&A, 651, A71 g g g Laskar, J. 1993, Celest. Mech. Dyn. Astron., 56, 191 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Desidera, S., Chauvin, G., Bonavita, M., et al. 2021, A&A, 651, A Szuszkiewicz, E., & Podlewska-Gaca, E. 2012, Origins Life Evol. Biosphere, 42, 113 Thommes, E. W. 2005, ApJ, 626, 1033 Vigan, A., Fontanive, C., Meyer, M., et al. 2021, A&A, 651, A72 Wahhaj, Z., Milli, J., Romero, C., et al. 2021, A&A, 648, A26 Wang, J. J., Graham, J. R., Dawson, R., et al. 2018, AJ, 156, 192 Wertz, O., Absil, O., Gómez González, C. A., et al. 2017, A&A, 598, A83 Wilner, D. J., MacGregor, M. A., Andrews, S. M., et al. 2018, ApJ, 855, 56 Wright, J. T., Veras, D., Ford, E. B., et al. 2011, ApJ, 730, 93 Zuckerman, B., Rhee, J. H., Song, I., & Bessell, M. S. 2011, ApJ, 732, 61 Zurlo, A., Vigan, A., Mesa, D., et al. 2014, A&A, 572, A85 Zurlo, A., Vigan, A., Galicher, R., et al. 2016, A&A, 587, A57 Szuszkiewicz, E., & Podlewska-Gaca, E. 2012, Origins Life Evol. Biosphere, 42, 113 Szuszkiewicz, E., & Podlewska-Gaca, E. 2012, Origins Life Evol. Biosphere, 42, 113 Thommes, E. W. 2005, ApJ, 626, 1033 Vigan, A., Fontanive, C., Meyer, M., et al. 2021, A&A, 651, A72 Wahhaj, Z., Milli, J., Romero, C., et al. 2021, A&A, 648, A26 Wang, J. J., Graham, J. R., Dawson, R., et al. 2018, AJ, 156, 192 Wertz, O., Absil, O., Gómez González, C. A., et al. 2017, A&A, 598, A83 Wilner, D. J., MacGregor, M. A., Andrews, S. M., et al. 2018, ApJ, 855, 56 Wright, J. T., Veras, D., Ford, E. B., et al. 2011, ApJ, 730, 93 Zuckerman, B., Rhee, J. H., Song, I., & Bessell, M. S. 2011, ApJ, 732, 61 Zurlo, A., Vigan, A., Mesa, D., et al. 2014, A&A, 572, A85 Zurlo, A., Vigan, A., Galicher, R., et al. 2016, A&A, 587, A57 y p Masset, F. 2000, A&AS, 141, 165 (2015) 2013.88 – – – – -537 1 782 2 -370 1 -539 1 -381 2 -30 0 Wang et al. (2018)a 2014.53 – – – – – – – – -400 4 -512 4 -389 1 -22 2 IFS, This work 2014.53 1570 3 704 3 -521 3 790 9 -391 2 -530 2 -387 2 -10 3 IRDIS, This work 2014.54 1560 13 725 13 -540 13 799 13 -400 11 -534 11 -387 11 3 11 Konopacky et al. (2016) 2014.62 – – – – – – – – -396 1 -524 2 -389 1 -17 2 IFS, This work 2014.70 1569 4 707 2 -519 1 794 2 -397 1 -530 2 – – – – Wang et al. (2018)1 2014.93 1575 2 702 4 -511 2 799 2 -400 2 -523 2 -385 3 12 2 Wertz et al. (2017) 2014.93 1574 3 701 2 -514 3 798 4 -399 4 -525 4 -389 8 11 4 IRDIS, This work Appendix A: Astrometric points used in the analysis Table A.1: List of the astrometric points. Date Planet b Planet c Planet d Planet e ∆RA σ∆RA ∆Dec σ∆Dec ∆RA σ∆RA ∆Dec σ∆Dec ∆RA σ∆RA ∆Dec σ∆Dec ∆RA σ∆RA ∆Dec σ∆Dec Ref. 1998.83 1411 9 986 9 – – – – – – – – – – – – Lafrenière et al. (2009) 1998.83 1418 22 1004 20 -837 26 483 23 133 35 -533 34 – – – – Soummer et al. (2011) 2002.54 1481 23 919 17 – – – – – – – – – – – – Fukagawa et al. (2009) 2004.53 1471 6 884 6 -739 6 612 6 – – – – – – – – Konopacky et al. (2016) 2005.54 1496 5 856 5 -713 5 630 5 -87 10 -578 10 – – – – Currie et al. (2012) 2007.58 1504 3 837 3 -683 4 671 4 -179 5 -588 5 – – – – Konopacky et al. (2016) 2007.81 1500 7 836 7 -678 7 676 7 -175 10 -589 10 – – – – Konopacky et al. (2016) 2008.52 1527 4 799 4 -658 4 701 4 -208 4 -582 4 – – – – Marois et al. (2008) 2008.61 1527 2 801 2 -657 2 706 2 -216 2 -582 2 – – – – Marois et al. y p Masset, F. 2000, A&AS, 141, 165 (2008) 2008.71 1516 4 818 4 -663 3 693 3 -202 4 -588 4 – – – – Konopacky et al. (2016) 2008.89 1532 20 796 20 -654 20 700 20 -217 20 -608 20 – – – – Hinz et al. (2010) 2009.02 – – – – -612 30 665 30 – – – – – – – – Hinz et al. (2010) 2009.58 1526 4 797 4 -639 4 712 4 -237 3 -577 3 -306 7 -211 7 Konopacky et al. (2016) 2009.62 1536 10 785 10 – – – – – – – – – – – – Currie et al. (2011) 2009.70 1538 30 777 30 -634 30 697 30 -282 30 -590 30 – – – – Hinz et al. (2010) 2009.76 1535 20 816 20 -636 40 692 40 -270 70 -600 70 – – – – Bergfors et al. (2011) 2009.77 1532 7 783 7 -627 7 716 7 -241 7 -586 7 -306 7 -217 7 Currie et al. (2011) 2009.83 1524 10 795 10 -636 9 720 9 -251 7 -573 7 -310 9 -187 9 Konopacky et al. (2016) 2009.84 1540 19 800 19 -630 13 720 13 -240 14 -580 14 -304 10 -196 10 Galicher et al. (2011) 2010.53 1532 5 783 5 -619 4 728 4 -265 4 -576 4 -323 6 -166 6 Konopacky et al. (2016) 2010.55 1547 6 757 9 -606 6 725 6 -269 6 -580 6 -329 6 -178 6 Currie et al. (2014) 2010.83 1535 15 766 15 -607 12 744 12 -296 13 -561 13 -341 16 -143 16 Konopacky et al. (2016) 2011.55 1541 5 762 5 -595 4 747 4 -303 5 -562 5 -352 8 -130 8 Konopacky et al. (2016) 2011.79 1579 11 734 11 -561 13 752 13 -299 13 -563 13 -326 13 -119 13 Esposito et al. (2013) 2011.86 1546 11 725 11 -578 13 767 13 -320 13 -549 13 -382 16 -127 16 Esposito et al. (2013) 2012.55 1545 5 747 5 -578 5 761 5 -339 5 -555 5 -373 8 -84 8 Konopacky et al. (2016) 2012.82 1549 4 743 4 -572 3 768 3 -346 4 -548 4 -370 9 -76 9 Konopacky et al. (2016) 2012.83 1558 6 729 9 -557 6 763 6 -343 6 -555 6 -371 6 -80 6 Currie et al. y p Masset, F. 2000, A&AS, 141, 165 2022, AJ, 163, 52 23 George Mason University, Department of Physics & Astronomy, MS 3F3, 4400 University Drive, Fairfax, VA 22030, USA Soummer, R., Hagan, J. B., Pueyo, L., et al. 2011, ApJ, 741, 55 g y p Soummer, R., Pueyo, L., & Larkin, J. 2012, ApJ, 755, L28 F3, 4400 University Drive, Fairfax, VA 22030, USA Soummer, R., Pueyo, L., & Larkin, J. 2012, ApJ, 755, L28 24 European Southern Observatory, Alonso de Cordova 3107, Casilla 19001 Vitacura, Santiago 19, Chile Su, K. Y. L., Rieke, G. H., Stapelfeldt, K. R., et al. 2009, ApJ, 705, 314 Sudol, J. J., & Haghighipour, N. 2012, ApJ, 755, 38 A133, page 23 of 25 A&A 666, A133 (2022) Appendix A: Astrometric points used in the analysis Table A.1: List of the astrometric points. Date Planet b Planet c Planet d Planet e ∆RA σ∆RA ∆Dec σ∆Dec ∆RA σ∆RA ∆Dec σ∆Dec ∆RA σ∆RA ∆Dec σ∆Dec ∆RA σ∆RA ∆Dec σ∆Dec Ref. 1998.83 1411 9 986 9 – – – – – – – – – – – – Lafrenière et al. (2009) 1998.83 1418 22 1004 20 -837 26 483 23 133 35 -533 34 – – – – Soummer et al. (2011) 2002.54 1481 23 919 17 – – – – – – – – – – – – Fukagawa et al. (2009) 2004.53 1471 6 884 6 -739 6 612 6 – – – – – – – – Konopacky et al. (2016) 2005.54 1496 5 856 5 -713 5 630 5 -87 10 -578 10 – – – – Currie et al. (2012) 2007.58 1504 3 837 3 -683 4 671 4 -179 5 -588 5 – – – – Konopacky et al. (2016) 2007.81 1500 7 836 7 -678 7 676 7 -175 10 -589 10 – – – – Konopacky et al. (2016) 2008.52 1527 4 799 4 -658 4 701 4 -208 4 -582 4 – – – – Marois et al. (2008) 2008.61 1527 2 801 2 -657 2 706 2 -216 2 -582 2 – – – – Marois et al. (2008) 2008.71 1516 4 818 4 -663 3 693 3 -202 4 -588 4 – – – – Konopacky et al. (2016) 2008.89 1532 20 796 20 -654 20 700 20 -217 20 -608 20 – – – – Hinz et al. y p Masset, F. 2000, A&AS, 141, 165 (2010) 2009.02 – – – – -612 30 665 30 – – – – – – – – Hinz et al. (2010) 2009.58 1526 4 797 4 -639 4 712 4 -237 3 -577 3 -306 7 -211 7 Konopacky et al. (2016) 2009.62 1536 10 785 10 – – – – – – – – – – – – Currie et al. (2011) 2009.70 1538 30 777 30 -634 30 697 30 -282 30 -590 30 – – – – Hinz et al. (2010) 2009.76 1535 20 816 20 -636 40 692 40 -270 70 -600 70 – – – – Bergfors et al. (2011) 2009.77 1532 7 783 7 -627 7 716 7 -241 7 -586 7 -306 7 -217 7 Currie et al. (2011) 2009.83 1524 10 795 10 -636 9 720 9 -251 7 -573 7 -310 9 -187 9 Konopacky et al. (2016) 2009.84 1540 19 800 19 -630 13 720 13 -240 14 -580 14 -304 10 -196 10 Galicher et al. (2011) 2010.53 1532 5 783 5 -619 4 728 4 -265 4 -576 4 -323 6 -166 6 Konopacky et al. (2016) 2010.55 1547 6 757 9 -606 6 725 6 -269 6 -580 6 -329 6 -178 6 Currie et al. (2014) 2010.83 1535 15 766 15 -607 12 744 12 -296 13 -561 13 -341 16 -143 16 Konopacky et al. (2016) 2011.55 1541 5 762 5 -595 4 747 4 -303 5 -562 5 -352 8 -130 8 Konopacky et al. (2016) 2011.79 1579 11 734 11 -561 13 752 13 -299 13 -563 13 -326 13 -119 13 Esposito et al. (2013) 2011.86 1546 11 725 11 -578 13 767 13 -320 13 -549 13 -382 16 -127 16 Esposito et al. (2013) 2012.55 1545 5 747 5 -578 5 761 5 -339 5 -555 5 -373 8 -84 8 Konopacky et al. (2016) 2012.82 1549 4 743 4 -572 3 768 3 -346 4 -548 4 -370 9 -76 9 Konopacky et al. (2016) 2012.83 1558 6 729 9 -557 6 763 6 -343 6 -555 6 -371 6 -80 6 Currie et al. (2014) 2013.79 1545 22 724 22 -542 22 784 22 -382 16 -522 16 -373 13 -17 13 Konopacky et al. (2016) 2013.81 1562 8 713 13 -538 6 784 13 -377 7 -538 11 -394 11 -36 17 Maire et al. y p Masset, F. 2000, A&AS, 141, 165 (2014) 2013.79 1545 22 724 22 -542 22 784 22 -382 16 -522 16 -373 13 -17 13 Konopacky et al. (2016) 2013.81 1562 8 713 13 -538 6 784 13 -377 7 -538 11 -394 11 -36 17 Maire et al. (2015) 2013.88 – – – – -537 1 782 2 -370 1 -539 1 -381 2 -30 0 Wang et al. (2018)a 2014.53 – – – – – – – – -400 4 -512 4 -389 1 -22 2 IFS, This work 2014.53 1570 3 704 3 -521 3 790 9 -391 2 -530 2 -387 2 -10 3 IRDIS, This work 2014.54 1560 13 725 13 -540 13 799 13 -400 11 -534 11 -387 11 3 11 Konopacky et al. (2016) 2014.62 – – – – – – – – -396 1 -524 2 -389 1 -17 2 IFS, This work 2014.70 1569 4 707 2 -519 1 794 2 -397 1 -530 2 – – – – Wang et al. (2018)1 2014.93 1575 2 702 4 -511 2 799 2 -400 2 -523 2 -385 3 12 2 Wertz et al. (2017) 2014.93 1574 3 701 2 -514 3 798 4 -399 4 -525 4 -389 8 11 4 IRDIS, This work 5 A133, page 24 of 25 A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. y p Masset, F. 2000, A&AS, 141, 165 ( ) 2014.93 1574 4 701 3 -512 3 798 4 -400 4 -523 4 -390 7 12 4 IRDIS, This work 2014.93 1573 3 701 3 -512 3 797 4 -403 4 -524 4 -383 8 11 4 IRDIS, This work 2015.51 – – – – – – – – -424 4 -509 3 -391 1 33 2 IFS, This work 2015.51 1579 1 694 1 -498 1 806 1 -417 1 -517 1 -383 9 33 5 IRDIS, This work 2015.58 1580 5 689 3 -495 2 806 2 -419 2 -516 1 -386 1 36 1 IRDIS, This work 2015.65 1569 11 666 7 -482 5 813 6 -436 11 -510 12 – – – – Sepulveda & Bowler (2022) 2015.74 – – – – – – – – -420 4 -513 4 -392 1 39 3 IFS, This work 2015.74 1580 1 688 1 -494 1 811 1 -426 1 -512 1 -382 9 50 5 IRDIS, This work 2016.72 – – – – -466 1 821 2 -453 1 -499 1 -376 2 81 0 Wang et al. y p Masset, F. 2000, A&AS, 141, 165 (2018)1 2016.88 – – – – – – – – -464 1 -486 2 -382 2 94 6 IFS, This work 2016.88 1589 2 666 1 -464 2 824 2 -454 2 -489 2 -378 4 90 2 IRDIS, This work 2017.45 – – – – – – – – -473 2 -476 2 -377 1 115 3 IFS, This work 2017.45 1591 1 653 1 -449 1 835 1 -472 2 -482 2 -373 3 118 2 IRDIS, This work 2017.78 – – – – – – – – -480 2 -478 2 -372 1 129 2 IFS, This work 2017.78 1595 1 647 1 -441 1 839 1 -480 1 -477 1 -369 1 128 1 IRDIS, This work 2017.78 – – – – – – – – -492 5 -463 6 -370 2 135 3 IFS, This work 2017.78 1595 1 647 1 -441 1 839 1 -480 1 -477 1 -371 2 128 2 IRDIS, This work 2018.46 – – – – – – – – -495 1 -460 2 -360 2 162 2 IFS, This work 2018.46 1601 1 635 1 -424 1 848 1 -497 2 -463 2 -358 2 156 2 IRDIS, This work 2018.63 – – – – – – – – -509 2 -452 3 -361 2 166 2 IFS, This work 2018.63 1601 2 632 3 -421 1 850 1 -502 1 -461 1 -357 1 162 2 IRDIS, This work 2018.63 – – – – – – – – -503 2 -456 2 -359 1 167 2 IFS, This work 2018.63 1600 1 632 1 -421 1 851 1 -502 2 -458 1 -358 2 163 1 IRDIS, This work 2018.66 – – – – – – – – – – – – -358 0 163 0 GRAVITY Collaboration (2019) 2019.83 – - – – – – – – -527 3 -432 3 -337 3 210 3 IFS, This work 2019.83 1606 2 615 2 -392 2 875 2 -532 3 -425 2 -338 2 215 2 IRDIS, This work 2019.84 – – – – – – – – -528 1 -435 2 -337 2 208 2 IFS, This work 2019.84 1611 1 611 1 -388 2 870 2 -530 2 -430 1 -335 1 210 1 IRDIS, This work 2020.75 1620 1 591 3 -364 2 883 1 -551 1 -415 4 -315 3 242 5 LUCI, This work 2021.64 – – – – – – – – -569 1 -390 2 -292 2 276 2 IFS, This work 2021.64 1626 1 578 2 -339 2 890 2 -563 2 -391 2 -287 4 272 2 IRDIS, This work Table A.1: (continued) 2014.93 1574 4 701 3 -512 3 798 4 -400 4 -523 4 -390 7 12 4 IRDIS, This work 2014.93 1573 3 701 3 -512 3 797 4 -403 4 -524 4 -383 8 11 4 IRDIS, This work 2015.51 – – – – – – – – -424 4 -509 3 -391 1 33 2 IFS, This work 2015.51 1579 1 694 1 -498 1 806 1 -417 1 -517 1 -383 9 33 5 IRDIS, This work 2015.58 1580 5 689 3 -495 2 806 2 -419 2 -516 1 -386 1 36 1 IRDIS, This work 2015.65 1569 11 666 7 -482 5 813 6 -436 11 -510 12 – – – – Sepulveda & Bowler (2022) 2015.74 – – – – – – – – -420 4 -513 4 -392 1 39 3 IFS, This work 2015.74 1580 1 688 1 -494 1 811 1 -426 1 -512 1 -382 9 50 5 IRDIS, This work 2016.72 – – – – -466 1 821 2 -453 1 -499 1 -376 2 81 0 Wang et al. y p Masset, F. 2000, A&AS, 141, 165 (2018)1 2016.88 – – – – – – – – -464 1 -486 2 -382 2 94 6 IFS, This work 2016.88 1589 2 666 1 -464 2 824 2 -454 2 -489 2 -378 4 90 2 IRDIS, This work 2017.45 – – – – – – – – -473 2 -476 2 -377 1 115 3 IFS, This work 2017.45 1591 1 653 1 -449 1 835 1 -472 2 -482 2 -373 3 118 2 IRDIS, This work 2017.78 – – – – – – – – -480 2 -478 2 -372 1 129 2 IFS, This work 2017.78 1595 1 647 1 -441 1 839 1 -480 1 -477 1 -369 1 128 1 IRDIS, This work 2017.78 – – – – – – – – -492 5 -463 6 -370 2 135 3 IFS, This work 2017.78 1595 1 647 1 -441 1 839 1 -480 1 -477 1 -371 2 128 2 IRDIS, This work 2018.46 – – – – – – – – -495 1 -460 2 -360 2 162 2 IFS, This work 2018.46 1601 1 635 1 -424 1 848 1 -497 2 -463 2 -358 2 156 2 IRDIS, This work 2018.63 – – – – – – – – -509 2 -452 3 -361 2 166 2 IFS, This work 2018.63 1601 2 632 3 -421 1 850 1 -502 1 -461 1 -357 1 162 2 IRDIS, This work 2018.63 – – – – – – – – -503 2 -456 2 -359 1 167 2 IFS, This work 2018.63 1600 1 632 1 -421 1 851 1 -502 2 -458 1 -358 2 163 1 IRDIS, This work 2018.66 – – – – – – – – – – – – -358 0 163 0 GRAVITY Collaboration (2019) 2019.83 – - – – – – – – -527 3 -432 3 -337 3 210 3 IFS, This work 2019.83 1606 2 615 2 -392 2 875 2 -532 3 -425 2 -338 2 215 2 IRDIS, This work 2019.84 – – – – – – – – -528 1 -435 2 -337 2 208 2 IFS, This work 2019.84 1611 1 611 1 -388 2 870 2 -530 2 -430 1 -335 1 210 1 IRDIS, This work 2020.75 1620 1 591 3 -364 2 883 1 -551 1 -415 4 -315 3 242 5 LUCI, This work 2021.64 – – – – – – – – -569 1 -390 2 -292 2 276 2 IFS, This work 2021.64 1626 1 578 2 -339 2 890 2 -563 2 -391 2 -287 4 272 2 IRDIS, This work A. Zurlo et al.: Dynamical analysis of HR 8799 from SPHERE and LUCI astrometry. Table A.1: (continued) A133, page 25 of 25

https://openalex.org/W2512634839

https://www.arkat-usa.org/get-file/19944/

English

Synthesis and natural occurrence of (Z/E)-β-and γ-curcumen-12-ol

ARKIVOC

cc-by

Abstract: (Z/E)-β-Curcumen-12-ol (Z/E)-(1) was synthesized via Birch reduction of acid 6 starting from α- curcumene (5). An olefin isomerization of 1 is the key step in the synthesis of (Z/E)-γ-curcumen- 12-ol (Z/E)-(2). Sesquiterpene alcohol (E)-1 was found for the first time in nature as a minor constituent of different Santalum species by using the synthetic sample as reference. Keywords: β-/γ-Curcumen-12-ol, monocyclic sesquiterpenes, isomerization, Wittig olefination Norbert A. Brauna and Dietrich Spitznerb,* aSymrise Pte. Ltd., Scent & Care, Asia Pacific, 226 Pandan Loop, Singapore 128412, Singapore bUniversität Stuttgart, Institut für Organische Chemie, Pfaffenwaldring 55, D-70569 Stuttgart, Germany E-mail: [email protected] E-mail: [email protected] E-mail: [email protected] Dedicated to Professor Atta-ur-Rahman on the occasion of his 65th birthday Dedicated to Professor Atta-ur-Rahman on the occasion of his 65th birthday ARKIVOC 2007 (vii) 273-279 ARKIVOC 2007 (vii) 273-279 Issue in Honor of Prof. Atta-ur-Rahman Introduction β- and γ-Curcumen-12-ol (1)/(2) are rare sesquiterpene alcohols, probably due to their highly reactive diene ring systems and may be considered therefore as biogenetic precursors for more complex sesquiterpene skeletons, e.g. tricyclic helifolenals 3 or italicenols 4.1 Although, the enzymes involved have not been identified, it can be speculated that either a step-wise ionic mechanism or a concerted intramolecular cycloaddition is responsible for the formation of such terpenes.1,2 ISSN 1424-6376 ©ARKAT USA, Inc. Page 273 Page 273 Issue in Honor of Prof. Atta-ur-Rahman ARKIVOC 2007 (vii) 273-279 R R1 R R1 3 R=CHO, R1=Me R R1 R R1 1 (Z)-1/2: R=Me, R 1=CH2OH (E)-1/2: R=CH 2OH, R1=Me 2 4 R=CH2OH, R1=Me Figure 1. Structures of β- and γ-curcumen-12-ol (1)/(2), helifolenals 3 and italicenols 4. R R1 2 R R1 1 1 (Z)-1/2: R=Me, R 1=CH2OH (E)-1/2: R=CH 2OH, R1=Me 2 2 3 R=CHO, R1=Me R R1 R R1 4 R=CH2OH, R1=Me 3 R=CHO, R1=Me 4 R=CH2OH, R1=Me gure 1. Structures of β- and γ-curcumen-12-ol (1)/(2), helifolenals 3 and italicenols (Z)- and (E)-γ-Curcumen-12-ole (Z/E)-(2) were first characterized by Weyerstahl et al. in the essential oil of Pulicaria gnaphalodes (Vent.) Boiss.1,3 Isomer (E)-2 is also a constituent in Brazilian lantana oil (Lantana camara L.).4 The isomeric 1,4-diene (Z)-β-curcumen-12-ol (Z)-(1) was recently isolated for the first time in Western Australian sandalwood oil (Santalum spicatum (R.Br.) A. DC.) together with the corresponding 1,3-diene (Z)-2.5 Both sesquiterpene alcohols (Z)-1 and (Z)-2 were also detected in East Indian (S. album L.)6 and New Caledonian sandalwood oil (S. austrocaledonicum Vieill. var. austrocaledonicum)7. However, it is interesting to note that in contrast to 2, isomer (E)-1 has not been found in nature and neither a synthesis of terpene alcohol 1 nor of 2 has been described in literature so far. Here, we report a short synthesis of β-and γ-curcumenol (1)/(2) and the identification of (E)-1 in various sandalwood oils. Results and Discussion The synthesis of (Z/E)-β-curcumen-12-ol (Z/E)-(1) commenced from acid 6, which was obtained from α-curcumene = ar-curcumene (5), a main constituent of curcuma oil (Curcuma xanthorrhiza Roxb., Zingiberaceae), following a procedure of Weinreb et al.8 ©ARKAT USA, Inc. Page 274 ISSN 1424-6376 Page 274 ARKIVOC 2007 (vii) 273-279 Issue in Honor of Prof. Atta-ur-Rahman CO2H MeO2C b) d) 10 c) HOH2C 5 6 7 (X=OH) 8 (X=NMeOMe) (E)-1 9 (X=H) (E)-2 a) f) g) X O HOH2C e) 1 6 10 11 13 14 15 Scheme 1. (a) cat. OsO4, Jones reagent, 84% 8b; (b) Li, liquid NH3, 92%; (c) BOP, NEt3, NHMeOMe x HCl, DMF, 96%; (d) LiAlH4, Et2O; (e) Ph3P+CHMeCO2MeBr–, NEt3, CH2Cl2, 53% (over 2 steps); (f) LiAlH4, Et2O, 83%; (g) KOtBu-toluene, microwave, 100°C, 64%. 6 5 HOH2C (E)-1 (E)-2 g) HOH2C 1 6 10 11 13 14 15 10 Scheme 1. (a) cat. OsO4, Jones reagent, 84% 8b; (b) Li, liquid NH3, 92%; (c) BOP, NEt3, NHMeOMe x HCl, DMF, 96%; (d) LiAlH4, Et2O; (e) Ph3P+CHMeCO2MeBr–, NEt3, CH2Cl2, 53% (over 2 steps); (f) LiAlH4, Et2O, 83%; (g) KOtBu-toluene, microwave, 100°C, 64%. Benzene derivative 6 was reduced under Birch conditions (Li in liquid NH3/THF)9 to form 1,4-cyclohexadiene 7 in 92% yield. Acid 7 was then transformed into the corresponding Weinreb amide 8 (96%)10 using BOP in DMF as coupling agent11. Reduction of 8 with LiAlH4 in Et2O gave aldehyde 9, which was reacted without further purification with phosphonium salt [Ph3P+CHMeCO2MeBr–] to obtain α,β-unsaturated ester (E)-10 (53% for two steps, ratio: E/Z = 95:5). Consecutive reduction of ester (E)-12 with LiAlH4 yielded 83% of alcohol (E)-1. Its (E)- geometry at the C-10/C-11 π bond was confirmed by 1H-NMR: irradiation at δ = 5.4 ppm (olefinic H’s) gave a NOE signal at δ = 4.0 ppm (CH2OH). The corresponding (Z)-1 can be analogously prepared from aldehyde 9 by a (Z)-selective Wittig type olefination.12 Synthetic alcohol (Z)-1 was identical in all respects (1H, 13C-NMR, GC-MS, RI) to the natural product (Z)-1 described by Braun et al. from Western Australian sandalwood oil.5 Enantiomerically pure β-curcumenol (1) may be prepared starting from enantiopure aldehyde 9 following the protocol of Ogasawara et al.13 With the synthetic sample of (E)-1 in hand, we reinvestigated various commercially available sandalwood oils and were able to detect this so far unknown isomer for the first time as a new minor constituent in nature. Results and Discussion GC-MS and retention indices (RI) on two different columns [RI = 2513 (DB-Wax) and RI = 1745 (DB-1)] of synthetic and natural (E)-1 fully matched. However, in all three Santalum species (Z)-15-7 is the predominant isomer compared to (E)-1: S. spicatum (7.2%/1.6%), S. album (2.0%/1.4%) and in S. austrocaledonicum Vieill. var. austrocaledonicum (1.1%/0.12%). ©ARKAT USA, Inc. Page 275 Page 275 ISSN 1424-6376 Issue in Honor of Prof. Atta-ur-Rahman ARKIVOC 2007 (vii) 273-279 The key step for the synthesis of (Z/E)-γ-curcumenol (Z/E)-(2) is a microwave-assisted14 olefin isomerization15 of (Z/E)-1 under basic conditions (KOtBu/toluene). As a model compound we used 1,4-diene 7, which was rearranged into the corresponding 1,3-diene in 62% yield (31% of diene 7 recovered). In contrast (E)-1 led to a 4:1 mixture of (E)-2 and (E)-1 together with a minor amount of aromatized product [(E)-Nuciferol]. For further separation the crude product can be transformed into the corresponding acetates under standard conditions (Ac2O, NEt3, DMAP, CH2Cl2) and separated by column chromatography using AgNO3 impregnated silica gel.16 However, (Z/E)-2 is chemically much more sensitive to decomposition and/or air-oxidation compared to (Z/E)-1 as described before.1,5 In summary, we have demonstrated a short total synthesis (6 steps, 33% over all yield) of (Z/E)-β-curcumen-12-ol (Z/E)-(1) and its transformation into (Z/E)-γ-curcumen-12-ol (Z/E)-(2) by microwave assisted olefin isomerization. Furthermore, we have shown (E)-β-curcumen-12-ol (E)-(1) being a new natural product present in different sandalwood oils. This is a further example, that natural product synthesis and structure elucidation are closely connected.17 General Procedures All reagents were commercial products (Fluka, Aldrich or Lancaster) and were used as received. THF and Et2O were freshly distilled from Na/benzophenone. All other solvents were distilled prior to use. Reactions involving air and/or moisture sensitive reagents were conducted under an argon atmosphere, and the glassware was oven dried (140°C) and purged with argon. All reactions were monitored by analytical TLC (silica gel 60 F254), Merck, Darmstadt, Germany. Preparative column chromatography: silica gel 60 (63-200 µm), Macherey & Nagel, Düren, Germany. NMR: Unity INOVA Varian 300 spectrometer (1H: 300 MHz; 13C: 75.48 MHz) in CDCl3 if not otherwise stated with TMS as internal standard; chemical shifts (δ in ppm and coupling constants (J) in Hz. IR: Perkin-Elmer Paragon 1000 FT-IR spectrometer; wave number (ν) in cm-1. MS: Finnigan MAT 8200 in EI mode (70 eV); data in m/z (%). GC-MS: Hewlett Packard 5973N; columns: DB-Wax or DB-1 (20m x 0.18mm x 0.18µm film thickness, carrier gas: He) programmed from 60°C to 220°C at 9°C/min, mass spectrometer operating at 70 eV ionization energy (EI mode). 4-(4-Methyl-phenyl)pentanoic acid (6). Prepared according to literature, starting from α- curcumene = ar-curcumene (5)8b [isolated from curcuma oil (Curcuma xanthorrhiza Roxb., Zingiberaceae) (15-18%) or ginger oil (Zingiber officinale Roscoe, Zingiberaceae) (9-13%)]. Concentration of 5 can be increased in ginger oil by aromatization of zingiberene (>25%) and sesquiphellandrene (11-15%) prior to oxidative cleavage.8b IR (neat): 2959 (br), 1702. 1H NMR: 10.8 (br, 1H), 7.12-7.04 (m, 4H), 2.76-2.62 (m, 1H), 2.32 (s, 3H), 2.26-2.20 (m, 2H), 1.95-1.83 Page 276 ©ARKAT USA, Inc. ©ARKAT USA, Inc. ©ARKAT USA, Inc. Page 276 ISSN 1424-6376 ARKIVOC 2007 (vii) 273-279 Issue in Honor of Prof. Atta-ur-Rahman (m, 2H), 1.30 (d, J = 7 Hz, 3H). 13C NMR: 180.4 (s, C-10), 143.3 (s), 136.0 (s), 129.5 (d), 127.1 (d), 39.1 (d, C-7), 33.2, 32.6 (t, C-8, C-9), 22.5 (q, Me), 21.3 (q, Me). (m, 2H), 1.30 (d, J = 7 Hz, 3H). 13C NMR: 180.4 (s, C-10), 143.3 (s), 136.0 (s), 129.5 (d), 127.1 (d), 39.1 (d, C-7), 33.2, 32.6 (t, C-8, C-9), 22.5 (q, Me), 21.3 (q, Me). 4-(4-Methyl-1,4-cyclohexadienyl)pentanoic acid (7). A solution of 6 (1.5 g, 7.8 mmol) in anhydrous THF (35 mL) was added slowly with stirring to liquid ammonia (100 mL) at -70°C. To the clear solution Li metal (2.5 g, 357 mmol) was added in small pieces. General Procedures After 30 min EtOH (5 mL) was added to the deep blue mixture and stirring was continued for 1h. The reaction was quenched with EtOH and the ammonia was allowed to distill off over night. The reaction mixture was concentrated in vacuum, the remaining aqueous phase cooled with ice and acidified with 1M HCl and extracted with tert. butylmethyl ether (5 x 50 mL). The combined organic phases were washed with water (4 x 30 mL), concentrated and the remaining oil distilled to yield 1.40 g (92%) of 7. Colorless, viscous oil. Bp (Kugelrohr) 110-120°C/0.01 Torr. IR (film): 2954, 1515. 1H NMR: 5.51 and 5.47 (br s, 2H), 2.8-2.5 (m, 4H), 2.4-2.2 (m, 2H), 1.68 (s, 3H), 1.8-1.6 (m, 2H), 1.04 (d, J = 7 Hz, 3H). 13C NMR: 180.4 (s, C-10), 137.8 (s, C-6), 131.5 (s, C-3), 119.3, 118.8 (d, C-1, C-4), 40.4 (d, C-7), 32.4 (t, C-2), 31.8 (t, C-9), 29.6 (t, C-8), 26.5 (t, C-5), 23.2 (q, Me), 19.6 (q, Me). GC-MS: 194 (15) [M]+, 134 (10), 132 (15), 105 (40), 93 (50), 91 (100)77 (45). HRMS m/z 194.1306 [M]+, (calcd. for C12H18O2 194.1307). ( ) [ ] , ( 12 18 2 ) N-Methoxy, N-methyl 4-(4-methyl-1,4-cyclohexadienyl)pentanoic amide (8): To an ice-cold solution of acid 7 (0.21 g, 1.1 mmol) in DMF (5 mL) and 1 mL NEt3 was added BOP (0.58 g, 1.3 mmol) under stirring. After 30 min N,O-dimethylhydroxylamine hydrochloride (0.20 g, 2 mmol) was added in one portion. The reaction mixture was diluted with H2O (20 mL) and extracted with petroleum ether (5x30 mL). The combined organic phases were filtered through silica gel (30 g) concentrated and the remaining oil distilled to give 2.50 g (96%) of 8. Colorless oil. Bp (Kugelrohr) 110-120°C/0.002 Torr. IR (film): 1662. 1H NMR: 5.47-5.43 (br m, 2H, 2- and 5-H), 3.66 (s, 3H, OMe), 3.16 (s, 3H, NMe), 2.57 (br s, 4H, 1- and 4-H), 2.4-2.2 (m, 2H, 9-H), 2.14 (m, 7-H), 1.62 (br s, 5H, 8-H, olefin. Me), 1.04 (d, J = 6.9 Hz, 3H, Me); 13C NMR: 175.1 (s, C- 10), 138.4 (s, C-6), 131.4 (s, C-3), 118.9, 118.8 (d, C-1, C-4), 61.4 (q, OMe), 40.6 (d, C-7), 32.5 (q, NMe), 31.8 (t, C-2), 30.3 (t, C-9), 29.6 (t, C-8), 26.7 (t, C-5), 23.2 (q, Me), 19.8 (q, Me). General Procedures The mixture can be separated after transformation into the corresponding acetates on silica gel/10% AgNO3.16 (E)-2 is very sensitive to decomposition. General Procedures 1H NMR: 5.4-5.36 (m, 3H), 4.0 (s, 2H), 2.6 (br s, 4H), 2.2-2.0 (m, 1H), 2.0-1.9 (m, 2H), 1.68 (s, 3H), 1.66 (s, 3H), 1.6-1.2 (m, 4H), 1.05 (d, J = 7 Hz, 3H). 13C NMR: 139.0 (s, C-6), 134.8 (s, C-11), 131.6 (s, C-3), 126.9 (d, C-10), 119.0 and 118.2 (d, C-1, C-4), 69.4 (t, C-12), 40.5 (d, C-7), 34.8 (t, C-8), 31.9 and 26.8 (t, C-2, C-5), 25.9 (t, C-9), 23.2 (q, C-15), 19.8 (q, C-14), 13.9 (q, C-13). GC-MS: 220 (0.5) [M]+, 202 (2), 187 (6), 159 (5), 145 (23), 132 (68), 119 (100), 105 (34), 93 (47), 91 (38), 77 (21), 68 (7), 55 (10), 43 (10). HRMS m/z 220.1827 [M]+, (calcd. for C15H24O 220.1827). (10), 119 (100), 91 (45). HRMS m/z 248.1761 [M] , (calcd. for C16H24O2 248.1763). (E)-2-Methyl-6-(4-methyl-cyclohexa-1,4-dienyl)hept-2-en-1-ol = (E)-β-curcumen-12-ol (E)- (1). To an ice-cold solution of ester 10 (0.30 g, 1.2 mmol) in Et2O (50 mL) was added LiAlH4 (0.20 g, 5.3 mmol) in small portions with stirring. 0.1M HCl (10 mL) was added after 30 min, the organic phase separated and the aqueous extracted with Et2O (3x20 mL). The combined organic phases were filtered through silica gel (30 g). The filtrate was concentrated and the oily residue distilled to yield 0.22 g (83%) of alcohol (E)-1. Colorless oil. Bp (Kugelrohr) 90- 100°C/0.03 Torr. E/Z = 95:5. IR (film): 3327 (br). 1H NMR: 5.4-5.36 (m, 3H), 4.0 (s, 2H), 2.6 (br s, 4H), 2.2-2.0 (m, 1H), 2.0-1.9 (m, 2H), 1.68 (s, 3H), 1.66 (s, 3H), 1.6-1.2 (m, 4H), 1.05 (d, J = 7 Hz, 3H). 13C NMR: 139.0 (s, C-6), 134.8 (s, C-11), 131.6 (s, C-3), 126.9 (d, C-10), 119.0 and 118.2 (d, C-1, C-4), 69.4 (t, C-12), 40.5 (d, C-7), 34.8 (t, C-8), 31.9 and 26.8 (t, C-2, C-5), 25.9 (t, C-9), 23.2 (q, C-15), 19.8 (q, C-14), 13.9 (q, C-13). GC-MS: 220 (0.5) [M]+, 202 (2), 187 (6), 159 (5), 145 (23), 132 (68), 119 (100), 105 (34), 93 (47), 91 (38), 77 (21), 68 (7), 55 (10), 43 (10). HRMS m/z 220.1827 [M]+, (calcd. for C15H24O 220.1827). ( ), ( ), ( ) [ ] , ( 16 24 2 ) (E)-2-Methyl-6-(4-methyl-cyclohexa-1,4-dienyl)hept-2-en-1-ol = (E)-β-curcumen-12-ol (E)- (1). To an ice-cold solution of ester 10 (0.30 g, 1.2 mmol) in Et2O (50 mL) was added LiAlH4 (0.20 g, 5.3 mmol) in small portions with stirring. General Procedures 0.1M HCl (10 mL) was added after 30 min, the organic phase separated and the aqueous extracted with Et2O (3x20 mL). The combined organic phases were filtered through silica gel (30 g). The filtrate was concentrated and the oily residue distilled to yield 0.22 g (83%) of alcohol (E)-1. Colorless oil. Bp (Kugelrohr) 90- 100°C/0.03 Torr. E/Z = 95:5. IR (film): 3327 (br). 1H NMR: 5.4-5.36 (m, 3H), 4.0 (s, 2H), 2.6 (br s, 4H), 2.2-2.0 (m, 1H), 2.0-1.9 (m, 2H), 1.68 (s, 3H), 1.66 (s, 3H), 1.6-1.2 (m, 4H), 1.05 (d, J = 7 Hz, 3H). 13C NMR: 139.0 (s, C-6), 134.8 (s, C-11), 131.6 (s, C-3), 126.9 (d, C-10), 119.0 and 118.2 (d, C-1, C-4), 69.4 (t, C-12), 40.5 (d, C-7), 34.8 (t, C-8), 31.9 and 26.8 (t, C-2, C-5), 25.9 (t, C-9), 23.2 (q, C-15), 19.8 (q, C-14), 13.9 (q, C-13). GC-MS: 220 (0.5) [M]+, 202 (2), 187 (6), 159 (5), 145 (23), 132 (68), 119 (100), 105 (34), 93 (47), 91 (38), 77 (21), 68 (7), 55 (10), 43 (10). HRMS m/z 220.1827 [M]+, (calcd. for C15H24O 220.1827). (E)-2-Methyl-6-(4-methyl-cyclohexa-1,3-dienyl)hept-2-en-1-ol = (E)-γ-curcumen-12-ol (E)- (2). A mixture of (E)-1 (0.1 g, 0.45 mmol) and KOtBu (0.2 g) in degassed toluene (4 mL) was heated with stirring under Ar atmosphere in a microwave oven at 100°C for 2h. The cooled yellow reaction mixture was diluted with Et2O (20 mL), the organic phase washed with H2O (5 mL) and brine (5 mL) and filtered through a pad of silica gel (20 g). The filtrate was concentrated and the remaining oil distilled via Kugelrohr (100°C/0.03 Torr) to yield 0.08 g (80%) of a 4:1 mixture of (E)-2 and (E)-1. The mixture can be separated after transformation into the corresponding acetates on silica gel/10% AgNO3.16 (E)-2 is very sensitive to decomposition. (E)-2-Methyl-6-(4-methyl-cyclohexa-1,3-dienyl)hept-2-en-1-ol = (E)-γ-curcumen-12-ol (E)- (2). A mixture of (E)-1 (0.1 g, 0.45 mmol) and KOtBu (0.2 g) in degassed toluene (4 mL) was heated with stirring under Ar atmosphere in a microwave oven at 100°C for 2h. The cooled yellow reaction mixture was diluted with Et2O (20 mL), the organic phase washed with H2O (5 mL) and brine (5 mL) and filtered through a pad of silica gel (20 g). The filtrate was concentrated and the remaining oil distilled via Kugelrohr (100°C/0.03 Torr) to yield 0.08 g (80%) of a 4:1 mixture of (E)-2 and (E)-1. General Procedures GC- MS: 237 (3) [M]+, 236 (5), 222 (5), 206 (2), 188 (7), 175 (35), 159 (33), 158 (15), 157 (19), 144 (19), 143 (13), 134 (32), 119 (59), 105 (56), 103 (48), 93 (29), 91 (100), 79 (33), 77 (47), 73 (28), 61 (26), 58 (26). ESI-HRMS m/z 260.1624 [M+Na]+, (calcd. for C14H23NO2+Na+ 260.1621; C14H23NO2 237.1729). Methyl (E)-2-methyl-6-(4-methyl-cyclohexa-1,4-dienyl)hept-2-enoate (10). To a solution of amid 8 (0.54 g, 2.3 mmol) in Et2O (10 mL) at 5°C was added LiAlH4 (0.20 g, 5.3 mmol) in small portions with stirring. Stirring was continued for 30 min and then ethyl acetate (3 mL) was added followed by 1M HCl (5 mL). The organic phase was separated and the aqueous extracted with Et2O (3x 25 mL). The combined organic phases were washed with H2O (2x10 mL), filtered through silica gel (30 g) and the filtrate was concentrated to obtain aldehyde 9, which was used without further purification for the consecutive Wittig reaction. Crude aldehyde 9 was dissolved in CH2Cl2 (20 mL) and Ph3P+CHMeCO2MeBr– (1.10 g, 2.6 mmol) was added under stirring followed by NEt3 (0.5 mL). The reaction mixture was concentrated after 18h, the residue ©ARKAT USA, Inc. ©ARKAT USA, Inc. ISSN 1424-6376 Page 277 Issue in Honor of Prof. Atta-ur-Rahman ARKIVOC 2007 (vii) 273-279 dissolved in Et2O (50 mL), the solution decanted from the precipitate and filtered through silica gel (40 g). The filtrate was concentrated and the oily residue distilled to yield 0.30 g (53% for two steps) of 10. Colorless oil. E/Z = 95:5. Bp (Kugelrohr) 110-120°C/0.001 Torr. IR (film): 1713, 1261. 1H NMR: 6.76 (q, J = 6 Hz, 1H, 10-H), 5.45 (m, 2H), 3.74 (s, 3H, ester Me), 2.59 (br s, 4H), 2.05-2.26 (m, 4H), 1.82 (s, 3H, 13-H), 1.68 (s, 3H, 15-H), 1.6-1.35 (m, H), 1.03 (d, J = 7 Hz, 3H, 14-H). 13C NMR: 169.0 (s, C-12), 143.1 (s, C-11), 138.4 (s, C-6), 131.6 (s, C-3), 127.6 (d, C-10), 118.8, 118.7 (d, C1, C-4), 51.9 (OMe), 40.6 (d, C-7), 33.8 (t, C-8), 31.8, 27.0, 26.6 (t, C-2, C-5, C-9) 23.2 (q, C-15), 19.8 (q, C-14), 12.6 (q, C-13). GC-MS: 248 (2) [M]+, 216 (10), 119 (100), 91 (45). HRMS m/z 248.1761 [M]+, (calcd. for C16H24O2 248.1763). dissolved in Et2O (50 mL), the solution decanted from the precipitate and filtered through silica gel (40 g). General Procedures The filtrate was concentrated and the oily residue distilled to yield 0.30 g (53% for two steps) of 10. Colorless oil. E/Z = 95:5. Bp (Kugelrohr) 110-120°C/0.001 Torr. IR (film): 1713, 1261. 1H NMR: 6.76 (q, J = 6 Hz, 1H, 10-H), 5.45 (m, 2H), 3.74 (s, 3H, ester Me), 2.59 (br s, 4H), 2.05-2.26 (m, 4H), 1.82 (s, 3H, 13-H), 1.68 (s, 3H, 15-H), 1.6-1.35 (m, H), 1.03 (d, J = 7 Hz, 3H, 14-H). 13C NMR: 169.0 (s, C-12), 143.1 (s, C-11), 138.4 (s, C-6), 131.6 (s, C-3), 127.6 (d, C-10), 118.8, 118.7 (d, C1, C-4), 51.9 (OMe), 40.6 (d, C-7), 33.8 (t, C-8), 31.8, 27.0, 26.6 (t, C-2, C-5, C-9) 23.2 (q, C-15), 19.8 (q, C-14), 12.6 (q, C-13). GC-MS: 248 (2) [M]+, 216 (10), 119 (100), 91 (45). HRMS m/z 248.1761 [M]+, (calcd. for C16H24O2 248.1763). dissolved in Et2O (50 mL), the solution decanted from the precipitate and filtered through silica gel (40 g). The filtrate was concentrated and the oily residue distilled to yield 0.30 g (53% for two steps) of 10. Colorless oil. E/Z = 95:5. Bp (Kugelrohr) 110-120°C/0.001 Torr. IR (film): 1713, 1261. 1H NMR: 6.76 (q, J = 6 Hz, 1H, 10-H), 5.45 (m, 2H), 3.74 (s, 3H, ester Me), 2.59 (br s, 4H), 2.05-2.26 (m, 4H), 1.82 (s, 3H, 13-H), 1.68 (s, 3H, 15-H), 1.6-1.35 (m, H), 1.03 (d, J = 7 Hz, 3H, 14-H). 13C NMR: 169.0 (s, C-12), 143.1 (s, C-11), 138.4 (s, C-6), 131.6 (s, C-3), 127.6 (d, C-10), 118.8, 118.7 (d, C1, C-4), 51.9 (OMe), 40.6 (d, C-7), 33.8 (t, C-8), 31.8, 27.0, 26.6 (t, C-2, C-5, C-9) 23.2 (q, C-15), 19.8 (q, C-14), 12.6 (q, C-13). GC-MS: 248 (2) [M]+, 216 (10), 119 (100), 91 (45). HRMS m/z 248.1761 [M]+, (calcd. for C16H24O2 248.1763). (E)-2-Methyl-6-(4-methyl-cyclohexa-1,4-dienyl)hept-2-en-1-ol = (E)-β-curcumen-12-ol (E)- (1). To an ice-cold solution of ester 10 (0.30 g, 1.2 mmol) in Et2O (50 mL) was added LiAlH4 (0.20 g, 5.3 mmol) in small portions with stirring. 0.1M HCl (10 mL) was added after 30 min, the organic phase separated and the aqueous extracted with Et2O (3x20 mL). The combined organic phases were filtered through silica gel (30 g). The filtrate was concentrated and the oily residue distilled to yield 0.22 g (83%) of alcohol (E)-1. Colorless oil. Bp (Kugelrohr) 90- 100°C/0.03 Torr. E/Z = 95:5. IR (film): 3327 (br). Acknowledgements The authors are indebted to Birgit Kohlenberg, Symrise GmbH & Co. KG, Holzminden, Germany, and to Iris Klaiber, Universität Hohenheim, Stuttgart, Germany, for GC-MS analysis and helpful discussions. We are especially grateful to Klaus-Dieter Protzen, Paul Kaders GmbH, Hamburg, Germany, for generous gift of curcuma and ginger oil. Page 278 Page 278 ©ARKAT USA, Inc. ISSN 1424-6376 ISSN 1424-6376 ©ARKAT USA, Inc. ARKIVOC 2007 (vii) 273-279 Issue in Honor of Prof. Atta-ur-Rahman References 1. Weyerstahl, P.; Wahlburg, H.-C.; Marschall, H.; Rustaiyan, A. Liebigs Ann. Chem. 1993, 1117. 1. Weyerstahl, P.; Wahlburg, H.-C.; Marschall, H.; Rustaiyan, A. Liebigs Ann. Chem. 1993, 1117. 2. Braun, N. A.; Meier, M.; Kohlenberg, B.; Valder, C.; Neugebauer, M. J. Essent. Oil Res. 2003, 15, 381. 2. Braun, N. A.; Meier, M.; Kohlenberg, B.; Valder, C.; Neugebauer, M. J. Essent. Oil Res. 2003, 15, 381. 3. Weyerstahl, P.; Marschall, H.; Wahlburg, H.-C.; Christiansen, C.; Rustaiyan, A.; Mirdjalili, F. Flav. Fragr. J. 1999, 14, 121. stahl, P.; Marschall, H.; Eckardt, A.; Christiansen, C. Flav. Fragr. J. 1999, 14, 15. 4. Weyerstahl, P.; Marschall, H.; Eckardt, A.; Christiansen, C. Flav. Fragr. J. 5. Valder, C.; Neugebauer, M.; Meier, M.; Kohlenberg, B.; Hammerschmidt, F.-J.; Braun, N. A. J. Essent. Oil Res. 2003, 15, 178. 6. Braun, N. A.; Meier, M. EuroCosmetics 2004, 12(1), 22. 7. Braun, N. A.; Meier, M.; Hammerschmidt, F.-J. J. Essent. Oil Res. 2005, 17, 477 8. (a) Henry, J. R.; Weinreb, S. M. J. Org. Chem. 1993, 58, 4745. (b) Chavan, S. P.; Dhondge, V. D.; Patil, S. S.; Rao, Y.; Tripura, S.; Govande, C. A. Tetrahedron: Asymmetry 1997, 8, 2517. 9. (a) Davisson, V. J.; Poulter, C. D. J. Am. Chem. Soc. 1993, 115, 1245. (b) Birch, A. J. Pure Appl. Chem. 1996, 68, 553. 10. Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815. 11. (a) Nagafuji, P.; Cushman, M. J. Org. Chem. 1996, 61, 4999. (b) We thank Prof. Dr. J. Podlech, TU Karlsruhe, Germany, for discussion. y 12. Braun, N. A.; Bürkle, U.; Feth, M. P.; Klein, I. Spitzner, D. Eur. J. Org. Chem. 1998, 1569. A.; Bürkle, U.; Feth, M. P.; Klein, I. Spitzner, D. Eur. J. Org. Chem. 1998, 1569. 12. Braun, N. A.; Bürkle, U.; Feth, M. P.; Klein, I. Spitzner, D. Eur. J. Org. Chem. 13. Takano, S.; Goto, E.; Ogasawara, K. Tetrahedron Lett. 1982, 23, 5567. 14. (a) Lidström, P.; Tierney, J.; Wathey, B.; Westman, J. Tetrahedron 2001, 57, 9225. (b) Kappe, C. O. Angew. Chem. 2004, 116, 6408 and Angew. Chem. Int. Ed. 2004, 43, 6250. 15. Hubert, A. J.; Reimlinger, H. Synthesis 1969, 97. 15. Hubert, A. J.; Reimlinger, H. Synthesis 1969, 97. 16. Williams, C. M.; Mander, L. N. Tetrahedron 2001, 57, 425. 17. Nicolaou, K. C.; Snyder, S. A. Angew. Chem. 2005, 117, 1036 and Corrigendum Angew. Chem. 2005, 117, 2086; Angew. Chem. Int. References Ed. 2005, 44, 1012 and Corrigendum Angew. Chem. Int. Ed. 2005, 44 2050. ISSN 1424-6376 ©ARKAT USA, Inc. Page 279 Page 279

https://openalex.org/W2142660902

https://uwe-repository.worktribe.com/preview/968346/EPAfinal.pdf

English

Comparing Alternative Media in North and South: The Cases of IFIWatchnet and Indymedia in Africa

Environment & planning. A

public-domain

Comparing alternative media in North and South: The cases of IFIWatchnet and Indymedia in Africa Comparing alternative media in North and South: The cases of IFIWatchnet and Indymedia in Africa Keywords Alternative media, independent media, „global north‟ and „global south‟, Internet, networks, IFIwatchnet, Indymedia Centre in Africa (IMCA), social movements, non- governmental public action (NGPA) Comparing alternative media in North and South: The cases of IFIWatchnet and Indymedia in Africa Authors: Fabian Frenzel Bristol Business School University of the West of England, Bristo Frenchay Campus Coldharbour Lane Bristol BS16 1QY Tel: +44 (0) 117 32 83414 Email: [email protected] Pennie Quinton Email: [email protected] London UK André Spicer Email: [email protected] IROB, Warwick Business School, University of Warwick, Coventry CV4 7AL, UK André Spicer Email: [email protected] IROB, Warwick Business School, University of Warwick, Coventry CV4 7AL, UK Sian Sullivan Email: [email protected] Department of Geography, Environment and Development Studies School of Social Sciences, History and Philosophy Birkbeck, University of London 26 Russell Square London WC1B 5DQ Zoe Young 1 Email: [email protected] London UK Acknowledgements The authors would like to thank research participants from IFIwatchnet and IMCAfrica for their contributions to the research process and the anonymous reviewers for their valuable comments. The research underpinning this paper emerged out of a project entitled „Alternative Media and Public Action: Organising the Global Alternative Networks‟, which was funded by the UK‟s Economic and Social Research Council (Grant: RES-155-25-0029). Abstract Alternative media form an important part of the global mediascape. Research on this phenomenon is however often drawn upon studies in the „global North‟. In this paper we discuss alternative media in the „global South‟. We do this by exploring two case studies of co-operation between Northern and Southern partners: „IFIwatchnet‟ and „Indymedia Centre in Africa (IMCA)‟. We highlight how Northern and Southern partners differed in identity, organizational forms, and accountability. We find that Northern partners were oriented to more „marginal‟ identities, fluid organizational structures, and informal structures of accountability. In contrast, Southern activists articulated more „mainstream‟ identities, relied on more structured forms, and linked to formalized modes of accountability. The result was often significant clashes over what it meant to be an alternative media, how it should be organized and how people should be held to account. 2 2 This meant North-South co-operation was often fraught with struggle. These difficulties remind us of the limitations of creating global co-operation through seeking to spread modes of activist organization developed in the North, which emphasize autonomy, networks, fluidity and in some instances, direct action. Introduction The arrival and establishment of low-cost media platforms based on Information Communication Technologies (ICTs) has created a rich ecology of media falling outside state or corporate ownership. This has been variously labeled radical media (Downing 1984), citizen media (Rodriguez 2001) or alternative media (Atton 2002). We prefer the term alternative media, which broadly involves „media production which challenges, at least implicitly, actual concentration of media power, whatever form these concentrations may take in different locations‟ (Couldry & Curran 2003, p.7). Despite an increasing awareness of alternative media organizations, research has largely focused on alternative media in the wealthy „global North‟. This provides relatively benign circumstances for alternative media networks to emerge: there are plentiful resources and state repression or censorship is often not particularly harsh. But, what do alternative media look like in the „global South‟ where state repression and censorship is more intense and resources are less plentiful than in the developed world? Little work has been done so far to consider this question. Existing comparative studies of media indicate that media organizations in relatively low income countries and/or in situations constrained by sustained conflict have quite a radically different experience to 3 3 those in more developed contexts (Frenzel & Sullivan 2009). But at the same time, there is a widespread assumption that the proliferation of ICTs based media would allow for alternative media organizations to develop easily in the „global South‟. This assumption has prompted investment in ICT projects by development NGOs and international donors in the last 20 years. Such tendencies have been questioned for „uncritically positing new media technology as deterministic of social progress‟ (Mudhai et al. 2009, p.1). Southern alternative media organizations often rely on significant support from „Northern‟ NGOs, donors, or funding. This creates unique challenges for Southern alternative media organizations. Rather than only challenging the power of large-scale media organization, they also have to relate to and in some cases to resist the power exercised by their Northern partners and donors. In this paper we address the lack of research into Southern alternative media by exploring differences between alternative media organizations in the „North‟ and the „South‟. To do this, we have conducted research with two alternative media networks: namely IFIWatchnet and Indymedia Africa. These two networks bridge the „North‟ and „South‟ not simply by establishing links between nodes in the North and the South. Introduction Rather they attempted to create more substantial co-operation between multiple Northern and Southern nodes. They hoped this would facilitate the democratization and growth of the respective networks. However we found that North-South co-operation often unveiled significant differences about how formal the organization should be: Northern media activists had a preference for highly decentralized and autonomous modes of organizing. Southern activists usually focused on more formalized structures and processes that could IFIWatchnet and Indymedia Africa. These two networks bridge the „North‟ and „South‟ not simply by establishing links between nodes in the North and the South. Rather they attempted to create more substantial co-operation between multiple Northern and Southern nodes. They hoped this would facilitate the democratization and growth of the respective networks. However we found that North-South co-operation often unveiled significant differences about how formal the organization should be: Northern media activists had a preference for highly decentralized and autonomous modes of organizing. Southern activists usually focused on more formalized structures and processes that could 4 4 garner funding from large foreign NGOs and aid agencies. This led sometimes to conflicts, which were often difficult to resolve because a lack of a set of shared assumptions about the identity, form and structures of accountability in each group associated with alternative media. This suggests that attempts to build North-South co- operation by transferring the model of alternative media with an emphasis on „open source‟ and „autonomist‟ principles is rather naïve. Instead, North-South alternative media co-operation appears to involve struggle between very different sets of assumptions about what alternative media is and how it is organized. To make this point, we begin by reviewing existing studies of alternative media organizations, noting that this literature has rarely considered the important differences which might exist between alternative media in „Northern‟ and „Southern‟ contexts. We then outline our comparative case-based methodology for studying these differences. Next we look in more detail at our two case organizations of North-South co-operation: 1. IFIWatchnet, a civil society network established to link the work of groups critically „watching‟ the activities of International Financial Institutions (IFIs) such as the International Monetary Fund (IMF) and the World Bank (WB), and in which we focus on the shift of the network‟s „headquarters‟ from „North‟ to „South‟; and 2. „Indymedia Centre in Africa (IMCA)‟, initiated as part of the global „Indymedia‟ network. Introduction After outlining the history of these two cases, we focus on their identities, their organizational forms, and their structures of funding and accountability. We then move to a discussion where these two cases are compared, and we draw out the differences between the Northern and Southern partner‟s assumptions. We conclude by suggesting some 5 5 important areas for future comparative research on alternative media organizations facing clashing assumptions between Northern and Southern partners. We also draw out some implications for cross-context collaborations to sustain effective global alternative media- producing networks that are sensitive to these differences. Alternative Media Organizations Alternative media has been a neglected area of study, with few studies of this thriving sector existing prior to the 1990s (for some exceptions see Halloran 1970; Gitlin 1980; Downing 1984). In the past twenty years, however, there has been increasing recognition of a long history of alternative media and the important role it plays in the contemporary media landscape (Coyer et al. 2007). Because of the diversity of forms of alternative media, it is difficult to identify a set of stable characteristics that define it. Some argue that it is precisely because of the participatory and dynamic nature of alternative media that it is so difficult to tie it down in any meaningful sense (Gumucio-Dagron 2004). This is because alternative media appears to always be in the process of evolving and over- flowing any strict boundaries around what it could, or indeed should, be. Despite these concerns, there have been a number of attempts to enumerate a cluster of common aspects associated with alternative media. These include: interactivity between producers and consumers, collective production and a focus on everyday life Despite these concerns, there have been a number of attempts to enumerate a cluster of common aspects associated with alternative media. These include: interactivity between producers and consumers, collective production and a focus on everyday life (Enzensberger 1974); the production of small-scale media involving horizontal (i.e. relatively non-hierarchical) patterns of communication (McQuail 1986); a rejection of dominant political values, coupled with democratic or collectivist means of producing (Enzensberger 1974); the production of small-scale media involving horizontal (i.e. l i l hi hi l) f i i (M Q il 1986) j i f relatively non-hierarchical) patterns of communication (McQuail 1986); a rejection of dominant political values, coupled with democratic or collectivist means of producing relatively non-hierarchical) patterns of communication (McQuail 1986); a rejection of dominant political values, coupled with democratic or collectivist means of producing 6 6 media; and a commitment to innovative creation of content (O‟Sullivan 1994). Alternative media is also associated with innovations in media use and distribution, being characterized by „distributive use‟ and alternative forms of transmission, transformed social use, such as breaking down conventional producer-consumer relationships, and transformed communication processes, through an emphasis on networked horizontal links (Atton 2002). Alternative Media Organizations Alternative media is also associated with innovations in media use and distribution, being characterized by „distributive use‟ and alternative forms of transmission, transformed social use, such as breaking down conventional producer-consumer relationships, and transformed communication processes, through an emphasis on networked horizontal links (Atton 2002). From these different lists of alternative media characteristics, it is possible to identify some common themes. The first is that alternative media typically have a radically different collective identity to other forms of media, involving a shared sense of what is specific, unique and enduring to the organization or network (Whetten 2006). While the specific form this takes may differ quite significantly, most alternative media organizations typically have an identity based around an overriding concern for producing social change (Gumucio-Dagron 2004): from radical utopian desires to more modest aesthetic goals; and from claims for economic justice and social accountability, to reactionary right wing desires. Alternative media has often evolved from grievances of social movement activists who find their view misrepresented in mainstream media (Cresswell 1996; Routledge 1997). The intent of alternative media thus is not simply to garner a large audience. Rather, it involves an attempt to create some kind of social and cultural change. The second characteristic that seems to be unique to alternative media producers is the organizational form they adopt. An organizational form is made up of goals of the 7 organization, its authority relations, the technology it uses, and the markets it seeks to serve (Hannan & Freeman 1984; Scott 1995). Alternative media tends to emphasize the desirability of social change (as discussed above), organizing through more participatory and voluntary divisions of labour, using locally specific or „tactical‟ media, and blurring the boundaries between producer and consumer. A central thread is the importance of participation as a core organizing principle, with a consistent theme in the literature being the active involvement of ordinary people or staff of campaigning organizations who otherwise are often marginalized in the production of media content. This involves encouraging contributions from active citizens, not being attached to a political party, a focus on and association with social movements, and an emphasis on „prefigurative politics‟ (Downing 1984, p.17). The final distinctive aspect of alternative media organizations is their structure of accountability. This involves the answerability of an organization (i.e. who it must justify its actions to and how), and the enforcement of standards (Schedler 1999). Alternative Media Organizations Many alternative media organizations would say that they are ultimately answerable to the communities that they set out to serve. These communities might be formally represented through a board or some other kind of governance mechanism typical to an NGO with intra-organizational hierarchies. Others might reject this kind of accountability, and instead seek to create accountability through consensus decision-making processes and direct democracy. Many alternative media organizations thus seek to break down structured links that typically exist between media producers and consumers, governors and the governed, through championing strongly participatory modes of decision-making The final distinctive aspect of alternative media organizations is their structure of accountability. This involves the answerability of an organization (i.e. who it must justify its actions to and how), and the enforcement of standards (Schedler 1999). Many 8 8 such as consensus processes. Such emphasis on radical democracy may be emphasized and enshrined in standards or code of practices. such as consensus processes. Such emphasis on radical democracy may be emphasized and enshrined in standards or code of practices. Alternative media organizations tend to differ significantly from more mainstream media organizations in terms of their oppositional identities, horizontal organizational forms, and participatory structures of accountability. However, we should note that the difference between alternative media organizations and other more mainstream forms is far from neat (Downing 2001; Kim & Hamilton 2006), Nevertheless, many of the radical characteristics of alternative media described above have themselves been further radicalized following the rise of the ICTs (Lovink 2002; Meikle 2002; Van de Donk 2004; Atton 2002). This has permitted the appearance of a whole new set of now well known forms of alternative media which are largely web based, including the global independent media network or „Indymedia‟ (Pickerill 2007; Downing 2001), OhmyNews in South Korea (Kim & Hamilton 2006), and others. Many of these networked organizations foster identities that are radically boundary-less and open, influenced strongly by discourses associated with the open publishing and open source software movements and associated to „autonomous geographies‟ (Pickerill & Chatterton 2006). The assumption has been that ICTs permit the radical involvement and broadcasting of marginalized views and voices (Spicer & Perkmann 2008). In addition, online ventures are considered to reframe and radicalize the ways in which these organizations operate, through placing further emphasis on networked and „glocal‟ organizational forms (Sullivan 2008). Alternative Media Organizations Bennett (2003), for instance, points out the importance of „SPIN‟ organizational characteristics in online-based social movements: emphasizing The assumption has been that ICTs permit the radical involvement and broadcasting of marginalized views and voices (Spicer & Perkmann 2008). In addition, online ventures are considered to reframe and radicalize the ways in which these organizations operate, through placing further emphasis on networked and „glocal‟ organizational forms (Sullivan 2008). Bennett (2003), for instance, points out the importance of „SPIN‟ organizational characteristics in online-based social movements: emphasizing 9 Segmentation (with fluid boundaries between hubs or nodes), Polycentricism (or multiple hubs), Integration (through horizontal structures of multiple or rhizomatic paths of communication), and Networking (effected by high degrees of connectivity). In addition, many online-based social movements have sought to push participatory accountability structures even further by actively changing the relationship of users to content. This has entailed a radical blurring of boundaries between users and producers through experimentation with open content and open editing processes, exemplified by the Indymedia slogan „Don‟t hate the media, be the media!‟ (Downing 2001). Segmentation (with fluid boundaries between hubs or nodes), Polycentricism (or multiple hubs), Integration (through horizontal structures of multiple or rhizomatic paths of communication), and Networking (effected by high degrees of connectivity). In addition, many online-based social movements have sought to push participatory accountability structures even further by actively changing the relationship of users to content. This has entailed a radical blurring of boundaries between users and producers through experimentation with open content and open editing processes, exemplified by the Indymedia slogan „Don‟t hate the media, be the media!‟ (Downing 2001). The language of participation, networks, openness and engagement is now ubiquitous within these media-producing organizations, The principles of Indymedia for example are also made explicit in the so called „principles of unity‟, a description of a radical democratic organizational identity, form and accountability that is meant to enable the growth of the network without compromising local autonomy (Kidd 2003). However conflicts between the local and the global have continued to appear as codifications and standards may be interpreted in different ways while powers to enforce a certain understanding may be limited. Discussing a conflict over funding from the Ford Foundation in the early years of the Indymedia Network, Picard (2006a) has pointed towards the problems that may arise when principles of local autonomy and network coherence clash. Alternative Media Organizations The „principles of unity‟ could not prevent a very different reading of whether it was ok to receive funding from the Ford Foundation. So far, it might be argued, this general problem has only been addressed in the contexts 10 of a fairly limited set of studies and alternative media-producing situations. Indeed the study of alternative media is often focused in Northern contexts (Hamilton & Atton 2001) where the material conditions are relatively comparable. The little work that does focus on Southern contexts indicates that there indeed are some important differences between alternative media in „North‟ and „South‟ (e.g. Downing et al. 1995; Park & Curran 2000; Frenzel & Sullivan 2009; Kim & Hamilton 2006; Gumucio-Dagron 2004). Furthermore, Picard‟s (2006a) study of the fate of Ford Foundation funding for Indymedia suggests that these differences may shape how Northern and Southern partners co-operate. Given this relative paucity of work that researches alternative media specific to Southern contexts, in this paper we ask whether and how alternative media networks in the North and South differ. In particular, we are interested the potential differences between the identities, organizational forms, and structures of accountability expressed by alternative media organizations that are attempting to cooperate and collaborate in global alternative media networks. Methodology To explore differences between alternative media in the „North‟ and „South‟, and how these played out during North-South co-operation, we adopt a comparative case study methodology. This involves the selection, analysis and systematic comparison of cases that we might expect to differ in theoretically significant ways (as described above). To explore differences between alternative media in the „North‟ and „South‟, and how these played out during North-South co-operation, we adopt a comparative case study methodology. This involves the selection, analysis and systematic comparison of cases that we might expect to differ in theoretically significant ways (as described above). Single case studies have been favoured in the study of alternative media research: proving effective in investigating the dynamics and specificities of the work of alternative proving effective in investigating the dynamics and specificities of the work of alternative 11 media in particular settings; in revealing rich information regarding each case; and sometimes in generating new concepts and theories (Eisenhardt 1989). Here, however, we are interested in producing a relatively systematic analysis of how concepts and theories might generate and explain context-dependent differences in alternative media organizations operating as nodes within global media networks. A comparative case method involves the systematic analysis of the characteristics and dynamics specific to cases, coupled with analysis of the similarities and differences between cases along a range of relevant dimensions (Ragin 1992). Such comparative approaches have proved useful in identifying differences in models of public broadcasting (Kueng-Shankleman 2000), and are likely to be similarly useful for exploring differences in alternative media. In order to explore differences between alternative media in „Northern‟ and „Southern‟ contexts, we selected two global networks that involved both Northern and Southern partners. These are IFIWatchnet, focusing on the shift of the network‟s „headquarters‟ from „North‟ to „South‟, and the Indymedia Centre Africa (IMCA), which is part of the global „Indymedia‟ network. We have tried to ensure that these two media networks are sufficiently similar to warrant comparison: they use the same medium (the Internet), and share broadly similar left-leaning political orientations. Each involves substantial interaction between Northern and Southern actors. As we describe, a couple of years after its founding, the headquarters of IFIWatchnet moved from the North (London) to the South (Montevideo), whilst continuing to serve a global audience and involving activists from multiple locations. Methodology On the other hand, the founding conferences of IMCA discussed in this paper were all based in the „South‟ (in Senegal, Mali and Kenya). Nevertheless, 12 thus provide extremely interesting and potentially comparable examples of co-operation between Northern and Southern contexts. thus provide extremely interesting and potentially comparable examples of co-operation between Northern and Southern contexts. To compare these cases, we sought first to construct a comprehensive picture of each organization through extensive documentation, coupled with first-hand involvement and observation. In some cases this involved narrative accounts of organizational histories; in others, projects were documented through a collection of their output (e.g. exchanges on public domain e-lists). We also conducted semi-structured interviews with participants in each of these cases. The interviewees were chosen because of their intense involvement in the case study projects. Two co-authors (Zoe Young and Fabian Frenzel) were involved over the longer-term as active creators of aspects of IFIwatchnet and IMCA respectively, whilst also conducting „action research‟ regarding these projects. Furthermore participants in both cases were involved in the reviewing of this paper in line with the action research methodology adopted here (Kemmis & McTaggart 2005; Kindon et al. 2007). In analyzing the extensive data generated for each case, we began by developing a descriptive account of each organization. This involved broadly tracing the history of each of the cases, documenting how they developed, the core participants involved, and some of the central issues faced by each organization and their associated network. We then developed a more analytical account of each of the cases, based on the three key dimensions of alternative media, which we located in the literature: their identity, their organizational form, and their structures of accountability. These three dimensions were then used in a comparison of the similarities and differences between these organizations. 13 In what follows, we detail the results of this analysis. In what follows, we detail the results of this analysis. In what follows, we detail the results of this analysis. IFIWATCHNET IFIwatchnet.org was established in London in 2002 through the Bretton Woods Project (BWP). The IFIWatchnet homepage eventually presented news, campaigns, reports, and links to member groups, video, and more. This information came mainly from pre- existing network hubs such as the Bank Information Centre in Washington, Central and Eastern European Bankwatch Network in Prague and the NGO Forum on the Asian Development Bank in Manila. These organizations were invited to participate in IFIWatchnet to share information and raise the profile of their work on International Financial Institutions (IFIs), and in particular to establish a more visible and systematic presence on the web for research, perspectives and campaigns from Southern civil society. In October 2002 BWP applied to the Ford Foundation for funding for a one year pilot project, and in early 2003 „ifiwatchnet.org‟ was established with a shared calendar and e- mail discussion „D-group‟ that became the de facto governance structure for the nascent IFIwatchnet. The design was simple enough to be easy to access on dial up connections in the South. The technical platform chosen for IFIwatchnet was Actionapps, an open- source content management system (CMS). With support from the Ford Foundation, a coordinator was employed for two days a week, supported by staff of BWP and also the Bank Information Center. In 2005 the Ford Foundation continued its support for IFIWatchnet with a second two-year phase to be hosted in an organization based in the South. Several networks were put forward as 14 potential new hosts, and the Instituto Tercer Mundo (ITeM) in Uruguay was selected from among the interested organizations. This decision was partly made because the majority of organizations in the network were from the North, and very few in South America. One interviewee explained “ITeM are better resourced, with more web access and people with the paid time to get involved, and provide materials online”. The management of the IFIWatchnet.org website was shifted from BWP – a small, horizontally organized Northern research and lobbying network – to ITeM, a more hierarchically-organized and information-technology focused Southern NGO which had not until then been a member of IFIWatchnet. Initially, there was much optimism associated with the move, not least because BWP were happy to hand over technical aspects of the work to a more specialized team. However, the transfer process was hindered by the fact that nobody visited from either organization in the initial process. IFIWATCHNET The substantial role of the part time core coordinator, more or less invisible from outside the network, was not effectively shifted into the new host. Her role in relationship building, managing informal interactions and facilitating ownership was not effectively recreated or replaced within ITeM, and regional animators became somewhat neglected. As one interviewee said, “like any network, it was about people and personalities who could sell an idea, or push a certain idea, or a certain way of thinking”. It was this side of the organization of the network that was now missing or at least not very emphasized. potential new hosts, and the Instituto Tercer Mundo (ITeM) in Uruguay was selected from among the interested organizations. This decision was partly made because the majority of organizations in the network were from the North, and very few in South America. One interviewee explained “ITeM are better resourced, with more web access and people with the paid time to get involved, and provide materials online”. The management of the IFIWatchnet.org website was shifted from BWP – a small, and people with the paid time to get involved, and provide materials online”. The management of the IFIWatchnet.org website was shifted from BWP – a small, horizontally organized Northern research and lobbying network – to ITeM, a more hierarchically-organized and information-technology focused Southern NGO which had not until then been a member of IFIWatchnet. Initially, there was much optimism associated with the move, not least because BWP were happy to hand over technical aspects of the work to a more specialized team. However, the transfer process was hindered by the fact that nobody visited from either organization in the initial process. h b i l l f h i di l i i ibl f id horizontally organized Northern research and lobbying network – to ITeM, a more hierarchically-organized and information-technology focused Southern NGO which had not until then been a member of IFIWatchnet. Initially, there was much optimism associated with the move, not least because BWP were happy to hand over technical aspects of the work to a more specialized team. However, the transfer process was hindered by the fact that nobody visited from either organization in the initial process. Instead, emphasis was now placed on maintaining and updating the technical infrastructure such as the web platform. IFIWATCHNET In 2006, ITeM developers shifted the database from the original Actionapps software to the open source Drupal content management system (CMS). The decision to invest time and resources in this shift was perceived 15 within ITeM as technical, responding to the growing profile and utility of Drupal. The decision involved consultation with the Association of Progressive Communications (APC) and the IFIWatchnet steering committee, but not with IFIwatchnet‟s animators who mostly used the CMS. within ITeM as technical, responding to the growing profile and utility of Drupal. The decision involved consultation with the Association of Progressive Communications (APC) and the IFIWatchnet steering committee, but not with IFIwatchnet‟s animators who mostly used the CMS. IFWatchnet‟s funding was due to run out in 2007. A year‟s extension to the Ford Foundation grant was allowed for ITEM to explore further funding opportunities. Suggestions included subscriptions by participating groups and networks, and selling appropriate ethical advertising. These proposals did not get far, ostensibly because member organizations had not really seen the point of a network that expected substantial extra work by their staff and showed only limited progress towards information sharing. Organizational Identity Several key aspects of the political identity motivating IFIwatchnet can be identified. Central for its Northern founders was the idea of a networked information sharing structure that could be enabled by ICTs and used to enhance counter-hegemonic struggle to influence development discourses. Positioned as challengers of the power of dominant donors to shape public banking institutions‟ policies and discourses, the founders‟ political backgrounds can be located in the global justice movement focusing on critique of the international finance sector, environmental degradation, un-payable „third world‟ debt and neoliberal structural adjustment. Several key aspects of the political identity motivating IFIwatchnet can be identified. Central for its Northern founders was the idea of a networked information sharing structure that could be enabled by ICTs and used to enhance counter-hegemonic struggle to influence development discourses. Positioned as challengers of the power of dominant donors to shape public banking institutions‟ policies and discourses, the founders‟ political backgrounds can be located in the global justice movement focusing on critique of the international finance sector, environmental degradation, un-payable „third world‟ debt and neoliberal structural adjustment. With the move of IFWatchnet from its former base in London to a new location in the 16 South, some changes in the nature of the project were apparent. While there was great overlap in an understanding of IFIwatchnet as counter-hegemonic globally, some more intangible elements of this identity came into conflict with parts of the new host organization. To some degree this may have been because ITeM‟s technical work processes were more formalized than they had been in BWP. In addition, since most ITeM staff were only peripherally involved in international IFIwatching communities, they did not attend the World Bank/IMF annual meetings or engage in the cross-network policy communications that are BWP‟s everyday work. The result was that without strategic efforts to bridge this gap, communications with those networks became less intimate, collaborative and effective. As a result “quite often, things don‟t happen”, as one interviewee expressed it, which was “a very slow, frustrating way to work”. So while the move introduced a more „professionalized‟ identity, it also led to a sense of distance between managers of the networking tools and those directly campaigning on IFIs. Staff in ITeM claimed independence in the site management process. Organizational Identity Some interviewees told us that the content of IFIwatchnet became simply another facet of ITeM‟s strategic technical work; a website to promote certain political perspectives and generate grant funding. ITeM‟s sense of independence in terms of claiming the right to manage the network in a particular way was partly informed by a deep-seated anti- colonial identity amongst some staff in Uruguay. In one interview reference was made, without the prompting by the interviewer, to British colonial involvement in Uruguay‟s past, which the interviewee at ITeM used to explain some of the communication problems the network had been experiencing. 17 Organizational Form When IFIwatchnet initially emerged in London, it had a fairly flat organizational structure. This informality was a practical way of working to a low budget, with a few people collaborating on a wide range of tasks, embodying oversight at the same time as communications and some of the more traditionally „low status‟ tasks involved. Building networks and communities for information-sharing and coalition-building was seen as a productive political act in and of itself. The aim was to produce „warm‟ atmospheres of supportive and responsive collaboration among an „adhocracy‟ of those committed to the networking project. This was frequently contrasted with the „colder‟ feeling of more formal structures. It was felt that mutual understanding between core people involved in delivery of a broad network like IFIWatchnet was essential. This included respect for different organizational set ups and how they shape participation, individual motivations and assumptions. Before the move to ITeM, informality was embodied both in the way the project was formed in London and also in the networking with partners across the world by the Northern founders of the project. The adoption of ICTs was intended as media sharing infrastructure to enhance this informal development of networks, and was based in a social and political convergence prior and parallel to the technical networking. As the project moved to Uruguay, arguably a different, more instrumental and technical understanding of the role of ICTs emerged, exemplified by the management of the change of CMS. Importantly, the problem did not lie in the new CMS, but rather in the 18 assumption that decisions about CMS are merely technical and hence do not require consultation and building on others‟ experience. As one interviewee said, during the move to the South “a lot of the sense that this is a network not a technical platform” got lost. Attitudes towards ICTs arguably indicate different organizational forms. For another interviewee this was not about a “North-South split”, as she expressed it. However, she continued to say that “if you want it [the network] to be more representative, then you need to put more work into supporting and encouraging people ... and that sort of implies more people involved really; and if there are more people involved then I think you do need to have somebody who‟s pushing that, and driving that, and keeping people motivated and so on”. Organizational Form With the move from the North to the South this precise engagement with people became de-emphasized, as the network became more technically focused and less engaged with network building activities and informal interaction. assumption that decisions about CMS are merely technical and hence do not require consultation and building on others‟ experience. As one interviewee said, during the move to the South “a lot of the sense that this is a network not a technical platform” got lost. Attitudes towards ICTs arguably indicate different organizational forms. For another interviewee this was not about a “North-South split”, as she expressed it. However, she continued to say that “if you want it [the network] to be more representative, then you need to put more work into supporting and encouraging people ... and that sort of implies more people involved really; and if there are more people involved then I think you do need to have somebody who‟s pushing that, and driving that, and keeping people motivated and so on”. With the move from the North to the South this precise engagement with people became de-emphasized, as the network became more technically focused and less engaged with network building activities and informal interaction. The movement drew out a clash between a more informal organizational form with a focus on networking championed by Northern activists and a more formalized one with a focus on technical issues championed by Southern activists. This clash became more evident on a number of occasions. One instance was when some Northern activists collaborated with each other to explore possibilities to further develop Actionapps, the CMS that was used for the IFIWatchnet website at that time. One activist told us that “ITeM then got involved, hysterically as always, sending a message to xxx telling him that he‟s fucked up things”. This example of an overheated discussion, is an indication of the strains in the network at that time. 19 Accountability Accountability IFNWatchnet‟s move to the South was proposed by its initiators to the Ford Foundation as an indicator of their intent to maximize Southern involvement in IFIWatchnet. In other words, a project based on critique of global financial institutions established in the „North‟ ideally should not be based in Northern centres of finance capital like London. It was felt that by locating in the South, the network could practice their responsibility and broader accountability to the global movement they served. However, this move had unforeseen implications. When the network was based in the North, systems of accountability were largely based on dense networked relationships. Accountabilities and responsibilities were negotiated and usually founded upon personal trust. This system of accountability was relatively informal in nature. However, when the networked moved to Montevideo, it became far more formalized in nature and specific task responsibilities became more circumscribed and specified. As a result tasks that did not fit clearly with circumscribed responsibilities could be disregarded. Similarly, when Northern activists intervened in others areas of responsibility (such as happened with the CMS exampled mentioned above), there was a strong negative reaction from the Southern partners. What this suggests is that different understandings of what it meant to be accountable and responsible for an activity or outcome gave rise to some important clashes. The clashes between the more informally goal oriented Northern and more formally process oriented Southern groups evolved in a way which limited functionality of the 20 network as a whole to such an extent that many in the originating project discontinued their engagement. It is unknowable to what extent outcomes might have been different had IFIWatchnet relocated to an organization with a profile similar to ITeM‟s – technically focused, more hierarchical – but still based in London; nor whether more ideal co-ordination and technical systems would have resulted in more users engaging with the network. As one interviewee from the North said, “we are all suffering from the same capacity problems, not really finding sufficient time for working on IFIWatchnet”. That is, there are capacity issues at work that do not necessarily have anything to do with North-South relations. However, as our research on IFIWatchnet shows, the move of the network management role from North to South did create a range of problems for this alternative media organization. Indymedia Centre Africa (IMCA) Indymedia is a global network of Independent Media Centres (IMCs) founded in Seattle in the context of the emerging global justice movement (GJM) and its „coming out party‟ which closed the Ministerial Meeting of the World Trade Organization in November 1999. It is bound up with a broader desire to foster a „globalization from below‟, or what has been described as „grassroots globalization‟ (Appadurai 2001). Indymedia represents itself as a tool for the networking of struggles and the convergence of activists across the world against neo-liberal globalization, as well as a social movement to democratize the production of news media (Mamadouh 2004a; Pickard 2006; Pickerill 2007). After a period of rapid „organic‟ growth effecting the proliferation of network nodes in various locations in the first months and years of its existence, Indymedia became concerned 21 about the radical North-South imbalance of its growth (Halleck 2002). In 2003 the whole continent of Africa was represented by only four IMCs, while the UK alone had thirteen (Mamadouh 2004b, Frenzel & Sullivan 2009). Mirroring the global „digital divide‟, the lack of IMCs in Africa was understood as a challenge to Indymedia by some of its members. Indymedia‟s claim to „globalize from below‟ suffered from the conspicuous dominance of Northern nodes in the network. At the same time ICT based media promised to have a great potential to support processes of democratization in Africa. In 2003 following an initiative of activists from the IMC Ambazonia (Cameroon), an IMC run by Diaspora activists in Europe, four existing African IMCs, together with members from across the global network, founded IMC Africa (IMCA). IMCA was inspired by the idea to foster the development of Indymedia network nodes in Africa through support, knowledge exchange and networking. Initially a network based on an email listserv, IMCA soon started organizing for a conference in Senegal, to be held in March 2004. The aim of the conference was to bring together journalists and media activists from Africa, and to train them in the use of ICTs as independent media activists. This was intended to initiate the foundation of more IMCs in various places in Africa. about the radical North-South imbalance of its growth (Halleck 2002). In 2003 the whole continent of Africa was represented by only four IMCs, while the UK alone had thirteen (Mamadouh 2004b, Frenzel & Sullivan 2009). Indymedia Centre Africa (IMCA) Mirroring the global „digital divide‟, the lack of IMCs in Africa was understood as a challenge to Indymedia by some of its members. Indymedia‟s claim to „globalize from below‟ suffered from the conspicuous dominance of Northern nodes in the network. At the same time ICT based media about the radical North-South imbalance of its growth (Halleck 2002). In 2003 the whole continent of Africa was represented by only four IMCs, while the UK alone had thirteen (Mamadouh 2004b, Frenzel & Sullivan 2009). Mirroring the global „digital divide‟, the lack of IMCs in Africa was understood as a challenge to Indymedia by some of its members. Indymedia‟s claim to „globalize from below‟ suffered from the conspicuous dominance of Northern nodes in the network. At the same time ICT based media promised to have a great potential to support processes of democratization in Africa. In 2003 following an initiative of activists from the IMC Ambazonia (Cameroon), an IMC run by Diaspora activists in Europe, four existing African IMCs, together with members from across the global network, founded IMC Africa (IMCA). IMCA was inspired by the idea to foster the development of Indymedia network nodes in Africa through support, knowledge exchange and networking. Initially a network based on an email listserv, IMCA soon started organizing for a conference in Senegal, to be held in March 2004. The aim of the conference was to bring together journalists and media activists from Africa, and to train them in the use of ICTs as independent media activists. This was intended to initiate the foundation of more IMCs in various places in Africa. Within the global Indymedia network this idea was controversial. The GJM had one of its origins in the movements against neo-liberal development policies imposed under the „Washington Consensus‟. The GJM was inspired by the idea to go „beyond development‟ (Escobar 1992; Esteva & Prakash 1998). Indymedia activists consequently showed some reservations to embark on a project that tried to „bring‟ the idea of Indymedia to Africa, being highly conscious about the colonial undertone of this approach. To operate with 22 metaphors like „open space‟, used frequently in ICT and GJM discourses, in post-colonial contexts had already been criticized as „double colonization‟ (Chesher 1994; Frenzel & Sullivan 2009; Flatz 1999). Indymedia Centre Africa (IMCA) Significantly, a funding request to Indymedia global for delegates‟ plane tickets was blocked by the IMC Uruguay on the grounds that linking the creation of new IMCs to a transfer of funds from the global Indymedia would create „dependent‟ rather than „independent media‟ (Indymedia 2004). The project had support from the existing African IMCs and from many other IMC and went ahead without the financial support from Indymedia global. The activists in IMCA organized their first conference in Dakar in 2004 and two subsequent conferences in 2006 and 2007 in Bamako and Nairobi. From the outset, IMCA faced the challenge of a lack of equivalent resources in the African context. To combat this, IMCA adopted models more closely associated with conventional NGO funding structures. Funding was sought and found through donors active in ICT and North-South cooperation, which were used to finance travel costs of African participants. IMCA also often relied on a broader set of local partners, involving NGOs, development institutions and universities. This caused a series of controversies between Northern and Southern participants in the IMCA and with the broader Indymedia network. In the 2006 IMCA conference in Bamako, Mali, some local organizing work in preparation of the conference was done from the offices of a media NGOs called Geek corps, whose operations were financed by USAid, the US government development agency. This involvement caused strong concerns among Indymedia members from Latin-America: 23 “I was very disturbed with the information that IMC Bamako room was co-financed by USAID (see story below from liege.indymedia). USAID was one of the main US agencies financing ideological apparatus during Latin American dictatorships in the 60s and 70s (Indymedia 2006). Despite a level of misunderstanding causing this particular conflict (IMC Africa did not receive financial support from USAid), this conflict had a base in very different organizational identities of IMCs in Latin America and West Africa. Local organizers in Mali had very little concern about working with USAid, as the French government‟s post-colonial influence in this part of Africa was considered much more problematic than US influence. Despite the increasing understanding and reflection within IMCA and Indymedia regarding particularities of the African context, conflicts occurred repetitively over organizational patterns and funding during its time existence often infringing on the working of the project. Indymedia Centre Africa (IMCA) As we attempt to do in the section that follows, these features of IMCA‟s history enables us to tease out some strands relevant for possible further analysis, reflection and comparison. It will look particularly at the ways these conflicts emerged during IMCA conferences between Northern and Southern participants. Organizational Identity Participants agreed to a framework in which the conferences were to be radically democratic, aspiring to relatively non-hierarchical organizational and communications structures, following not simply the „principles of unity‟ but more implicit, everyday 24 understandings of politics. For some Northern participants this included the politics of house-keeping, cooking, washing dishes and other tasks which are frequently gendered as part of a feminized domestic sphere. As some of the Southern (female) participants wryly expressed, „African men‟ were not used to the idea of doing their dishes, while Northern male participants were likely to at least pay lip service to the importance of these activities. The shared rejection of global capitalism and its consequences did not extent to a shared discourse on politics of the everyday. In other instances political cultures clashed as activists from the North who self-identified as „queer‟, met openly homophobic participants from the South. One Northern participant expressed in an interview: “(some) people have been guided by this idea that gayness, queerness, homosexuality that is bad, that is wrong that is evil. So that has been really hard, that has been hard.” understandings of politics. For some Northern participants this included the politics of house-keeping, cooking, washing dishes and other tasks which are frequently gendered as part of a feminized domestic sphere. As some of the Southern (female) participants wryly expressed, „African men‟ were not used to the idea of doing their dishes, while Northern male participants were likely to at least pay lip service to the importance of these activities. The shared rejection of global capitalism and its consequences did not extent to a shared discourse on politics of the everyday. In other instances political cultures clashed as activists from the North who self-identified as „queer‟, met openly homophobic participants from the South. One Northern participant expressed in an interview: “(some) people have been guided by this idea that gayness, queerness, homosexuality that is bad, that is wrong that is evil. So that has been really hard, that has been hard.” understandings of politics. For some Northern participants this included the politics of house-keeping, cooking, washing dishes and other tasks which are frequently gendered as part of a feminized domestic sphere. Organizational Identity As some of the Southern (female) participants wryly expressed, „African men‟ were not used to the idea of doing their dishes, while Northern male participants were likely to at least pay lip service to the importance of these understandings of politics. For some Northern participants this included the politics of house-keeping, cooking, washing dishes and other tasks which are frequently gendered as part of a feminized domestic sphere. As some of the Southern (female) participants wryly expressed, „African men‟ were not used to the idea of doing their dishes, while Northern male participants were likely to at least pay lip service to the importance of these activities. The shared rejection of global capitalism and its consequences did not extent to a shared discourse on politics of the everyday. In other instances political cultures clashed as activists from the North who self-identified as „queer‟, met openly homophobic participants from the South. One Northern participant expressed in an interview: “(some) people have been guided by this idea that gayness, queerness, homosexuality that is bad, that is wrong that is evil. So that has been really hard, that has been hard.” Some of the perceptions of African poverty held by Northern participants clashed considerably with Southern perspectives. Some Northern participants pressed the group for close relations with „slum dwellers‟ and were actively searching for projects that engaged with slums. In Nairobi there was a direct conflict when some Northern participants questioned the gated community type of location of the conference and asked for a relocation of the project “into the slums”. For African participants in the project, it involved “glorifying poverty and glorifying ghetto life” which expressed the naivety of Northern participants. One participant from Nairobi explained this point in more depth: “. . in the western context if you wear very very dirty jeans, you have a cap with Che Guevara on it, and you hang out in these abandoned houses then you really feel you are separate from society and you are doing something.(…) in Nairobi Some of the perceptions of African poverty held by Northern participants clashed considerably with Southern perspectives. Some Northern participants pressed the group for close relations with „slum dwellers‟ and were actively searching for projects that engaged with slums. In Nairobi there was a direct conflict when some Northern for close relations with „slum dwellers‟ and were actively searching for projects that engaged with slums. Organizational Identity In Nairobi there was a direct conflict when some Northern participants questioned the gated community type of location of the conference and asked for a relocation of the project “into the slums”. For African participants in the project, it involved “glorifying poverty and glorifying ghetto life” which expressed the naivety of Northern participants. One participant from Nairobi explained this point in more depth: “. . in the western context if you wear very very dirty jeans, you have a cap with Che Guevara on it and you hang out in these abandoned houses then you really 25 you are never going to find an abandoned house to hang out in and if you want to see real change (…) (y)ou then need to work closely with Kenya Indymedia to understand this whole notion of poverty(…)” The request to relocate to the slums arguably indicated an antipathy towards working with African activists who were middle class. It seemed that the Northern participants were actively searching to find in their African partners the images of Africa they brought with them. Organizational Form Following the experiences of difficulties in aligning organizational form between different contexts, IMCA came to the conclusion that participants first needed to learn about forms of horizontal decision making. In an email exchange during the preparation process of the Nairobi IMCA conference an organizer said: “Since at the core of the whole work of INDYMEDIA IS THE CONCEPT OF HORIZINTAL/PARTICIPATIVE DEMOCRACY, X will run workshops before the 13th. The hope is that the skills acquired in how to organize along consensus lines could be put to practice in all the workshops in tech skills.” (Indymedia 2007, Capitalization in original) The focus on horizontality led to some Northern participants assuming powerful roles of teachers and instructors in the convergence. By doing this, IMCA arguably replicated developmental approaches whereby an advanced Northern mode of organization is to be diffused into Africa by Northern experts. One interviewee reflected on the consequences of this approach: 26 “I think to an extent a lot of us in that first week had our work clearly cut out, but some of us didn‟t really have it clearly cut out. And then at that foundational level “I think to an extent a lot of us in that first week had our work clearly cut out, but some of us didn‟t really have it clearly cut out. And then at that foundational level 26 they pretty much established themselves in that mentality of receiving. I think you can bear me witness that that seems to have endured pretty much up to the very last day.” As Northern teachers attempt to instruct Southern participants about horizontal and participatory organizational forms, there was a sense that they some Southern participants became „spectators‟ in the process. Accountability The fact that most of the funding for IMCA came from the North might be not surprising. This however led to imbalances between participants, even within the relatively informal setting of IMCA. Some Southern participants often asked for resources from the group to pay for mobile phone costs, transport expenses, and copy costs. For many Northerners these requests were questionable and even „selfish‟, because they were seen as private consumption needs rather than needs for the project. This notion of selfishness is important as it belies Northerner‟s assumption that political work should be „unselfish‟ or charitable. The radically differing understandings of what constituted legitimate use of funds for political work made it difficult to operate within formally horizontal structures like a finance committee. One Nairobi activist thought that for many African participants certain organizers appeared “like gatekeepers of the money. (…) There are resources there but there are people standing in the way so we have to push them so we can reach these resources.” In the process, activists from the North became conscious about the material basis of their „unselfishness‟. A Northern activist said “So here I am from the US, I am a white female, and I am coming with a lot of tools, a lot of electronics, a lot of 27 media and so that just sets up a whole series of dynamics and conversations and things to sort through.” media and so that just sets up a whole series of dynamics and conversations and things to sort through.” Discussion In the case of IFIwatchnet, many of the Northern founders of IFIWatchnet understood it to be a global network with a relatively informal and non-hierarchical organizational form with informal trust based forms of accountability. In contrast, the new Southern hosts understood IFIWatchnet to be a technical platform that should be organized in a more hierarchical way in order to comply with more formal modes of accountability. In the case of IMCA, we found that Northern participants saw it as a radical political movement that was informally or horizontally organized and should be based on passionate political accountability. In contrast, Southern participants saw IMCA as more of an non-government organization (NGO) that involve more formal structures which has restrictive forms of accountability controlled by Northern participants. Although there are significant differences in both cases, there are some striking aspects of similarity. Northern participants appear to seek alternative media as having a radical identity, being organized in a more informal and horizontal way and having more trust or commitment based forms of accountability. In contrast Southern participants tended to emphasize a more „mainstream‟ identity, formalized modes of organization, and modes of accountability more tied to foundations and donors. The organizational characteristics which we found among Northern participants very much reflects other accounts of alternative media participants who see themselves as agents of radical social change, 28 work with decentralization and flexible networks, and use trust-based forms of accountability (eg. Atton 2002). However, Southern participants approached alternative media in a very different way that could be said to more closely resemble „mainstream‟ media. work with decentralization and flexible networks, and use trust-based forms of accountability (eg. Atton 2002). However, Southern participants approached alternative media in a very different way that could be said to more closely resemble „mainstream‟ media. There are at least three possible reasons we can identify for these differences between the Northern and Southern activists orientation to alternative media. The first reason that comes up clearly in each of the cases is the differential access to resources. Northern activists tended to rely on a broader range of sources of funding and other resources, such as private means of support, help from friends, paid part-time jobs, as well as donor funding. Discussion In contrast, Southern activists frequently relied heavily – often exclusively – on formal sources of funding, such as foundations, and developed an organizational identity that was more akin to formal organizational and accountability structures. The second reason for this difference is the differing activist cultures that media activists work within. The Northern activists tend to be part of activist cultures that emphasize informal trust relations, broad participation, and radical democratic involvement based on temporary projects. In contrast, many of the Southern activists were immersed in activist sub-cultures which tended to operate through more formal NGO structures. These two cultures were reflected in how activists sought to build up alternative media organizations. The final reason for these divergences in Northern and Southern orientation to alternative media is the different political opportunity structures which media activists face in the North and South. Most of the Northern media activists faced relative benign opportunity structures that are characteristic of liberal societies that 29 provided them significant space for airing their views and engaging in political action. In contrast, Southern activists typically faced more difficult opportunity structures that may have limited political action and the airing of their political voices. The different assumptions about identity, form, and accountability in conjunction with the different patterns of resource distribution, activist culture and political opportunity structure created significant barriers for co-operation between Northern and Southern activists. In the case of IFIWatchnet, there was significant clash around how the network should be managed and whether more attention should be paid to building a technical platform or fostering network interaction. Failure to resolve this ultimately led to many participants losing interest in the project and moving onto new networks and media sharing tools. In the IMCA case, the differences created some conflicts around where events were held, how they should be run, and who should have access to resources. Although there was not the kind of loss of interest we found in the IFIWatchnet case, there were ongoing concerns that infused the project. Conclusion In this paper, we have investigated the difference between alternative media in the North and South and how these differences influence co-operation. Focusing on IFIWatchnet and IMCA, we found that Northern participants were broadly oriented towards more „horizontal‟ modes of organizing while Southern participants tended to draw on more 30 „vertical‟ or hierarchical modes of organizing. This suggests we need to move away from broad claims that attribute particular organizational characteristics to alternative media throughout, independent of context. From our small sample, it appears that understanding alternative media in the South requires the investigation of a range of organizational identities, forms and structures of accountability. More broadly, this recognition of diversity offers a clear rebuke to recent techno-utopian visions of the Internet that tend to see it as creating a natural drive towards horizontal forms of social interaction and collaboration (eg. Shirky, 2008). Based on our current study, it appears that the network modes of organizing are actually primarily the providence of Northern activists. In the South, more „vertical‟ models of organization appear to have been used to put new ICTs to work. The different assumptions of Northern and Southern participants had a profound influence on co-operation. The lack of a shared identity, organizational form and accountability can produce misunderstandings and conflicts little known within local organizing contexts of „autonomous geographies‟ (Pickerill & Chatterton 2006). We highlighted how conflicts over projects, the distribution of funding, and the broader survival of the network were shaped by clashing assumption and modes of organizing. This meant that North-South co-operation was often underpinned by deeper struggles around what each of the organizations were and how they should operate. Furthermore, there was a propensity to overlook differences in alternative media networks among participants. We assume that Northern activists tend to understand their preferred organizational forms as being democratically advanced and „naturally‟ facilitating global 31 involvement and co-operation. However these „horizontal‟ structures clash with the locally specific needs of many Southern participants to garner resources through funding from more formal NGOs and foundations. What this suggests is that far from being a natural basis for facilitating spontaneous co-operation, the horizontal structures championed by many media activists may actually be a barrier to co-operation in some cases. Equally „vertical‟ modes of organization that are encouraged by various factors in the South can certainly be an important barrier to co-operation. Conclusion But Northern activists clinging to the assumption that horizontal modes of organizing are naturally superior at facilitating co-operation may hinder co-operation as well. High levels of commitment to the ideals of Horizonalism can mean that Northern alternative media organizations overlook the particularity of their contexts. The result is that co-operation might reproduce developmental logics and unequal power relations between Northern and Southern participants. An outcome akin to what has been criticized as „double colonization‟ (Chesher 1994; Flatz 1999) might result from this. While it is important and somewhat inevitable that global networking of alternative media continues, this process may face significant difficulties. In particular these difficulties may arise out of different sets of assumption of what alternative media organization is and how it should be organized. We should be clear that our sample is limited to two networks that involve Northern and Southern activists. This makes our broader claims about North-South differences somewhat initial. In order to consider whether our findings are more general, it would be vital to examine other North-South co-operation in alternative media. By considering other cases, it would be possible to 32 explore the dynamics of co-operation and conflict. Future studies should reflect upon the issues that came to the forefront in this study to complement the research presented here. In this way, research can play an active role in enhancing and improving future co- operations. Alternative media will have to join into the ongoing learning processes of meetings like the World Social Forum (WSF) where many of the issues discussed have been a constant feature of organization (Boehm et al. 2005; Wallerstein 2004; Keraghel & Sen 2004). We would argue that it is only by learning from these problematic encounters that real progress towards higher degrees of global co-operation can be achieved. References Appadurai A, 2001 Globalization (Duke University Press, Durham NC) Atton C, 2002 Alternative media (Sage, London; Thousand Oaks, CA) Bennett WL, 2003 "New Media Power: The Internet and Global Activism", in Contesting media power : alternative media in a networked world Eds N Couldry, J Curran (Rowman & Littlefield, Lanham MD) Boehm S, Sullivan S, Reyes O, 2005. The Organisation and Politics of Social Forums, ephemera: theory & politics in organization http://www.ephemeraweb.org/journal/5-2/5-2index.htm Chesher C, 1994 "Colonizing Virtual Reality" http://cultronix.eserver.org/chesher/ Couldry N, Curran J, eds 2003 Contesting media power : alternative media in a networked world (Rowman & Littlefield , Lanham MD) Coyer K, Dowmunt T, Fountain A, 2007 The alternative media handbook (Routledge, London; New York) Cresswell T, 1996 In place out of place : geography, ideology, and transgression, (University of Minnesota Press, Minneapolis) Downing J, 2001 Radical media : rebellious communication and social movements (Sage Publications, Thousand Oaks CA) 33 Downing J, 1984 Radical media : the political experience of alternative communication 1st ed. (South End Press, Boston MA) Downing J, Mohammadi A, Sreberny A, eds 1995 Questioning the media : a critical introduction 2nd ed (Sage Publications, Thousand Oaks CA) Eisenhardt K, 1989, "Building Theories from Case Study Research" Academy of Management Review 14(4) 532-545 Enzensberger H, 1974 The consciousness industry on literature, politics and the media (Seabury Press: New York) Escobar A, 1992, "Imagining a Post-Development Era? Critical Thought, Development and Social Movements" Third World and Post-Colonial Issues Social Text 31/32 20-56 Esteva G, Prakash M S, 1998, "Beyond development, what?" Development in Practice, 8(3) 280-296 Flatz C, 1999, "Africa in the fast lane of the highway development?" [Orginal German]. Available at: http://translate.google.co.uk/translate?js=y&prev=_t&hl=en&ie=UTF- 8&layout=1&eotf=1&u=http%3A%2F%2Fwww.heise.de%2Ftp%2Fr4%2Fartikel %2F5%2F5459%2F1.html&sl=auto&tl=en [Accessed June 25, 2010]. Frenzel F, Sullivan S, 2009, "ICTs and Democratic Development in the Project IMC Africa" in African Media and the Digital Public Sphere Eds O F Mudhai, F Banda, W Tettey (Palgrave Macmillian, Basingstoke ; New York) Gitlin T, 1980 The whole world is watching : mass media in the making & unmaking of the New Left (University of California Press Berkeley) Gumucio-Dagron A, 2004 "The Long and winding road to alternative media" in The SAGE handbook of media studies ed J Downing (Sage, Thousand Oaks CA) Halleck D, 2002, “Indymedia: Building an International Activist Internet Network” International Symposium of Interactive Design. Available at: http://newmedia.yeditepe.edu.tr/pdfs/isimd_04/12.pdf [Accessed October 31, 2010]. References Halloran J, 1970 Demonstrations and communication : a case study (Penguin Books, Harmondsworth) 34 Hamilton J, Atton C, 2001, "Theorizing Anglo-American Alternative Media: toward a contextual history and analysis of US and UK scholarship" Media History 7(2) 119-135 Hannan M, Freeman J, 1984, "Structural Inertia and Organizational Change" American Sociological Review 49(2) 149-160 Kemmis S, McTaggart R, 2005, "Participatory action research" in The SAGE handbook of qualitative research eds N K Denzin, Y S Lincoln (Sage Publications, Thousand Oaks) 350-375 Indymedia, 2004, “[IMC-Seattle] FYI [Imc-finance] Proposal 2004-B Dakar Conference request BLOCKED”, Archive of Indymedia Mailing Lists, <http://archives.lists.indymedia.org/imc-seattle/2004-March/004732.html> Indymedia, 2006, “[Imc-communication] USAID”, Archive of Indymedia Mailing Lists, <http://lists.indymedia.org/pipermail/imc-communication/2006-January/0121-y7.html> Indymedia, 2007 “[Imc-africa] Information from Nairobi” Archive of Indymedia Mailing Lists, <http://lists.indymedia.org/pipermail/imc-africa/2006-October/1005-lr.html> Keraghel C, Sen J, 2004, "Explorations in open space. The World Social Forum and cultures of politics" International Social Science Journal 56(182) 483-493 Kidd D, 2003,” The Independent Media Center: A new model” Journal of the World Association for Christian Communication, <http://www.waccglobal.org/en/20034- indymedia/619-The-Independent-Media-Center-A-new-model.html>. Kim E, Hamilton J, 2006, "Capitulation to capital? References OhmyNews as alternative media" Media, Culture & Society 28(4) 541-560 Kindon S L, Pain R, Kesby M, Eds 2007, Participatory action research approaches and methods : connecting people, participation and place (Routledge, London ; New York) Kueng-Shankleman L, 2000 Inside the BBC and CNN : managing media organisations (Routledge, London) Lovink G, 2002 Dark fiber : tracking critical Internet culture (MIT Press, Cambridge MA) Mamadouh V, 2004a, "Internet, Scale and the Global Grassroots" Tijdschrift voor Economische en Sociale Geografie 95(5) 482-497 35 Mamadouh V, 2004b, “The North South digital divide in transnational grassroots networks: Open publishing and the Indymedia Network”, International Conference ICTs & Inequalities : the digitial divides, http://irene.asso.free.fr/digitaldivides/papers/MAMADOUH_Virginie.pdf McQuail D, 1986 New media politics : comparative perspectives in Western Europe (SAGE Publications, London) Meikle G, 2002 Future active : media activism and the Internet (Pluto Press/Routlegde, New York NY) Mudhai O F, Banda F, Tettey W, Eds, 2009 African Media and the Digital Public Sphere (Palgrave Macmillian, Basingstoke; New York) O‟Sullivan T, ed, 1994 Key concepts in communication and cultural studies 2nd ed (Routledge, London; New York) Park M, Curran J, 2000 De-Westernizing media studies (Routledge, London ;New York) Pickard V W, 2006a, “United yet autonomous: Indymedia and the struggle to sustain a radical democratic network”, Media, Culture & Society 28 (3) 315 -336 Pickard V W, 2006b, "Assessing the Radical Democracy of Indymedia: Discursive, Technical, and Institutional Constructions" Critical Studies in Media Communication 23(1) 19-38 Pickerill J, 2007, "„Autonomy online‟: Indymedia and practices of alter-globalisation" Environment and Planning A 39(11) 2668 – 2684 Pickerill J, Chatterton P, 2006, "Notes towards autonomous geographies: creation, resistance and self-management as survival tactics" Progress in Human Geography 30(6) 730-746 Ragin C C, 1992, " „Casing‟ and the process of social inquiry" in What is a case? References : exploring the foundations of social inquiry Eds C C Ragin, H S Becker (Cambridge University Press, Cambridge; New York) 217-226 , 2001 Fissures in the mediascape : an international study of citizens’ media, (Hampton Press, Cresskill NJ) Routledge P, 1997, "The imagineering of resistance: Pollock Free State and the practice of postmodem politics" Transactions of the Institute of British Geographers 22(3) 359-376 36 Schedler A, 1999, "Conceptualizing accountability" in The self-restraining state : power and accountability in new democracies ed A Schedler (Rienner, Boulder CO) Scott W, 1995 Institutions and organizations (Sage, Thousand Oaks CA) Shirky C, 2008 Here comes everybody: the power of organizing without organizations. (Penguin Press, New York) Spicer A, Perkmann M, 2008 "Translating an organizational form: The case of indymedia" paper presented at McGill University, Canada, June 2008; copy available from Andre Spicer, Warwick Business School, Coventry Sullivan S, 2008, "Conceptualising glocal organisation: from rhizome to E=mc2 in becoming post-human" In Metaphors of globalization : mirrors, magicians and mutinies eds M. Kornprobst et al (Palgrave Macmillan, Baskingstoke van de Donk W, ed, 2004 Cyberprotest : new media, citizens, and social movements (Routledge, London; New York) Wallerstein I, 2004, "The dilemmas of open space: the future of the WSF" International Social Science Journal 56(182) 629-637 Whetten D, 2006, "Albert and Whetten Revisited: Strengthening the Concept of Organizational Identity" Journal of Management Inquiry 15(3) 219-234 Schedler A, 1999, "Conceptualizing accountability" in The self-restraining state : power and accountability in new democracies ed A Schedler (Rienner, Boulder CO) Scott W, 1995 Institutions and organizations (Sage, Thousand Oaks CA) Shirky C, 2008 Here comes everybody: the power of organizing without organizations. (Penguin Press, New York) Shirky C, 2008 Here comes everybody: the power of organizing without organizat (Penguin Press, New York) Wallerstein I, 2004, "The dilemmas of open space: the future of the WSF" International Social Science Journal 56(182) 629-637 Whetten D, 2006, "Albert and Whetten Revisited: Strengthening the Concept of Organizational Identity" Journal of Management Inquiry 15(3) 219-234 37 37

https://openalex.org/W2094628462

https://zenodo.org/record/1270173/files/article.pdf

English

Color matching and illumination estimation for urban scenes

public-domain

1Historic images are from atlanta history center. Abstract Photographs taken of the same scene often look very dif- ferent, due to various conditions such as the time of day, the camera characteristics, and subsequent processing of the image. Prime examples are the countless photographs of urban centers taken throughout history. In this paper we present an approach to match the appearance between pho- tographs that removes effects such as different camera set- tings, illumination, fading ink and paper discoloration over time, and digitization artifacts. Global histogram matching techniques are inadequate for appearance matching of com- plex scenes where background, light, and shadow can vary drastically, making correspondence a difficult problem. We alleviate this correspondence problem by registering pho- tographs to 3D models of the scene. In addition, by estimat- ing the calendar date and time of day, we can additionally remove the effect of drastic lighting and shadow differences between the photographs. We present results for the case of urban scenes, and show that our method allows for re- alistic visualizations by blending information from multiple photographs without color-matching artifacts. Figure 1. We want to visualize urban scenes evolving over time, given a collection of images taken by different cameras from dif- ferent viewpoints at different times. In the above example scene, not only have the images been captured under different conditions, but the geometry has also changed over the years 1966 to 1978. applied structure from motion to recover camera poses and a 3D point cloud from 2D images correspondences. Most recently, Schindler et al. [13] expanded the scope towards modeling time-varying 3D structures from historical pho- tographs in the 4D Cities project. Color Matching and Illumination Estimation for Urban Scenes Mingxuan Sun, Grant Schindler, Greg Turk, and Frank Dellaert Georgia Institute of Technology {cynthia,schindler,turk,dellaert}@cc.gatech.edu Figure 1. We want to visualize urban scenes evolving over time, given a collection of images taken by different cameras from dif- ferent viewpoints at different times. In the above example scene, not only have the images been captured under different conditions, but the geometry has also changed over the years 1966 to 1978. 1. Introduction We are interested in color-matching images of urban scenes for the purpose of visualizing them across space and time. In the urban context, where there is a richness of his- torical images, we are specifically interested in the visual- ization of both large and small temporal changes in a city from any chosen viewpoint. Examples include observing the changes of the evolving cityscape over many decades, and observing a cityscape changing from sunrise to sunset on an arbitrary day. Unfortunately, visualizing an urban scene using images taken with different cameras at different times of day, at different times of the year, and in different historical eras is challenging due to many photometric inconsistencies be- tween images. Figure 1 shows an example of five such images1, illustrating the difficulties. In traditional image- based rendering, the scene is relatively well controlled and the image collections are structured, which means the scene is captured densely by a set of images that are uniformly distributed around the scene in a relatively short period of time [3, 18, 17, 6]. Most previous work assumes constant scene brightness and small geometry changes over a short time, which can hardly be applied to large-scale dynamic scenes like cities over a hundred years. Modeling and interactively visualizing large-scale 3D urban scenes from images has been a very successful ap- proach, as evidenced by commercial programs such as Google Earth and Microsoft’s Virtual Earth, which speak to the public imagination. The Facade project by Debevec et al. [4] introduced view-dependent texture mapping for vir- tual view synthesis, while Photo Tourism [14] successfully 1566 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops U.S. Government work not protected by U.S. copyright 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops U.S. Government work not protected by U.S. copyright (a) Source Image (b) Target Image (c) Global Statistics (d) Local Statistics Figure 2. First Row: Left: source image (through red filter), Right: target image (no filter). Bottom row: Left: color matching based on global color statistics, which does not transfer local color cor- rectly. Right: color matching based on local color checker corre- spondences. (b) Target Image (a) Source Image In order to visualize a scene at any time from an arbi- trary viewpoint, we would like to recover the reflectance properties (e.g. albedo) of structures in the scene. This task is made difficult by changes in illumination, shadow- ing, 3D structure, and camera color properties. When deal- ing with a limited number of historical images captured by dramatically different cameras at unknown dates and times, it is especially hard to infer the photometry of the scene. Our goal is to recover a uniformly-shaded, shadow-free and photometrically-consistent scene albedo from all the im- ages. To achieve our goal, the underlying geometry and sun position for each image are needed to efficiently extract and combine textures from images. (b) Target Image (a) Source Image (c) Global Statistics (d) Local Statistics In this paper, to visualize a scene from an entire collec- tion of images, we select one photo as the target color tone, and perform color matching of all the other photos to this target, using scene geometry to find corresponding facades across multiple images for local color matching, and using of an estimated sun position for each photo to correct for the varying amounts of sunlight that fall on each facade. (c) Global Statistics (d) Local Statistics Figure 2. First Row: Left: source image (through red filter), Right: target image (no filter). Bottom row: Left: color matching based on global color statistics, which does not transfer local color cor- rectly. Right: color matching based on local color checker corre- spondences. We first address the color matching problem in section 2, where we find that global color matching methods are insufficient. In our urban scene application, where sky ap- pears in most of the outdoor images, global color mapping over the entire image can not transfer the local object col- ors consistently between images. Instead we perform local color mapping by identifying facades of the same building that are similarly lit in both images. 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops U.S. Government work not protected by U.S. copyright Thus, we introduce the geometry model in section 3, which we use in our local color matching technique. across each corresponding region. A third approach [10], used in the video conferencing domain, learns a color map- ping function for human faces and then applies this map- ping to the whole image. Some other approaches [2, 7, 16] learn an consistent color from multiple images or multiple parts of an image for 3D scenes. In addition, the problem of learning a good color mapping function from a large image database [9] has been studied outside the 4D scene visual- ization context. Based on the scene geometry, we propose a novel method to estimate sun position from images with unknown time stamps in section 4. Furthermore, we can estimate the date and time of an image if we know the approximate ge- ographical position of the scene in the image. And then in section 5, with the estimated sun position for each im- age, we can identify the shading and shadow information in the images and generate a consistent set of object albedos. This information also allows us to match object reflectances between photos, which lets us perform better color transfer between images. Finally in section 6, by adding the shading and shadow information from a virtual sun, we can visualize the scene from an arbitrary view at an arbitrary time. The simplest method of performing color transfer is his- togram matching using global image statistics in each color channel. This method matches the mean and variance of the source image S to those of the target image D in each color channel c. For each channel, let (mS,σS) denote the mean and variance of the source image S and (mD,σD) for the tar- get image D. The color value is normalize ranging from 0 to 1. The color transferring function f(x), which map the pixel value x of the source image S towards target image D in each channel c should satisfy these constraints: f(mS) = f(mD) f ′(mS) = σD/σS (1) (1) 2. Color Matching Using Color Statistics The simplest such mapping is the linear mapping [12]: The simplest such mapping is the linear mapping [12]: We are interested in performing color matching between images for the purpose of visualizing a 4D urban scene from historical images. Color transfer is a widely used method applied to images or videos to make the color tone of one image look like the other. Some techniques [12] transfer global color image statistics (mean and variance) between two images, while others [15] segment the images into sev- eral color clusters and then transfer color independently f(x) = σD σS (x−mS)+mD (2) (2) When portions of the scene are very dark or overly bright, the linear mapping will lead to saturation at lower and upper intensities. To avoid this, we fit a piecewise cubic spline function that satisfy the constrains in 1. In addition, we limit f(0) = 0, f(1) = 1 and constrain the derivatives of the 1567 Figure 3. Facades of the same albedo that are lit in both imag share histograms of similar shape, the color statistics of which c be used to form the color mapping function. Top image: Le facade in shadow, Right: facade lit. Bottom image: Left: faca lit, Right: facade in shadow. Each histogram shows red, green, a blue channel information (from top to bottom) for the individu facade regions. Figure 4. Time-Dependent Geometry. Scene geometry in both 1930 (left) and 1972 (right) is overlaid on two photos from their recovered 3D viewpoints. Observe that the geometry changes dra- matically over the years. Our illumination estimation and color matching relies on such a time-varying scene description to accu- rately model shadows and establish corresponding regions across images. Figure 4. Time-Dependent Geometry. Scene geometry in both 1930 (left) and 1972 (right) is overlaid on two photos from their recovered 3D viewpoints. Observe that the geometry changes dra- matically over the years. Our illumination estimation and color matching relies on such a time-varying scene description to accu- rately model shadows and establish corresponding regions across images. of the checker in the image to obtain the mapping function and then map the whole image, the results in the lower right of Figure 2 shows that color matching by using local corre- spondences works better. 2. Color Matching Using Color Statistics In the context of outdoor urban scene photos, in addi- tion to the geometry correspondence between images, we should also take shading and shadow information into ac- count because the color statistics of the same facade will change dramatically from when it is being lit to when it is in shadow. For example, figure 3 shows the histogram of two facades of one building in two different images. The correspondence of the same facade no longer describes the overall color changes for most of the scene. Only the fa- cades of the same albedo that are lit in both images indicate the overall color statistics changes between the two images. Figure 3. Facades of the same albedo that are lit in both images share histograms of similar shape, the color statistics of which can be used to form the color mapping function. Top image: Left: facade in shadow, Right: facade lit. Bottom image: Left: facade lit, Right: facade in shadow. Each histogram shows red, green, and blue channel information (from top to bottom) for the individual facade regions. Figure 3. Facades of the same albedo that are lit in both images share histograms of similar shape, the color statistics of which can be used to form the color mapping function. Top image: Left: facade in shadow, Right: facade lit. Bottom image: Left: facade lit, Right: facade in shadow. Each histogram shows red, green, and blue channel information (from top to bottom) for the individual facade regions. Figure 3. Facades of the same albedo that are lit in both images share histograms of similar shape, the color statistics of which can be used to form the color mapping function. Top image: Left: facade in shadow, Right: facade lit. Bottom image: Left: facade lit, Right: facade in shadow. Each histogram shows red, green, and blue channel information (from top to bottom) for the individual facade regions. Based on this observation, we color match between im- ages by using the lit facades that are present in both im- ages. To building up the spatial correspondences between facades in different images, we need the underlying geom- etry which will be introduced in section 3. Given the geom- etry model, we can further estimate the illumination in the outdoor scene in section 4 and take the scene illumination into account for color matching. 2. Color Matching Using Color Statistics two ends in a way similarly to [10]: f ′(0) = 0.5∗mD mS , f ′(1) = 0.5∗1−mD 1−mS (3) (3) However, using global image statistics cannot transfer local color correctly when background and lighting vary drastically between images. To illustrate the problem, we captured a scene containing a color checker with a cam- era using different color filters, and performed histogram matching between the images. Figure 2 (lower left) shows the results of performing global histogram matching for the color checker datasets. 4. Estimating Sun Direction from Images To account for strong illumination changes in outdoor urban scenes, we take an extra step and attempt to recover the dominant lighting in the scene by estimating the direc- tion of the sun. Most of our historic images do not have exact time stamps, but the buildings do show shading con- trast and shadows. We make use of the shadows cast by buildings in an image to identify the sun direction when the image was captured. Figure 6 illustrates the underlying ge- ometry which lets us infer the light direction from the 2D projection of the shadows in the 3D scene. Estimating the sun direction from images is important since it is the most dominant light source. Thus sun position provides consid- erable information for image-based modeling and rendering applications. Applications such as [19] use time-stamped images collected intensively through out a day from a cal- ibrated camera to compute the sun position and radiance changes, used to recover the reflectance properties in the scene. Similarly, Jacobs et. al [5] infer the geographical lo- cation of a static camera by examining how image intensity varies throughout the day. There are methods for estimating general lighting sources from cast shadows such as [1]. We define the time-dependent geometry or 4D model S4D as a set of 3D models with associated temporal information. Each 3D model S3D is represented as a set of 3D vertices {v j|j ∈1..N} and a set of polygons {fk|k ∈1..K} connect- ing these vertices. We do not attempt to model dynami- cally changing objects such as cars, or planes, etc. Instead, we aim to represent large-scale discrete changes such as buildings appearing and then disappearing at discrete times. Hence, the 4D model S4D we propose is simply a set of 3D models S3D r , each having an associated time interval tr, i.e., S4D Δ= {(S3D r ,tr)|r ∈1..R}, with R the number of 3D models in the scene. We build the 4D model and recover camera motion M from images I via structure from motion (SFM). We per- form SFM with manually matched correspondences to re- cover 3D positions of building corners which we then man- ually connect into solid polygonal building models S3D r . For each photograph we know the year in which it was captured Y, and for each building we know the years of construction and demolition tr from historical records. 3. Time-Dependent Geometry and Motion Though not shown above, individ- ual facade correspondences are established in the same manner. Figure 5. For each photograph (left), a corresponding building segmentation (right) is derived from time-varying geometry and recovered camera motion. Color coding shows the identity of each building and demonstrates how we establish corresponding regions across the two images, despite being captured years apart and from different viewpoints. Though not shown above, individ- ual facade correspondences are established in the same manner. 3. Time-Dependent Geometry and Motion In order to perform color matching, we require scene geometry for two reasons: (1) to estimate scene illumina- tion from shadows cast by the geometry, and (2) to estab- lish correspondences between image regions in two distinct photographs of the same scene that are taken from differ- ent viewpoints. Traditionally, a 3D model of the scene ge- ometry could be used for both the illumination and corre- spondence tasks (e.g. [8], which registers images to static 3D building models in order to remove haze). However, because we are matching photographs across decades of time during which new buildings are constructed and de- molished, we must make use of time-varying 3D geometry which we refer to as a 4D model. Instead of matching the color statistics of the whole im- age, we use color matching with local correspondences to transfer color between images. The mapping function in equation 2 can be learned by using the mean/variance of local correspondences (e.g. the color checker) in both im- ages, and this mapping can then be applied to every pixel in the whole image. For example, we use the color statistics 1568 Figure 5. For each photograph (left), a corresponding building segmentation (right) is derived from time-varying geometry and recovered camera motion. Color coding shows the identity of each building and demonstrates how we establish corresponding regions across the two images, despite being captured years apart and from different viewpoints. Though not shown above, individ- ual facade correspondences are established in the same manner. Figure 6. 3D cast shadows of buildings help identify the sun direc- tion, where the sun is modeled as a distant light source. Different sun positions will change the shadow areas dramatically, and we will observe different projected 2D shadows in the image accord- ingly. Figure 6. 3D cast shadows of buildings help identify the sun direc- tion, where the sun is modeled as a distant light source. Different sun positions will change the shadow areas dramatically, and we will observe different projected 2D shadows in the image accord- ingly. Figure 5. For each photograph (left), a corresponding building segmentation (right) is derived from time-varying geometry and recovered camera motion. Color coding shows the identity of each building and demonstrates how we establish corresponding regions across the two images, despite being captured years apart and from different viewpoints. 4. Estimating Sun Direction from Images age can be identified as lit by sun or not by the following ray-tracing procedure: first find a pixel’s 3D position by backprojecting the pixel onto the aligned 3D geometry, and then shoot a ray from this 3D point along the sun direction. If the ray is not occluded by other geometry, then the pixel is lit, otherwise it is in shadow. We then compute the sum of squared differences (SSD) between pixels in Ip and Im. To bias the search towards likely sun positions, we do not directly search the azimuth-elevation space but instead search over calendar date and time of day. For a specific geographical location, not all positions of the sun (θ,α) are possible due to the Earth’s movement relative to sun. The date within a year and the time of day uniquely determine a position of the sun with respect to a given location [11], so we use data/time instead of angles. Figure 7 shows the results for estimating sun directions by sampling the date and time of day for each image. In these experiments, we sampled once every 10 days out of a possible 365, and we sample time once per hour in 24 hours. We observe that when the sampling over day and time, the sun probability map is symmetric with respect to June 22, because the summer and winter solstices divide every year into two symmetric halves. Every day between June 22 and December 22 will have a corresponding day in the first half of the year when the sun takes the same appar- ent path through the sky. So in the date and time estimation, there is always an ambiguity between the two dates which correspond to the same elevation θ and azimuth α of the sun. As such, we can speed up the sun position estimation process by evaluating only half a year’s sun positions. For every date/time combination, we evaluate the corre- sponding sun direction (θ,α) by a simple sum of square dif- ferences (SSD) criterion. We denote the predicted shadow image as Ip and the real, measured shadow region in the images as Im. Both Ip and Im are binary images where the value at each pixel indicates whether the pixel is lit (Ip(x,y) = 1) or shadowed (Ip(x,y) = 0) in the original pho- tograph. 4. Estimating Sun Direction from Images (a) Image (b) Day/Time Probability (c) Measured Shadows (d) Predicted Shadows (d) Predicted Shadows (c) Measured Shadows (c) Measured Shadows Figure 7. Illumination Estimation. For an image with unknown time and date (a), we measure a shadow by thresholding the ground region in the image (c) and search over times of day and days of the year for a sun position that predicts the same shadow- ing pattern. The shadow regions predicted by the estimated most likely sun position are shown in (d). By comparing measured and predicted shadows, we generate a probability distribution (b) over days of the year (horizontal axis from left to right is Jan. 1 to Dec. 31) and times of day (vertical axis from top to bottom is 0 to 23 hours). We predict that the image is taken at Jan. 23, 10 am. Figure 7. Illumination Estimation. For an image with unknown time and date (a), we measure a shadow by thresholding the ground region in the image (c) and search over times of day and days of the year for a sun position that predicts the same shadow- ing pattern. The shadow regions predicted by the estimated most likely sun position are shown in (d). By comparing measured and predicted shadows, we generate a probability distribution (b) over days of the year (horizontal axis from left to right is Jan. 1 to Dec. 31) and times of day (vertical axis from top to bottom is 0 to 23 hours). We predict that the image is taken at Jan. 23, 10 am. Figure 8. Top: two historic images with unknown dates and times. Bottom: two photos with exif tags, both taken at around 3pm Nov. 9, 2003. We predict the time of day at 3pm and day of year with most probability at two peaks (Nov. 6 and Feb. 9 of the year). position that best predicts the cast shadows. The shadow in the image can be measured using intensity thresholding in regions of interests. For our initial experiments, an ap- proximate ground shadow map is obtained by thresholding the ground region in the image given the building segmen- tation. One can further manually mark a region of interest with better accuracy. 4. Estimating Sun Direction from Images Note that such 4D models can also be constructed using the more automated approach presented in [13]. Our goal is to find the best sun position given the image. We model the sun as a parallel light source defined by ele- vation θ and azimuth α (clockwise from north). In order for the sun to properly interact with our geometry, it is impor- tant that we align the local coordinate system in which we reconstructed the 3D model with a global coordinate system on the Earth’s surface; knowing the true physical location of any three points in the scene produces the required rotation, translation, and scaling from local to global coordinates. As an example, in the context of an urban scene con- sisting of buildings, our reconstructed 4D model is shown in Figure 4. Using this 4D model, each individual building and facade can be identified in each image. Figure 5 shows a building-based segmentation of the images, accounting for visibility and occlusion. The whole model (S4D,M,Y) that is obtained serves as a strong geometry proxy for sun direc- tion estimation and virtual view visualization. The main idea is to use the geometry as a sundial: we find cast shadows in the scene, and then search for the sun 1569 (a) Image (b) Day/Time Probability (c) Measured Shadows (d) Predicted Shadows Figure 7. Illumination Estimation. For an image with unknown time and date (a), we measure a shadow by thresholding the ground region in the image (c) and search over times of day and days of the year for a sun position that predicts the same shadow- ing pattern. The shadow regions predicted by the estimated most likely sun position are shown in (d). By comparing measured and predicted shadows, we generate a probability distribution (b) over days of the year (horizontal axis from left to right is Jan. 1 to Dec. 31) and times of day (vertical axis from top to bottom is 0 to 23 hours). We predict that the image is taken at Jan. 23, 10 am. Figure 8. Top: two historic images with unknown dates and times. Bottom: two photos with exif tags, both taken at around 3pm Nov. 9, 2003. We predict the time of day at 3pm and day of year with most probability at two peaks (Nov. 6 and Feb. 9 of the year). 5. Color Matching with Geometry and Shading Correspondences When adjusting the color of the source image (a) to match the target (b), our local, lit facade correspondence technique (d) produces more accurate results than the traditional global color matching method (c). Figure 11. When adjusting the color of the source image (a) to match the target (b), our local, lit facade correspondence technique (d) produces more accurate results than the traditional global color matching method (c). Figure 9. For each photograph (left), a corresponding building seg- mentation with shading and shadow information is derived from time-varying geometry, recovered camera motion and sun direc- tions (middle). Shadow-free, shading corrected albedo texture are extracted (right) for color matching. 4. Estimating Sun Direction from Images Given the underlying scene geometry, for each sun direction (θ,α), we can easily predict the shadow map Ip. Given the geometry and camera pose, each pixel in the im- We verify this method on photographs with ground-truth 1570 (a) Source Image (b) Target Image (c) Global Color Match (d) Local + Lit Color Match Figure 10. When adjusting the color of the source image (a) to match the target (b), our local, lit facade correspondence technique (d) produces more accurate results than the traditional global color matching method (c). (b) Target Image (a) Source Image dates and times as shown in Figure 8. For each image, given the estimated sun position at a date and time, we can create a shading/shadow map for the scene at that time. We show how to make use of these results during color matching and virtual view synthesis in the following two sections. 5. Color Matching with Geometry and Shading Correspondences (b) Target Image (a) Source Image (d) Local + Lit Color Match (c) Global Color Match As discussed in section 2, we perform color matching between images using the statistics of lit facades that are in correspondence across images. The spatial correspon- dences of facades are identified using the underlying geom- etry. The sun position allows us to identify lit facades and to normalize the amounts of lighting falling on the facades in each image. This helps us to extract the uniformly lit fa- cades from each image that can be used as textures for scene visualization. (c) Global Color Match (d) Local + Lit Color Match Figure 10. When adjusting the color of the source image (a) to match the target (b), our local, lit facade correspondence technique (d) produces more accurate results than the traditional global color matching method (c). As shown in Figure 9, for each photo, we first select the facades that are lit by the sun. Then assuming the facade is Lambertian, we can determine the object albedo by divid- ing the per-pixel intensities on a given facade by the cosine between the facade’s normal and the direction to the sun. Then the mean and variance of the extracted shadow-free and shading corrected textures will be the constraints in 1 to form the color mapping function. (b) Target Image (a) Source Image (a) Source Image (b) Target Image (c) Global Color Match (d) Local + Lit Color Match Figure 11. When adjusting the color of the source image (a) to match the target (b), our local, lit facade correspondence technique (d) produces more accurate results than the traditional global color matching method (c). We then perform color transfer between the images based the color statistics of the extracted object albedos in both images. The mapping results are shown in Figure 10 and 11. We can see that the facades are better matched be- tween images when we use the local correspondences. (a) Source Image (b) Target Image Figure 9. For each photograph (left), a corresponding building seg- mentation with shading and shadow information is derived from time-varying geometry, recovered camera motion and sun direc- tions (middle). Shadow-free, shading corrected albedo texture are extracted (right) for color matching. (c) Global Color Match (d) Local + Lit Color Match Figure 11. 6. Virtual View Synthesis at a Virtual Time and Results One of our main goals is to visualize the reconstructed urban scene from different viewpoints and lighting condi- tions. In order to construct high quality object albedos (tex- 1571 tures) for the various building facades, we select the texture detail for a given facade from the image that best shows that particular building face. This means that different fa- cades come from different photos, making it necessary to match colors across these photos. To generate object albe- dos for the entire collection of images, we select one photo as the target color tone, and perform color matching of all the other photos to this target. We then make use of the known sun position for each photo to correct for the vary- ing amounts of sunlight that fall on each facade, making a Lambertian assumption about the facade reflectance. Note that we only select facades that are lit by the sun in order to generate facade albedos. Given the source images in Fig- ure 1, Figure 12 (top) shows how poorly mismatched the facades are if color matching and illumination are ignored. Figure 12 (bottom) shows the color matched and illumina- tion corrected versions of the facade albedos. Note that the models in this scene are not lit by any virtual lights, so that the image only shows the per-object albedos. applicable to non-Lambertian surfaces, such as buildings that have glass-dominated facades. Working with scenes that contain such non-Lambertian objects is a challenging area for future research. In conclusion, visualizing large-scale dynamic scenes from historic images is an exciting but challenging task. For a large-scale 3D scene with many occlusions, an accurate estimation of geometry is already difficult due to the accu- racy of image feature matching and SFM methods. When extended to 4D, additional techniques are required to iden- tify the changing structures and the time when each image is taken. We have shown above that as the scene changes through time, the photometry of the scene is even harder to estimate due to the unknown camera photometry model and illumination conditions. We have presented novel illumina- tion estimation and facade-based color matching techniques that form a valuable contribution toward solving the difficult problem of 4D scene visualization. With the color matched and illumination corrected fa- cade albedos, we can now create new views of the recon- structed urban scene. 6. Virtual View Synthesis at a Virtual Time and Results These views may be from any angle, with arbitrary illumination, and at any historic time that is represented in our 4D urban model. To synthesize the scene on a given date and time, we first compute the sun position at that time and then use this information to correctly illu- minate each object. To demonstrate this, we have created a synthetic time-lapse video of an urban scene from sunrise to sunset on Nov. 6, and snapshots from this video are showed in Figure 13. Note that the path of the virtual sun is a close match to the true sun path for that given date. Figure 12. Top: blindly combined textures. Bottom: shadow-free, uniform-shaded, color matched object albedos. 7. Conclusion We have demonstrated an approach that uses estimation of sun position and correspondences of facades across pho- tographs to build and visualize 4D models of urban scenes. The illumination estimation and facade correspondences are vital for performing color matching across photographs that were taken with un-calibrated cameras in different years or decades. Our method allows us to create the facade tex- tures for 4D models of these urban scenes, and this makes it possible to visualize such a scene from any viewpoint, date and time. Moreover, our tools also allow us to give time-of- year estimates of undated photographs and to perform color matching between dissimilar photos. Figure 12. Top: blindly combined textures. Bottom: shadow-free, uniform-shaded, color matched object albedos. There are a few limitations to our approach that we plan to explore in the future. First, we have largely ignored the effects of sky illumination apart from the sun, and account- ing for this should give improved results. Related to this, our illumination correction makes the assumption that the urban scene is directly lit by the sun, and as such does not apply to photos taken on heavily clouded days. In addition, our illumination estimation and albedo determination is not Figure 12. Top: blindly combined textures. Bottom: shadow-free, uniform-shaded, color matched object albedos. 1572 Figure 13. Virtual views at two different times of day (Nov. 6 Top: 10 am. Bottom: 5 pm. Note the changes in illumination an shadowing due to sun position. References Conf. on Computer Vision and Pattern Recognition (CVPR), pages 822–827, 2005. [7] R. Kawakami, K. Ikeuchi, and R. Tan. Consistent surface color for texturing large objects in outdoor scenes. In Intl. Conf. on Computer Vision (ICCV), volume 2, 2005. [8] J. Kopf, B. Neubert, B. Chen, M. Cohen, D. Cohen-Or, O. Deussen, M. Uyttendaele, and D. Lischinski. Deep photo: Model-based photograph enhancement and viewing. ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia 2008), to appear(to appear):to appear, 2008. [9] J. Lalonde and A. Efros. Using color compatibility for as- sessing image realism. In Intl. Conf. on Computer Vision (ICCV), pages 1–8, 2007. [10] Z. Liu, C. . Zhang, and Z. Zhang. Learning-based perceptual image quality improvement for video conferencing. In IEEE Intl. Conf. on Multimedia and Expo(ICME), pages 1035– 1038, 2007. Figure 13. Virtual views at two different times of day (Nov. 6). Top: 10 am. Bottom: 5 pm. 7. Conclusion Note the changes in illumination and shadowing due to sun position. [11] I. Reda and A. Andreas. Solar position algorithm for solar ra- diation application. Technical Report NREL/TP-560-34302, National Renewable Energy Laboratory, 2003. [12] E. Reinhard, M. Ashikhmin, B. Gooch, and P. Shirley. Color transfer between images. IEEE COMPUTER GRAPHICS AND APPLICATIONS, pages 34–41, 2001. [13] G. Schindler, F. Dellaert, and S. Kang. Inferring temporal or- der of images from 3D structure. In IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2007. [14] N. Snavely, S. Seitz, and R. Szeliski. Photo tourism: Explor- ing photo collections in 3D. In SIGGRAPH, pages 835–846, 2006. [15] Y. Tai, J. Jia, and C. Tang. Local color transfer via proba- bilistic segmentation by expectation-maximization. In IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), volume 1, pages 747–754, 2005. Figure 13. Virtual views at two different times of day (Nov. 6). Top: 10 am. Bottom: 5 pm. Note the changes in illumination and shadowing due to sun position. [16] A. Troccoli and P. Allen. Building illumination coherent 3d models of large-scale outdoor scenes. Intl. J. of Computer Vision, 78(2-3):261–280, 2008. References [17] S. Vedula, S. Baker, and T. Kanade. Image-based spatio- temporal modeling and view interpolation of dynamic events. ACM Transactions on Graphics, 24(2):240 – 261, April 2005. [1] A. Balan, M. Black, H. Haussecker, and L. Sigal. Shining a light on human pose: On shadows, shading and the estima- tion of pose and shape. In Intl. Conf. on Computer Vision (ICCV), pages 1–8, 2007. [18] B. Wilburn, N. Joshi, V. Vaish, E.-V. Talvala, E. Antunez, A. Barth, A.Adams, M.Horowitz, and M.Levoy. High per- formance imaging using large camera arrays. ACM Transac- tions on Graphics, 24(3):765 – 776, 2005. [2] N. Bannai, A. Agathos, and R. Fisher. Fusing multiple color images for texturing models. In 3D Data Processing Visual- ization and Transmission (3DPVT), pages 558–565, 2004. [19] Y. Yu and J. Malik. Recovering photometric properties of architectural scenes from photographs. In SIGGRAPH, pages 207–217, 1998. [3] C.Zitnick, S.B.Kang, M.Uyttendaele, S.Winder, and R.Szeliski. High-quality video view interpolation using a layered representation. ACM Transactions on Graphics, 23(3):600–608, 2004. [4] P. Debevec, C. Taylor, and J. Malik. Modeling and render- ing architecture from photographs: A hybrid geometry- and image-based approach. SIGGRAPH, 30:11–20, 1996. [5] N. Jacobs, S.Satkin, N. Roman, R. Speyer, and R. Pless. Ge- olocating static cameras. In Intl. Conf. on Computer Vision (ICCV), pages 1–6, 2007. [6] J.P.Pons, R.Keriven, and O.Faugeras. Modelling dynamic scenes by registering multi-view image sequences. In IEEE 1573

https://openalex.org/W2016097399

https://bmcdevbiol.biomedcentral.com/counter/pdf/10.1186/1471-213X-11-40

English

SOX10 directly modulates ERBB3 transcription via an intronic neural crest enhancer

BMC developmental biology

cc-by

© 2011 Prasad et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Background: The ERBB3 gene is essential for the proper development of the neural crest (NC) and its derivative populations such as Schwann cells. As with all cell fate decisions, transcriptional regulatory control plays a significant role in the progressive restriction and specification of NC derived lineages during development. However, little is known about the sequences mediating transcriptional regulation of ERBB3 or the factors that bind them. Results: In this study we identified three transcriptional enhancers at the ERBB3 locus and evaluated their regulatory potential in vitro in NC-derived cell types and in vivo in transgenic zebrafish. One enhancer, termed ERBB3_MCS6, which lies within the first intron of ERBB3, directs the highest reporter expression in vitro and also demonstrates epigenetic marks consistent with enhancer activity. We identify a consensus SOX10 binding site within ERBB3_MCS6 and demonstrate, in vitro, its necessity and sufficiency for the activity of this enhancer. Additionally, we demonstrate that transcription from the endogenous Erbb3 locus is dependent on Sox10. Further we demonstrate in vitro that Sox10 physically interacts with that ERBB3_MCS6. Consistent with its in vitro activity, we also show that ERBB3_MCS6 drives reporter expression in NC cells and a subset of its derivative lineages in vivo in zebrafish in a manner consistent with erbb3b expression. We also demonstrate, using morpholino analysis, that Sox10 is necessary for ERBB3_MCS6 expression in vivo in zebrafish. Conclusions: Taken collectively, our data suggest that ERBB3 may be directly regulated by SOX10, and that this control may in part be facilitated by ERBB3_MCS6. SOX10 directly modulates ERBB3 transcription via an intronic neural crest enhancer Megana K Prasad1†, Xylena Reed1, David U Gorkin1, Julia C Cronin2, Anthony R McAdow3, Kristopher Chain3, Chani J Hodonsky4, Erin A Jones6, John Svaren7, Anthony Antonellis4,5, Stephen L Johnson3, Stacie K Loftus2, William J Pavan2 and Andrew S McCallion1,8* Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Open Access Open Access * Correspondence: [email protected] † Contributed equally 1McKusick-Nathans Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA Full list of author information is available at the end of the article Background pigmentary and otic defects, enteric hypoganglionosis, and cardiac malformations [2,3]. Despite significant recent progress in the identification of key signaling pathways and transcription factors involved in NC induction, the hierarchical relationships between these pathways and factors are not well understood. The neural crest (NC) is a transient, multipotent and migratory population of cells present in early vertebrate development. NC cells arise from the lateral folds of the neural plate at neurulation and give rise to a multitude of cell types including pigment cells, neurons and glia of the peripheral nervous system, craniofacial skeleton and cartilage, and adrenal medullary cells [1]. Defects in NC development underlie several human diseases such as Waardenburg syndrome, Hirschsrpung disease and DiGeorge syndrome, which present a spectrum of phe- notypes including craniofacial defects, ocular, One critical gene in this network is Erbb3, a receptor tyrosine kinase that belongs to the epidermal growth factor (EGF) receptor family. Other members of the family include receptors such as Egfr1, Erbb2 and Erbb4 [4]. The structure of Erbb3 includes an extracellular domain, which interacts with ligands Neuregulin1 and Neuregulin2, and a cytoplasmic domain. Unlike other members of the EGF receptor family, the cytoplasmic domain of Erbb3 lacks tyrosine kinase activity [5,6]. Erbb3 is therefore thought to heterodimerize with Erbb2, which lacks a cognate receptor but possesses * Correspondence: [email protected] † Contributed equally 1McKusick-Nathans Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA Full list of author information is available at the end of the article Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 2 of 14 [10]. However, it is not known if this regulation is direct or indirect because the regulatory topography of Erbb3 has not yet been uncovered, and nor have any Sox10 responsive elements been demonstrated at the Erbb3 locus. tyrosine kinase activity, in order to activate downstream pathways [7,8]. Upon activation, Erbb3 triggers several downstream pathways such as PI3K, MAPK, protein kinase C, Jak-STAT and PLCg [4]. Erbb3 plays important roles in the development of the NC and its derivative tissues. Consistent with these observations, expression analyses in mouse and zebrafish have detected Erbb3/erbb3b transcripts in pre-migratory and migratory NC, and various NC derivatives includ- ing, cranial ganglia, posterior lateral line ganglia and Schwann cells [9-14]. Expression has also been observed in non-NC derived populations including the brain, olfactory lobes and myotome [10,14]. Background Furthermore, mice harbouring targeted mutant alleles of Erbb3 exhibit defects in the formation of NC and its derivatives. Homozygous null mice display a dramatic reduction in numbers of Schwann cells, hypoplastic cardiac cushion mesenchyme and cardiac valves, defects in cranial gang- lia formation and cerebellar hypoplasia [9,13]. Erbb3 deficient mice also demonstrate a lack of chromaffin cells in the adrenal medulla [9]. Furthermore, targeted deletions of Nrg1 and Erbb2, a ligand and a binding partner of Erbb3 respectively, result in NC cell migra- tion defects [15]. More recently, several in vitro and zeb- rafish studies have revealed a potential role for Erbb3 in melanocyte development and in melanoma [16-18]. Consistent with these data, Erbb3 has been implicated in other neoplasias including breast cancer and lung cancer [19]. However, what factors act upstream of Erbb3 to direct its expression and function in the variety of cell types derived from NC cells is not well under- stood. Even less well understood is the genomic sequence basis of the transcriptional regulatory control that facilitates the cell fate decisions and homeostatic maintenance of these cells during development In this study, we address the transcriptional regulation of ERBB3 during neural crest development. We use sequence constraint as a metric predictive of function to identify putative regulatory elements at the human ERBB3 locus. We test the regulatory potential of these elements in vitro in two cell lines representative of neural crest derived tissues. We also test each element for epigenetic marks consistent with enhancer activity. We verify enhancer activity of each element in vivo using zebrafish transgenesis, identifying three transcrip- tional enhancers of ERBB3. One intronic enhancer termed ERBB3_MCS6 displays the strongest enhancer activity in vitro and high enrichment for H3K4me1. We demonstrate that this enhancer is responsive to and dependent on Sox10 for its regulatory behaviour. Simi- larly, we demonstrate that Sox10 is necessary and suffi- cient for activation of the endogenous Erbb3. Furthermore, we also demonstrate that ERBB3_MCS6 directs expression in NC cells and their derivative lineages in vivo in zebrafish, and that the requirement for Sox10 for its regulation is maintained in vivo as well. Therefore, we conclude that SOX10 likely contributes to the transcriptional modulation of ERBB3 acting at least in part directly via ERBB3_MCS6. Results Multi-species conserved sequences (MCS) at the ERBB3 locus direct reporter expression in Erbb3 expressing cell lines and show epigenetic marks consistent with enhancer activity Multi-species conserved sequences (MCS) at the ERBB3 locus direct reporter expression in Erbb3 expressing cell lines and show epigenetic marks consistent with enhancer activity One factor proposed to regulate Erbb3 is Sox10, a member of the SoxE family of transcription factors [20]. Sox10 is indispensible for proper neural crest develop- ment and loss of Sox10 leads to reductions of specific NC derivatives including peripheral neurons and mela- nocytes [21,22]. Also, SOX10 mutations have been dis- covered in several patients with Waardenburg syndrome and syndromic Hirschsprung disease [23]. However, data generated to date linking Sox10 and Erbb3 have been largely correlative. Both Sox10 and Erbb3 share overlapping expressions patterns and deletions of Sox10 and Erbb3 affect overlapping cell populations such as Schwann cells, cranial ganglia and sympathetic neurons. In mice where the endogenous Sox10 locus has been replaced by a LacZ cassette, Erbb3 expression is initiated in premigratory NC but is lost once the cells begin migrating [10], suggesting Erbb3 regulation by Sox10. Furthermore Erbb3 transcript levels increase upon over- expression of Sox10 in Neuro2A neuroblastoma cells As in previous studies, we used sequence constraint to identify putative regulatory enhancers of ERBB3 [24,25]. Briefly, we identified evolutionarily conserved non-cod- ing sequences at the human ERBB3 locus using Phast- cons (http://genome.ucsc.edu/) [26], selecting eight sequences for functional analyses (LOD 30-196; Meth- ods; Figure 1A). We amplified and cloned each of the sequences upstream of an E1B minimal promoter driv- ing a luciferase reporter. We tested the regulatory potential of each of the sequences to drive transcription via luciferase assays in two NC derived cell lines, cul- tured mouse melanocytes (melan-a) and cultured rat Schwann cells (S16), both of which express Erbb3 (Fig- ure 1B, C). Of the eight MCS elements amplified and tested, five (ERBB3_MCS2, MCS3, MCS4, MCS6 and MCS7) directed luciferase reporter expression at levels significantly higher than the promoter only luciferase construct in both cell lines (p≤0.003) (Figure 1B, C). Furthermore, ERBB3_MCS6 showed the greatest Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 3 of 14 Figure 1 Multi-species conserved sequences at the ERBB3 drive reporter activity in neural crest derived lines and exhibit epigenetic marks consistent with enhancer activity. (A) Schematic representation of the eight multi-species conserved sequences at the human ERBB3 locus. (B) Luciferase activity of ERBB3_MCS1-8 in melan-a cells. (C) Luciferase activity of ERBB3_MCS1-8 in S16 cells. (D) Luciferase activity of ERBB3_MCS1-8 in NIH3T3 cells. Multi-species conserved sequences (MCS) at the ERBB3 locus direct reporter expression in Erbb3 expressing cell lines and show epigenetic marks consistent with enhancer activity Black bar indicates enrichment upon H3K4me1 ChIP whereas grey bar indicates enrichment with non-specific IgG Error bars indicate standard deviation of Figure 1 Multi-species conserved sequences at the ERBB3 drive reporter activity in neural crest derived lines and exhibit epigenetic marks consistent with enhancer activity. (A) Schematic representation of the eight multi-species conserved sequences at the human ERBB3 locus. (B) Luciferase activity of ERBB3_MCS1-8 in melan-a cells. (C) Luciferase activity of ERBB3_MCS1-8 in S16 cells. (D) Luciferase activity of ERBB3 MCS1-8 in NIH3T3 cells. All luciferase values are normalized to internal renilla control and are shown as fold-change in reporter activity as Figure 1 Multi-species conserved sequences at the ERBB3 drive reporter activity in neural crest derived lines and exhibit epigenetic marks consistent with enhancer activity. (A) Schematic representation of the eight multi-species conserved sequences at the human ERBB3 locus. (B) Luciferase activity of ERBB3_MCS1-8 in melan-a cells. (C) Luciferase activity of ERBB3_MCS1-8 in S16 cells. (D) Luciferase activity of ERBB3_MCS1-8 in NIH3T3 cells. All luciferase values are normalized to internal renilla control and are shown as fold-change in reporter activity as compared to the promoter only construct (pe1b) with standard deviation. (E) Real-time PCR results of Chromatin Immunoprecipitation. Black bar indicates enrichment upon H3K4me1 ChIP whereas grey bar indicates enrichment with non-specific IgG. Error bars indicate standard deviation of technical replicates in real-time PCR. Elements MCS1-8 represent the mouse orthologs of ERBB3_MCS elements. C refers to a region 3 Mb upstream of Erbb3, which was used as a negative control. increase in luciferase activity- approximately 23-fold greater than the promoter only construct (p = 0.002) in melan-a cells and approximately 150-fold greater than the promoter only construct in S16 cells. Interestingly, although ERBB3_MCS5 overlaps ERBB3_MCS4, it demonstrates lower luciferase activity than MCS4 in melan-a cells. This suggests that there may be transcrip- tional repressor elements in ERBB3_MCS5 that are causing it to drive lower reporter expression in the luci- ferase assay. When the eight ERBB3_MCS elements were tested in a cell line that does not express Erbb3 (NIH3T3), none directed higher luciferase activity than the promoter only construct (Figure 1D). Therefore, this suggests that ERBB3_MCS 2, MCS3, MCS4, MCS6 and increase in luciferase activity- approximately 23-fold greater than the promoter only construct (p = 0.002) in melan-a cells and approximately 150-fold greater than the promoter only construct in S16 cells. Multi-species conserved sequences (MCS) at the ERBB3 locus direct reporter expression in Erbb3 expressing cell lines and show epigenetic marks consistent with enhancer activity All luciferase values are normalized to internal renilla control and are shown as fold-change in reporter activity as compared to the promoter only construct (pe1b) with standard deviation. (E) Real-time PCR results of Chromatin Immunoprecipitation. Black bar indicates enrichment upon H3K4me1 ChIP whereas grey bar indicates enrichment with non-specific IgG. Error bars indicate standard deviation of technical replicates in real-time PCR. Elements MCS1-8 represent the mouse orthologs of ERBB3_MCS elements. C refers to a region 3 Mb upstream of Erbb3, which was used as a negative control. Figure 1 Multi-species conserved sequences at the ERBB3 drive reporter activity in neural crest derived lines and exhibit epigenetic marks consistent with enhancer activity. (A) Schematic representation of the eight multi-species conserved sequences at the human ERBB3 locus. (B) Luciferase activity of ERBB3_MCS1-8 in melan-a cells. (C) Luciferase activity of ERBB3_MCS1-8 in S16 cells. (D) Luciferase activity of ERBB3_MCS1-8 in NIH3T3 cells. All luciferase values are normalized to internal renilla control and are shown as fold-change in reporter activity as compared to the promoter only construct (pe1b) with standard deviation. (E) Real-time PCR results of Chromatin Immunoprecipitation. Black bar indicates enrichment upon H3K4me1 ChIP whereas grey bar indicates enrichment with non-specific IgG. Error bars indicate standard deviation of technical replicates in real-time PCR. Elements MCS1-8 represent the mouse orthologs of ERBB3_MCS elements. C refers to a region 3 Mb upstream of Erbb3, which was used as a negative control. served sequences at the ERBB3 drive reporter activity in neural crest derived lines and exhibit epigenetic i i (A) S h i i f h i h l i i d h h ERBB3 Figure 1 Multi-species conserved sequences at the ERBB3 drive reporter activity in neural crest derived lines and exhibit epigenetic marks consistent with enhancer activity. (A) Schematic representation of the eight multi-species conserved sequences at the human ERBB3 locus. (B) Luciferase activity of ERBB3_MCS1-8 in melan-a cells. (C) Luciferase activity of ERBB3_MCS1-8 in S16 cells. (D) Luciferase activity of ERBB3_MCS1-8 in NIH3T3 cells. All luciferase values are normalized to internal renilla control and are shown as fold-change in reporter activity as compared to the promoter only construct (pe1b) with standard deviation. (E) Real-time PCR results of Chromatin Immunoprecipitation. Multi-species conserved sequences (MCS) at the ERBB3 locus direct reporter expression in Erbb3 expressing cell lines and show epigenetic marks consistent with enhancer activity Interestingly, although ERBB3_MCS5 overlaps ERBB3_MCS4, it demonstrates lower luciferase activity than MCS4 in melan-a cells. This suggests that there may be transcrip- tional repressor elements in ERBB3_MCS5 that are causing it to drive lower reporter expression in the luci- ferase assay. When the eight ERBB3_MCS elements were tested in a cell line that does not express Erbb3 (NIH3T3), none directed higher luciferase activity than the promoter only construct (Figure 1D). Therefore, this suggests that ERBB3_MCS 2, MCS3, MCS4, MCS6 and MCS7 are potentially enhancers of ERBB3 expression in the NC. In order to determine whether these MCS elements were indeed enhancers within their genomic context, we assayed the presence of an epigenetic mark previously shown to be predictive of enhancer function. Since H3K4 monomethylation (H3K4me1) has been shown to be enriched at transcriptional enhancers, we used chro- matin immunoprecipitation (ChIP) to determine whether H3K4me1 marks were enriched at the mouse orthologs of ERBB3_MCS elements within the genomic context of melan-a cells [27]. We used an antibody against H3K4me1 and an isotype control IgG to immu- noprecipitate sheared crosslinked chromatin from these Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 4 of 14 known role in the regulation of ERBB3 expression [32-36]. The ERBB3 promoter harbours potential AP2 binding sites, causing speculation of the role of AP2 as an ERBB3 regulator [37]. cells. We analyzed immunoprecipitation using real-time PCR and calculated enrichment using the percent input method (Figure 1E). ERBB3_MCS1, MCS2 and MCS3 showed consistently low levels of enrichment upon immunoprecipitation with the H3K4me1 antibody as compared to the IgG control samples, whereas ERBB3_MCS4, MCS5, MCS6, MCS7 and MCS8 showed consistently high levels of enrichment upon immunopre- cipitation as compared to the IgG samples. A control sequence 3 Mb upstream of ERBB3 was not enriched in H3K4me1 samples as compared to IgG samples. ERBB3_MCS5, MCS7 and MCS8 showed the highest levels of enrichment as compared to the IgG samples. This contrasts with the low luciferase activity driven by these sequences in vitro, consistent with reports suggest- ing that H3K4me1 is not sufficient to distinguish between functionally active and inactive enhancers [28,29]. Elements ERBB3_MCS2, MCS3, MCS4 and MCS6 however did show increased luciferase activity as well as H3K4me1 marks, suggesting that these sequences may be functional endogenous enhancers of ERBB3 in NC-derived cell types. ERBB3_MCS6 harbours putative transcription factor binding sites for key neural crest transcription factors binding sites for key neural crest transcription factors Since ERBB3_MCS6 demonstrated the highest luciferase activity in vitro and showed enrichment for H3K4me1 consistent with a role for ERBB3_MCS6 as an enhancer of ERBB3, we focused our subsequent analyses on this element. To determine what pathways or transcription factors may be acting upon this element and hence potentially modulating Erbb3 expression, we queried three publicly available transcription factor identification databases (JASPAR, MATCH and TRANSFAC) to iden- tify putative transcription factor binding sites (TFBSs) within this sequence. To refine our search further we prioritized sites that were found by two or more TF databases, selecting those that corresponded to tran- scription factors with known roles in NC development (SOXE and AP2). We identified two SOXE sites (SOXE- 1 and SOXE-2) and one AP2 site (Additional file 1, Fig- ure S1). Members of the SOXE family, SOX8, SOX9 and SOX10, bind to the SOXE consensus site 5’- AACAAT-3’. The identification of SOXE family (SOX8, SOX9 and SOX10) motifs is consistent with the pre- dicted role for Sox10 in Erbb3 modulation. SOXE pro- teins typically bind the consensus site 5’-AACAAT-3’. Both SOXE-1 and SOXE-2 contain the 5’-ACAAT-3’ core sequence but differ from the consensus in that the 5’-most adenine is replaced by a cytosine. AP2 is a reti- noic acid inducible transcription factor that binds to the consensus sequence 5’-CCCCAGGC-3’ [30,31]. It is expressed in the NC and its derivatives including cranial and sensory ganglia and facial mesenchyme and has a Sox10 is necessary for ERBB3_MCS6 enhancer activity and Erbb3 expression Multi-species conserved sequences (MCS) at the ERBB3 locus direct reporter expression in Erbb3 expressing cell lines and show epigenetic marks consistent with enhancer activity SOXE2 is required for enhancer activity of ERBB3_MCS6 We set out to determine the importance of the SOXE and AP2 transcription factor families in modulating the activity of ERBB3_MCS6 by mutating predicted binding sites in the context of a pe1b promoter driving a lucifer- ase reporter. We then compared the luciferase activity of each mutant construct to that of the wild-type ERBB3_MCS6 sequence in the melan-a cell line, which expresses Sox10, Erbb3 and Ap2 (Figure 2A). Mutation of the SOXE2 but not the SOXE1 motif significantly reduced enhancer activity of ERBB3_MCS6 (Figure 2A, p = 0.002 for SOXE2). This observation is consistent with a suggested role for Sox10 in regulating Erbb3 in NC-derived populations [10]. However, since mutation of the SOXE-2 site does not completely abrogate ERBB3_MCS6, this suggests that additional factors may also be responsible for the activity of ERBB3_MCS6. By contrast, mutation of the AP2 site does not significantly reduce luciferase activity of ERBB3_MCS6 across repli- cate experiments. However, we cannot exclude the pre- sence of a non-canonical AP2 site that may be required for ERBB3_MCS6 activity independent of or in addition to this site. Sox10 is necessary for ERBB3_MCS6 enhancer activity and Erbb3 expression Since mutation of the SOXE-2 binding site caused a sig- nificant reduction in ERBB3_MCS6 enhancer activity, we asked whether Sox10 may be acting through ERBB3_MCS6 to regulate Erbb3 in NC populations. We used siRNA (SMARTpool, Dharmacon) to transiently knock down Sox10 gene product in melan-a cells and assayed the effect of the knockdown on ERBB3_MCS6 activity via luciferase assays, as before. Consistent with significantly reduced ERBB3_MCS6 activity (Figure 2B; p = 0.05). Sox10 knockdown was confirmed via Western blot as shown in Figure 2C. We also determined the effect of Sox10 knockdown on endogenous Erbb3 pro- tein levels in melan-a cells. As shown in figure 2D, knockdown of Sox10 causes a decrease of endogenous Erbb3 protein levels. Similarly, we interfered with Sox10 protein function in S16 cells using a previously charac- terized construct expressing a dominant negative mutant form of Sox10 (E189X) under the control of a CMV promoter [38]. The E189X mutant protein has an intact DNA binding domain but lacks a transactivation domain. Upon transfection of S16 cells with this mutant cDNA of Sox10, there was a significant decrease in luci- ferase activity of ERBB3_MCS6 (Figure 2E). Taken Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 5 of 14 Figure 2 Sox10 is necessary and sufficient for the activity of ERBB3_MCS6 in vitro. (A) Luciferase activities of wild-type ERBB3_MCS6 and TFBS mutations in ERBB3_MCS6 in melan-A cell line. The position of each TFBS in ERBB3_MCS6 is shown on top and the corresponding mutation in each TFBS is shown next to the luciferase value for that construct. * indicates statistical significance. (B) Luciferase activity of WT ERBB3_MCS6 in melan-A cells in mock-transfected cells and in cells with transient Sox10 and Ap2 knockdown. (C) Western blot to confirm knockdown of Sox10 and Ap2 protein upon siRNA treatment. Tubulin antibody was used as a loading control. (D) Western blot showing Erbb3 protein levels in melan-a cells upon transient transfection with Sox10 and Ap2 siRNA or mock-transfected cells. Tubulin antibody was used as a loading control. (E) Luciferase activity of ERBB3_MCS6 upon knockdown of Sox10 in S16 cells using a dominant negative SOX10 mutant (E189X) under a CMV prmoter. (F) Luciferase assay of WT and SOXE-2m ERBB3_MCS6 in Neuro2A cells when transiently co-transfected with an empty expression vector (pcDNA.31) and Sox10 cDNA. Cell lysates were collected 24 hours post transfection. Sox10 is necessary for ERBB3_MCS6 enhancer activity and Erbb3 expression (G) Luciferase assay of WT ERBB3_MCS6 in Neuro2A cells when transiently co-transfected with equal amounts of WT and Sox10-ΔSTP cDNA either individually or in combination. Cell lysates Figure 2 Sox10 is necessary and sufficient for the activity of ERBB3_MCS6 in vitro. (A) Luciferase activities of w Figure 2 Sox10 is necessary and sufficient for the activity of ERBB3_MCS6 in vitro. (A) Luciferase activities of wild-type ERBB3_MCS6 and TFBS mutations in ERBB3_MCS6 in melan-A cell line. The position of each TFBS in ERBB3_MCS6 is shown on top and the corresponding mutation in each TFBS is shown next to the luciferase value for that construct. * indicates statistical significance. (B) Luciferase activity of WT ERBB3_MCS6 in melan-A cells in mock-transfected cells and in cells with transient Sox10 and Ap2 knockdown. (C) Western blot to confirm knockdown of Sox10 and Ap2 protein upon siRNA treatment. Tubulin antibody was used as a loading control. (D) Western blot showing Erbb3 protein levels in melan-a cells upon transient transfection with Sox10 and Ap2 siRNA or mock-transfected cells. Tubulin antibody was used as a loading control. (E) Luciferase activity of ERBB3_MCS6 upon knockdown of Sox10 in S16 cells using a dominant negative SOX10 mutant (E189X) under a CMV prmoter. (F) Luciferase assay of WT and SOXE-2m ERBB3_MCS6 in Neuro2A cells when transiently co-transfected with an empty expression vector (pcDNA.31) and Sox10 cDNA. Cell lysates were collected 24 hours post transfection. (G) Luciferase assay of WT ERBB3_MCS6 in Neuro2A cells when transiently co-transfected with equal amounts of WT and Sox10-ΔSTP cDNA either individually or in combination. Cell lysates were collected 24 hours post transfection. All luciferase values are normalized to renilla internal control and shown as fold-change compared to promoter only construct (pe1B) with standard deviation. (H) Real-time PCR of Erbb3 transcript levels upon expression of WT and Sox10-ΔSTP cDNA individually and in combination. Values are normalized to 18s internal control and are shown as fold-change compared to promoter only construct (pcDNA3.1) with standard error. (I) Real-time PCR of ChIP against Sox10 in untreated S16 cells and in S16 cells treated with Sox10 morpholino. Black bar indicates enrichment upon Sox10 ChIP whereas grey bar indicates enrichment with non-specific IgG. Error bars indicate standard deviation of technical replicates in real-time PCR. Figure 2 Sox10 is necessary and sufficient for the activity of ERBB3_MCS6 in vitro. Sox10 is necessary for ERBB3_MCS6 enhancer activity and Erbb3 expression (A) Luciferase activities of wild-type ERBB3_MCS6 and TFBS mutations in ERBB3_MCS6 in melan-A cell line. The position of each TFBS in ERBB3_MCS6 is shown on top and the corresponding mutation in each TFBS is shown next to the luciferase value for that construct. * indicates statistical significance. (B) Luciferase activity of WT ERBB3_MCS6 in melan-A cells in mock-transfected cells and in cells with transient Sox10 and Ap2 knockdown. (C) Western blot to confirm knockdown of Sox10 and Ap2 protein upon siRNA treatment. Tubulin antibody was used as a loading control. (D) Western blot showing Erbb3 protein levels in melan-a cells upon transient transfection with Sox10 and Ap2 siRNA or mock-transfected cells. Tubulin antibody was used as a loading control. (E) Luciferase activity of ERBB3_MCS6 upon knockdown of Sox10 in S16 cells using a dominant negative SOX10 mutant (E189X) under a CMV prmoter. (F) Luciferase assay of WT and SOXE-2m ERBB3_MCS6 in Neuro2A cells when transiently co-transfected with an empty expression vector (pcDNA.31) and Sox10 cDNA. Cell lysates were collected 24 hours post transfection. (G) Luciferase assay of WT ERBB3_MCS6 in Neuro2A cells when transiently co-transfected with equal amounts of WT and Sox10-ΔSTP cDNA either individually or in combination. Cell lysates were collected 24 hours post transfection. All luciferase values are normalized to renilla internal control and shown as fold-change compared to promoter only construct (pe1B) with standard deviation. (H) Real-time PCR of Erbb3 transcript levels upon expression of WT and Sox10-ΔSTP cDNA individually and in combination. Values are normalized to 18s internal control and are shown as fold-change compared to promoter only construct (pcDNA3.1) with standard error. (I) Real-time PCR of ChIP against Sox10 in untreated S16 cells and in S16 cells treated with Sox10 morpholino. Black bar indicates enrichment upon Sox10 ChIP whereas grey bar indicates enrichment with non-specific IgG. Error bars indicate standard deviation of technical replicates in real-time PCR. Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 6 of 14 together, these experiments demonstrate that Sox10 is necessary for the expression of Erbb3 and for enhancer activity of ERBB3_MCS6, consistent with its predicted role in modulating Erbb3. Knockdown of Ap2 in melan- a cells also led to a significant reduction in ERBB3_MCS6 transcriptional activity (Figure 2B). Furthermore, Ap2 knockdown also reduced endogenous Erbb3 protein levels (Figure 2D), consistent with pre- vious evidence for a role for AP2 in regulating ERBB3 [35,36]. Sox10 is necessary for ERBB3_MCS6 enhancer activity and Erbb3 expression nucleotides replace the putative activation domain. The resulting protein inhibits the activity of wild-type Sox10 in a dominant negative fashion. Co-transfection of ERBB3_MCS6 with Sox10-ΔSTP does not result in ERBB3_MCS6 activity (Figure 2G). Additionally, ΔSTP also compromises the transactivation of ERBB3_MCS6 by wild-type Sox10 as seen in cells co-transfected with both the wild-type and the mutant forms of Sox10. This implies that ERBB3_MCS6 plays an important role in the regulation of Erbb3 by Sox10. If ERBB3_MCS6 were indeed an important link in the regulation of Erbb3 expression by Sox10, we would expect the overexpression of wild-type Sox10, but not the Sox10-ΔSTP mutant, to increase endogenous tran- script levels of Erbb3. In fact, when we examined the levels of Erbb3 transcript in Neuro2A cells upon Sox10 overexpression, we found that it was seven-fold higher than Erbb3 transcript levels in the absence of Sox10 (Figure 2H). Furthermore, Sox10-ΔSTP was unable to increase Erbb3 transcript levels and it also impeded the ability of the wild-type protein to increase Erbb3 tran- script levels. An increase in total Sox10 transcript levels upon overexpression was verified via real-time PCR (Additional file 3, Figure S3). Therefore, this implies that Sox10 is sufficient for the transcriptional activation of Erbb3. Sox10 is sufficient for ERBB3_MCS6 enhancer activity and Erbb3 expression Sox10 is sufficient for ERBB3_MCS6 enhancer activity and Erbb3 expression Given that Sox10 is necessary for ERBB3_MCS6 enhan- cer activity, we next tested whether Sox10 is also suffi- cient for ERBB3_MCS6 activity. We transiently co- transfected mouse neuroblastoma cells (Neuro2A) with the ERBB3_MCS6 luciferase vector and with either a construct expressing wild-type Sox10 cDNA under the control of a CMV promoter (pcDNA3.1) or with an empty expression vector. Neuro2A cells were selected due to low expression levels of Sox10 and Erbb3 and because this cell line has been previously used to study the interaction between Sox10 and Erbb3 [10]. In the presence of Sox10, the transcriptional activity of ERBB3_MCS6 increased almost 25-fold, compared to cells co-transfected with the empty vector (Figure 2F). In the absence of Sox10, ERBB3_MCS6 does not display significant transcriptional activity in Neuro2A cells. Further, consistent with a role for Sox10 in directly reg- ulating MCS6, mutation of the SOXE-2 site abrogates the increase in transcriptional activity. Similar results were also seen when human SOX10 cDNA was used in co-transfection experiments in NIH3T3 cells (Additional file 2, Figure S2). On the other hand, AP2 was unable to significantly transactivate ERBB3_MCS6 in NIH3T3 cells (Additional file 2, Figure S2). Also, when added in combination with SOX10, it was unable to increase tran- scriptional activity of ERBB3_MCS6 in a manner that is more than additive when compared to SOX10 and AP2 independent transactivation, further suggesting that AP2 and SOX10 do not interact to regulate ERBB3_MCS6. Sox10 physically binds ERBB3_MCS6 This sug- gests that ERBB3_MCS1, MCS3 and MCS6 are all NC- directed enhancers of ERBB3. Sox10 protein levels in S16 via siRNAs caused a loss of enrichment of ERBB3_MCS6 upon Sox10 immunopreci- pitated in these cells. Collectively, this implies that Sox10 directly binds ERBB3_MCS6. ERBB3_MCS6 directs reporter expression in a neural crest specific manner reminiscent of erbb3b expression in zebrafish In order to confirm that ERBB3_MCS6 is an enhancer of ERBB3 in vivo, we generated stable transgenic zebra- fish lines of ERBB3_MCS6 driving an eGFP reporter using a Tol2 based zebrafish transgenesis assay [39]. Briefly, we cloned ERBB3_MCS6 upstream of a cfos minimal promoter driving enhanced green fluorescent protein (eGFP). We then injected this sequence into 1-2 cell zebrafish embryos, screened G0 embryos for vector integration, and evaluated their offspring for germ line reporter expression. Reporter expression was assayed in four independent transgenic lines and compared to pre- viously published patterns of endogenous Erbb3/erbb3b expression [10-12,14]. Consistent with its strong in vitro regulatory potential, ERBB3_MCS6 drove reporter expression in NC populations as early as 24 hours post fertilization (hpf) in transgenic zebrafish in a manner reminiscent of erbb3b expression (Figure 3A, B). Expres- sion was noted in cranial neural crest, migratory and pre-migratory crest, all regions where Erbb3/erbb3b is known to be expressed (Figure 3A, B) [11]. At 48 hpf, expression was noted in the dorsal root ganglia, poster- ior lateral line ganglia and in Schwann cell precursors, consistent with known Erbb3/erbb3b expression (Figure 3C) [12,13]. Finally, at 72 hpf expression was noted in cells that are consistent in shape and position with mature oligodendrocytes (Figure 3D) [40]. Therefore, ERBB3_MCS6 drives reporter expression in vivo in neural crest cells and its derivative tissues in a manner consistent with erbb3b expression. Two other sequences, ERBB3_MCS1 and MCS4 also drove consistent expres- sion in multiple founders in NC-derived populations consistent with Erbb3/erbb3b expression (Additional file 5, Figure S5). Although ERBB3_MCS1 did not show sig- nificant enhancer activity in vitro, or H3K4me1 marks, it demonstrated expression in vivo in a manner consis- tent with Erbb3 expression. This sequence directed reporter expression starting at 48 hpf in the mesence- phalon, olfactory bulbs, cranial ganglia and pharyngeal arches, all regions of demonstrated Erbb3 expression [14,41]. ERBB3_MCS1 also directed expression in the midbrain-hindbrain boundary and in the hindbrain, regions where Erbb3 function has been shown to be essential [9] (Additional file 5, Figure S5A, B, C). Furthermore, this element directed expression in cells lining the anterior lateral line neurons and the epiphysis (data not shown). Consistent with its strong in vitro Sox10 physically binds ERBB3_MCS6 y y We next used electrophoretic mobility shift assays to determine if ERBB3_MCS6 physically binds protein in a manner dependent on the SOXE-2 motif. We generated 50 bp probes that span the SOXE-2 binding site and labeled the probes with biotin. We then incubated the probes with nuclear extract derived from melan-a cells and subjected the reactions to gel electrophoresis. As seen in Additional file 4, Figure S4, a factor in the melan-a nuclear extract bound the probe and retarded its migration in the gel, causing a shift. This binding was successfully competed away by using increasing amounts of unlabeled probe. To determine whether this binding was specific to the SOXE-2 binding site, we next mutated the SOXE-2 binding site and incubated unlabeled mutant probe with nuclear extract. As is seen in lane 5 of Additional file 4, Figure S4, the mutant cold probe was unable to compete away the binding, thus indicating that the gel shift seen is specific to binding at the SOXE-2 binding site. In order to assay whether Sox10 directly binds to ERBB3_MCS6, we used chroma- tin immunoprecipitation followed by quantitative PCR using an antibody directed against Sox10 in S16 cells. As shown in Figure 2I, the rat ortholog of ERBB3_MCS6 is enriched in Sox10 immunoprecipitated samples but not in IgG immunoprecipitated samples. This binding is specific to Sox10 as knocking down To further investigate the role of Sox10 in regulating ERBB3_MCS6, we next determined whether established disease causing mutations in Sox10 might in part med- iate their regulation of Erbb3 via ERBB3_MCS6. To this end, we used a Sox10 mutant cDNA expression con- struct (c.928_929ins), hereby referred to as Sox10-ΔSTP to transactivate ERBB3_MCS6. This mutant was first discovered in the Dom mouse model of Hirschsrpung disease [20,21] and has a single nucleotide (Guanine) insertion at nucleotide 929 of the coding sequence, lead- ing to a translational frameshift. The resulting protein possesses an intact DNA binding domain but 99 novel Page 7 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 regulatory potential and H3K4me1 marks, ERBB3_MCS4 directed reporter expression in vivo. ERBB3_MCS4 drove eGFP expression in vivo, beginning at 24 hpf in the mesencephalon, the hindbrain and the pharyngeal arches. It also directed expression in the myotome consistent with endogenous Erbb3 expression [10,11] (Additional file 5, Figure S5D, E, F). SOXE-2 motif is necessary for the activity of ERBB3_MCS6 in vivo in zebrafish To determine the contribution of the SOXE-2 motif to the proper expression of ERBB3_MCS6 in vivo, we gen- erated stable transgenic zebrafish lines with ERBB3_MCS6 harbouring a mutation in the SOXE-2 motif. We identified three independent stable transgenic zebrafish lines with the SOXE-2m ERBB3_MCS6 con- struct directing eGFP expression. eGFP expression was inconsistent among these lines. This may either be a direct result of position effect based on the position of integration of the vector, or mutation of the SOXE-2 motif may make the sequence more amenable to posi- tion effect due to loss of proper regulatory control. In one of the three lines, eGFP expression was not seen at 24 hpf whereas in the remaining two lines eGFP expres- sion was qualitatively similar to expression directed by the WT ERBB3_MCS6 construct at 24 hpf (Figure 3E, F). Additionally, mutation of this site may yield quanti- tative differences in eGFP expression between the WT and mutant constructs. Technical reasons such as varia- tion in transgene copy number and position effect pre- clude the use of this assay to evaluate such differences. Therefore, we cannot exclude the possibility that addi- tional elements within ERBB3_MCS6 may be sufficient for its expression in NC in the early embryo. However, at 48 hpf and 72 hpf, expression driven by the mutant construct was qualitatively different across all founders as compared to expression directed by the WT con- struct. Expression was lost from dorsal root ganglia cells and Schwann cells in 48 hpf SOXE-2m ERBB3_MCS6 transgenic embryos. Expression domains were gained within these mutants in the notochord and cells of the ventral spinal cord (Figure 3G and data not shown). Furthermore, at 72 hpf expression was not seen in oligo- dendroglial cells in SOXE-2m ERBB3_MCS6 transgenic embryos (Figure 3H). However, at both 48 hpf and 72 hpf, expression in the ectomesenchymal derivatives of the NC such as in the pharyngeal arches was maintained in the mutant transgenic fish (data not shown). This is in agreement with a known role for Sox10 in non-ecto- mesenchymal but not in ectomesenchymal NC popula- tions in the zebrafish [42]. This suggests that additional elements within ERBB3_MCS6 are sufficient for the Page 8 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Figure 3 ERBB3_MCS6 drives reporter expression in vivo in a pattern similar to erbb3b and under the control of sox10. Discussion Since the sox10 morpholino pro- duces embryos with a range of phenotypes, in embryos that show a weaker phenotype, we noticed a reduced number of eGFP expressing premigratory NC cells and highly reduced numbers of migrating crest at 24 hpf, consistent with a role for Sox10 in regulation ERBB3_MCS6 (Additional file 6, Figure S6C, D). eGFP expression in the cranial NC was affected to a lesser degree in these embryos (Additional file 6, Figure S6C). However, eGFP expression in 24 hpf embryos was noted in only 6/130 injected embryos (4.6%), which is in sharp contrast to the 27/60 uninjected (45%) embryos that showed eGFP expression at 24 hpf. By 48 hpf, ERBB3_MCS6 driven eGFP expression was noted in morpholino injected embryos in cells that are consistent in position with premigratory NC (Figure 3P). Expres- sion was also noted in some migratory NC, however, these cells are highly disorganized as compared to the eGFP expressing cells seen in uninjected embryos con- sistent with migratory crest previously described for this model (Figure 3L, P) [42,43]. eGFP expression persists in these non-migrating NC cells upto 96hpf (data not shown) but is not seen in NC derived cells Therefore Using a combination of sequence constraint, in vitro luciferase and chromatin immunoprecipitation assays, and in vivo zebrafish transgenesis, we have identified three functional enhancers of ERBB3- ERBB3_MCS1, MCS4 and MCS6. Although ERBB3_MCS1 did not sig- nificantly drive luciferase activity in vitro or exhibit enrichment of H3K4me1 marks, it drove reporter expression in transgenic zebrafish embryos in a manner consistent with endogenous erbb3b expression. Further- more, although elements ERBB3_MCS2, MCS3 and MCS7 showed increased luciferase activity in vitro, they did not exhibit NC-directed enhancer activity in vivo. This suggests that although in vitro assays are useful in assaying the regulatory potential of enhancer sequences, they represent a single stage and cell type in develop- ment as compared to the full spectrum of NC deriva- tives that can be assayed in vivo and can potentially overlook regulatory sequences, which may be active at different stages or cell types or may attribute activity to sequences that are not biologically functional. Similarly, comparison of H3K4me1 marks and enhancer activity in the elements tested demonstrates that H3K4me1 is not sufficient to distinguish between active and inactive enhancers. Discussion Although ERBB3 is an important gene in the develop- ment and differentiation of a range of NC-derived popu- lations, not much is known about the sequences and factors modulating its transcriptional output. Sox10, however, has been shown to influence Erbb3 levels [10], although, whether this regulation is direct or indirect was unknown. In this study, we identify novel regulatory enhancers of ERBB3 that direct expression in NC derived cell lines and exhibit epigenetic marks consistent with enhancer activity. One of these enhancers, ERBB3_MCS6 directs the strongest reporter expression in vitro and a broad overlap with endogenous erbb3b expression in vivo in zebrafish. We demonstrate that ERBB3_MCS6 is dependent upon and is responsive to Sox10 both in vitro and in vivo. Furthermore, we also demonstrate that Sox10 is both necessary and sufficient in vitro for endogenous Erbb3 expression. Finally, we demonstrate that Sox10 physically binds to ERBB3_MCS6, suggesting that SOX10 mediates its effect on ERBB3, at least in part, directly through ERBB3_MCS6. In order to determine whether Sox10 is required for the enhancer activity of ERBB3_MCS6 in vivo similar to its requirement in vitro, we used morpholinos against sox10 in the stable transgenic ERBB3_MCS6 zebrafish to assay the effect of sox10 depletion on ERBB3_MCS6 activity. Using a previously published translation-block- ing morpholino against the zebrafish sox10 gene pro- duct, we knocked down levels of sox10 in the ERBB3_MCS6 transgenic zebrafish [43]. Depletion of sox10 in zebrafish produces a phenocopy of the colour- less (sox10-/-) zebrafish mutant phenotype, which has been studied in detail [42]. Although NC cells form upon sox10 depletion, they accumulate in the premigra- tory position and fail to migrate and differentiate appro- priately into their subsequent lineages [42,43]. Thus, if sox10 were necessary for ERBB3_MCS6 expression, we would expect to see a loss of ERBB3_MCS6 driven reporter expression in premigratory NC as well as a subsequent loss of expression in NC derived cell types such as Schwann cells and DRG due to the loss of these populations. Consistent with our expectation, upon knocking down sox10, we noted a complete loss of ERBB3_MCS6 activity in the developing NC in one-day old embryos (Figure 3N,O). This is in sharp contrast to the ERBB3_MCS6 driven reporter expression seen in the premigratory and migratory NC of uninjected embryos (Figure 3J,K). SOXE-2 motif is necessary for the activity of ERBB3_MCS6 in vivo in zebrafish BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 9 of 14 ERBB3_MCS6 in vivo, we used morpholinos analysis in zebrafish. ERBB3_MCS6 in vivo, we used morpholinos analysis in zebrafish. this suggests that sox10 is necessary for the timely and proper expression of ERBB3_MCS6 in neural crest cells and its derivative lineages in vivo in zebrafish. SOXE-2 motif is necessary for the activity of ERBB3_MCS6 in vivo in zebrafish (A-D) Expression pattern of ERBB3_MCS driving eGFP in G1 transgenic 24-72hpf zebrafish embryos. Arrows indicate tissues where expression was noted in multiple founders. (E-H) eGFP expression pattern seen in multiple founders of SOXE-2m ERBB3_MCS6 transgenic fish. Although expression is similar to WT ERBB3_MCS6 at 24hpf, expression differs at later stages of development. (I-P) Results of morpholino knockdown of sox10 in ERBB3_MCS6 transgenic embryos. (I-L) Normal expression of ERBB3_MCS6 in uninjected transgenic embryos at 24-48hpf. (M-P) Loss of reporter expression in embryos injected with a Sox10 morpholino at 24hpf and appearance of disorganized GFP expressing NC cells at 48hpf. Abbreviations: cranial neural crest (CNC), migratory crest (MC), pre-migratory crest (PMC), dorsal root ganglia (DRG), posterior lateral line (PLL), Schwann cells (SC), oligodendrocytes (OD) and notochord (NC). Figure 3 ERBB3_MCS6 drives reporter expression in vivo in a pattern similar to erbb3b and under the control of sox10. (A-D) Expression pattern of ERBB3_MCS driving eGFP in G1 transgenic 24-72hpf zebrafish embryos. Arrows indicate tissues where expression was noted in multiple founders. (E-H) eGFP expression pattern seen in multiple founders of SOXE-2m ERBB3_MCS6 transgenic fish. Although expression is similar to WT ERBB3_MCS6 at 24hpf, expression differs at later stages of development. (I-P) Results of morpholino knockdown of sox10 in ERBB3_MCS6 transgenic embryos. (I-L) Normal expression of ERBB3_MCS6 in uninjected transgenic embryos at 24-48hpf. (M-P) Loss of reporter expression in embryos injected with a Sox10 morpholino at 24hpf and appearance of disorganized GFP expressing NC cells at 48hpf. Abbreviations: cranial neural crest (CNC), migratory crest (MC), pre-migratory crest (PMC), dorsal root ganglia (DRG), posterior lateral line (PLL), Schwann cells (SC), oligodendrocytes (OD) and notochord (NC). derivatives. However, based on the diminution and not complete abrogation of ERBB3_MCS6 activity in vitro, we expect that mutation of this site may produce quan- titative effects on the expression of ERBB3_MCS6, which cannot be assayed by zebrafish transgenesis. Therefore, to address the role of Sox10 in regulation expression of this sequence in ectomesenchymal deriva- tives of the NC. Therefore, the SOXE-2 motif is impor- tant for the proper expression of ERBB3_MCS6 in non- ectomesenchymal neural crest derivatives such as dorsal root ganglia and Schwann cells, but not in premigratory and migratory crest cells or ectomesenchymal NC Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Prasad et al. Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Page 10 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 assay, this data suggests that other motifs within ERBB3_MCS6 may be sufficient for its expression in early NC and ectomesenchymal derivatives. However, using additional in vitro and in vivo assays, we estab- lished an important role for Sox10 in regulating ERBB3_MCS6 via the SOXE-2 motif. Knockdown of Sox10 abrogates ERBB3_MCS6 activity, whereas ectopic expression of Sox10 induces ERBB3_MCS6 response, demonstrating that Sox10 is both necessary and suffi- cient for ERBB3_MCS6 transcriptional activity. Similarly, we also show that Sox10 is both necessary and sufficient for Erbb3 expression. This is consistent with previous data suggesting that Erbb3 is under the control of Sox10 [10]. Using gel-shift and chromatin immunoprecipitation assays, we also demonstrate direct binding of Sox10 to ERBB3_MCS6, most likely mediated through the SOXE- 2 site. Finally, we use morpholino analysis in ERBB3_MCS6 transgenic zebrafish to demonstrate a requirement for Sox10 in the proper expression of this enhancer. Therefore, our data suggest that ERBB3_MCS6 may play a critical role in mediating the regulation of ERBB3 by SOX10. assay, this data suggests that other motifs within ERBB3_MCS6 may be sufficient for its expression in early NC and ectomesenchymal derivatives. However, using additional in vitro and in vivo assays, we estab- lished an important role for Sox10 in regulating ERBB3_MCS6 via the SOXE-2 motif. Knockdown of Sox10 abrogates ERBB3_MCS6 activity, whereas ectopic expression of Sox10 induces ERBB3_MCS6 response, demonstrating that Sox10 is both necessary and suffi- cient for ERBB3_MCS6 transcriptional activity. Similarly, we also show that Sox10 is both necessary and sufficient for Erbb3 expression. This is consistent with previous data suggesting that Erbb3 is under the control of Sox10 [10]. Using gel-shift and chromatin immunoprecipitation assays, we also demonstrate direct binding of Sox10 to ERBB3_MCS6, most likely mediated through the SOXE- 2 site. Finally, we use morpholino analysis in ERBB3_MCS6 transgenic zebrafish to demonstrate a requirement for Sox10 in the proper expression of this enhancer. Therefore, our data suggest that ERBB3_MCS6 may play a critical role in mediating the regulation of ERBB3 by SOX10. These data are consistent with recent reports that sug- gest that additional epigenetic marks such as H3K27ac may be necessary to distinguish between H3Kem1 marked active and poised enhancers [28,29]. p [ ] The intronic enhancer ERBB3_MCS6 demonstrated high transcriptional activity in vitro and also exhibited H3K4me1 marks. Conclusions In conclusion, we have identified three transcriptional enhancers at the ERBB3 locus, whose expression domains overlap NC-derived populations. We show that one such enhancer (ERBB3_MCS6) directs reporter expression broadly in NC cells and, like ERBB3, is directly responsive to Sox10, implying that Sox10 may act through this enhancer to regulate ERBB3 transcrip- tion in the NC. Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Furthermore, it directed expression almost exclusively, and broadly, in developing NC popu- lations. The role of AP2 in regulating ERBB3_MCS6 remains unclear. Mutation of the AP2 binding site does not significantly reduce transcriptional activity of ERBB3_MCS6, suggesting that it is not required for the activity of ERBB3_MCS6. Furthermore, ectopic expres- sion of AP2 does not increase transcriptional activity of ERBB3_MCS6 (Additional file 2, Figure S2). However, knockdown of Ap2 caused a significant decrease in tran- scriptional activity of ERBB3_MCS6 (Figure 2B). There- fore, although AP2 may be necessary for the activity of ERBB3_MCS6, it is not sufficient for transcriptional activity of ERBB3_MCS6, suggesting that it may act in concert with other transcription factors to mediate its effect on ERBB3_MCS6. Based on the results of our assays, it does not seem that AP2 interacts with SOX10 to regulate ERBB3_MCS6, although we cannot exclude the possibility that the contribution of AP2 to the regula- tion of ERBB3_MCS6 may be rate limited by and second- ary to regulation by SOX10. Therefore, further examination will be required to determine how AP2 may regulate ERBB3_MCS6 and what, if any, co-factors may mediate this regulation. Since AP2 is known to regulate ERBB3, it is not surprising that we see a decrease in Erbb3 protein levels upon Ap2 knockdown in melan-a cells [35,36], however whether this decrease is mediated in part via ERBB3_MCS6 will require further investiga- tion. Similarly, it will also be interesting to determine what, if any, other TFs may be involved in regulating ERBB3 via ERBB3_MCS6 by expanding the search for functionally important sequence motifs therein. Discussion ERBB3_MCS5, MCS7 and MCS8 showed very high levels of H3K4me1, but did not direct tran- scriptional activity in vitro or in vivo, suggesting that additional epigenetic or transcriptional marks may be necessary to distinguish them from active enhancers. Page 10 of 14 Page 10 of 14 Methods Fish husbandry and transgene and morpholino injections Zebrafish were bred and maintained as previously described [44,45]. Microinjections were carried out as previously described [39]. Briefly, eGFP expression vec- tors were injected into 1-2 cell stage embryos (n≥300). Reporter expression was assayed between 24hpf-5dpf and embryos with consistent GFP expression were selected and raised to adulthood and founders were identified. For morhpholino experiments, previously published sox10 morpholino was ordered from Gene Tools (Philomath, OR) [43]. 6.6 ng of the morpholino was injected into each embryo by microinjection. Embryos were analyzed and imaged using a Carl Zeiss Lumar V12 Stereo microscope with AxioVision version 4.8 software. Transgenic lines for ERBB3_MCS1, ERBB3_MCS4 and ERBB3_MCS3 are listed at the Zeb- rafish International Resource Center (ZIRC) (http://zeb- rafish.org) under allele designations JH112, JH113 and JH114 respectively. However, by focusing our investigation on the role of SOX10 in regulating ERBB3_MCS6, we show that it is directly responsive to SOX10. Although there were two potential Sox10 binding sites in ERBB3_MCS6, only one of them, SOXE-2 was required in our in vitro assays. This site is perfectly conserved across mammals and mutation of the site causes a loss of Sox10 responsive- ness of ERBB3_MCS6. Furthermore, mutation of the SOXE-2 site affects the expression of ERBB3_MCS6 in non-ectomesenchymal NC derivatives in zebrafish, but not in ectomesenchymal derivatives of NC or in early premigratory and migratory NC. Although quantitative differences in expression of ERBB3_MCS6 due to the SOXE-2 mutation cannot be ruled out as they are below the threshold of detection by the zebrafish transgenesis Page 11 of 14 Page 11 of 14 Page 11 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Western blot The E189X Sox10 cDNA cloned into a pCMV promo- ter was a kind gift from Ken Inoue, Jim Lupski and Michael Wegner [23,38]. Cells were trypsinized and washed twice in 1XPBS and resuspended in 2X Incomplete Lamelli buffer and passed through a QiaShredder (Invitrogen, Carlsbad, CA) to obtain whole cell lysate. Protein was quantified and run on a 10% Mini Protean TGX gel (Biorad, Hercules, CA), transferred onto a nitrocellulose membrane and blocked overnight in 5% non-fat dry milk block. Sox10 antibody was used at a dilution of 1.5 ng/ul (MAB2864, R&D Bio- systems, Minneapolis, MN), AP2 antibody was used at Luciferase assay melan-a cells were plated in 24-well plates 24 hours prior to transfection at a density of 4 × 10^4 cells/well. 400 ng of the luciferase reporter plasmids were cotrans- fected with 8 ng of CMV-RL renilla expression vector (Promega, Madison, WI). 48 hours after transfection, cell lysate was collected and assayed using the Dual- Luciferase Reporter Assay System (Promega, Madison, WI). For siRNA knockdown, 200 ng of luciferase repor- ter vectors were cotransfected with 5 pmol of each siRNA pool to a total of 10 pmol/well of a 24-well plate. Where appropriate, scrambled siRNA was added to maintain a final concentration of 10 pmol/well of siRNA. Conserved sequences were identified at the human ERBB3 locus and upstream and downstream intergenic regions (chr12:54,724,274-54,784,370) using the Phast- Cons custom track on the UCSC Genome Browser (http://genome.ucsc.edu) on genome build hg18. We used the 17-way MutiZ alignment to identify the most conserved sequences within the introns of the gene and within intergenic regions surrounding the gene. The coordinates of the MCSs and the primers used to amplify them are shown in Additional file 7, Table S1. Fasta format DNA sequence of ERBB3_MCS6 from the genome browser was used to query MatInspector [46], MATCH 1.0 and Jaspar databases for identifiable TFBS using default settings. Neuro2A cells were plated in 24-well plates 24 hours prior to transfection at a density of 5 × 10^4 cells/well. 400 ng of luciferase reporter plasmids were cotrans- fected with 200 ng of Sox10-pcDNA3.1 and ΔSTP- Sox10-pcDNA3.1. Where appropriate, empty pcDNA3.1 vector was added to maintain a final concentration of 800 ng/well of DNA. 8 ng of CMV-RL were added to each well. Cell lysate was collected 24 hours after trans- fection and assayed as mentioned above. Luciferase assays were carried out using a Tecan GENiosPro machine. All assays were performed in triplicate and repeated in at least two independent experiments. siRNA knockdown siRNA knockdown ON-TARGETplus SMARTpool siRNA was ordered against mouse Sox10 and Tfap2a from Dharmacon (Lafayette, CO). Knockdown was achieved by using 5 pmol of each siRNA in a 24-well transfection format. Vector construction and mutagenesis Expression vectors were constructed using Gateway Technology (Invitrogen, Carlsbad, CA). The desired genomic regions were amplified by with attB-flanked primers and recombined into the pDONR221 vector. Successful recombination was confirmed by sequencing. Next, entry clones were recombined into the destination vectors pLGF-E1b for luciferase assays and pT2cfosGW for zebrafish injections [25,47]. For luciferase assays in S16 cells, 1 × 10^4 cells were plated in 96-well plates 24 hours prior to transfection with luciferase vectors. Each transfection reaction included 200 ng of experimental and control luciferase expression constructs and 2 ng of a renilla expression construct to control for transfection efficiency and cell viability. Cells lysates were collected 48 hours after transfection and luciferase assays were carried out with the Dual-Luciferase Assay System (Promega, Madison, WI) and analyzed on a Glomax Multi-Detection System (Promega, Madison, WI). ERBB3_MCS6 was mutated using the QuikChange II XL Site-directed mutagenesis kit (Stratagene, La Jolla, CA). Mutagenesis primers were designed to change the potential transcription factor binding sites (TFBS) to Hpa1 restriction sites using the QuikChange Primer Design tool. Primers are included in Additional file 8, Table S2. Sox10-pcDNA3.1 and Sox10-ΔSTP-pcDNA3.1 were cloned using Sox10-pCMV and Sox10-ΔSTP-pCMV as templates. Coding sequence for human SOX10 and AP2 for transactivation experiments was amplified by PCR from I.M.A.G.E clone MGC-3510 and MGC-22117 respectively and cloned into pcDNA3.1 (Invitrogen, Carlsbad, CA) using In-Fusion PCR Advantage Cloning Kit (Clontech, Mountain View, CA). The PCR primers used were designed using In-Fusion Primer Design tool and are shown in Additional file 9, Table S3. Successful cloning was verified by sequencing. Electrophoretic mobility shift assay Probes spanning the SOXE-2 binding site were designed (Additional file 10, Table S4) and labeled with the Biotin 3’ End DNA Labeling Kit (Pierce, Rockford, IL) accord- ing to manufacturers instructions. Nuclear extract was made from melan-a cells using the NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce, Rockford, IL). EMSAs were performed using the LightShift Chemilu- minescent EMSA Kit (Pierce, Rockford, IL). Briefly, 12.5 fmol of labeled probe was incubated with nuclear extract in the presence of binding buffer and 1 ug of poly (dI.dC) in a 20 ul reaction for 10 min at room tem- perature. For competition assays, 500 and 1000 molar fold excess of unlabeled probe was added. Products were run on precast polyacrylamide gels (4-20% or 7.5%) (Biorad, Hercules, CA) and signal was developed using the LightShift Chemiluminescent kit. Cell culture and transfection Immortalized melanocytes (melan-a) and immortalized Schwann cells were maintained as described [48,49]. NIH 3T3 cells were grown in 10% FCS in DMEM under standard conditions. Neuro2A cells were grown in 10% FCS in MEM under standard conditions. Page 12 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Tris-HCl, 1.2 mM EDTA, 167 mM NaCl, 0.01% SDS, 1.1% Triton X-100, and 1 × Roche Complete, EDTA- free protease inhibitor). Sonication was performed with a Bioruptor (Diagenode, Denville, NJ) with the following settings: high output; 30 second disruption; 30 second cooling; total sonication time of 35 min with addition of fresh ice and cold water to water bath every 10 minutes. 2 ug of antibody specific to H3K4me1 (ab8895, Abcam Cambridge, MA) and non-specific IgG (ab46540, Abcam Cambridge, MA) was used for immunoprecipitation. IP wash conditions were also adjusted from the above referenced protocol, as follows: Each IP was washed twice with low salt wash buffer (0.1% SDS, 1% Triton X- 100, 2 mM EDTA, 20 mM Tris-HCl, 150 mM NaCl), twice with high salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, 500 mM NaCl), twice with cold LiCl wash buffer (0.25M LiCl, 1% IGE- PAL CA630, 1% deoxycholic acid (sodium salt), 1 mM EDTA, 10 mM Tris-HCl), and rinsed once with PBS, pH 7.4. Immunoprecipitated DNA was analyzed by real- time PCR using SYBR Green (ABI, Carlsbad, CA) and primers shown in Additional file 11, Table S4 and per- formed and analyzed on an Opticon2 (Biorad, Hercules, CA) using the % Input method. Primer sequences for PCR are given in supplementary table 5. Tris-HCl, 1.2 mM EDTA, 167 mM NaCl, 0.01% SDS, 1.1% Triton X-100, and 1 × Roche Complete, EDTA- free protease inhibitor). Sonication was performed with a Bioruptor (Diagenode, Denville, NJ) with the following settings: high output; 30 second disruption; 30 second cooling; total sonication time of 35 min with addition of fresh ice and cold water to water bath every 10 minutes. 2 ug of antibody specific to H3K4me1 (ab8895, Abcam Cambridge, MA) and non-specific IgG (ab46540, Abcam Cambridge, MA) was used for immunoprecipitation. NA extraction, cDNA synthesis and Real-time PCR Neuro2A cells were plated in 6-well dish at a density of 2.5 × 10^5 cells/well using Lipofectamine 2000 (Invitro- gen, Carlsbad, CA). A total of 2 ng of Sox10-pcDNA3.1 and ΔSTP-Sox10-pcDNA3.1 were transfected into the cells and RNA was collected 24 hours later using the RNeasy Mini Kit (Qiagen, Valencia, CA) using manufac- turers instructions. Subsequently, cDNA was synthesized using the SuperScriptIII First Strand Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA). Real-time PCR was then performed in triplicate using the Universal Gene Expression Master Mix (ABI, Carlsbad, CA) and the PrimeTime qPCR Assay designed against mouse Sox10 and ErbB3. Real-time PCR and analysis were per- formed on the Opticon2 (Biorad, Hercules, CA). ChIP in S16 cells was performed as previously described [51]. The antibodies used were anti-Sox10 (sc- 17342X, Santa Cruz Biotechnology, Santa Cruz, CA) and control anti-goat IgG (sc-2808, Santa Cruz Biotech- nology, Santa Cruz, CA). ChIP was analyzed in duplicate by quantitative PCR and analyzed by the % Input method. Primers used for PCR are shown in Additional file 11, Table S4. Cell culture and transfection IP wash conditions were also adjusted from the above referenced protocol, as follows: Each IP was washed twice with low salt wash buffer (0.1% SDS, 1% Triton X- 100, 2 mM EDTA, 20 mM Tris-HCl, 150 mM NaCl), twice with high salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, 500 mM NaCl), twice with cold LiCl wash buffer (0.25M LiCl, 1% IGE- PAL CA630, 1% deoxycholic acid (sodium salt), 1 mM EDTA, 10 mM Tris-HCl), and rinsed once with PBS, pH 7.4. Immunoprecipitated DNA was analyzed by real- time PCR using SYBR Green (ABI, Carlsbad, CA) and primers shown in Additional file 11, Table S4 and per- formed and analyzed on an Opticon2 (Biorad, Hercules, CA) using the % Input method. Primer sequences for PCR are given in supplementary table 5. 1:500 (ab52222, Abcam, Cambrigde, MA), ErbB3 anti- body was used at 1:500 (sc285, Santa Cruz Biotechnol- ogy, Santa Cruz, CA) and anti-tubulin at 1:3000 (CP06, Calbiochem, San Diego, CA). HRP-conjugated secondary was used and antibody binding was visualized using SuperSignal West Dura Extended Duration Substrate (34076, Thermo Fischer, Rockford, IL). Membranes were stripped using Restore PLUS Western Blot Strip- ping Buffer (46430, Thermo Fischer, Rockford, IL 1:500 (ab52222, Abcam, Cambrigde, MA), ErbB3 anti- body was used at 1:500 (sc285, Santa Cruz Biotechnol- ogy, Santa Cruz, CA) and anti-tubulin at 1:3000 (CP06, Calbiochem, San Diego, CA). HRP-conjugated secondary was used and antibody binding was visualized using SuperSignal West Dura Extended Duration Substrate (34076, Thermo Fischer, Rockford, IL). Membranes were stripped using Restore PLUS Western Blot Strip- ping Buffer (46430, Thermo Fischer, Rockford, IL References References 1. Basch ML, Garcia-Castro MI, Bronner-Fraser M: Molecular mechanisms of neural crest induction. Birth Defects Res C Embryo Today 2004, 72(2):109-123. 2. Hou L, Pavan WJ: Transcriptional and signaling regulation in neural crest stem cell-derived melanocyte development: do all roads lead to Mitf? Cell Res 2008, 18(12):1163-1176. Additional file 6: Figure S6- Range of eGFP phenotypes in ERBB3_MCS6 transgenic fish upon sox10 morpholino injection. (A-B) eGFP expression driven by ERBB3_MCS6 in uninjected fish at 24hpf. Expression is noted in cranial neural crest (CNC), premigratory NC (PMC) and migratory crest (MC) (C-D) Fewer eGFP positive CNC (C) and PMC cells seen in sox10 morpholino injected transgenic embryos, and significantly reduced numbers of MC (D). 3. McCallion AS, Chakravarti A: EDNRB/EDN3 and Hirschsprung disease type II. Pigment Cell Res 2001, 14(3):161-169. 3. McCallion AS, Chakravarti A: EDNRB/EDN3 and Hirschsprung disease type II. Pigment Cell Res 2001, 14(3):161-169. 4. Britsch S: The neuregulin-I/ErbB signaling system in development and disease. Adv Anat Embryol Cell Biol 2007, 190:1-65. 5. Carraway KL, Sliwkowski MX, Akita R, Platko JV, Guy PM, Nuijens A, Diamonti AJ, Vandlen RL, Cantley LC, Cerione RA: The erbB3 gene product is a receptor for heregulin. J Biol Chem 1994, 269(19):14303-14306. Additional file 7: Table S1- Coordinates of the ERBB3_MCS8 elements (human genome build hg18) and primers used for PCR amplification of each element. 6. Guy PM, Platko JV, Cantley LC, Cerione RA, Carraway KL: Insect cell- expressed p180erbB3 possesses an impaired tyrosine kinase activity. Proc Natl Acad Sci USA 1994, 91(17):8132-8136. 6. Guy PM, Platko JV, Cantley LC, Cerione RA, Carraway KL: Insect cell- expressed p180erbB3 possesses an impaired tyrosine kinase activity. Proc Natl Acad Sci USA 1994, 91(17):8132-8136. Additional file 8: Table S2- Primers used for site-directed mutagenesis of ERBB3_MCS6. 7. Carraway KL, Soltoff SP, Diamonti AJ, Cantley LC: Heregulin stimulates mitogenesis and phosphatidylinositol 3-kinase in mouse fibroblasts transfected with erbB2/neu and erbB3. J Biol Chem 1995, 270(13):7111-7116. Additional file 9: Table S3- Primers used for cloning WT and mutant sox10 cDNA, AP2 and SOX10 cDNA. Additional file 10: Table S4- Probes used for EMSA assay. 8. Wallasch C, Weiss FU, Niederfellner G, Jallal B, Issing W, Ullrich A: Heregulin- dependent regulation of HER2/neu oncogenic signaling by heterodimerization with HER3. EMBO J 1995, 14(17):4267-4275. Additional file 11: Table S5- Primers used for qPCR analysis of ChIP assay. 9. References Erickson SL, O’Shea KS, Ghaboosi N, Loverro L, Frantz G, Bauer M, Lu LH, Moore MW: ErbB3 is required for normal cerebellar and cardiac development: a comparison with ErbB2-and heregulin-deficient mice. Development 1997, 124(24):4999-5011. Author details 1 13. Riethmacher D, Sonnenberg-Riethmacher E, Brinkmann V, Yamaai T, Lewin GR, Birchmeier C: Severe neuropathies in mice with targeted mutations in the ErbB3 receptor. Nature 1997, 389(6652):725-730. 1McKusick-Nathans Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA. 2Genetic Disease Research Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland 20892, USA. 3Department of Genetics, Washington University School of Medicine, St. Louis, Missouri 63110, USA. 4Department of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48109, USA. 5Department of Neurology, University of Michigan Medical School, Ann Arbor, Michigan 48109, USA. 6Program in Cellular and Molecular Biology; University of Wisconsin-Madison, Madison, Wisconsin 53705, USA. 7Department of Comparative Biosciences, University of Wisconsin-Madison, Madison, Wisconsin 53705, USA. 8Department of Molecular and Comparative Pathobiology, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA. 1McKusick-Nathans Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA. 2Genetic Disease Research Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland 20892, USA. 3Department of Genetics, Washington University School of Medicine, St. Louis, Missouri 63110, USA. 4Department of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48109, USA. 5Department of Neurology, University of Michigan Medical School, Ann Arbor, Michigan 48109, USA. 6Program in Cellular and Molecular Biology; University of Wisconsin-Madison, Madison, Wisconsin 53705, USA. 7Department of Comparative Biosciences, University of Wisconsin-Madison, Madison, Wisconsin 53705, USA. 8Department of Molecular and Comparative Pathobiology, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA. 14. Meyer D, Yamaai T, Garratt A, Riethmacher-Sonnenberg E, Kane D, Theill LE, Birchmeier C: Isoform-specific expression and function of neuregulin. Development 1997, 124(18):3575-3586. 15. Britsch S, Li L, Kirchhoff S, Theuring F, Brinkmann V, Birchmeier C, Riethmacher D: The ErbB2 and ErbB3 receptors and their ligand, neuregulin-1, are essential for development of the sympathetic nervous system. Genes Dev 1998, 12(12):1825-1836. 16. Gordon-Thomson C, Jones J, Mason RS, Moore GP: ErbB receptors mediate both migratory and proliferative activities in human melanocytes and melanoma cells. Melanoma Res 2005, 15(1):21-28. 17. Buac K, Xu M, Cronin J, Weeraratna AT, Hewitt SM, Pavan WJ: NRG1/ERBB3 signaling in melanocyte development and melanoma: inhibition of differentiation and promotion of proliferation. Pigment Cell Melanoma Res 2009, 22(6):773-784. Acknowledgements h h ld l k g The authors would like to thank the Avramopoulos lab for their kind gift of Neuro2A cells and Seneca Bessling for her help with zebrafish husbandry. We would also like to thank Zachary Stine for critical scientific discussion and reading of the manuscript and also Grzegorz Burzynski for critical reading of the manuscript. Funding for this work comes from NIH/NIGMS and NIH/NINDS grants to ASM (GM071648 and NS062972), a NIGMS grant to SJ (GM056988), a NIH grant to JS (HD41590), a NIH/NINDS grant to AA (NS073748) and intramural NIH grants to WJP. 10. Britsch S, Goerich DE, Riethmacher D, Peirano RI, Rossner M, Nave KA, Birchmeier C, Wegner M: The transcription factor Sox10 is a key regulator of peripheral glial development. Genes Dev 2001, 15(1):66-78. g The authors would like to thank the Avramopoulos lab for their kind gift of Neuro2A cells and Seneca Bessling for her help with zebrafish husbandry. We would also like to thank Zachary Stine for critical scientific discussion and reading of the manuscript and also Grzegorz Burzynski for critical reading of the manuscript. Funding for this work comes from NIH/NIGMS and NIH/NINDS grants to ASM (GM071648 and NS062972), a NIGMS grant to SJ (GM056988), a NIH grant to JS (HD41590), a NIH/NINDS grant to AA (NS073748) and intramural NIH grants to WJP. The authors would like to thank the Avramopoulos lab for their kind gift of Neuro2A cells and Seneca Bessling for her help with zebrafish husbandry. We would also like to thank Zachary Stine for critical scientific discussion and reading of the manuscript and also Grzegorz Burzynski for critical reading of the manuscript. Funding for this work comes from NIH/NIGMS 11. Lyons DA, Pogoda HM, Voas MG, Woods IG, Diamond B, Nix R, Arana N, Jacobs J, Talbot WS: erbb3 and erbb2 are essential for schwann cell migration and myelination in zebrafish. Curr Biol 2005, 15(6):513-524. and NIH/NINDS grants to ASM (GM071648 and NS062972), a NIGMS grant to SJ (GM056988), a NIH grant to JS (HD41590), a NIH/NINDS grant to AA (NS073748) and intramural NIH grants to WJP. g y 12. Honjo Y, Kniss J, Eisen JS: Neuregulin-mediated ErbB3 signaling is required for formation of zebrafish dorsal root ganglion neurons. Development 2008, 135(15):2615-2625. Additional material Expression is noted in cranial neural crest (CNC), premigratory NC (PMC) and migratory crest (MC) (C-D) Fewer eGFP positive CNC (C) and PMC cells seen in sox10 morpholino injected transgenic embryos, and significantly reduced numbers of MC (D). Additional file 7: Table S1- Coordinates of the ERBB3_MCS8 elements (human genome build hg18) and primers used for PCR amplification of each element. Additional file 8: Table S2- Primers used for site-directed mutagenesis of ERBB3_MCS6. Additional file 9: Table S3- Primers used for cloning WT and mutant sox10 cDNA, AP2 and SOX10 cDNA. Additional file 10: Table S4- Probes used for EMSA assay. Additional file 11: Table S5- Primers used for qPCR analysis of ChIP assay. in melan-a nuclei at the SOXE-2 site. Nuclear extract binds free probe (Lane 1) and shifts it upwards (Lane 2). Addition of 500X (Lane 3) and 1000X (Lane 4) molar excess of unlabeled probes competes shift. Addition of cold unlabeled probe with a mutation in the SOXE-2 binding site does not compete away the shift (Lane 5). in melan-a nuclei at the SOXE-2 site. Nuclear extract binds free probe (Lane 1) and shifts it upwards (Lane 2). Addition of 500X (Lane 3) and 1000X (Lane 4) molar excess of unlabeled probes competes shift. Addition of cold unlabeled probe with a mutation in the SOXE-2 binding site does not compete away the shift (Lane 5). Additional material Additional file 1: Figure S1 - Identification of putative neural crest transcription factor binding sites in ERBB3_MCS6. Sequence of intronic neural crest enhancer ERBB3_MCS6 showing the location of putative transcription factor binding sites (TFBS) identified using multiple TFBS search programs. SOXE-2 adheres to the SOXE binding consensus sequence. Additional file 2: Figure S2 - Transcriptional transactivation of ERBB3_MCS6 by SOX10 and AP2. Luciferase assay of wild type ERBB3_MCS6 in NIH3T3 cells when transiently co-transfected with equal amounts of an empty expression vector (pcDNA.31), SOX10 cDNA or AP2 cDNA. Cell lysates were collected 24 hours post transfection. All values are normalized to a renilla internal control and shown as fold-change compared to the promoter only construct (pe1B) with standard deviation. Chromatin Immunoprecipitation (ChIP) and Real-time PCR ChIP was performed in melan-a cells as previously described [50] with some changes. Each ChIP experi- ment was performed with ~1 × 108 cells. Alternative lysis buffers to those in the referenced protocol were used, as follows: Lysis buffer 1 (5 mM PIPES, 85 mM KCl, 0.5% NP-40, and 1 × Roche Complete, EDTA-free protease inhibitor), lysis buffer 2 (50 mM Tris-HCl, 10 mM EDTA, 1% SDS, and 1 × Roche Complete, EDTA- free protease inhibitor), and lysis buffer 3 (16.7 mM Additional file 3: Figure S3 - Increase in Sox10 transcript levels upon Sox10 overexpression. Real-time PCR results showing an increase in WT and mutant Sox10 cDNA upon transient transfection of WT and Sox10-ΔSTP cDNA in Neuro2A cells. Values are normalized to an 18S internal control and shown as a fold-change compared to the promoter only construct (pcDNA3.1) with standard error. Additional file 4: Figure S4 - A nuclear protein within melan-a cells binds SOXE2. EMSA demonstrating binding of ERBB3_MCS6 to a protein Page 13 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 study and helped write the manuscript. All authors read and approved the final manuscript. study and helped write the manuscript. All authors read and approved the final manuscript. in melan-a nuclei at the SOXE-2 site. Nuclear extract binds free probe (Lane 1) and shifts it upwards (Lane 2). Addition of 500X (Lane 3) and 1000X (Lane 4) molar excess of unlabeled probes competes shift. Competing interests The authors declare that they have no competing interests. The authors declare that they have no competing interests. Additional file 5: Figure S5 - ERBB3_MCS1 and ERBB3_MCS4 drive reporter expression in vivo in a pattern similar to erbb3b. (A-J) Expression pattern of the indicated MCS driving eGFP in G1 transgenic 24-72hpf zebrafish embryos. Arrows indicate tissues where expression was noted in multiple founders. Abbreviations: mesencephalon (M), hindbrain (HB), olfactory bulb (OB), pharyngeal arches (PA), cranial ganglia, posterior lateral line ganglia (PLLg). Received: 25 May 2011 Accepted: 14 June 2011 Published: 14 June 2011 Additional material Addition of cold unlabeled probe with a mutation in the SOXE-2 binding site does not compete away the shift (Lane 5). Additional file 5: Figure S5 - ERBB3_MCS1 and ERBB3_MCS4 drive reporter expression in vivo in a pattern similar to erbb3b. (A-J) Expression pattern of the indicated MCS driving eGFP in G1 transgenic 24-72hpf zebrafish embryos. Arrows indicate tissues where expression was noted in multiple founders. Abbreviations: mesencephalon (M), hindbrain (HB), olfactory bulb (OB), pharyngeal arches (PA), cranial ganglia, posterior lateral line ganglia (PLLg). Additional file 6: Figure S6- Range of eGFP phenotypes in ERBB3_MCS6 transgenic fish upon sox10 morpholino injection. (A-B) eGFP expression driven by ERBB3_MCS6 in uninjected fish at 24hpf. Expression is noted in cranial neural crest (CNC), premigratory NC (PMC) and migratory crest (MC) (C-D) Fewer eGFP positive CNC (C) and PMC cells seen in sox10 morpholino injected transgenic embryos, and significantly reduced numbers of MC (D). Additional file 7: Table S1- Coordinates of the ERBB3_MCS8 elements (human genome build hg18) and primers used for PCR amplification of each element. Additional file 8: Table S2- Primers used for site-directed mutagenesis of ERBB3_MCS6. Additional file 9: Table S3- Primers used for cloning WT and mutant sox10 cDNA, AP2 and SOX10 cDNA. Additional file 10: Table S4- Probes used for EMSA assay. Additional file 11: Table S5- Primers used for qPCR analysis of ChIP assay. in melan-a nuclei at the SOXE-2 site. Nuclear extract binds free probe (Lane 1) and shifts it upwards (Lane 2). Addition of 500X (Lane 3) and 1000X (Lane 4) molar excess of unlabeled probes competes shift. Addition of cold unlabeled probe with a mutation in the SOXE-2 binding site does not compete away the shift (Lane 5). Additional file 5: Figure S5 - ERBB3_MCS1 and ERBB3_MCS4 drive reporter expression in vivo in a pattern similar to erbb3b. (A-J) Expression pattern of the indicated MCS driving eGFP in G1 transgenic 24-72hpf zebrafish embryos. Arrows indicate tissues where expression was noted in multiple founders. Abbreviations: mesencephalon (M), hindbrain (HB), olfactory bulb (OB), pharyngeal arches (PA), cranial ganglia, posterior lateral line ganglia (PLLg). Additional file 6: Figure S6- Range of eGFP phenotypes in ERBB3_MCS6 transgenic fish upon sox10 morpholino injection. (A-B) eGFP expression driven by ERBB3_MCS6 in uninjected fish at 24hpf. Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 21. Southard-Smith EM, Kos L, Pavan WJ: Sox10 mutation disrupts neural crest development in Dom Hirschsprung mouse model. Nat Genet 1998, 18(1):60-64. adrenal agenesis, neural crest defects and exencephaly in mice lacking Cited2, a new Tfap2 co-activator. Nat Genet 2001, 29(4):469-474. adrenal agenesis, neural crest defects and exencephaly in mice lacking Cited2, a new Tfap2 co-activator. Nat Genet 2001, 29(4):469-474. 42. Dutton KA, Pauliny A, Lopes SS, Elworthy S, Carney TJ, Rauch J, Geisler R, Haffter P, Kelsh RN: Zebrafish colourless encodes sox10 and specifies non-ectomesenchymal neural crest fates. Development 2001, 128(21):4113-4125. ( ) 22. Herbarth B, Pingault V, Bondurand N, Kuhlbrodt K, Hermans-Borgmeyer I, Puliti A, Lemort N, Goossens M, Wegner M: Mutation of the Sry-related Sox10 gene in Dominant megacolon, a mouse model for human Hirschsprung disease. Proc Natl Acad Sci USA 1998, 95(9):5161-5165. 23. Pingault V, Bondurand N, Kuhlbrodt K, Goerich DE, Prehu MO, Puliti A, Herbarth B, Hermans-Borgmeyer I, Legius E, Matthijs G, et al: SOX10 mutations in patients with Waardenburg-Hirschsprung disease. Nat G 1998 18(2) 171 173 22. Herbarth B, Pingault V, Bondurand N, Kuhlbrodt K, Hermans-Borgmeyer I, Puliti A, Lemort N, Goossens M, Wegner M: Mutation of the Sry-related Sox10 gene in Dominant megacolon, a mouse model for human Hirschsprung disease. Proc Natl Acad Sci USA 1998, 95(9):5161-5165. 43. Dutton K, Dutton JR, Pauliny A, Kelsh RN: A morpholino phenocopy of the colourless mutant. Genesis 2001, 30(3):188-189. 23. Pingault V, Bondurand N, Kuhlbrodt K, Goerich DE, Prehu MO, Puliti A, Herbarth B, Hermans-Borgmeyer I, Legius E, Matthijs G, et al: SOX10 mutations in patients with Waardenburg-Hirschsprung disease. Nat Genet 1998, 18(2):171-173. 44. Westerfield EM: The Zebrafish Book. A guide for the laboratory use of zebrafish (Danio rerio). University of Oregon Press; 2000. 45. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF: Stages of embryonic development of the zebrafish. Dev Dyn 1995, 203(3):253-310. 24. Antonellis A, Huynh JL, Lee-Lin SQ, Vinton RM, Renaud G, Loftus SK, Elliot G Wolfsberg TG, Green ED, McCallion AS, et al: Identification of neural crest and glial enhancers at the mouse Sox10 locus through transgenesis in zebrafish. PLoS Genet 2008, 4(9):e1000174. 46. Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A, Frisch M, Bayerlein M, Werner T: MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics 2005, 21(13):2933-2942. 25. Authors’ contributions MKP, SKL, WJP and ASM conceived of the study. MKP designed and performed the experiments, analyzed all data, and wrote the manuscript. XR and DG performed the ChIP experiment, JC participated in the luciferase assays, SKL contributed to the study design and data analysis, identified the MCS elements and cloned them into pe1B and pXIG vectors, ARM, KC and SJ participated in the creation and analysis of zebrafish transgenics, EAJ and JS did the chromatin immunoprecipitation assays in S16 cells, CJH and AA did the luciferase assays in S16 cells, WJP participated in the design and analysis of the study, ASM participated in the design and analysis of the MKP, SKL, WJP and ASM conceived of the study. MKP designed and performed the experiments, analyzed all data, and wrote the manuscript. XR and DG performed the ChIP experiment, JC participated in the luciferase assays, SKL contributed to the study design and data analysis, identified the MCS elements and cloned them into pe1B and pXIG vectors, ARM, KC and SJ participated in the creation and analysis of zebrafish transgenics, EAJ and JS did the chromatin immunoprecipitation assays in S16 cells, CJH and AA did the luciferase assays in S16 cells, WJP participated in the design and analysis of the study, ASM participated in the design and analysis of the 18. Budi EH, Patterson LB, Parichy DM: Embryonic requirements for ErbB signaling in neural crest development and adult pigment pattern formation. Development 2008, 135(15):2603-2614. 19. Baselga J, Swain SM: Novel anticancer targets: revisiting ERBB2 and discovering ERBB3. Nat Rev Cancer 2009, 9(7):463-475. 20. Kuhlbrodt K, Herbarth B, Sock E, Hermans-Borgmeyer I, Wegner M: Sox10, a novel transcriptional modulator in glial cells. J Neurosci 1998, 18(1):237-250. Page 14 of 14 Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 BMC Developmental Biology 2011 11:40. 30. Luscher B, Mitchell PJ, Williams T, Tjian R: Regulation of transcription factor AP-2 by the morphogen retinoic acid and by second messengers. Genes Dev 1989, 3(10):1507-1517. 31. Williams T, Admon A, Luscher B, Tjian R: Cloning and expression of AP-2, a cell-type-specific transcription factor that activates inducible enhancer elements. Genes Dev 1988, 2(12A):1557-1569. 32. Mitchell PJ, Timmons PM, Hebert JM, Rigby PW, Tjian R: Transcription factor AP-2 is expressed in neural crest cell lineages during mouse embryogenesis. Genes Dev 1991, 5(1):105-119. 33. Zhang J, Hagopian-Donaldson S, Serbedzija G, Elsemore J, Plehn- Dujowich D, McMahon AP, Flavell RA, Williams T: Neural tube, skeletal and body wall defects in mice lacking transcription factor AP-2. Nature 1996, 381(6579):238-241. 34. Schorle H, Meier P, Buchert M, Jaenisch R, Mitchell PJ: Transcription factor AP-2 essential for cranial closure and craniofacial development. Nature 1996, 381(6579):235-238. 35. Zhu CH, Huang Y, Oberley LW, Domann FE: A family of AP-2 proteins down-regulate manganese superoxide dismutase expression. J Biol Chem 2001, 276(17):14407-14413. 36. Zhu CH, Domann FE: Dominant negative interference of transcription factor AP-2 causes inhibition of ErbB-3 expression and suppresses malignant cell growth. Breast Cancer Res Treat 2002, 71(1):47-57. g g 37. Skinner A, Hurst HC: Transcriptional regulation of the c-erbB-3 gene in human breast carcinoma cell lines. Oncogene 1993, 8(12):3393-3401. 37. Skinner A, Hurst HC: Transcriptional regulation of the c-erbB-3 gene in human breast carcinoma cell lines. Oncogene 1993, 8(12):3393-3401. 38. Inoue K, Khajavi M, Ohyama T, Hirabayashi S, Wilson J, Reggin JD, Mancias P, Butler IJ, Wilkinson MF, Wegner M, et al: Molecular mechanism for distinct neurological phenotypes conveyed by allelic truncating mutations. Nat Genet 2004, 36(4):361-369. Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Fisher S, Grice EA, Vinton RM, Bessling SL, McCallion AS: Conservation of RET regulatory function from human to zebrafish without sequence similarity. Science 2006, 312(5771):276-279. 47. Antonellis A, Bennett WR, Menheniott TR, Prasad AB, Lee-Lin SQ, Green ED, Paisley D, Kelsh RN, Pavan WJ, Ward A: Deletion of long-range sequences at Sox10 compromises developmental expression in a mouse model of Waardenburg-Shah (WS4) syndrome. Hum Mol Genet 2006, 15(2):259-271. y 26. Siepel A, Bejerano G, Pedersen JS, Hinrichs AS, Hou M, Rosenbloom K, Clawson H, Spieth J, Hillier LW, Richards S, et al: Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res 2005, 15(8):1034-1050. 48. Sviderskaya EV, Hill SP, Evans-Whipp TJ, Chin L, Orlow SJ, Easty DJ, Cheong SC, Beach D, DePinho RA, Bennett DC: p16(Ink4a) in melanocyte senescence and differentiation. J Natl Cancer Inst 2002, 94(6):446-454. 27. Heintzman ND, Stuart RK, Hon G, Fu Y, Ching CW, Hawkins RD, Barrera LO, Van Calcar S, Qu C, Ching KA, et al: Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat Genet 2007, 39(3):311-318. 49. Toda K, Small JA, Goda S, Quarles RH: Biochemical and cellular properties of three immortalized Schwann cell lines expressing different levels of the myelin-associated glycoprotein. J Neurochem 1994, 63(5):1646-1657. 50. Lee TI, Johnstone SE, Young RA: Chromatin immunoprecipitation and microarray-based analysis of protein location. Nat Protoc 2006, 1(2):729-748. 28. Creyghton MP, Cheng AW, Welstead GG, Kooistra T, Carey BW, Steine EJ, Hanna J, Lodato MA, Frampton GM, Sharp PA, et al: Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc Natl Acad Sci USA 2010, 107(50):21931-21936. 51. Jones EA, Lopez-Anido C, Srinivasan R, Krueger C, Chang LW, Nagarajan R, Svaren J: Regulation of the PMP22 Gene through an Intronic Enhancer. J Neurosci 2011, 31(11):4242-4250. 51. Jones EA, Lopez-Anido C, Srinivasan R, Krueger C, Chang LW, Nagarajan R, 29. Rada-Iglesias A, Bajpai R, Swigut T, Brugmann SA, Flynn RA, Wysocka J: A unique chromatin signature uncovers early developmental enhancers in humans. Nature 2011, 470(7333):279-283. Svaren J: Regulation of the PMP22 Gene through an Intronic Enhancer. J Neurosci 2011, 31(11):4242-4250. doi:10.1186/1471-213X-11-40 Cite this article as: Prasad et al.: SOX10 directly modulates ERBB3 transcription via an intronic neural crest enhancer. BMC Developmental Biology 2011 11:40. doi:10.1186/1471-213X-11-40 Cite this article as: Prasad et al.: SOX10 directly modulates ERBB3 transcription via an intronic neural crest enhancer. doi:10.1186/1471-213X-11-40 Cite this article as: Prasad et al.: SOX10 directly modulates ERBB3 transcription via an intronic neural crest enhancer. BMC Developmental Biology 2011 11:40. Prasad et al. BMC Developmental Biology 2011, 11:40 http://www.biomedcentral.com/1471-213X/11/40 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 39. Fisher S, Grice EA, Vinton RM, Bessling SL, Urasaki A, Kawakami K, McCallion AS: Evaluating the biological relevance of putative enhancers using Tol2 transposon-mediated transgenesis in zebrafish. Nat Protoc 2006, 1(3):1297-1305. 40. Dutton JR, Antonellis A, Carney TJ, Rodrigues FS, Pavan WJ, Ward A, Kelsh RN: An evolutionarily conserved intronic region controls the spatiotemporal expression of the transcription factor Sox10. BMC Dev Biol 2008, 8:105. 41. Bamforth SD, Braganca J, Eloranta JJ, Murdoch JN, Marques FI, Kranc KR, Farza H, Henderson DJ, Hurst HC, Bhattacharya S: Cardiac malformations, 41. Bamforth SD, Braganca J, Eloranta JJ, Murdoch JN, Marques FI, Kranc KR, Farza H, Henderson DJ, Hurst HC, Bhattacharya S: Cardiac malformations,

https://openalex.org/W4393256712

https://journals.wlb-stuttgart.de/ojs/index.php/sh/article/download/11154/11026

German

Villa rustica von Bochingen

Schwäbische Heimat

cc-by

Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft leicht ist, folgt man der Germania des Tacitus, vor der römischen Besetzung im 1. Jahrhundert von Gallien her allerlei zweifelhaftes Volk, (dann wohl am ehesten keltische) Abenteurer und Flüchtlinge, die die Not verwegen gemacht hat, in das Dekumatland (agri decumates) eingedrungen, ein Gebiet, dessen Besitzverhältnisse ungeklärt waren. Römische Gutshöfe (villae rusticae) gibt es viele. In Baden-Württemberg sind es zwischen eintausend und zweitausend. Rund 1400 davon sind genau lo- kalisierbar. Die große Zahl nimmt nicht wunder, war doch diese Siedlungsart im ländlichen Raum die übliche zur Römerzeit. Daneben gab es im 1. bis 3. Jahrhundert n. Chr. die (Lager-)Dörfer (vici) und die Städte (municipia) oder stadtähnliche Ansied- lungen, die in aller Regel ihre Existenz einem Ka- stell der römischen (Hilfs-)Truppen verdankten. Ob es in dem damals dünn besiedelten Gebiet auch ein- heimische (keltische) Weiler oder Einzelhöfe gege- ben hat, ist noch nicht ganz geklärt. Man vermutet es, auch wenn dem entgegenzustehen scheint, daß die keltischen Helvetier im 1. Jahrhundert v. Chr. unter dem Druck der Germanen das Gebiet zwi- schen Donau und Rhein in Richtung Schweiz ver- lassen haben. Irgendwoher müssen jedenfalls die zahlreichen Erntehelfer, die Knechte und Mägde gekommen sein, die für den Betrieb der römischen Gutshöfe er- forderlich waren. Denn eine Villa rustica war kein Bauernhof, wie wir ihn aus dem Mittelalter und der Neuzeit kennen. Er glich eher einer Farm oder ei- nem ostelbischen Landgut. Lebensmittel wurden hier nicht nur für eine Bauernfamilie und ihr Ge- sinde erzeugt, sondern bewußt im großen Stil pro- duziert. Die Überschußproduktion wanderte dann in die Dörfer und Kastelle. Die Villae sicherten die Versorgung der am Limes stationierten Truppen und ihres Trosses an zivilen Dorfbewohnern, die im Handel, Handwerk und Dienstleistungsgewerbe tätig waren, kaum aber im Gartenbau oder in der Landwirtschaft. In der Helvetiereinöde, wie Klaudios Ptolemaios um 150 n. Chr. in seiner Geographie diese Region in Süd- deutschland beschrieb, mögen einzelne keltische Bewohner vom Stamme der Helvetier oder der Ver- bigener und wohl auch die Vindeliker, die südlich der Donau siedelten, zurückgeblieben sein. Viel- Die Bodensee-Autobahn (am rechten Bildrand) und die Straße nach Oberndorf (die in derMitte waagrecht durchs Bild führt) bilden den Rahmen für das Areal der Villa rustica von Böchingen. Die ersten Häuser dieses Oberndorfer Teilorts sind links oben zu sehen. Der römische Gutshofsetzt sich links der Straße fort, die senkrecht durchs Bild läuft. Hauptgebäude und Badhaus liegen rechts neben dieser Straße. Schwäbische Heimat 96/3 Die nördliche Raum- zeile des Badegebäudes wird freigelegt. Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft Die Bodensee-Autobahn (am rechten Bildrand) und die Straße nach Oberndorf (die in derMitte waagrecht durchs Bild führt) bilden den Rahmen für das Areal der Villa rustica von Böchingen. Die ersten Häuser dieses Oberndorfer Teilorts sind links oben zu sehen. Der römische Gutshofsetzt sich links der Straße fort, die senkrecht durchs Bild läuft. Hauptgebäude und Badhaus liegen rechts neben dieser Straße. Die Bodensee-Autobahn (am rechten Bildrand) und die Straße nach Oberndorf (die in derMitte waagrecht durchs Bild führt) bilden den Rahmen für das Areal der Villa rustica von Böchingen. Die ersten Häuser dieses Oberndorfer Teilorts sind links oben zu sehen. Der römische Gutshofsetzt sich links der Straße fort, die senkrecht durchs Bild läuft. Hauptgebäude und Badhaus liegen rechts neben dieser Straße. 228 Schwäbische Heimat 96/3 Lauffen am Neckar, wo eine große Rebflurbereini- gung erfolgte, und schließlich jener Gutshof im badi- schen Büßlingen am Hochrhein. Die villa von Lauffen ist als Freilichtmuseum hergerichtet worden und er- freut sich regen Besucherinteresses. Die Rekonstruk- tionszeichnung des Gutshofs von Hoheneck, in dem schon Oscar Paret gegraben hatte, dient bis heute als Modell für eine Villa rustica in Obergermanien. Tausend römische Gutshöfe und keiner vollständig erforscht Obwohl die römische Siedlungsform der Villa ru- stica nicht gerade selten und schon seit langem be- kannt ist, blieben die Kenntnisse über sie beschei- den. Zwar haben in etwa zehn Prozent der villae schon einmal Ausgrabungen stattgefunden, doch geschah dies meist im 19. Jahrhundert oder in der ersten Hälfte des 20. Jahrhunderts. Dementspre- chend unzureichend sind die Grabungsmethoden und daher unbefriedigend die Grabungsergebnisse gewesen. Die Nachforschungen galten damals vor allem den Mauerzügen des Zentralgebäudes, in dem der Gutsherr oder Pächter gewohnt hatte. Sel- ten nur erstreckten sie sich auch auf das Bad, ein- zelne Nebengebäude und Teile der Hofmauer. Dar- über hinausgehende Untersuchungen scheiterten oft auch am Geld, denn Flächengrabungen sind teurer und zeitraubender als die Untersuchung nur einzel- ner Objekte. Aufwand und Nutzen werden heute anders in Relation gesetzt. Wer Forschung betreiben und mehr erfahren will, als bereits bekannt ist, muß eine ganze Villa rustica und ihr Umfeld freilegen. So war es naheliegend, daß die Archäologen des Landesdenkmalamts den Entschluß faßten, die seit langem bekannte villa von Oberndorf-Bochingen zur Gänze auszugraben. Den Startschuß hatte je- doch die Stadt Oberndorf gegeben, die darauf be- stand, genau hier, im Grabungsschutzgebiet, ein großes Gewerbegebiet anzulegen und neue Arbeits- plätze zu schaffen. Schwäbische Heimat 96/3 Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft An den Kosten dieser Grabung darf sie sich nun in erheblichem Umfang beteiligen. Seit zwei Jahren untersucht Grabungsleiter Dr. C. Sebastian Sommer mit Grabungstechniker Tho- mas Schlipf und einer großen Mannschaft von Gra- bungshelfern, vielfach Arbeitslosen, die Überreste dieser Villa rustica, die in verschiedener Hinsicht neue Erkenntnisse zu liefern verspricht. Der Guts- hof liegt neben der Autobahntrasse und zählt mit seiner mauerumgebenen Hoffläche von 3,2 Hektar zu den großen Anlagen im Lande. Fünf Grabungs- kampagnen dürften deshalb für die komplette Aus- grabung im Bereich des ersten Abschnitts des Bau- gebiets erforderlich sein. Der überregionalen Be- deutung angemessen wird die Ausgrabung sehr sorgfältig gemacht. Neben der Luftbildarchäologie kommen zur Prospektion auch geophysikalische Untersuchungen auf sieben Hektar Fläche und So war es naheliegend, daß die Archäologen des Landesdenkmalamts den Entschluß faßten, die seit langem bekannte villa von Oberndorf-Bochingen zur Gänze auszugraben. Den Startschuß hatte je- doch die Stadt Oberndorf gegeben, die darauf be- stand, genau hier, im Grabungsschutzgebiet, ein großes Gewerbegebiet anzulegen und neue Arbeits- plätze zu schaffen. An den Kosten dieser Grabung darf sie sich nun in erheblichem Umfang beteiligen. Vollständig erforscht ist bisher noch keine einzige der weit über tausend villae rusticae im Land. Sehr weit- gehend ausgegraben sind fünf römische Gutshöfe: Der in Sontheim/Brenz und jener bei Bondorf im Kreis Böblingen, wo für die Bodensee-Autobahn die kleeblattförmige Anschlußstelle Rottenburg geschaf- fen wurde, die genau auf dem Gelände einer Gutsan- lage geplant war, die villa von Ludwigsburg-Hohen- eck, wo ein Neubaugebiet entstand, und die von 229 Schwäbische Heimat 96/3 er römische Gutshofvon Böchingen mit dem Herrenhaus nach dem Umbau um 200. Rot ausgezogen sind die bisher ausgegrabe- nen Mauerzüge, schwarz der sicher erschlossene und gepunktet der vermutete Mauerverlauf. Grün markiert sind die als Ganzes umgekippten Wände eines Wirtschaftsgebäudes. Straßen, Wege und moderne Bebauung ist blau wiedergegeben. Der römische Gutshofvon Böchingen mit dem Herrenhaus nach dem Umbau um 200. Rot ausgezogen sind die bisher ausgegrabe- nen Mauerzüge, schwarz der sicher erschlossene und gepunktet der vermutete Mauerverlauf. Grün markiert sind die als Ganzes umgekippten Wände eines Wirtschaftsgebäudes. Straßen, Wege und moderne Bebauung ist blau wiedergegeben. Phosphatuntersuchungen zur Anwendung. Das Re- ferat Photogrammetrie des Landesdenkmalamts ist für die exakte Dokumentation eingeschaltet wor- den. Die Villa rustica von Böchingen ist zum Fall für die Wissenschaft geworden. über einer feuchten Niederung liegende Gutshof war von einer Hofmauer umgeben. An der Nord- ostseite, also zur Autobahn hin, verbarg sich die Umfassungsmauer zunächst unter einem Feldweg. Schwäbische Heimat 96/3 Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft Fläche, wo umbiegt, so daß das gesamte Areal berechnet werden kann. Nur mit einer Ecke ist bisher das Hauptgebäude des Gutshofes erfaßt. Die Grabungsbefunde lassen auf eine Eckrisalit-Villa mit Porticus schließen, den gängigen Bautyp im Lande. An der Schaufront gen Südwesten flankierten zwei vorspringende turmar- tige Bauelemente (Risalite) eine offene lange Wan- delhalle (die Porticus) mit einem von Säulen getra- genen Dach. Der Eingang führte in der Regel über eine Treppe oder Rampe in die loggiaähnliche Porti- cus. In den turmartigen 6 mal 6 Meter großen Risa- liten waren die Wohnräume untergebracht, die zum Teil heizbar waren. Nach hinten schlossen sich, um einen Atrium-artigen Hof gelegen, die Schlaf- und sonstigen Räume an. Ein Keller befand sich unter einem Risaliten. Der Bochinger Eckrisalit-Villa ging ein älteres Gebäude voraus. Dieses scheint eine zehn Meter lange Halle besessen zu haben. Das Vor- gängergebäude war in seiner Orientierung um 90 Grad nach Südosten gedreht. Westlich der Villa rustica von Böchingen verläuft die alte Römerstraße von Rottweil nach Sulz am Neckar. Auf Luftbildern ist zu erkennen, daß sie hier einen leichten Bogen macht. Bei der Nachschau konnten die Archäologen im Boden jedoch kaum noch Spu- ren der Straße finden; sie war in den vergangenen Jahren schon weitgehend dem Ackerbau zum Opfer gefallen. Ein Stichsträßchen schloß die Villa ans überörtliche Verkehrsnetz an. Gute Verbindungen waren für römische Gutshöfe wichtig, denn schließ- lich mußten die hier erzeugten landwirtschaftlichen Überschüsse mit Ochsengespannen zu den staatli- chen Lagerhäusern und auf die Märkte in Dörfern und Städten gekarrt werden. Der Staat zog Natural- steuern ein, und die Soldaten benötigten Fourage, - ganz zu schweigen von den Dörflern und Stadtbe- wohnern, die sich nicht selbst ernähren konnten. Abwasserkanal und Estrichboden einer Kaltwasserwanne (Bildmitte, rechts) im Badhaus. Neben dem Herrenhaus fand sich das Badege- bäude, das stets zu einer Villa rustica gehörte. Von dem Bad sind bisher vier Räume untersucht, von denen einer heizbar war und ein weiterer das Kalt- wasserbecken enthielt. Die Funktion der beiden an- deren Räume ist noch nicht gesichert. Der südliche Bereich des Badegebäudes ist noch gar nicht ausge- graben. So kann vorerst nur die West-Ost-Ausdeh- nung gemessen werden. Sie beträgt zehn Meter. Die nördliche Raumzeile scheint später angebaut wor- den zu sein. Frischwasser erhielt das Bad aus einer Quelle im Norden durch eine hölzerne Deicheilei- tung zugeführt. Das Abwasser floß durch einen steinernen Kanal ab. Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft Der Steinschutt der eingestürzten Mauer ist im Mit- telalter oder in der Neuzeit als willkommenes Stein- bett für einen allwettertauglichen Weg verwendet worden. Suchschnitte lösten dann das Rätsel. Die Hofmauer ist auf dieser Seite auf eine Länge von 240 Metern untersucht, einschließlich der Ostecke und eines kurzen Stücks Fortsetzung nach Südwesten. Bei der Auswahl und der Reihenfolge der zu unter- suchenden Teilflächen richten sich die Archäologen nach den Erfordernissen der Flächenerschließung für das Gewerbegebiet. Erst am Ende der Ausgra- bung werden sich deshalb die Teilflächen zu einem Ganzen zusammensetzen lassen, können dann um- fassende und neue Erkenntnisse gewonnen werden. Vorerst sind die Ergebnisse noch lückenhaft und lassen nur bedingt weiterführende Schlüsse zu. Merkwürdigerweise weitet sich das Fundament der Hofmauer auf den letzten 40 Metern bis zur Ost- ecke immer weiter zur Hofinnenseite hin aus, wo- bei die Dicke der aufgehenden, noch drei Steinlagen hoch erhaltenen Mauer aber gleich bleibt. Ob sich hier statische Probleme der mit gut zwei Meter Höhe anzunehmenden Hofmauer erkennen lassen oder eine zweite Bauphase der Mauer, ist den Ar- Hofmauer unterm Feldweg versteckt Hauptgebäude eine Eckrisalit-Villa Der an einem sanften Südosthang in der Flur Breite 230 Schwäbische Heimat 96/3 Schwäbische Heimat 96/3 chäologen noch nicht klar. Untersucht ist auch die Fläche, wo die Mauer im Westen umbiegt, so daß das gesamte Areal berechnet werden kann. Nur mit einer Ecke ist bisher das Hauptgebäude des Gutshofes erfaßt. Die Grabungsbefunde lassen auf eine Eckrisalit-Villa mit Porticus schließen, den gängigen Bautyp im Lande. An der Schaufront gen Südwesten flankierten zwei vorspringende turmar- tige Bauelemente (Risalite) eine offene lange Wan- delhalle (die Porticus) mit einem von Säulen getra- genen Dach. Der Eingang führte in der Regel über eine Treppe oder Rampe in die loggiaähnliche Porti- cus. In den turmartigen 6 mal 6 Meter großen Risa- liten waren die Wohnräume untergebracht, die zum Teil heizbar waren. Nach hinten schlossen sich, um einen Atrium-artigen Hof gelegen, die Schlaf- und sonstigen Räume an. Ein Keller befand sich unter einem Risaliten. Der Bochinger Eckrisalit-Villa ging ein älteres Gebäude voraus. Dieses scheint eine zehn Meter lange Halle besessen zu haben. Das Vor- gängergebäude war in seiner Orientierung um 90 Grad nach Südosten gedreht. chäologen noch nicht klar. Untersucht ist auch die Fläche, wo die Mauer im Westen umbiegt, so daß das gesamte Areal berechnet werden kann. sen Pflichten der Reichsbewohner, denen ansonsten Kulte um ihre eigenen keltischen Götter durchaus erlaubt waren. Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft Es diente dabei zugleich als Wasserspülung für eine an der Nordseite des Bades gelegene Toilette. Dort fanden die Archäologen die Hälfte eines ausgeschnittenen Sitzsteins. In einem quadratischen Steinfundament von 1,5 Me- tern Seitenlänge, das westlich vor den beiden Ge- bäuden zum Vorschein kam, vermuten die Ausgrä- ber die Standfläche für eine Jupitergigantensäule. Solche Säulen, ihren Namen haben sie von der krö- nenden Statue eines Jupiters, der über einen am Bo- den liegenden Giganten hinwegreitet, sind typische Kultsäulen der romanisierten, einheimisch-kelti- schen Bevölkerung, die häufig in villae, aber auch in anderen Siedlungen im Obergermanien aufge- stellt wurden. Verehrt wird damit der oberste Him- melsgott, Jupiter, der manchmal auch Attribute des Kaisers trägt. Die Anbetung der Kapitolinischen Trias mit Jupiter an der Spitze und die Verehrung des vergöttlichten Kaisers gehörten zu den religiö- Abwasserkanal und Estrichboden einer Kaltwasserwanne (Bildmitte, rechts) im Badhaus. Schwäbische Heimat 96/3 231 von zwei Metern Höhe und 80 Zentimetern Breite. Alle Bögen sind mit Schilfsandsteinplatten gebildet, während das Mauerwerk aus Muschelkalkstein be- steht. Vom Giebeldach stammen rund 4,5 Tonnen Zie- gelschutt. Leider ist der Giebel des Hauses weder an der Nord- noch an der Südseite erhalten, weil hier die Steine ins Hausinnere beziehungsweise auf die be- nachbarte Hofmauer fielen und sich dabei aus dem Verband lösten. So ist die genaue Form nicht mehr abzulesen. Ein Walmdach kann das Gebäude jedoch nicht gehabt haben, und aus technischen Gründen muß das Satteldach flach gewesen sein, mit einer Nei- gung von weniger als 35 Grad. Nur dann nämlich bleiben die großen römischen Flach- oder Leistenzie- gel (tegulae) durch ihr Gewicht auf der Dachschräge Hegen und rutschen nicht herab. Bei deutlich steilerer Neigung hätten die Ziegel festgenagelt werden müs- sen. Damit kein Regenwasser ins Dach eindringen konnte, überdeckten halbrunde Hohlziegel (imbices) die nach oben stehenden Randleisten der nebenein- ander verlegten tegulae. Senkrechtes flach gelegt - ein merkwürdiger Mauerfall Senkrechtes flach gelegt - ein merkwürdiger Mauerfall Eine Überraschung erlebten die Ausgräber, als sie sich einem 10 mal 15 Meter großen Steingebäude in der Ostecke der Gutsanlage zuwandten. Nachdem ein Meer von Steinen freigelegt war, zeigte es sich, daß sie von den Wänden des Gebäudes stammten. Drei der vier Mauern des Hauses waren als Ganzes umgekippt. Was einst senkrecht stand, war flach ge- legt, Stein um Stein, dicht an dicht. Die umgeklappte West- und Südwand lagen nun außerhalb des Ge- bäudes, die Nordwand war in das Bauwerk hinein- gekippt. Dieter Kapff Die Villa rustica von Böchingen - Ein Fall für die Wissenschaft Erstaunlich ist, daß die Steine dieser zwei- schalig konstruierten Mauern beim Fallen im Ver- band geblieben sind. Dieser merkwürdige Mauerfall gibt nun den Wissenschaftlern die Möglichkeit, ein- mal ganz genau zu sehen, wie ein römisches Ge- bäude tatsächlich gebaut war. Gewöhnlich sind nämlich die Mauerreste in nur noch geringer Höhe erhalten und müssen dann mit viel Phantasie und nach anderswo vollständiger überkommenen Vor- bildern zu haushohen Wänden ergänzt und rekon- struiert werden. Dank besonderer Umstände ist hier römische Architektur am Stück überliefert - zwar nicht in der Vertikalen, aber in der Horizontalen. Mit Erde verstopft ist derMauerdurchlaß für die Wasserleitung ins Bad. Erstaunliche Architektur auf dem Lande Das Wirtschaftsgebäude der Villa rustica war ver- mutlich zweigeschossig - die 0,9 Meter breiten und 0,5 Meter tiefen Fundamente der Mauern lassen darauf schließen. Die Hausecken waren sehr sorg- fältig gemauert. Im Innenraum gab es keine Tra- gestützen, das Bauwerk war auch nicht in Räume untergliedert. In der nördlichen Hälfte lag eine Feu- erstelle. Von West nach Ost querte eine Deichellei- Am eindrucksvollsten erhalten ist die Längswand auf der Westseite, die einst, das kann man nun mit Si- cherheit sagen, 7,5 Meter hoch gewesen war. Mitten drin hatte sie ein drei Meter breites und mehr als 5,5 Meter hohes Tor mit halbrundem Torbogen. Rechts und links des Tores waren zwei rundbogige Fenster Mit Erde verstopft ist derMauerdurchlaß für die Wasserleitung ins Bad. Mit Erde verstopft ist derMauerdurchlaß für die Wasserleitung ins Bad. 232 232 Schwäbische Heimat 96/3 tung für Frischwasser das Gebäude und endete außerhalb der Hofmauer. Welchem Zweck das Haus gedient hat, ist unklar. Ein Getreidespeicher war es nicht, schließt Dr. C. Sebastian Sommer eine Deu- tungsmöglichkeit aus. Vielleicht ein Stall mit Vieh- tränke? Die Auswertung von Phosphatuntersuchun- gen, die eine Phosphatanreicherung durch Mist nachweisen und damit eine Nutzung als Stall erhär- ten könnten, liegt noch nicht vor. bisher nur aus Lebach im Kreis Saarlouis und aus Bietigheim-Bissingen. In Lebach war 1989 ein 23 mal 12 Meter messendes Wirtschaftsgebäude freige- legt worden, dessen Wände ebenfalls umgekippt waren. Freilich liegt dort nur die Steinfassade, die einer Fachwerkwand außen vorgeblendet war, am Boden. In Böchingen handelt es sich dagegen um eine zweischalige Steinmauer. Auch in Lebach wa- ren die Längswände durch rundbogige 6,5 Meter hohe und 2,5 Meter breite Tore sowie durch Fenster im Format 0,9 mal 2 Meter gegliedert. Das Wirtschaftsgebäude in der östlichen Hofecke führt den Wissenschaftlern vor Augen, daß bei den Römern selbst Zweckgebäude auf dem Lande mas- siv aus Stein gebaut sein konnten und nicht nur, wie im Mittelalter, als Fachwerkkonstruktionen er- richtet wurden. Auch die kunstvolle Einrahmung von Tür und Fenstern und, an anderer Stelle, die Verwendung von profilierten Konsolsteinen, wie man sie bisher nur bei städtischer Architektur oder allenfalls am Herrenhaus einer Villa rustica erwar- tet hätte, führt zu ganz neuen Erkenntnissen über den Bau von Nebengebäuden. Bereits vor zehn Jahren hatte Dr. Ingo Stork vom Landesdenkmalamt in Bietigheim-Bissingen ein rö- misches Heeresproviantamt weitgehend ausgegra- ben, eine 250 mal 200 Meter große Anlage über dem Enztal, die zunächst als Villa rustica angesehen worden war. Erstaunliche Architektur auf dem Lande Hier war ein 40 mal 14 Meter messen- des Lagerhaus in einem Großbrand untergegangen. Unter dem Gewicht des Daches und durch die hit- zebedingte Ausdehnung der inneren Mauerschale war die 40 Meter lange Nordwand nach außen um- gekippt. Die untersten 44 Steinreihen der äußeren Mauerschale, also nicht die ganze Wand des einst zweistöckigen Gebäudes, trafen die Archäologen noch im Verband erhalten an. Da jedoch der oberste Teil der Wand fehlte, konnte die Gebäudehöhe nur geschätzt werden. Nordwestlich zur Mitte des Hofareals hin ist ein weiteres Steingebäude an seiner südlichen Ecke an- gegraben worden: Es mißt 15 mal 25 Meter. Auch hier deutet sich an, daß alle vier Wände, wie bei einem Kartenhaus, en bloc nach außen gekippt sind. Es ist vermutlich ebenfalls ein Wirtschaftsge- bäude und wird im nächsten Jahr ausgegraben. Bei einem fünften Steingebäude waren die Mauern aus- gebrochen. Welchem Zweck es diente, muß offen bleiben.Gebäude mit geschlossen in die Horizontale gelegten Hauswänden sind selten. Man kennt sie Der mit Holz ver- schalte, viereckige Brunnenschacht, in dessen Ecken noch die dunklen Abdrücke kräftiger Pfosten zu sehen sind. Die Südostecke des römischen Gutshofs mit dem nicht parallel zur Hofmauer ausge- richteten Wirtschafts- gebäude - vom Flug- zeug aus gesehen. Gutshöfe erschließen das Hinterland Die Ausgrabung der Villa rustica von Oberndorf- Bochingen steht noch am Anfang. Es ist noch nicht Schwäbische Heimat 96/3 233 Der mit Holz ver- schalte, viereckige Brunnenschacht, in dessen Ecken noch die dunklen Abdrücke kräftiger Pfosten zu sehen sind. Der mit Holz ver- schalte, viereckige Brunnenschacht, in dessen Ecken noch die dunklen Abdrücke kräftiger Pfosten zu sehen sind. Der mit Holz ver- schalte, viereckige Brunnenschacht, in dessen Ecken noch die dunklen Abdrücke kräftiger Pfosten zu sehen sind. Schwäbische Heimat 96/3 233 233 Schwäbische Heimat 96/3 Schwäbische Heimat 96/3 Die Südostecke des römischen Gutshofs mit dem nicht parallel zur Hofmauer ausge- richteten Wirtschafts- gebäude - vom Flug- zeug aus gesehen. märkte, dort konnte man sich mit Waren eindecken, die man auf dem Hof nicht selbst herstellen konnte. Die Entfernungen zwischen den ländlichen Nieder- lassungen richteten sich auch nach der für den Gutshof benötigten Wirtschaftsfläche und nach den topographischen Gegebenheiten. Für die villa in Böchingen rechnet man mit ein bis zwei Quadratki- lometer Wirtschaftsareal, also hundert oder zwei- hundert Hektar. möglich, sich ein abschließendes Urteil zu bilden und ein gültiges Bild von der römischen Gutsan- lage zu zeichnen. Unter dem Vorbehalt, daß nur der aktuelle Erkenntnisstand wiedergegeben werden kann, soll dennoch versucht werden, die Geschichte der Villa rustica nachzuzeichnen. Nachdem die römischen Truppen vom Hochrhein nach Norden und weiter ins Dekumatland vorge- drungen waren, ist das Gebiet um Rottweil zur Etappe geworden. Mit der Schaffung der Provinz Obergermanien im Jahre 86 n. Chr. sind auch die zi- vilen Verwaltungsstrukturen entstanden, die es er- laubten, auf dem durch die Eroberung zum kaiserli- chen Besitz gewordenen Land Zivilpersonen anzu- siedeln. Dies konnte durch Erwerb oder vor allem durch Vergabe in Erbpacht an Kolonisten aus Italien und Gallien oder an Veteranen erfolgen, also an aus dem Militärdienst ehrenhaft entlassene Legionäre. Die breitgefächerte und landschaftsprägende Er- schließung des besetzten Landes mit villae rusticae war auch für das wirtschaftliche Wachstum der Pro- vinz wichtig. Die Aufsiedelung mit Landgütern un- terschiedlicher Größe, sie konnten bis zu hundert Hektar umfassen, folgte meist den Römerstraßen, die das Militär zum raschen Vorrücken angelegt hatte. So reihen sich in wenigen Kilometern Ab- stand die römischen Gutshöfe auf. In dichtbesiedel- ten Gebieten liegen sie in Sichtnähe. Der nächste Nachbar der Bochinger Villa rustica ist nur 1,5 Kilo- meter entfernt. Zum vicus in Sulz sind es rund sie- ben, zum municipium Arae Flaviae (Rottweil) etwa sechzehn Kilometer. Schwäbische Heimat 96/3 Gutshöfe erschließen das Hinterland Ein steigender Militärhaushalt und fal- lende Erzeugerpreise gehören dazu; weiterhin eine deutliche Bevölkerungsvermehrung, deren Folgen im landwirtschaftlichen Bereich durch einen wach- senden Arbeitskräftemangel verschärft wurden, so- wie Landflucht und Höfesterben. gentum eines Römers - Grundbesitz war schon da- mals eine beliebte Vermögensanlage - verwaltete. Auch das Badegebäude ist durch einen Anbau er- weitert worden. Die geringe Zahl der Gebäude - bisher sind nur vier Steinhäuser bekannt, die im übrigen nicht alle gleichzeitig existiert haben müssen - und der schwere, teilweise nasse Boden lassen die Archäolo- gen vermuten, daß man sich in der Villa rustica mit Weidewirtschaft und Viehzucht abgegeben hat und nicht - jedenfalls nicht überwiegend - mit dem Ackerbau, wie in der Mehrzahl der villae sonst. Mit der Aufzucht von Zugtieren und mit der Fleisch- und Milchproduktion war damals durchaus For- tüne zu machen. Ob in diesem Zusammenhang das Wirtschaftsgebäude in der Ostecke der Gutsanlage als Stall gedeutet werden kann - eventuell mit Un- terkunft für die Hirten - oder als Remise, ist noch nicht entschieden. Die große Nähe zur Hofmauer und die mit dem Mauerverlauf nicht parallelen Fluchten des Gebäudes könnten darauf hindeuten, daß das Gebäude älter ist als die Hofmauer. Aller- dings zeigten die Ausgrabungen, daß der Bauhori- zont bei beiden der gleiche ist. Vor den Alamannen, die erstmals um 233 n. Chr. den Limes überrannten und raubend, mordend und brandschatzend durch das bis dahin ruhige Hinter- land streiften, flüchteten viele Gallorömer in si- chere, grenzferne Regionen oder hinter die Mauern der Städte. Ein Wiederaufbau der zerstörten Ge- bäude war kostspielig und nur schwer zu bewerk- stelligen. Der Staat forderte immer höhere Steuern, um seine ausufernden Aufgaben im Riesen-Impe- rium zwischen Nordsee und Rotem Meer finanzie- ren zu können. Dazu kam die schleichende Geld- entwertung. So ist der Silbergehalt der Denare ver- ringert worden, um aus der gleichen Edelmetall- menge mehr Münzen und damit höheren Gewinn schlagen zu können. Nun rächten sich auch die Sünden wider die Natur, die mit der intensiven Bewirtschaftung, ja Ausbeu- tung der Landschaft durch die Römer verbunden sind. Der Raubbau in den Wäldern, wo in großen Mengen Holz geschlagen wurde für Bauten, Ver- hüttung und Energiegewinnung, vor allem aber für die effektiven, jedoch brennstoffintensiven Heizun- gen in Wohnräumen, Bädern und Schreibstuben, Niedergang im 3. Jahrhundert Niedergang im 3. Jahrhundert Noch interessanter als die Geschichte des Baus ist die Geschichte vom Zerfall der Gutsanlage, - auch wenn erst einzelne Details davon erkennbar sind. Schwäbische Heimat 96/3 Die am Stück um- gekippte Westmauer des Wirtschaftsgebäu- des mit den Tür- und Fensterbögen. Gutshöfe erschließen das Hinterland Dort lagen die Hauptabsatz- Nachdem die römischen Truppen vom Hochrhein nach Norden und weiter ins Dekumatland vorge- drungen waren, ist das Gebiet um Rottweil zur Etappe geworden. Mit der Schaffung der Provinz Obergermanien im Jahre 86 n. Chr. sind auch die zi- vilen Verwaltungsstrukturen entstanden, die es er- laubten, auf dem durch die Eroberung zum kaiserli- chen Besitz gewordenen Land Zivilpersonen anzu- siedeln. Dies konnte durch Erwerb oder vor allem durch Vergabe in Erbpacht an Kolonisten aus Italien und Gallien oder an Veteranen erfolgen, also an aus dem Militärdienst ehrenhaft entlassene Legionäre. Die hölzerne Verschalung am Grunde eines Brun- nenschachtes auf dem villa-Areal in Böchingen hat es den Dendrochronologen erlaubt, die Bauzeit des Römerbrunnens auf die Zeit ums Jahr 97 n. Chr. festzulegen. Damals wird wohl der erste Gutshof an dieser Stelle gebaut worden sein. Sicher umfaßte er nur Holz- oder Fachwerkhäuser. Spätestens um die Mitte des 2. Jahrhunderts ist ein Herrenhaus aus Stein samt einem kleinen Badegebäude errichtet worden. Zur ersten Hofanlage dürften die römi- schen Brandgräber gehört haben, die später beim Bau der erweiterten Hofmauer im Osten zerstört worden sind. Die breitgefächerte und landschaftsprägende Er- schließung des besetzten Landes mit villae rusticae war auch für das wirtschaftliche Wachstum der Pro- vinz wichtig. Die Aufsiedelung mit Landgütern un- terschiedlicher Größe, sie konnten bis zu hundert Hektar umfassen, folgte meist den Römerstraßen, die das Militär zum raschen Vorrücken angelegt hatte. So reihen sich in wenigen Kilometern Ab- stand die römischen Gutshöfe auf. In dichtbesiedel- ten Gebieten liegen sie in Sichtnähe. Der nächste Nachbar der Bochinger Villa rustica ist nur 1,5 Kilo- meter entfernt. Zum vicus in Sulz sind es rund sie- ben, zum municipium Arae Flaviae (Rottweil) etwa sechzehn Kilometer. Dort lagen die Hauptabsatz- Man kann annehmen, daß das erste steinerne Guts- haus mit einem angebauten Keller noch nicht dem Typ der repräsentativen Porticus-Villa mit Eckrisali- ten entsprochen hat, sondern schlichter gebaut war. Ein Umbau im letzten Drittel des 2. Jahrhunderts oder vielleicht erst zu Beginn des 3. Jahrhunderts trug dem wachsenden Wohlstand des Hofeigentü- mers oder Pächters Rechnung. Der Mann könnte durchaus ein Einheimischer, ein Kelte, gewesen sein, der es zu etwas gebracht hatte oder der das Ei- 234 Schwäbische Heimat 96/3 Schwäbische Heimat 96/3 Die am Stück um- gekippte Westmauer des Wirtschaftsgebäu- des mit den Tür- und Fensterbögen. zweiten Viertel des 3. Jahrhunderts ausbreitete. Überall zeichnete sich damals der wirtschaftliche Niedergang ab. Mehrere Ursachen haben dazu bei- getragen. Gutshöfe erschließen das Hinterland Die villa von Böchingen blieb nicht unberührt von der Rezession, die sich im ganzen Römerreich im 235 Schwäbische Heimat 96/3 Eine Schicht zerbro- chener Dachziegel kam zutage, als die Archäologen die Steine der umgekipp- ten Mauern des Wirt- schaftsgebäudes ent- fernt hatten. Noch mit Steinen bedeckt sind die rechtwinkligen Vermessungsstege. Deutlich treten nun der Grundriß des Wirtschaftsgebäudes und der der Hofmauer in der Südostecke der Villa rustica hervor. Deutlich treten nun der Grundriß des Wirtschaftsgebäudes und der der Hofmauer in der Südostecke der Villa rustica hervor. wenden zu können. Nur ein Drittel der römischen Leistenziegel, die für die Dachdeckung nötig gewe- sen waren, sind übriggeblieben, als das Dach schließ- lich herunterbrach. Wissenschaftler haben dies aus der Menge des Ziegelschutts errechnet. Dann ver- ging noch einige Zeit, bis die Ostmauer des nun für Wind und Wetter offenen Gebäudes einstürzte. Auch die Hofmauer, die nun ein viel zu großes Areal ein- friedete, bröckelte altersschwach ab und wurde nicht mehr repariert. Als Ruine hat das Wirtschaftsge- bäude noch einige Jahrzehnte weiterexistiert. Im Südwesten schwemmte der Regen eine zehn Zenti- meter dicke Erdschicht auf den Ziegelschutt. Danach erst, viel später, kippten die übrigen Wände des Bochinger Wirtschaftsgebäudes am Stück um. Dr. C. Sebastian Sommer hat für diesen Fall keine andere Erklärung, als daß die Wände bei einem Erdbeben ins Wanken geraten und schließlich um- gekippt sind. Das ist gut möglich, denn am östli- chen Schwarzwaldrand gibt es einige kleine Erdbe- benherde, die auch in der späten Römerzeit aktiv gewesen sein konnten. Man wird von einem loka- len Herd ausgehen müssen, der keine weitreichen- den Folgen gehabt hat, das heißt, der nicht über hundert und mehr Kilometer hinweg die Erde be- ben ließ und Gebäude zum Einsturz brachte. Pro- fessor Götz Schneider vom Institut für Geophysik der Universität Stuttgart hält deshalb ein Beben im bekannten Hohenzollerngraben als Ursache des Bochinger Falls für unwahrscheinlich. Ganz ausge- schlossen sei auch eine durch das wenden zu können. Nur ein Drittel der römischen Leistenziegel, die für die Dachdeckung nötig gewe- sen waren, sind übriggeblieben, als das Dach schließ- lich herunterbrach. Wissenschaftler haben dies aus der Menge des Ziegelschutts errechnet. Dann ver- ging noch einige Zeit, bis die Ostmauer des nun für Wind und Wetter offenen Gebäudes einstürzte. Auch die Hofmauer, die nun ein viel zu großes Areal ein- friedete, bröckelte altersschwach ab und wurde nicht mehr repariert. Als Ruine hat das Wirtschaftsge- bäude noch einige Jahrzehnte weiterexistiert. Schwäbische Heimat 96/3 Gutshöfe erschließen das Hinterland Ein Blick auf die geologische Karte des Landes läßt Störungen im Aufbau des Untergrundes erkennen, die ganz in der Nähe der Villa rustica von Böchin- gen verlaufen. Die bedeutendste Störung im Schichtaufbau ist der Harthauser Graben, der bis auf zwei Kilometer an Böchingen heranreicht. Auch «kleine Ereignisse», wie sie die Erdbebenforscher nennen, können Auslöser für Effekte sein, wie etwa Hangbewegungen, die dann zum Einsturz der Wände führen konnten. Gerade an leichten Hängen und in Talauen, die wie Vergrößerer wirken, das ist un- ter Forschern seit langem bekannt, sind die Schäden viel größer als auf den Höhen. Diese Lagebeschrei- bung trifft auf den römischen Gutshof genau zu. Daß das Gebiet um Rottweil heute noch nicht zur Ruhe gekommen ist, zeigte ein kleines Beben im Ja- nuar 1995. Damals war ein lokaler Bebenherd wie- der aktiv geworden. Das Epizentrum, von dem die Stoßwellen ausgingen, lag sechzehn Kilometer südöstlich von Böchingen. Erdbeben, das im nordschweizerischen Augst am Hochrhein, dem römischen Augusta Raurica, von Archäologen nachgewiesen und in den Zeitraum um 260 n. Chr. datiert wurde. Wann das Beben stattfand, das die Wände des Wirt- schaftsgebäudes direkt oder indirekt umgelegt hat, ist nicht festzustellen. Berichte über Erdbeben aus dieser Zeit existieren nicht. Wer hätte auch solche, ohnehin wohl nur lokal spürbaren Ereignisse auf- zeichnen sollen, in einer Zeit des Niedergangs und der allgemeinen Unruhen? Die Erdbebenverzeich- nisse des Geologischen Landesamts beginnen erst um die Jahrtausendwende. Ein Blick auf die geologische Karte des Landes läßt Störungen im Aufbau des Untergrundes erkennen, die ganz in der Nähe der Villa rustica von Böchin- gen verlaufen. Die bedeutendste Störung im Schichtaufbau ist der Harthauser Graben, der bis auf zwei Kilometer an Böchingen heranreicht. Auch «kleine Ereignisse», wie sie die Erdbebenforscher nennen, können Auslöser für Effekte sein, wie etwa Hangbewegungen, die dann zum Einsturz der Wände führen konnten. Gerade an leichten Hängen und in Talauen, die wie Vergrößerer wirken, das ist un- ter Forschern seit langem bekannt, sind die Schäden viel größer als auf den Höhen. Diese Lagebeschrei- bung trifft auf den römischen Gutshof genau zu. Daß das Gebiet um Rottweil heute noch nicht zur Ruhe gekommen ist, zeigte ein kleines Beben im Ja- nuar 1995. Damals war ein lokaler Bebenherd wie- der aktiv geworden. Das Epizentrum, von dem die Stoßwellen ausgingen, lag sechzehn Kilometer südöstlich von Böchingen. So bleibt manches an dem Fall im Dunkeln. Gutshöfe erschließen das Hinterland Im Südwesten schwemmte der Regen eine zehn Zenti- meter dicke Erdschicht auf den Ziegelschutt. dieser Raubbau führte zu Erosionserscheinungen, zu Überschwemmungen und zur Klimaverschlech- terung. Schlimme Mißernten waren die Folge. Auch in guten Jahren gaben die ausgelaugten Böden nicht mehr soviel her wie einst. Schädlingsbefall und Viehseuchen kamen hinzu. Kaum ein Gutsherr, der damals nicht den Gürtel enger schnallen mußte; manch einen trieben diese Schicksalsschläge an den Rand der Existenz. Arbeitskräfte konnten nicht mehr entlohnt und mußten entlassen werden, ganze Gebäude wurden aufgegeben, weil man sie für die geringer gewordene Produktion nicht mehr benötigte oder auch nur, weil man sie nicht mehr weiter unterhalten konnte. Der Gutsherr zog sich mit seiner Familie in kleinere, weniger komfortable Gemächer zurück. Das Badegebäude blieb leer und kalt, mindestens wurde die Nutzung eingeschränkt. In der Bochinger villa läßt sich diese neue Sparsam- keit in Ansätzen nachweisen. Den Archäologen fiel auf, daß aus dem Badegebäude die Heizanlage aus- gebaut worden war. Die «ausgeschlachteten» Hy- pokaustpfeiler der Unterbodenheizung wurden of- fenbar an anderer Stelle dringender benötigt. Danach erst, viel später, kippten die übrigen Wände des Bochinger Wirtschaftsgebäudes am Stück um. Dr. C. Sebastian Sommer hat für diesen Fall keine andere Erklärung, als daß die Wände bei einem Erdbeben ins Wanken geraten und schließlich um- gekippt sind. Das ist gut möglich, denn am östli- chen Schwarzwaldrand gibt es einige kleine Erdbe- benherde, die auch in der späten Römerzeit aktiv gewesen sein konnten. Man wird von einem loka- len Herd ausgehen müssen, der keine weitreichen- den Folgen gehabt hat, das heißt, der nicht über hundert und mehr Kilometer hinweg die Erde be- ben ließ und Gebäude zum Einsturz brachte. Pro- fessor Götz Schneider vom Institut für Geophysik der Universität Stuttgart hält deshalb ein Beben im bekannten Hohenzollerngraben als Ursache des Bochinger Falls für unwahrscheinlich. Ganz ausge- schlossen sei auch eine Schadenswirkung durch das Die Erde bebte und verursachte den «Bochinger Fall» Die Erde bebte und verursachte den «Bochinger Fall» Das Wirtschaftsgebäude im Osten des Areals ist ebenfalls nach langer Nutzungszeit aufgegeben wor- den. Hierzu hat man sogar das Dach abgedeckt, um die noch guten Ziegel an anderer Stelle wiederver- Schwäbische Heimat 96/3 Schwäbische Heimat 96/3 236 Erdbeben, das im nordschweizerischen Augst am Hochrhein, dem römischen Augusta Raurica, von Archäologen nachgewiesen und in den Zeitraum um 260 n. Chr. datiert wurde. Gutshöfe erschließen das Hinterland Sicher ist hingegen, daß das Wirtschaftsgebäude im Guts- hof von Böchingen nicht einem Brand und nicht dem Wüten der Alamannen zum Opfer gefallen ist, sondern von den Römern längst aufgegeben war, als es in Teilen zusammenbrach und später ganze Wände der Ruine umkippten. Die Ausgrabungen in dem römischen Gutshof wer- den noch manches Rätsel für die Archäologen be- reithalten, aber wohl auch viele bisher offene Fragen beantworten. Römische Architektur und das Sied- lungswesen, wirtschaftliche und soziale Aspekte der Zeit können hier studiert werden. Das Wissen über das Phänomen Villa rustica in Obergermanien wird am Ende deutlich größer sein. Schicht um Schicht arbeiten sich die Ausgräber in der Südostecke des römischen Gutshofs tiefer. Alle Befunde werden sorgfältig zeichnerisch dokumentiert. Vorne im Bild derZiegelschutt vom Dach des Wirtschaftsgebäudes. Schicht um Schicht arbeiten sich die Ausgräber in der Südostecke des römischen Gutshofs tiefer. Alle Befunde werden sorgfältig zeichnerisch dokumentiert. Vorne im Bild derZiegelschutt vom Dach des Wirtschaftsgebäudes. Schwäbische Heimat 96/3 Schwäbische Heimat 96/3 237

https://openalex.org/W3099194188

https://bmchealthservres.biomedcentral.com/track/pdf/10.1186/s12913-020-05851-w

English

The Integrated Tracking, Referral, and Electronic Decision Support, and Care Coordination (I-TREC) program: scalable strategies for the management of hypertension and diabetes within the government healthcare system of India

BMC health services research

cc-by

Patel et al. BMC Health Services Research (2020) 20:1022 https://doi.org/10.1186/s12913-020-05851-w Patel et al. BMC Health Services Research (2020) 20:1022 https://doi.org/10.1186/s12913-020-05851-w STUDY PROTOCOL Open Access The Integrated Tracking, Referral, and Electronic Decision Support, and Care Coordination (I-TREC) program: scalable strategies for the management of hypertension and diabetes within the government healthcare system of India Shivani A. Patel1*† , Hanspria Sharma2†, Sailesh Mohan3,4, Mary Beth Weber1, Devraj Jindal3, Prashant Jarhyan3, Priti Gupta3, Rakshit Sharma2, Mumtaj Ali3, Mohammed K. Ali1,5, K. M. Venkat Narayan1, Dorairaj Prabhakaran3,4, Yashdeep Gupta6, Ambuj Roy7 and Nikhil Tandon6 The Integrated Tracking, Referral, and Electronic Decision Support, and Care Coordination (I-TREC) program: scalable strategies for the management of hypertension and diabetes within the government healthcare system of India Shivani A. Patel1*† , Hanspria Sharma2†, Sailesh Mohan3,4, Mary Beth Weber1, Devraj Jindal3, Prashant Jarhyan3, Priti Gupta3, Rakshit Sharma2, Mumtaj Ali3, Mohammed K. Ali1,5, K. M. Venkat Narayan1, Dorairaj Prabhakaran3,4, Yashdeep Gupta6, Ambuj Roy7 and Nikhil Tandon6 Abstract Background: Hypertension and diabetes are among the most common and deadly chronic conditions globally. In India, most adults with these conditions remain undiagnosed, untreated, or poorly treated and uncontrolled. Innovative and scalable approaches to deliver proven-effective strategies for medical and lifestyle management of these conditions are needed. Methods: The overall goal of this implementation science study is to evaluate the Integrated Tracking, Referral, Electronic decision support, and Care coordination (I-TREC) program. I-TREC leverages information technology (IT) to manage hypertension and diabetes in adults aged ≥30 years across the hierarchy of Indian public healthcare facilities. The I-TREC program combines multiple evidence-based interventions: an electronic case record form (eCRF) to consolidate and track patient information and referrals across the publicly-funded healthcare system; an electronic clinical decision support system (CDSS) to assist clinicians to provide tailored guideline-based care to patients; a revised workflow to ensure coordinated care within and across facilities; and enhanced training for physicians and nurses regarding non-communicable disease (NCD) medical content and lifestyle management. The program will be implemented and evaluated in a predominantly rural district of Punjab, India. The evaluation will employ a quasi-experimental design with mixed methods data collection. Evaluation indicators assess changes in the continuum of care for hypertension and diabetes and are grounded in the Reach, Effectiveness, Adoption Implementation, and Maintenance (RE-AIM) framework. Data will be triangulated from multiple sources, including (Continued on next page) * Correspondence: [email protected] †Shivani A. Patel and Hanspria Sharma contributed equally to this work. 1Department of Global Health, Emory University, 1518 Clifton Rd NE / Rm 7037, Atlanta, USA Full list of author information is available at the end of the article * Correspondence: [email protected] †Shivani A. Patel and Hanspria Sharma contributed equally to this work. 1Department of Global Health, Emory University, 1518 Clifton Rd NE / Rm 7037, Atlanta, USA Full list of author information is available at the end of the article * Correspondence: [email protected] †Shivani A. Patel and Hanspria Sharma contributed equally to this work. 1Department of Global Health, Emory University, 1518 Clifton Rd NE / Rm 7037, Atlanta, USA Full list of author information is available at the end of the article ( p p g ) community surveys, health facility assessments, stakeholder interviews, and patient-level data from the I-TREC program’s electronic database. p p g community surveys, health facility assessments, stakeholder interviews, and patient-level data from the I-TREC program’s electronic database. Discussion: I-TREC consolidates previously proven strategies for improved management of hypertension and diabetes at single-levels of the healthcare system into a scalable model for coordinated care delivery across all levels of the healthcare system hierarchy. Findings have the potential to inform best practices to ultimately deliver quality public-sector hypertension and diabetes care across India. Trial registration: The study is registered with Clinical Trials Registry of India (registration number CTRI/2020/01/ 022723). The study was registered prior to the launch of the intervention on 13 January 2020. The current version of protocol is version 2 dated 6 June 2018. Keywords: Hypertension, Diabetes, Health system, Information technology, mHealth, Implementation science, Quality improvement, Continuum of care Contributions to the literature disproportionately high in rural settings [11, 12], where the majority of the Indian population resides.
The Government of India, beholden to its population of 1.3 billion, has developed an electronic “NCD Portal” that consists of an electronic case record form (eCRF) to manage non-communicable diseases (NCDs) within the government sector. The I-TREC program builds on the eCRF by integrating a proven-effective clinical decision support system for hypertension and diabetes care, accompanied by clinical training, to assist with patient management.
The Government of India, beholden to its population of 1.3 billion, has developed an electronic “NCD Portal” that consists of an electronic case record form (eCRF) to manage non-communicable diseases (NCDs) within the government sector. The I-TREC program builds on the eCRF by integrating a proven-effective clinical decision support system for hypertension and diabetes care, accompanied by clinical training, to assist with patient management. India’s rural healthcare system is currently organized as a hierarchy of facilities that range from relatively lower-skilled personnel supported by simple infrastruc- ture at the village level to relatively higher-skilled personnel supported by sophisticated infrastructure at the district level. This model attempts to maximize geo- graphical coverage by allowing for “up referrals” and “down referrals” across levels of the healthcare system so that the demand of the individual patient can be met by appropriate resources, such as skilled human resources, infrastructure and services. The referral linkages be- tween these institutions, while theoretically in place, are not implemented efficiently or cohesively. In practice, patients access any level of the healthcare system con- venient for them, resulting in a mismatch between pa- tient needs and resource availability. Challenges to the system are compounded by the heterogeneity of treat- ment guidelines, diagnostic modalities, and medications [13–17] needed at all levels of health care to appropri- ately serve the growing population with NCDs alongside the large population seeking care for maternal and child health and infectious diseases [18–21].
We describe the evaluation protocol for the I-TREC multi-component strategy to improve diabetes and hypertensions care at all levels of the four-tier healthcare system in India.
Lessons learned may inform optimal approaches to improve healthcare processes and health outcomes within the public sector healthcare system in India and in other similar settings. © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Page 2 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Background upon the tremendous need and political will to identify optimal and scalable approaches to expand successful care models to manage blood pressure and diabetes within the public healthcare system, we developed the Integrated Tracking, Referral, Electronic decision sup- port, and Care coordination (I-TREC) program. We de- scribe the components of the I-TREC program and its evaluation design. Setting and target population g g p p I-TREC was developed as a collaboration between the All India Institute of Medical Sciences, New Delhi (AIIMS), the Centre for Chronic Disease Control (CCDC), and Emory University. For over a decade, these three institutions have collaboratively developed and tested the combination of information technology (IT), enhanced personnel training, and workflow alterations to improve the quality of care for diabetes and hyperten- sion in diverse settings across India [22–29]. Most of these prior efforts focused on a single level of the health- care system and relied on research staff to implement the intervention. In I-TREC, however, our goal was to develop and evaluate a coordinated package of tested tools and provider training approaches that catered to functions and personnel available in each type of the 4- tier healthcare system (see Table 1). We further sought to embed the program within the infrastructural scaf- folding provided by the Government of India in the interest of future scalability. I-TREC was thus designed to be implemented by personnel and using resources (medications, diagnostics) already present within the public healthcare system. Background Hypertension and diabetes together affect over 275 mil- lion Indians and their families [1]. These conditions are rising rapidly in all regions of India, commonly co-occur [2–4], and are associated with several adverse health outcomes—such as higher rates of death, myocardial in- farction, stroke, blindness, kidney failure. Yet, both hypertension and diabetes are treatable such that timely and appropriate therapy mitigates associated morbidity due to complications and premature mortality. While lack of diagnosis is among the major obstacles to seeking appropriate treatment [5, 6], treatment outcomes even after diagnosis are far from ideal. Less than half of indi- viduals who have hypertension and diabetes in the com- munity are aware of their condition [7, 8], and only 20– 25% achieve adequate blood pressure [9] or blood glu- cose control [10]. Under-diagnosis, under-treatment, and poor control for both hypertension and diabetes are Recognizing the growing burden of hypertension and diabetes across all segments of the population, the Min- istry of Health & Family Welfare, Government of India, has taken the initiative to integrate screening and man- agement of these conditions into primary care under its National Health Mission (NHM) and the National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Disease, and Stroke (NPCDCS). A major component of the government strategy is to encourage universal screening for hypertension and diabetes of adults aged ≥30 years in the community and subsequent referral of potential cases to higher level facilities. Consequently, the expected volume of adults seeking care for hypertension and diabetes at govern- ment health facilities is anticipated to surge. Building Patel et al. BMC Health Services Research (2020) 20:1022 Page 3 of 12 Page 3 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Government of Punjab (use of IT tools and altered work flow to deliver routine care); TATA Trusts (conduct training of healthcare workers and technical assistance); and Dell Technologies (development of software and IT infrastructure). Monitoring and evaluation activities for I-TREC will be conducted by AIIMS, CCDC, and Emory University. I-TREC will be implemented in Mukandpur block of Shaheed Bhagat Singh Nagar district, Punjab, India, and evaluated through comparison of program in- dictors with those observed in the neighboring Sujjon block in the same district (See Fig. 1). The program and comparison locations were selected based on consult- ation with the Punjab Department of Health and Family Welfare. Program components Many barriers to optimal hypertension and diabetes care can be alleviated through IT-based quality improvement strategies. Clinical decision support software can provide up-to-date guidance to clinicians [30] to manage hyper- tension and diabetes following standard treatment proto- cols. Electronic health records can ensure that access to historical patient data and course of illnesses are avail- able to clinicians to guide clinical decisions at whichever facility the patient enters. IT tools can also help make referral linkages between the different levels of health- care more transparent, efficient, and effective by suggest- ing referral thresholds to clinicians, notifying facilities of referred patients, and maintaining a record of recom- mendations to refer the patient to. Finally, digitized sys- tems to track and monitor case management can incentivize improved health provider performance. Motivated by the potential benefits IT tools offer clini- cians and health systems, the I-TREC program includes: an electronic case record form (eCRF) to consolidate The primary implementation partners for the program include the Department of Health and Family Welfare, Table 1 Intervention components Healthcare facility level I-TREC components Available Staff Tasks and functions Level 1: Village Sub-Centre eCRF ANM Universal screening of hypertension and diabetes for adults ages ≥30 y in the community Enrol community members into the NCD portal Adults with blood pressure ≥140/90 or random blood glucose ≥140 mg/dl referred to medical officer at nearest facility (level 2) for confirmation and initiation of treatment Level 2: Primary Health Centre eCRF+CDSS SN, MO Confirmatory diagnosis of hypertension and diabetes of suspected cases referred from sub- centre Routine management of adults with stable hypertension and diabetes Generate and update eCRF Use CDSS to develop treatment plan and determine need for up-referral Level 3: Community Health Centre eCRF+CDSS SN, MO Run dedicated NCD clinics Routine management of adults with stable hypertension and diabetes Generate and update eCRF Use CDSS to develop treatment plan and determine need for up- or down-referral Level 4: District Hospital eCRF+CDSS SN, MO Secondary care available for all health conditions and complications Management of medically complex patients with hypertension and diabetes Generate and update eCRF Use CDSS to develop treatment plan and determine need for up- or down-referral ANM Auxiliary nurse midwife, CDSS Clinical decision support system, eCRF Electronic case record form, MO Medical officer (physician), SN Staff nurse Page 4 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Fig. Program components 1 Geographical location of the study setting. Figure 1 was created by the authors using a map taken from Map Data© 2020 Google Fig. 1 Geographical location of the study setting. Figure 1 was created by the authors using a map taken from Map Data© 2020 Google prompts for triggering referrals across health facilities as needed; a revised workflow to ensure coordinated care within and across facilities; and enhanced training for and track patient information; an electronic clinical deci- sion support system (CDSS) for clinicians to provide tai- lored guideline-based care to patients with in-built Page 5 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Page 5 of 12 related to the etiology, behavioral counseling, and med- ical management of NCDs following established NPCD CS training manuals. The content training is provided over a full day in separate sessions for auxiliary nurse midwives [ANMs], staff nurses, and medical officers. The sessions include training on effective techniques for delivering behavioral and lifestyle advice counselling. Unlike routine training, this refresher training includes innovative learning methods, such as case studies and role playing to enhance trainee engagement to assure improved comprehension and retention of behavioral and lifestyle counselling approaches. In addition, staff nurses and medical officers in the program block receive IT training on the use of the eCRF and CDSS, specific to their level of expertise and the level of healthcare facility in which they are employed. clinicians regarding NCD medical content and lifestyle management (See Table 1). Each component of this inte- grated system is described below. Patient flow under I-TREC Figure 2 depicts patient flow within and across facilities in the I-TREC program. Following NHM and NPCDCS recommendations, all adults aged ≥30 years are eligible for universal screening of hypertension and diabetes in the community and opportunistic screening in health fa- cilities by government health providers. At the village- level Sub-Centre, the lowest level of the healthcare facil- ity hierarchy, the ANM is tasked with screening adults to identify suspected cases of hypertension and diabetes in the community. ANMs enter screening results into the “ANM portal” using a tablet-based application. Adults who are suspected to have hypertension and/or diabetes are referred to the nearest Primary Health Centre, the second level of the healthcare hierarchy, for diagnosis and treatment (see Table 1 and Fig. 2). Clinical decision support system (CDSS) pp y The CDSS generates customized evidence-based treat- ment advisories for patients with hypertension and dia- betes. The treatment advisories are based on up-to-date national and international guidelines that were further vetted by our expert clinical investigators, and tailored to each level of health facility (primary, secondary or ter- tiary), for example, by taking into consideration the local availability of medications and diagnostic capability. The CDSS algorithms provide the clinician with an instantan- eous advisory regarding medication titration based on pa- tient history and current clinical examination as inputted into the eCRF. The attending clinician has the option of rejecting, partially accepting, or fully accepting the advis- ory to generate a final treatment plan. In addition to the treatment plan, the CDSS has in-built prompts to refer patients up or down the healthcare facility hierarchy to direct patients to the most appropriate level of care for ongoing disease management. The treatment plan, referral instructions, and lifestyle advice specific to the patient be- come a part of the patient’s eCRF and are also printed out on paper and given to the patient. At Primary Health Centres and above—namely Com- munity Health Centres and the District Hospital—nurses generate and update the patient eCRF through a web- based application on a computer tablet. At the time the eCRF is first generated, nurses record the patient’s clin- ical history. At future visits, the eCRF is updated with ongoing examination data so that the CDSS is respon- sive to the patient’s health status at a given visit. Patient data entered into the eCRF are uploaded to a secure cloud-based server once per day, and these data are synced and retrievable at all facilities to inform the CDSS and assist with clinical decisions. After the initial eCRF review and update by the nurse, patients with confirmed hypertension or diabetes are instructed to see the medical officer, who is aided by the CDSS to manage these conditions. The CDSS algorithms are tailored to the expertise, medications, and diagnostic tests available at each level of facility and customized to the clinical history of the patient over all past and present contacts with the health system. Medically com- plex patients, such as those who are resistant to therapy, Electronic case record form (eCRF) The eCRF is the Government of India’s digitized health rec- ord focused on compiling patient data relevant to chronic diseases. The eCRF allows nurses to enter patient demo- graphic information, medical history, physical examina- tions, and laboratory investigations into an electronic form through a web-based “NCD portal.” Patient data are then stored in a cloud server hosted by National Informatics Centre, Government of India. The eCRF ensures that ne- cessary patient medical history will be available seamlessly up and down the healthcare facility hierarchy, reduces re- dundant data entry when the same patient seeks care at dif- ferent facilities, and allows clinicians to track patient health information over time across visits. Simultaneously, these data serve as inputs for the CDSS to provide guideline- based recommendations to clinicians to optimize medical and lifestyle management and referral of patients. The eCRF itself was developed by a committee of experts across India, including members of the I-TREC investigator team (NT and AR). We chose to build upon the Government of India’s eCRF to align our program with the national effort to incorporate IT into the management of NCDs in the public sector. Ethics The evaluation protocol was reviewed and approved by the ethics committee at All India Institute of Medical Sciences (AIIMS), New Delhi, India (IEC-361/ 07.07.2017). Given that the Government of Punjab will be implementing I-TREC, and role of research partners is limited to program design, training, and evaluation, this study was deemed to be observational. The role of researchers at Emory University, Atlanta was deemed Not Human Subjects Research (IRB00098808). Partici- pants from whom our research team collects data for evaluation purposes will provide written informed con- sent following procedures approved by the AIIMS Ethics Committee. Once the patient achieves a stable clinical state, she or he will be referred back “down” to the lowest level of health facility (PHC or CHC) that is suitable for routine management of stable disease and dispensation of appro- priate medication. Outcomes Program indicators are guided by the Reach, Effective- ness, Adoption Implementation, and Maintenance (RE-AIM) framework [31]. The principal endpoints for evaluation are shown in Table 2, and focus on the domains of reach, effectiveness, adoption, and imple- mentation of the program components. Reach and ef- fectiveness will be assessed using a combination of community-based data and facility-based data. For ex- ample, the proportion of adults ages 30 and older in the community who are screened for hypertension is a measure of reach that will be obtained through a representative community survey. A second measure of reach is the number of patients seeking care for hypertension and diabetes who have an eCRF, which will be measured through health facility data. Simi- larly, effectiveness will be assessed through health outcomes (e.g., reductions in mean blood pressure and/or mean blood glucose) among patients attending Healthcare provider training Healthcare providers employed in both the program and comparison blocks receive refresher content training Page 6 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Fig. 2 Patient flow under the I-TREC program. Panel a shows patient flow across facilities and Panel b shows patient flow within facilities. Figure 2 was developed by the authors may be referred further “up” the referral hierarchy. The referral algorithms take into account a patient’s full clin- ical history and current health profile, including the number of medications currently prescribed, treatment response based on laboratory investigations, and comor- bidities. For example, if a patient with diabetes under treatment at a PHC has uncontrolled hyperglycemia des- pite being on the maximum tolerated dose of three oral hypoglycaemic agents, the CDSS will trigger a referral to a CHC. Similarly, if a patient with hypertension under treatment at a CHC level has uncontrolled blood pres- sure despite being on the maximum tolerated dose of two antihypertensive drugs, the CDSS will trigger a re- ferral to the District Hospital. Routine care Adults residing in the comparator block will continue to receive the usual care by local physicians and nurses using paper-based record systems and without the assist- ance of the CDSS. Health facility and patient assessments program facilities and also among adults with hyper- tension and/or diabetes in the community. Adoption metrics focus on healthcare provider utilization of the eCRF and CDSS tools. Implementation measures focus on quantifying the proportion of patients who receive care through the eCRF and CDSS tools. Fi- nally, maintenance will be assessed through qualitative research with stakeholders within the health system to understand views of sustainability. y Pre- and post-program health facility and patient assess- ments include 1) a health facility form; 2) patient flow mapping; and 3) patient out-of-pocket cost of care sur- veys. Health facility forms will be completed at all 52 gov- ernment health facilities (21 in program and 30 in control and the common District Hospital) to describe the infra- structure, facility personnel and salaries, availability of medications, and availability of diagnostics and laboratory investigations. Health facility form completion requires a combination of observational checklists and structured in- terviews with administrators. Patient flow mapping entails identifying and following patients with hypertension and diabetes through typical visits to map the typical work- flow, diagnostic and prescription practices, and duration of visits. Together, the health facility form and patient flow mapping will provide data to describe the resources (time and costs) associated with typical healthcare visits for pa- tients with hypertension and diabetes with and without the I-TREC program. Patient cost surveys will be used to Evaluation Design We will employ a two-group pre-post quasi-experimental design to conduct a mixed methods evaluation of the I- TREC program in Punjab, India. While the I-TREC pro- gram is implemented within the health system by clini- cians, we will assess indicators at the levels of facilities, clinicians, patients, and communities. Patel et al. Evaluation Design BMC Health Services Research (2020) 20:1022 Page 7 of 12 Table 2 Key outcome indicators for the I-TREC evaluation RE-AIM domain Key indicators Reach Proportion of adults ages 30 and older in the community screened for hypertension by a government healthcare provider Proportion of adults ages 30 and older in the community screened for diabetes by a government healthcare provider Number of patients seeking care for hypertension and diabetes at a government health facility who have an eCRF Effectiveness Reduction in mean blood pressure in patients receiving care in program facilities Reduction in mean blood glucose in patients receiving care in program facilities Proportion who achieve blood pressure and blood glucose control among patients receiving care in program facilities Proportion who achieve blood glucose control among patients receiving care in program facilities Reduction in mean blood pressure in the community Reduction in mean blood glucose in the community Proportion of hypertension patients who achieve blood pressure control in the community Proportion of diabetes patients who achieve blood glucose control in the community Adoption Proportion of healthcare providers (by type) who log into the NCD portal Proportion of clinicians who fully or partially accept CDSS prompts Proportion of healthcare providers (by type) who report satisfaction with the eCRF+CDSS Implementation Percentage of hypertension patients who received guideline-based care through the eCRF+CDSS (of all registered patients with hypertension) Percentage of diabetes patients who received guideline-based care through the eCRF+CDSS (of all registered patients with diabetes) Percentage of hypertension patients who made repeat visits to health facility Percentage of diabetes patients who made repeat visits to health facility Percentage of “up-referral” cases who attend appointments Percentage of patients who were seen at a higher level facility that returned to the Sub-Centre for ongoing management (“closing the referral loop” and ensuring continuity of care) Percentage of patients tracked with multiple visits over the course of the program Mean time for data upload from each level of facility to central server Maintenance Views of program sustainability and barriers to sustaining and disseminating the program (qualitative) Table 2 Key outcome indicators for the I-TREC evaluation RE-AIM domain Key indicators Stakeholder perspectives Qualitative methods will include a combination of focus group discussions of the community members, key in- formant interviews with healthcare providers and in- depth interviews of patients to provide a richer inter- pretation of quantitative findings and explore the pro- cesses underlying the uptake and delivery of the I-TREC program. The qualitative research will be conducted be- fore, during, and after program implementation. Purpos- ive sampling will be done to recruit information-rich participants for interviews and focus group discussions. For all qualitative data analysis, the textual data (verba- tim transcripts created from digital recordings of inter- views and focus group discussions) will be reviewed to identify key themes and domains of interest. A code book will then be developed to reflect these domains and include both inductive (derived from the textual data) and deductive (based on literature and theory) codes. Inter-coder reliability will be assessed, and the codebook will be finalized and applied to the data. The codebook will include codes specific to each type of data collection and shared codes across participant type. A thematic analysis will be used to describe individual- or community-level views on discussion topics including program barriers and facilitators, community barriers, views of the healthcare system, and acceptability and feasibility of the program. The community survey sample size was determined to estimate differences in mean reduction in systolic blood pressure among those with diagnosed hypertension in the community. First, we computed the base sample size re- quired to detect a desired effect size of 5 mmHg given the SBP standard deviation of 18.5, power = .80 and α = .05, based on the mPower Heart Study [24]. We estimated that 168 individuals with hypertension would be needed to de- tect the anticipated effect size. Second, we estimated that we would require a sample size of 839 adults in the gen- eral population to identify 168 individuals with hyperten- sion, assuming prevalence of diagnosed hypertension of 20%. Third, we determined the optimal sample allocation for a multi-stage sampling design that would be time- efficient for field work and statistical precision. We as- sumed an intraclass correlation of systolic blood pressure of 0.018 based on village-level clustering of SBP in the DISHA study [32] (unpublished findings). Stakeholder perspectives After applying a 10% refusal rate based on our prior field studies in the re- gion, we determined that a cluster size of 50 adults per vil- lage distributed across 35 villages per block was optimal. This yielded a total sample size of 3508 for the community survey to be evenly split between the program and comparison blocks. Pre- and post-program quantitative cross-sectional community surveys The community-based evaluation component will assess whether the I-TREC program has an impact on blood pressure and blood glucose awareness, treatment seek- ing, and control among adults in the community. This evaluation component is critical to learning the real- world impact of the I-TREC program on community- level indicators of the care continuum (screening, treat- ment, control). Data collected in the program and com- parison blocks prior to the program will be compared with data collected from these same blocks after the program using identical procedures. This design allows us to assess and address several threats to validity, in- cluding lack of temporal order, comparability across the two blocks (leading to potential confounding by Sources of data for evaluation Data will be triangulated from multiple sources, includ- ing facility and patient assessments, stakeholder inter- views, community surveys, and patient-level data from the I-TREC electronic database. Where appropriate, data will be collected prior to the program launch and again following 36 months of the program. With the exception of the I-TREC eCRF-CDSS data, evaluation data will be collected by trained research staff. Patel et al. BMC Health Services Research (2020) 20:1022 Page 8 of 12 Page 8 of 12 obtain data on expenditures related to outpatient and in- patient health care utilization in the last 3 months to understand the cost incurred by patients to manage their disease. Administration of the pre- and post-program pa- tient cost surveys will contribute data to understand whether the program has any impact on patient expendi- tures related to hypertension and diabetes. Purposive sam- pling will be done to recruit patients for the patient flow mapping and cost surveys. population composition) and secular changes unrelated to our program that affect study endpoints (leading to potential confounding by external factors). Given that we will be sampling separate cross-sections of the popu- lation in each group and time point, we do not expect inference to be affected by population aging (maturation threats) over the 3-year program period. At baseline—prior to intervention—we employed a multi-stage cluster sampling design to obtain a represen- tative sample of adults aged 30 years and older in both blocks under study. Within each block, census data were used to select villages proportionate to population size and subsequently we conducted household mapping and listing to generate a sampling frame for households. Households were selected using systematic random sam- pling, and one adult man and woman from each house- hold were randomly selected using the Kish method to achieve the desired sample size. At endline, this same procedure will be repeated. Statistical analysis plan actively following patients for research visits. Rather, all adults residing in the I-TREC catchment area (i.e., resi- dents of Mukandpur block) will be exposed to the pro- gram and patient data will be collected every time a person chooses to receive care at a government health facility. Data from all patients visiting facilities in the program block captured in the Government of India eCRF will be de-identified and obtained by AIIMS throughout the program period for monitoring and evaluation purposes. y Quantitative data analysis will be performed using SAS, STATA, and R software. Descriptive analyses of the community-based data will examine socio-demographic characteristics, health indicators, and healthcare behav- iors of the program and comparison block samples at baseline and end-line. The quantitative evaluation of health and healthcare endpoints will focus on assessing changes in the continuum of care indicators and mean blood pressure in the community-based surveys. We will assess changes in baseline to end-line indicators of health outcomes (e.g., blood pressure) and changes in continuum of care indicators (e.g., proportion screened) for both the program and comparison blocks; see Table 2 for indicators. A simple difference-in-difference (DiD) [33] estimate for each indicator will be computed as The sample size for the facility-based evaluation is out of our control and contingent on the number of patients who seek care at government facilities. We therefore re- port the detectable effect size for longitudinal change in systolic blood pressure over time in patients at I-TREC facilities after setting power to 80% and α = .05. The pro- gram block, Mukandpur, has a population of 98,000. We expect that 30% of the local population will seek care at a government facility, 50% will be age-eligible (30 years and older) per the government guidelines for universal and opportunistic screening of hypertension and dia- betes, and 20% will test positive for hypertension, amounting to an estimated patient pool of 2940 adults. Assuming that 50% of all enrolled patients with hyper- tension and diabetes make repeat visits (enabling us examine changes in outcomes), we will be able to detect a 1.35 mmHg difference in SBP. DiD ¼ pg¼i;t¼1 - pg¼i;t¼0
 - pg¼c;t¼1 - pg¼c;t¼0
 where p indicates prevalence or mean of each indicator; g subscripts group (i = program; c = comparison); and t subscripts the time point of data (0 = pre-program; 1 = post-program). Health outcomes among patients receiving the I-TREC program Using patient health data from the eCRF, we will assess processes of care and changes in blood pressure and blood glucose outcomes over time among patients with hypertension or diabetes who seek care at I-TREC pro- gram facilities in a facility-based evaluation component. Because I-TREC is being integrated into the routine care in the program block under real-world conditions, we will not be assigning individual patients to treatment nor Patel et al. BMC Health Services Research (2020) 20:1022 Page 9 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Statistical analysis plan We will estimate log-binomial models (binary outcomes) or linear models (continuous out- comes, with log-transformation if needed) with robust variance to compute the DiD after accounting for com- positional characteristics of the community and cluster- ing of data within villages. For each outcome indicator separately, the following model will be estimated using individual-level data: Current status d Intervention development and pre-testing were com- pleted in August 2019. Healthcare providers in both blocks received training in December 2019, and the I- TREC program was launched in January 2020. Prior to the program launch, pre-program data collection, in- cluding facility and patient assessments, qualitative re- search, and community survey, were completed. Since March 2020, both intervention roll-out and research ac- tivities have been impacted by the COVID-19 pandemic. Specifically, government healthcare system resources have been diverted from NCD care to test and treat pa- tients with SARS-COV-2, national and local lockdown measures have forced intermittent closures of lower-tier health facilities and prevented field staff from conduct- ing routine monitoring activities. In addition, patient flow through facilities—when open—has generally de- clined, possibly due to fear of contracting the virus while seeking healthcare. Nevertheless, as and when healthcare facilities are operational, the intervention components are being implemented by nurses and physicians in the program facilities and program monitoring activities are underway. We expect intermittent disruptions to inter- vention implementation and monitoring activities to continue until the COVID-19 pandemic has been fully controlled. p y The I-TREC program and its evaluation have several strengths but also some limitations. Given the primacy of scalability, the role of research staff is limited to pro- gram design, training, monitoring and evaluation. There- fore, the context and conditions of implementation are beyond the control of investigators. For example, the availability of drugs, diagnostic investigations, and clini- cians are likely to impact the reach, effectiveness, and implementation of I-TREC but rest in the hands of the state government. Nevertheless, we expect variations in these structural elements to affect both the program and comparison blocks similarly. In order to maintain com- parability between the program and comparison block, neighboring blocks within the same district were chosen. However, this means that both blocks share the district hospital, which will have to be taken into account during the analysis. While we will be able to obtain patient data in the program block from the eCRF, no comparable source of data is available to use in the comparison block. Moreover, undue monitoring of the comparison block may inadvertently lead to compensation behaviors on the part of clinicians that could undermine our ability to measure performance differences across the two blocks. Process measures I h Process measures In the program block, we will examine measures of adoption and implementation of the IT tools, such as completeness of eCRF forms, acceptance (partial and full) and rejection of the CDSS advisories, time stamp of data entry, the initials of the enterer, and average num- ber of new records per day. In both the program and comparison block, data regarding the total number of patients recorded in the out-patient registry at the facil- ity, numbers screened for hypertension and diabetes, numbers receiving medication from the pharmacy, and numbers referred to higher level facilities will be col- lected through a combination of paper-based registries and routine NPCDCS reports. The I-TREC evaluation team will obtain these facility-level data using abstrac- tion forms without removing any paper records from premises. In addition, the I-TREC evaluation team will periodically conduct random, unannounced visits to dir- ectly observe the number of patients seeking care for hypertension and diabetes facilities in both blocks. Add- itional data on intervention fidelity measures (e.g., use of eCRF during health visit, measurement of blood pressure and blood glucose, provision of the I-TREC print out to the patient) will also be collected through patient exit in- terviews and the eCRF backend data. Outcome indicator  program group þpre −post indicator þprogram group x pre −post indicator þ age þ sex þeducation þ religion þmarital status þbelow poverty line þfacility type public versus private ð Þ: −post indicator þ age þ sex þeducation þ religion þmarital status The coefficient associated with the interaction term, “program group x pre-post indicator,” is the adjusted DiD estimate accounting for heterogeneity in socio- demographic characteristics. The model will be esti- mated using generalized estimating equations (GEE) to account for clustering of outcomes within the villages (i.e. village is the cluster variable specified for statistical analysis). Sub-group analyses will examine differences by gender and socioeconomic status. Data points recorded in the I-TREC system will be an- alyzed by month to examine trends over time and sea- sonality. We will also evaluate change patient outcomes over time (e.g., mean SBP change). While several of the I-TREC platform indicators are purely descriptive mea- sures of performance (e.g., mean time for data upload), Page 10 of 12 Page 10 of 12 Page 10 of 12 Patel et al. Acknowledgements We would like to acknowledge the contribution of our implementing partners Sunita Nadhamuni, Supriya Prabhakar, Sruti Sridhar and colleagues at Dell Technologies, Bangalore and Drs. Aman Singh and Prashant Pathak at TATA Trusts. Abbreviations l ANM: Auxiliary nurse midwives; AIIMS, CCDC: All India Institute of Medical Sciences, New Delhi (AIIMS), the Centre for Chronic Disease Control; CDSS: Clinical decision support system; DiD: Difference-in-difference; eCRF: Electronic case record form; IT: Information technology; I- TREC: Integrated Tracking, Referral, and Electronic Decision Support, and Care Coordination Program; NCD: Non-communicable disease; NHM: National Health Mission; NPCDCS: National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Disease, and Stroke; RE-AIM: Reach, Effectiveness, Adoption Implementation, and Maintenance This implementation research is intended to provide evidence of workable programs to manage chronic dis- eases in India and inform the evolving NPCDCS [34]. Given the similarities in the health system and epidemi- ologic transition between in India and other low- and middle-income countries, this research has additional scope to potentially inform best practices for manage- ment of hypertension and diabetes outside of India. In fact, members of our team (DJ, AV, DP) have collabo- rated with the World Health Organization (WHO)- Southeast Asia Regional Office and the Republic of Current status d The community survey, in part, is designed to mitigate these limitations by providing a well-designed comparison of healthcare processes and outcomes as ob- served in the program and comparison blocks. Process measures I h BMC Health Services Research (2020) 20:1022 other indicators of healthcare delivery may be compared between I-TREC facilities and comparison group facil- ities (e.g., patient volume). Maldives for the development of the “mPEN App,” which is a CDSS tool that draws on similar technologies as what is described here to implement the WHO Pre- vention of Essential NCDs package in primary health care in the Maldives. We envision that the I-TREC pro- gram and evaluation will provide opportunities for con- tinued cross-national collaborations and idea exchange to improve hypertension and diabetes care globally. Summary This is the first study in India evaluating the composite impact of a clinical decision support system integrated with the Government of India eCRF, combined with modified patient flow and enhanced healthcare provider training. It is also an important early effort to systematic- ally evaluate a program for integrated management of hypertension and diabetes at all levels of the public health- care system, starting from the sub-centre up to the district hospital. Lessons learned may inform optimal approaches to improve healthcare processes and health outcomes within the public sector healthcare system in India. Received: 30 September 2020 Accepted: 21 October 2020 p 21. Patel V, Chatterji S, Chisholm D, Ebrahim S, Gopalakrishna G, Mathers C, et al. Chronic diseases and injuries in India. Lancet Lond Engl. 2011;377:413–28. 22. Ajay VS, Tian M, Chen H, Wu Y, Li X, Dunzhu D, et al. A cluster-randomized controlled trial to evaluate the effects of a simplified cardiovascular management program in Tibet, China and Haryana, India: study design and rationale. BMC Public Health. 2014;14:924. Ethics approval and consent to participate 13. Hasan H, Zodpey S, Saraf A. Diabetologist’s perspective on practice of evidence based diabetes management in India. Diabetes Res Clin Pract. 2012;95:189–93. The evaluation protocol was reviewed and approved by the ethics committee at All India Institute of Medical Sciences (AIIMS), New Delhi, India (IEC-361/07.07.2017). The role of researchers at Emory University, Atlanta was deemed Not Human Subjects Research (IRB00098808) by the Emory University Institutional Review Board. Participants enrolled in the research study provided written informed consent. 14. Pakhare A, Kumar S, Goyal S, Joshi R. Assessment of primary care facilities for cardiovascular disease preparedness in Madhya Pradesh, India. BMC Health Serv Res. 2015;15:408. 14. Pakhare A, Kumar S, Goyal S, Joshi R. Assessment of primary care facilities for cardiovascular disease preparedness in Madhya Pradesh, India. BMC Health Serv Res. 2015;15:408. 15. Kaveeshwar SA, Cornwall J. The current state of diabetes mellitus in India. Australas Med J. 2014;7:45–8. 15. Kaveeshwar SA, Cornwall J. The current state of diabetes mellitus in India. Australas Med J. 2014;7:45–8. 16. Shah VN, Kamdar PK, Shah N. Assessing the knowledge, attitudes and practice of type 2 diabetes among patients of Saurashtra region, Gujarat. Int J Diabetes Dev Ctries. 2009;29:118–22. 16. Shah VN, Kamdar PK, Shah N. Assessing the knowledge, attitudes and practice of type 2 diabetes among patients of Saurashtra region, Gujarat. Int J Diabetes Dev Ctries. 2009;29:118–22. Availability of data and materials f 12. Prenissl J, Jaacks LM, Mohan V, Manne-Goehler J, Davies JI, Awasthi A, et al. Variation in health system performance for managing diabetes among states in India: a cross-sectional study of individuals aged 15 to 49 years. BMC Med. 2019;17:92. Data generated from this study will be made available to researchers upon request. Requests may be made to Dr. Patel and will be processed through the I-TREC Data Sharing and Publications Committee. 13. Hasan H, Zodpey S, Saraf A. Diabetologist’s perspective on practice of evidence based diabetes management in India. Diabetes Res Clin Pract. 2012;95:189–93. Consent for publication Not applicable. 17. Kotwani A, Ewen M, Dey D, Iyer S, Lakshmi PK, Patel A, et al. Prices & availability of common medicines at six sites in India using a standard methodology. Indian J Med Res. 2007;125:645–54. Funding 9. Anchala R, Kannuri NK, Pant H, Khan H, Franco OH, Di Angelantonio E, et al. Hypertension in India: a systematic review and meta-analysis of prevalence, awareness, and control of hypertension. J Hypertens. 2014;32:1170–7. This study is supported in part by the National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health (NIH), award number 5U01HL138635 under the Hypertension Outcomes for T4 Research within Lower Middle-Income Countries (Hy-TREC) program. The content is solely the responsibility of the authors and does not necessarily represent the offi- cial views of the NIH. The sponsor had no role in study design as presented here. SAP, KMVN, MKA, and MBW were supported in part by funding from NIDDK (P30DK111024). PG was supported by funding from the NIH Fogarty International Center (VUMC67452 under D43TW009337). 10. Deepa M, Grace M, Binukumar B, Pradeepa R, Roopa S, Khan HM, et al. High burden of prediabetes and diabetes in three large cities in South Asia: the Center for cArdio-metabolic risk reduction in South Asia (CARRS) study. Diabetes Res Clin Pract. 2015;110(2):172–82. 11. Prenissl J, Manne-Goehler J, Jaacks LM, Prabhakaran D, Awasthi A, Bischops AC, et al. Hypertension screening, awareness, treatment, and control in India: a nationally representative cross-sectional study among individuals aged 15 to 49 years. PLoS Med. 2019;16:e1002801. Author details 1D f Author details 1Department of Global Health, Emory University, 1518 Clifton Rd NE / Rm 7037, Atlanta, USA. 2All India Institute of Medical Sciences, New Delhi, India. 3Centre for Chronic Disease Control, New Delhi, India. 4Public Health Foundation of India, Gurgaon, India. 5Department of Family and Preventive Medicine, Emory University, Atlanta, USA. 6Department of Endocrinology & Metabolism, All India Institute of Medical Sciences, New Delhi, India. 7 19. Siegel KR, Patel SA, Ali MK. Non-communicable diseases in South Asia: contemporary perspectives. Br Med Bull. 2014;111:31–44. 19. Siegel KR, Patel SA, Ali MK. Non-communicable diseases in South Asia: contemporary perspectives. Br Med Bull. 2014;111:31–44. 20. Murray CJL, Barber RM, Foreman KJ, Ozgoren AA, Abd-Allah F, Abera SF, et al. Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition. Lancet. 2015; Available from: http://www.sciencedirect.com/science/article/pii/S014 067361561340X. Cited 2015 Sep 7. 7Department of Cardiology, All India Institute of Medical Sciences, New Delhi, India. Competing interests The authors declare that they have no competing interests. 18. Jamison DT, Summers LH, Alleyne G, Arrow KJ, Berkley S, Binagwaho A, et al. Global health 2035: a world converging within a generation. Lancet Lond Engl. 2013;382:1898–955. Authors’ contributions NT, SAP, SM, DP, KMVN, MKA, and MBW conceptualized the program and obtained funding. NT, SAP, MBW, and SM designed the baseline needs assessment and evaluation protocol, including the data analysis plans. SAP, HS, MBW, PG, PJ, RS, and MA developed the data collection protocols, and NT and SM provided additional review of protocols. NT, DP, AR, YG, and DJ, contributed to development of clinical algorithms and DJ oversaw the Page 11 of 12 Page 11 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Patel et al. BMC Health Services Research (2020) 20:1022 Patel et al. BMC Health Services Research (2020) 20:1022 development and integration of the CDSS. HS coordinated ethical approvals and field permissions. SAP and HS developed the first draft of the development and integration of the CDSS. HS coordinated ethical approvals and field permissions. SAP and HS developed the first draft of the development and integration of the CDSS. HS coordinated ethical approvals and field permissions. SAP and HS developed the first draft of the manuscript and coordinated revisions. All authors provided critical inputs to the manuscript and approve its final submission. study protocol for the Kerala diabetes prevention program. BMC Public Health. 2013;13:1035. study protocol for the Kerala diabetes prevention program. BMC Public Health. 2013;13:1035. manuscript and coordinated revisions. All authors provided critical inputs to the manuscript and approve its final submission. 8. Joshi SR, Saboo B, Vadivale M, Dani SI, Mithal A, Kaul U, et al. Prevalence of diagnosed and undiagnosed diabetes and hypertension in India--results from the screening India’s Twin Epidemic (SITE) study. Diabetes Technol Ther. 2012;14:8–15. References 1. Mohan S, Jarhyan P, Ghosh S, Venkateshmurthy NS, Gupta R, Rana R, et al. UDAY: a comprehensive diabetes and hypertension prevention and management program in India. BMJ Open. 2018;8 Available from: https://bmjopen.bmj.com/content/8/6/e015919. Cited 2020 Feb 2. 2 Cheung BMY Li C Diabetes and hypertension: is there a common 1. Mohan S, Jarhyan P, Ghosh S, Venkateshmurthy NS, Gupta R, Rana R, et al. UDAY: a comprehensive diabetes and hypertension prevention and management program in India. BMJ Open. 2018;8 Available from: https://bmjopen.bmj.com/content/8/6/e015919. Cited 2020 Feb 2. 1. Mohan S, Jarhyan P, Ghosh S, Venkateshmurthy NS, Gupta R, Rana R, et al. UDAY: a comprehensive diabetes and hypertension prevention and management program in India. BMJ Open. 2018;8 Available from: https://bmjopen.bmj.com/content/8/6/e015919. Cited 2020 Feb 2. 2. Cheung BMY, Li C. Diabetes and hypertension: is there a common metabolic pathway? Curr Atheroscler Rep. 2012;14:160–6. 3 Mancia G The association of hypertension and diabetes prevalence 23. Tian M, Ajay VS, Dunzhu D, Hameed SS, Li X, Liu Z, et al. A cluster-randomized, controlled trial of a simplified multifaceted management program for individuals at high cardiovascular risk (SimCard trial) in rural Tibet, China, and Haryana, IndiaCLINICAL PERSPECTIVE. Circulation. 2015;132:815–24. 2. Cheung BMY, Li C. Diabetes and hypertension: is there a comm metabolic pathway? Curr Atheroscler Rep. 2012;14:160–6. 3. Mancia G. The association of hypertension and diabetes: prevalence, cardiovascular risk and protection by blood pressure reduction. Acta Diabetol. 2005;42:s17–25. 24. Ajay VS, Jindal D, Roy A, Venugopal V, Sharma R, Pawar A, Kinra S, Tandon N, Prabhakaran D. Development of a Smartphone-Enabled Hypertension and Diabetes Mellitus Management Package to Facilitate Evidence-Based Care Delivery in Primary Healthcare Facilities in India: The mPower Heart Project. J Am Heart Assoc. 2016;5(12):e004343. https://doi.org/10.1161/JAHA. 116.004343. 4. Selby JV, Peng T, Karter AJ, Alexander M, Sidney S, Lian J, et al. High rates of co-occurrence of hypertension, elevated low-density lipoprotein cholesterol, and diabetes mellitus in a large managed care population. Am J Manag Care. 2004;10:163–70. 4. Selby JV, Peng T, Karter AJ, Alexander M, Sidney S, Lian J, et al. High rates of co-occurrence of hypertension, elevated low-density lipoprotein cholesterol, and diabetes mellitus in a large managed care population. Am J Manag Care. 2004;10:163–70. 25. Ali MK, Singh K, Kondal D, Devarajan R, Patel SA, Shivashankar R, et al. Effectiveness of a multicomponent quality improvement strategy to improve achievement of diabetes care goals: a randomized, controlled trial. Ann Intern Med. 2016;165:399–408. 5. References Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. Am J Public Health. 1999;89:1322–7. 31. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. Am J Public Health. 1999;89:1322–7. 32. Jeemon P, Narayanan G, Kondal D, Kahol K, Bharadwaj A, Purty A, et al. Task shifting of frontline community health workers for cardiovascular risk reduction: design and rationale of a cluster randomised controlled trial (DISHA study) in India. BMC Public Health. 2016;16:264. 32. Jeemon P, Narayanan G, Kondal D, Kahol K, Bharadwaj A, Purty A, et al. Task shifting of frontline community health workers for cardiovascular risk reduction: design and rationale of a cluster randomised controlled trial (DISHA study) in India. BMC Public Health. 2016;16:264. 33. Athey S, Imbens GW. Identification and inference in nonlinear difference-in- differences models. Econometrica. 2006;74:431–97. 33. Athey S, Imbens GW. Identification and inference in nonlinear difference-in- differences models. Econometrica. 2006;74:431–97. 34. National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular diseases and Stroke (NPCDCS). Ministry of Health and Family Welfare, Government of India. https://main.mohfw.gov.in/Major- Programmes/non-communicable-diseases-injury-trauma/Non- Communicable-Disease-II/National-Programme-for-Prevention-and-Control- of-Cancer-Diabetes-Cardiovascular-diseases-and-Stroke-NPCDCS. Accessed 29 Oct 2020. References Khatib R, Schwalm J-D, Yusuf S, Haynes RB, McKee M, Khan M, et al. Patient and healthcare provider barriers to hypertension awareness, treatment and follow up: a systematic review and meta-analysis of qualitative and quantitative studies. PLoS One. 2014;9:e84238. 6. Chow CK, Teo KK, Rangarajan S, et al. Prevalence, awareness, treatment, and control of hypertension in rural and urban communities in high-, middle-, and low-income countries. JAMA. 2013;310:959–68. 7. Sathish T, Williams ED, Pasricha N, Absetz P, Lorgelly P, Wolfe R, et al. Cluster randomised controlled trial of a peer-led lifestyle intervention program: 5. Khatib R, Schwalm J-D, Yusuf S, Haynes RB, McKee M, Khan M, et al. Patient and healthcare provider barriers to hypertension awareness, treatment and follow up: a systematic review and meta-analysis of qualitative and quantitative studies. PLoS One. 2014;9:e84238. 26. Jindal D, Gupta P, Jha D, Ajay VS, Goenka S, Jacob P, et al. Development of mWellcare: an mHealth intervention for integrated management of hypertension and diabetes in low-resource settings. Glob Health Action. 2018;11 Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6161589/. Cited 2019 Jul 27. 6. Chow CK, Teo KK, Rangarajan S, et al. Prevalence, awareness, treatment, and control of hypertension in rural and urban communities in high-, middle-, and low-income countries. JAMA. 2013;310:959–68. 7. Sathish T, Williams ED, Pasricha N, Absetz P, Lorgelly P, Wolfe R, et al. Cluster randomised controlled trial of a peer-led lifestyle intervention program: Page 12 of 12 Patel et al. BMC Health Services Research (2020) 20:1022 Patel et al. BMC Health Services Research (2020) 20:1022 27. Prabhakaran D, Jha D, Prieto-Merino D, Roy A, Singh K, Ajay VS, et al. Effectiveness of an mHealth-based electronic decision support system for integrated management of chronic conditions in primary care: the mWellcare cluster-randomized controlled trial. Circulation. 2018. https://doi.org/10.1161/CIRCULATIONAHA.118.038192. 28. Kowalski AJ, Poongothai S, Chwastiak L, Hutcheson M, Tandon N, Khadgawat R, et al. The INtegrating DEPrEssioN and Diabetes treatmENT (INDEPENDENT) study: design and methods to address mental healthcare gaps in India. Contemp Clin Trials. 2017;60:113–24. 29. Jindal D, Roy A, Ajay VS, Yadav SK, Prabhakaran D, Tandon N. Strategies for stakeholder engagement and uptake of new intervention: experience from state-wide implementation of mHealth technology for NCD care in Tripura, India. Glob Heart. 2019;14:165–72. 30. Gupta R, Yusuf S. Towards better hypertension management in India. Indian J Med Res. 2014;139:657–60. 30. Gupta R, Yusuf S. Towards better hypertension management in India. Indian J Med Res. 2014;139:657–60. 31. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

https://openalex.org/W2909748506

https://iris.uniroma1.it/bitstream/11573/1256759/2/Leotta_Surveying-human-habit_2019.pdf

English

Surveying Human Habit Modeling and Mining Techniques in Smart Spaces

Future internet

cc-by

Surveying Human Habit Modeling and Mining Techniques in Smart Spaces Francesco Leotta * , Massimo Mecella * , Daniele Sora and Tiziana Catarci Dipartimento di Ingegneria Informatica, Automatica e Gestionale Antonio Ruberti, Sapienza Università di Roma, 00185 Rome, Italy; [email protected] (D.S.); [email protected] (T.C.) * Correspondence: [email protected] (F.L.); [email protected] (M.M.)



 Received: 28 December 2018; Accepted: 16 January 2019; Published: 19 January 2019 Abstract: A smart space is an environment, mainly equipped with Internet-of-Things (IoT) technologies, able to provide services to humans, helping them to perform daily tasks by monitoring the space and autonomously executing actions, giving suggestions and sending alarms. Approaches suggested in the literature may differ in terms of required facilities, possible applications, amount of human intervention required, ability to support multiple users at the same time adapting to changing needs. In this paper, we propose a Systematic Literature Review (SLR) that classifies most influential approaches in the area of smart spaces according to a set of dimensions identified by answering a set of research questions. These dimensions allow to choose a specific method or approach according to available sensors, amount of labeled data, need for visual analysis, requirements in terms of enactment and decision-making on the environment. Additionally, the paper identifies a set of challenges to be addressed by future research in the field. Keywords: smart spaces; intelligent environments; survey; human habits; modeling future internet Review Surveying Human Habit Modeling and Mining Techniques in Smart Spaces Francesco Leotta * , Massimo Mecella * , Daniele Sora and Tiziana Catarci Dipartimento di Ingegneria Informatica, Automatica e Gestionale Antonio Ruberti, Sapienza Università di Roma, 00185 Rome, Italy; [email protected] (D.S.); [email protected] (T.C.) * Correspondence: [email protected] (F.L.); [email protected] (M.M.) Received: 28 December 2018; Accepted: 16 January 2019; Published: 19 January 2019



 Abstract: A smart space is an environment, mainly equipped with Internet-of-Things (IoT) technologies, able to provide services to humans, helping them to perform daily tasks by monitoring the space and autonomously executing actions, giving suggestions and sending alarms. Approaches suggested in the literature may differ in terms of required facilities, possible applications, amount of human intervention required, ability to support multiple users at the same time adapting to changing needs. In this paper, we propose a Systematic Literature Review (SLR) that classifies most influential approaches in the area of smart spaces according to a set of dimensions identified by answering a set of research questions. These dimensions allow to choose a specific method or approach according to available sensors, amount of labeled data, need for visual analysis, requirements in terms of enactment and decision-making on the environment. Additionally, the paper identifies a set of challenges to be addressed by future research in the field. gent environments; survey; human habits; modeling and communication technologies has many faces; while computing miniaturization of electronic devices increase year after year, their idespread adoption of embedded systems (e.g., appliances, sensors, puting devices (e.g., laptops, smartphones), thus, turning pervasive lity. Pervasive computing embodies a vision of computers seamlessly esponding to information provided by sensors in the environment, n from users [1]. At the same time, connecting all these computing rtefacts, using local and global network infrastructures has become at exploit these technologies represents a major characteristic of the emerging class of IoT-based applications. Smart homes and ples where pervasive computing could take advantage of ambient n in other scenarios where artificial intelligence—AI problems soon vel of adoption of commercial smart home systems is provided in [4]. d di f h “ ” h l i h deling and Mining Sora and Tiziana Catarci tionale Antonio Ruberti, Sapienza Università di [email protected] (T.C.) @diag.uniroma1.it (M.M.) ublished: 19 January 2019



 inly equipped with Internet-of-Things (IoT) ping them to perform daily tasks by monitoring g suggestions and sending alarms. Approaches uired facilities, possible applications, amount of ple users at the same time adapting to changing ure Review (SLR) that classifies most influential set of dimensions identified by answering a set hoose a specific method or approach according for visual analysis, requirements in terms of nt. Additionally, the paper identifies a set of field. urvey; human habits; modeling technologies has many faces; while computing ctronic devices increase year after year, their of embedded systems (e.g., appliances, sensors, laptops, smartphones), thus, turning pervasive ting embodies a vision of computers seamlessly ation provided by sensors in the environment, he same time, connecting all these computing and global network infrastructures has become ologies represents a major characteristic of the IoT-based applications. Smart homes and e computing could take advantage of ambient where artificial intelligence—AI problems soon mmercial smart home systems is provided in [4]. erm “smart” has a more general meaning than includes non-technological aspects such as the future internet future internet future internet future internet future internet Review Surveying Human Habit Modeling and Mining Techniques in Smart Spaces Francesco Leotta * , Massimo Mecella * , Daniele Sora and Tiziana Catarci Dipartimento di Ingegneria Informatica, Automatica e Gestionale Antonio Ruberti, Sapienza Università di Roma, 00185 Rome, Italy; [email protected] (D.S.); [email protected] (T.C.) * Correspondence: [email protected] (F.L.); [email protected] (M.M.) Received: 28 December 2018; Accepted: 16 January 2019; Published: 19 January 2019



 Abstract: A smart space is an environment, mainly equipped with Internet-of-Things (IoT) technologies, able to provide services to humans, helping them to perform daily tasks by monitoring the space and autonomously executing actions, giving suggestions and sending alarms. Approaches suggested in the literature may differ in terms of required facilities, possible applications, amount of human intervention required, ability to support multiple users at the same time adapting to changing needs. In this paper, we propose a Systematic Literature Review (SLR) that classifies most influential approaches in the area of smart spaces according to a set of dimensions identified by answering a set of research questions. These dimensions allow to choose a specific method or approach according to available sensors, amount of labeled data, need for visual analysis, requirements in terms of enactment and decision-making on the environment. Additionally, the paper identifies a set of challenges to be addressed by future research in the field. future internet Review Surveying Human Habit Modeling and Mining Techniques in Smart Spaces Francesco Leotta * , Massimo Mecella * , Daniele Sora and Tiziana Catarci Dipartimento di Ingegneria Informatica, Automatica e Gestionale Antonio Ruberti, Sapienza Università di Roma, 00185 Rome, Italy; [email protected] (D.S.); [email protected] (T.C.) * Correspondence: [email protected] (F.L.); [email protected] (M.M.) Received: 28 December 2018; Accepted: 16 January 2019; Published: 19 January 2019



 Abstract: A smart space is an environment, mainly equipped with Internet-of-Things (IoT) technologies, able to provide services to humans, helping them to perform daily tasks by monitoring the space and autonomously executing actions, giving suggestions and sending alarms. Approaches suggested in the literature may differ in terms of required facilities, possible applications, amount of human intervention required, ability to support multiple users at the same time adapting to changing needs. In this paper, we propose a Systematic Literature Review (SLR) that classifies most influential approaches in the area of smart spaces according to a set of dimensions identified by answering a set of research questions. These dimensions allow to choose a specific method or approach according to available sensors, amount of labeled data, need for visual analysis, requirements in terms of enactment and decision-making on the environment. Additionally, the paper identifies a set of challenges to be addressed by future research in the field. n Habit Modeling and Mining art Spaces Mecella * , Daniele Sora and Tiziana Catarci atica, Automatica e Gestionale Antonio Ruberti, Sapienza Università di g.uniroma1.it (D.S.); [email protected] (T.C.) iroma1.it (F.L.); [email protected] (M.M.) pted: 16 January 2019; Published: 19 January 2019



 an environment, mainly equipped with Internet-of-Things (IoT) rvices to humans, helping them to perform daily tasks by monitoring ecuting actions, giving suggestions and sending alarms. Approaches differ in terms of required facilities, possible applications, amount of bility to support multiple users at the same time adapting to changing e a Systematic Literature Review (SLR) that classifies most influential spaces according to a set of dimensions identified by answering a set mensions allow to choose a specific method or approach according f labeled data, need for visual analysis, requirements in terms of ng on the environment. Additionally, the paper identifies a set of uture research in the field. 1. Introduction The progress of information and communication technologies has many faces; while computing speed, reliability and level of miniaturization of electronic devices increase year after year, their costs decrease. This allows a widespread adoption of embedded systems (e.g., appliances, sensors, actuators) and of powerful computing devices (e.g., laptops, smartphones), thus, turning pervasive (or ubiquitous) computing into reality. Pervasive computing embodies a vision of computers seamlessly integrating into everyday life, responding to information provided by sensors in the environment, with little or no direct instruction from users [1]. At the same time, connecting all these computing devices together, as networked artefacts, using local and global network infrastructures has become easy. The rise of applications that exploit these technologies represents a major characteristic of the Internet-of-Things (IoT) [2]. Smart spaces represent an emerging class of IoT-based applications. Smart homes and offices are representative examples where pervasive computing could take advantage of ambient intelligence (AmI) more easily than in other scenarios where artificial intelligence—AI problems soon become intractable [3]. A study about the current level of adoption of commercial smart home systems is provided in [4]. This study reveals how people understanding of the term “smart” has a more general meaning than what we presented here as AmI; in particular, it also includes non-technological aspects such as the spatial layout of the house. Additionally, an automated behavior is considered as smart, especially from people without a technical background, only if it performs a task quicker than the user could do by himself. The research also reveals that interest in smart home systems is subject to a virtuous circle such that people experiencing benefits from their services feel the need of upgrading them. Future Internet 2019, 11, 23; doi:10.3390/fi11010023 www.mdpi.com/journal/futureinternet www.mdpi.com/journal/futureinternet Future Internet 2019, 11, 23 2 of 23 The universAAL specification [5] defines a smart space as “an environment centered on its human users in which a set of embedded networked artefacts, both hardware and software, collectively realize the paradigm of ambient intelligence” (AmI). The universAAL specification [5] defines a smart space as “an environment centered on its human users in which a set of embedded networked artefacts, both hardware and software, collectively realize the paradigm of ambient intelligence” (AmI). Many different definitions of AmI are provided in the literature, e.g., [6] introduces a set of distinguishing keywords characterizing AmI systems, namely: sensitivity, responsiveness, adaptivity, ubiquity and transparency. 1. Introduction The term sensitivity refers to the ability of an AmI system to sense the environment and, more generally, to understand the context it is interacting with. Strictly related to sensitivity are responsiveness and adaptivity, which denote the capability to timely act, in a reactive or proactive manner, in response to changes in the context according to user preferences (personalization). Sensitivity, responsiveness and adaptivity all contribute to the concept of context awareness. Finally, the terms ubiquity and transparency directly refer to concept of pervasive computing. AmI is obtained by merging techniques from different research areas [5] including artificial intelligence (AI) and human-computer interaction (HCI). The different proposed approaches largely differs in terms of models taken as input for both space dynamics and human behavior in it, how these models are extracted, what kind of input sensors they need, which are the constraints under which they are supposed to effectively work. The performance evaluation of all these approaches, especially in the academic field, is often conducted in controlled situations where some of the features of a real environment are hid by the need for repeatability of the experiment. In this paper, we propose a comparative framework for techniques and approaches for modeling and extracting models to be employed with modern smart spaces. Differently from other surveys in the same area, this work focuses on giving an overview of the approaches by enforcing a taxonomy with dimensions that are chosen in order to understand the suitability of a specific technique to a specific setting. The paper is organized as it follows. Section 2 introduces the terminology and basic notions of smart spaces needed to fruitfully follow the discussion. Section 3 explains the inclusion criteria for the literature review. Section 4 discusses sensor types and how they are considered in the relevant literature. Sections 5 and 6 discuss the different types of smart space models employed and how they are constructed. Section 7 compares our work with other similar literature surveys. Finally, Section 8 concludes the paper by providing final considerations. 2. Background Figure 1 depicts the closed loops that characterize a running smart space [7]. The main closed loop, depicted using solid arrows and shapes, shows how the knowledge of environment dynamics and of users behaviors and preferences is employed to interpret sensors output in order to perform appropriate actions on the environment. Sensor data is first analyzed to extract the current context, which is an internal abstraction of the state of the environment from the point of view of the AmI system. The extracted context is then employed to make decisions on the actions to perform on the controlled space. Actions related to these decisions modify the environment (both physical and digital) by means of actuators of different forms. Sensors can be roughly divided into physical ones, which provide direct measurements about the environment (e.g., humidity, brightness, temperature), the devices and the users, and cyber ones, which provide digital information, not directly related to physical phenomena, such as user calendars. A cyber sensor often provides information related to the presence of the user in the cyberspace (e.g., the calendar of an user, a tweet posted by him/her, etc.). The term sensor data encompasses raw (or minimally processed) data retrieved from both physical sensors and cyber sensors. We can imagine a smart space producing, at runtime, a sensor log containing raw measurements from available sensors. 3 of 23 Future Internet 2019, 11, 23 Cyber-Physical AmbientwIntelligence Runtime Learning Context Extraction Acting Sensing Decision Making Knowledge Smart Cyber-Physical Environment Figure 1. The ambient intelligence closed loop. Arrows denote information flow. Dashed lines are used to denote optional information flows. Acting Context Extraction Figure 1. The ambient intelligence closed loop. Arrows denote information flow. Dashed lines are used to denote optional information flows. Definition 1 (Sensor Log). Given a set S of sensors, a sensor log is a sequence of measurements of the kind ⟨ts, s, v⟩where ts is the timestamp of the measurement, s ∈S is the source sensor and v the measured value, which can be either nominal (categorical) or numeric (quantitative). Measurements can be produced by a sensor on a periodic base (e.g., temperature measurements) or whenever a particular event happens (e.g., door openings). As many of the algorithms proposed in the literature borrow the terminology of data mining, the sensor log could be conceived as a sequence of events instead of a sequence of measurements. Definition 2 (Event Log). 2. Background Given a set E = {e1, . . . , enE} of event types, an event sequence is a sequence of pairs ⟨e, t⟩, where e ∈E and t is an integer, the occurrence time of the event type e. Definition 2 (Event Log). Given a set E = {e1, . . . , enE} of event types, an event sequence is a sequence of pairs ⟨e, t⟩, where e ∈E and t is an integer, the occurrence time of the event type e. Definition 2 is more restrictive than Definition 1. Translating a sensor log into an event log could cause a loss of information especially if discretization of periodic sensor measurements is required. Authors in the field of smart spaces uses, sometimes as synonyms, a variety of terms to refer to the state of the environment and the tasks humans perform in it. For the rest of the article, we will use the following terminology: • Context: The state of the environment including the human inhabitants. This includes the output of sensors and actuators, but also the state of human inhabitants including the action/activities/habits he/she is performing. In this very comprehensive meaning, the term situation is sometimes used. • Action: Atomic interaction with the environment or a part of it (e.g., a device). Recognizing actions can be easy or difficult depending on the sensors installed. Certain methods only focuses on actions and they will not be part of our survey. In some cases methods to recognize activities and habits completely skip the action recognition phase, only relying on the raw measurements in the sensor log. • Activity: A sequence of actions (one in the extreme case) or sensor measurements/events with a final goal. In some cases an action can be an activity itself (e.g., ironing). Activities can be collaborative, including actions by multiple users and can interleave one each other. The granularity (i.e., the temporal extension and complexity) of considered activities cannot be precisely specified. According to the approach, tidying up a room can be an activity whereas others approaches may generically consider tidying up the entire house as an activity. In any case, some approaches may hierarchically define activities, where an activity is a combination of sub-activities. • Habit: A sequence or interleaving of activities that happen in specific contextual conditions (e.g., what the user does every morning between 08:00 and 10:00). 2. Background • Habit: A sequence or interleaving of activities that happen in specific contextual conditions (e.g., what the user does every morning between 08:00 and 10:00). Knowledge plays a central role in AmI systems. As it intervenes both for context extraction and decision-making, it takes the form of a set of models describing (i) users behavior, 4 of 23 Future Internet 2019, 11, 23 (ii) environment/device dynamics, and (iii) user preferences. Anyway, knowledge should not be considered as a static resource as both users behavior and preferences change over time. Vast majority of works in the area of ambient intelligence suppose the knowledge to be obtained off-line, independently from the system runtime. A second optional loop in Figure 1, depicted using dashed arrows, shows that the current context could be employed to update the knowledge by applying learning techniques at runtime. Noteworthy, AmI is not intended to be provided by a centralized entity, on the contrary, its nature is distributed with embedded devices and software modules, possibly unaware one of each other, contributing to its features. Recently introduced smart space appliances, such as the NEST thermostat (see https://nest.com/thermostats/nest-learning-thermostat), contain sensors, actuators and AmI features in a single small package. 3. Inclusion Criteria and Comparison Framework In order to conduct our analysis we took inspiration from SLR (Systematic Literature Review) guidelines. A SLR is a method to identify, evaluate and interpret relevant scientific works with respect to a specific topic. We designed a protocol for conducting the SLR inspired to the guidelines and policies presented in [8], in order to ensure that the results are repeatable and the means of knowledge acquisition are scientific and transparent. The necessary steps to guarantee compliance with the guidelines include (i) the formulation of the research questions; (ii) the definition of a search string; (iii) the selection of the data sources on which the search is performed; (iv) the identification of inclusion and exclusion criteria; (v) the selection of studies; (vi) the method of extracting data from the studies; and (vii) the analysis of the data. Our analysis covers the different aspects of habit mining by separately analyzing the features requested to the sensor log, the modeling phase, and the runtime phase. The modeling phase is in charge of creating the models of human habits and environmental dynamics, whereas the runtime phase covers the aspect related to how these models are employed at runtime to recognize the context and to act on the environment. These two phases can even overlap in the case the system is able to refine models at runtime (either in collaboration or not with the user) or even to completely create them from scratch at runtime. In any case, the runtime phase covers only the way models are employed whereas the modeling phase cover any phase that is related to model production or update. g p y p p p On the basis of the above premises, the following research questions RQ-x have been defi • RQ-A: Sensor measurements represent the traces of user behavior in the environment. This information is needed at runtime to understand the current context, but the specific available information must be known when the model is defined (through specification or learning). The current review will only take into account the sensors for which the method has been validated. Section 4 will present for each of the included papers the following information: • RQ-A: Sensor measurements represent the traces of user behavior in the environment. 3. Inclusion Criteria and Comparison Framework Section 6 expands the following analysis: 5 of 23 Future Internet 2019, 11, 23 – RQ-B2.1: for methods involving an, optionally partial, automatic construction of models (i.e., through learning), a training set consisting of a sensor log must be fed as input. At first, learning methods in the field of ambient intelligence can be classified according to the effort devoted to label the training set; – RQ-B2.1: for methods involving an, optionally partial, automatic construction of models (i.e., through learning), a training set consisting of a sensor log must be fed as input. At first, learning methods in the field of ambient intelligence can be classified according to the effort devoted to label the training set; g ; – RQ-B2.2: does the system consider the possibility of having multiple users? How many? RQ B2 3: in the latter case is some type of additional labeling needed? g ; – RQ-B2.2: does the system consider the possibility of having multiple users? How many RQ B2 3 i th l tt i t f dditi l l b li d d? Q y p y g p y – RQ-B2.3: in the latter case, is some type of additional labeling needed? y p y g p – RQ-B2.3: in the latter case, is some type of additional labeling needed? As terms in this specific research area are not yet standardized, we replaced the search string with a crawling-like procedure starting from a set of seed papers, recursively navigating papers through the “Cited by” feature of Google Scholar, and selecting influential works from the last 10 years. This kind of search was manually performed on Google Scholar, as it includes all the relevant sources of scientific papers available. We then computed the 33rd percentile over the number of citations per year. In general, using the number of citations per year promotes recent works, as they attract citations more than outdated ones, but as we are analyzing only works from the last 10 years, this approach allows to highlight the most influential works. Additionally, papers that extend works included by this criterion are included in the results. The application of this criteria allowed to identify 22 primary studies (cf. Table 1) that were included in the final SLR. For each work, in the following tables, a general description is provided in addition to some schematic information. Table 1. Selected studies. 3. Inclusion Criteria and Comparison Framework This information is needed at runtime to understand the current context, but the specific available information must be known when the model is defined (through specification or learning). The current review will only take into account the sensors for which the method has been validated. Section 4 will present for each of the included papers the following information: – RQ-A1: which sensors are taken into account? Q – RQ-A2: are the sensors only categorical or numerical? And in the latter case, RQ A3 hi h di i i i l d? Q RQ-A2: are the sensors only categorical or nu y g – RQ-A3: which discretization strategy is employed? • RQ-B1: Any proposed method has a way to represent models. Models can be represented using graphical or mathematical/logic formalisms. Some methods propose formalisms that are specifically designed for the particular approach. Other methods conversely employ standard formalisms from machine learning and data mining. Section 5 analyzes the following aspects about employed models: • RQ-B1: Any proposed method has a way to represent models. Models can be represented using graphical or mathematical/logic formalisms. Some methods propose formalisms that are specifically designed for the particular approach. Other methods conversely employ standard formalisms from machine learning and data mining. Section 5 analyzes the following aspects about employed models: – RQ-B1.1: the type of adopted model; h d f h – RQ-B1.1: the type of adopted model; RQ B1 2 h d i t f th d l b t d? I it h d bl ? Q yp p ; – RQ-B1.2: how does an instance of the model can be represented? Is it h RQ B1.2: how does an instance of the model can be represented? Is it human readable? – RQ-B1.3: which is the granularity at which the model works? p – RQ-B1.3: which is the granularity at which the model works? • RQ-B2: Whichever model type is adopted by a specific method, each method introduces a different way to construct the model. Section 6 expands the following analysis: • RQ-B2: Whichever model type is adopted by a specific method, each method introduces a different way to construct the model. 3. Inclusion Criteria and Comparison Framework Acronym Study Title AUG-ECA [9] Augusto, J. C., Liu, J., McCullagh, P., Wang, H., and Yang, J.-B. Management of uncertainty and spatio-temporal aspects for monitoring and diagnosis in a smart home. AUG-APUBS [10] Aztiria, A., Augusto, J. C., Basagoiti, R., Izaguirre, A., and Cook, D. J. Discovering frequent user–environment interactions in intelligent environments. CASAS-DISCOREC [11–13] Cook, D. J., Krishnan, N. C., and Rashidi, P. Activity discovery and activity recognition: A new partnership. Rashidi, P., Cook, D. J., Holder, L. B., and Schmitter-Edgecombe, M. Discovering activities to recognize and track in a smart environment. Rashidi, P. and Cook, D. J. Com: A method for mining and monitoring human activity patterns in home-based health monitoring systems. CASAS-HMM [14] Singla, G., Cook, D. J., and Schmitter-Edgecombe, M. Recognizing independent and joint activities among multiple residents in smart environments. CASAS-HAM [15] Rashidi, P. and Cook, D. J. Keeping the resident in the loop: Adapting the smart home to the user. CASAS-HMMNBCRF [16] Cook, D. J. Learning setting-generalized activity models for smart spaces. CASAS-SVM [17] Krishnan, N. C. and Cook, D. J. Activity recognition on streaming sensor data. CHEN-ONT [18] Chen, L., Nugent, C. D., and Wang, H. A knowledge-driven approach to activity recognition in smart homes. WANG-EP [19] Gu, T., Wang, L., Wu, Z., Tao, X., and Lu, J. A pattern mining approach to sensor-based human activity recognition. 6 of 23 Future Internet 2019, 11, 23 Table 1. Cont. Acronym Study Title RIB-PROB [20,21] Helaoui, R., Riboni, D., and Stuckenschmidt, H. A probabilistic ontological framework for the recognition of multilevel human activities. Riboni, D., Sztyler, T., Civitarese, G., and Stuckenschmidt, H. Unsupervised recognition of interleaved activities of daily living through ontological and probabilistic reasoning. NUG-EVFUS [22] Hong, X., Nugent, C., Mulvenna, M., McClean, S., Scotney, B., and Devlin, S. Evidential fusion of sensor data for activity recognition in smart homes. WANG-HIER [23] William, C. et al. Fast effective rule induction. KROS-CRF [24] Van Kasteren, T., Noulas, A., Englebienne, G., and Kröse, B. Accurate activity recognition in a home setting. REIG-SITUATION [25] Brdiczka, O., Crowley, J. L., and Reignier, P. Learning situation models in a smart home. YANG-NN [26] Yang, J.-Y., Wang, J.-S., and Chen, Y.-P. Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers. STIK-MISVM [27] Stikic, M., Larlus, D., Ebert, S., and Schiele, B. Weakly supervised recognition of daily life activities with wearable sensors. 3. Inclusion Criteria and Comparison Framework FLEURY-MCSVM [28] Fleury, A., Vacher, M., and Noury, N. Svm-based multimodal classification of activities of daily living in health smart homes: sensors, algorithms, and first experimental results. PALMES-OBJREL [29] Palmes, P., Pung, H. K., Gu, T., Xue, W., and Chen, S. Object relevance weight pattern mining for activity recognition and segmentation. MOORE-STRAWMAN [30] Kwapisz, J. R., Weiss, G. M., and Moore, S. A. Activity recognition using cell phone accelerometers. LES-PHI [31] Lester, J., Choudhury, T., and Borriello, G. A practical approach to recognizing physical activities. BUE-WISPS [32] Buettner, M., Prasad, R., Philipose, M., and Wetherall, D. Recognizing daily activities with rfid-based sensors. 4. Supported Sensors Sensing technologies have made significant progress on designing sensors with smaller size, lighter weight, lower cost, and longer battery life. Sensors can, thus, be embedded in an environment and integrated into everyday objects and onto human bodies without affecting users’ comfort. Nowadays, sensors do not only include those traditionally employed for home and building automation (e.g., presence detectors, smoke detectors, contact switches for doors and windows, network-attached and close circuit cameras) but also more modern units (e.g., IMU—Inertial Measurements Units such as accelerometer and gyroscopes, WSN nodes), which are growingly available as part of off-the-shelf products. Measured values are usually affected by a certain degree of uncertainty. Sensors have indeed their own technical limitations as they are prone to breakdowns, disconnections from the system and environmental noise (e.g., electromagnetic noise). As a consequence, measured values can be out of date, incomplete, imprecise, and contradictory with each other. Techniques for cleaning sensor data do exist [33], but uncertainty of sensor data may still lead to wrong conclusions about the current context, which in turn potentially lead to incorrect behaviors of the system. Formalisms employed for representing knowledge in AmI systems often need environmental variables to be binary or categorical. A wide category of sensors (e.g., temperature sensors) produce instead numerical values, making it necessary to discretize sensor data before they can be used for reasoning. 7 of 23 Future Internet 2019, 11, 23 Discretization methods in machine learning and data mining are usually classified according to the following dimensions [34]: Discretization methods in machine learning and data mining are usually classified according to the following dimensions [34]: • Supervised vs. Unsupervised. Unsupervised methods do not make use of class information in order to select cut-points. Classic unsupervised methods are equal-width and equal-frequency binning, and clustering. Supervised methods employ instead class labels in order to improve discretization results. • Static vs. Dynamic. Static discretization methods perform discretization, as a preprocessing step, prior to the execution of the learning/mining task. Dynamic methods instead carry out discretization on the fly. y • Global vs. Local. Global methods, such as binning, are applied to the entire n-dimensional space. Local methods, as the C4.5 classifier, produce partitions that are applied to localized regions of the instance space. A local method is usually associated with a dynamic discretization method. • Top-down vs. Bottom-up. 4. Supported Sensors Top-down methods start with an empty list of cut-points (or split-points) and keep on adding new ones to the list by splitting intervals as the discretization progresses. Bottom-up methods start with the complete list of all the continuous values of the feature as cut-points and remove some of them by merging intervals as the discretization progresses. • Direct vs. Incremental. Direct methods directly divide the range of a quantitative attribute in k intervals, where the parameter k is provided as input by the user. Conversely, incremental methods start from a simple discretization and improve it step by step in order to find the best value of k. An important aspect to take into account when evaluating the usage of a particular category of sensors is the final user acceptance. Wearable sensors or cameras usually provoke a certain level of embarrassment in the users that must be taken into account. Table 2 shows, for each of the selected papers, the answers to research questions A1, A2 and A3. A3 is always empty for methods only supporting discrete sensors. Table 2. Supported Sensors—RQ-A1: sensors type—RQ-A2: data produced type—RQ-A3: discretization strategy. Approach RQ-A1 RQ-A2 RQ-A3 AUG-ECA [9] Position Discrete - AUG-APUBS [10] Object sensor, motion sensor, and context sensor Mixed - CASAS-DISCOREC [11–13] motion + interaction (RFID) Discrete - CASAS-HMM [14] motion + interaction (RFID) Discrete - CASAS-HMMNBCRF [16] motion + interaction (RFID) Discrete - CASAS-HAM [15] motion + interaction( RFID) Discrete - CASAS-SVM [17] Passive InfraRed Discrete - CHEN-ONT [18] Dense sensed environment: RFID, PIR, context sensors (tilt, pressure) Mixed Sensor measurements continuously fed into the system. RIB-PROB [20,21] RFID + body sensor Discrete - WANG-EP [19] RFID + user’s movement, environmental information, user location, human-object interaction Mixed From analogical measurements, like acceleration data and temperature, statistical values are computed (e.g., mean, variance) NUG-EVFUS [22] PIR, contact, pressure mats Discrete - WANG-HIER [23] RFID + body sensors (IMOTE2) Discrete - KROS-CRF [24] large variety of sensors, from contact to humidity/temperature Mixed Temporal probabilistic model: sampling measurements creating sequences of observations Table 2. Supported Sensors—RQ-A1: sensors type—RQ-A2: data produced type—RQ-A3: discretization strategy. 8 of 23 Future Internet 2019, 11, 23 Table 2. Cont. 5. Model Types Knowledge is represented in AmI systems using models. The literature about representing models of human habits is wide. In this section, we will review the most adopted approaches, highlighting those formalisms that are human understandable, thus, being easy to validate by a human expert or by the final user (once the formalism is known). Bayesian classification techniques are based on the well known Bayes theorem P(H|X) = P(X|H)P(H) P(X) , where H denotes the hypothesis (e.g., a certain activity is happening) and X represents the set of evidences (i.e., the current value of context objects). As calculating P(X|H) can be very expensive, different assumptions can be made to simplify the computation. For example, naïve Bayes (NB) is a simple classification model, which supposes the n single evidences composing X independent (that the occurrence of one does not affect the probability of the others) given the situational hypothesis; this assumption can be formalized as P(X|H) = ∏n k=1 P(xk|H). The inference process within the naïve Bayes assumption chooses the situation with the maximum a posteriori (MAP) probability. y p p p y Hidden Markov Models (HMMs) represent one of the most widely adopted formalism to model the transitions between different states of the environment or humans. Here hidden states represent situations and/or activities to be recognized, whereas observable states represent sensor measurements. HMMs are a statistical model where a system being modeled is assumed to be a Markov chain, which is a sequence of events. A HMM is composed of a finite set of hidden states (e.g., st−1, st, and st+1) and observations (e.g., ot−1, ot, and ot+1) that are generated from states. HMM is built on three assumptions: (i) each state depends only on its immediate predecessor; (ii) each observation variable only depends on the current state; and (iii) observations are independent from each other. In a HMM, there are three types of probability distributions: (i) prior probabilities over initial state p(s0); (ii) state transition probabilities p(st|st−1); and (iii) observation emission probabilities p(ot|st). A drawback of using a standard HMM is its lack of hierarchical modeling for representing human activities. To deal with this issue, several other HMM alternatives have been proposed, such as hierarchical and abstract HMMs. In a hierarchical HMM, each of the hidden states can be considered as an autonomous probabilistic model on its own; that is, each hidden state is also a hierarchical HMM. 4. Supported Sensors Approach RQ-A1 RQ-A2 RQ-A3 REIG-SITUATION [25] Video and audio records Analogical Extraction of user position and speed (|⃗v|) YANG-NN [26] triaxial accelerometer on a wearable board Discrete - STIK-MISVM [27] Audio-video, object interaction, environmental conditions, appliances’ use, 3D accelerometers Mixed Statistical features computed over 30 s sliding window shifted in increments of 15 s. FLEURY-MCSVM [28] Many: position, microphones, contact, wearable Mixed Extraction of statistical features (e.g., number of events per microphone) PALMES-OBJREL [29] RFID Discrete - LES-PHI [31] Accelerometers + audio + barometric pressure Mixed BUE-WISPS [32] WISPS: UHF RFID + accelerometers Discrete - 5. Model Types HMMs generally assume that all observations are independent, which could possibly miss long-term trends and complex relationships. Conditional Random Fields—CRFs, on the other hand, eliminate the independence assumptions by modeling the conditional probability of a particular sequence of hypothesis, Y, given a sequence of observations, X; succinctly, CRFs model P(Y|X). Modeling the conditional probability of the label sequence rather than the joint probability of both the labels and observations P(X, Y), as done by HMMs, allows CRFs to incorporate complex features of the 9 of 23 Future Internet 2019, 11, 23 observation sequence X without violating the independence assumptions of the model. The graphical model representations of a HMM (a directed graph, Figure 2a) and a CRF (an undirected graph, Figure 2b) makes this difference explicit. In [24], a comparison between HMM and CRF is shown, where CRF outperforms HMM in terms of timeslice accuracy, while HMM outperforms CRF in terms of class accuracy. Wash Dishes Wash Dishes Eating Eating Meal Preparation Meal Preparation S01 S01 S02 S02 S03 S03 S04 S04 S05 S05 a11 a12 b12 a21 a22 a23 a31 a32 a33 a13 b21 b23 b32 (a) Wash Dishes Wash Dishes Eating Eating Meal Preparation Meal Preparation S01 S01 S02 S02 S03 S03 S04 S04 S05 S05 a11 a12 a23 a22 a24 a35 b12 b21 b23 b32 (b) Figure 2. Examples of HMM and CRF models. Ellipses represent states (i.e., activities). Rectangles represent sensors. Arrows between states are state transition probabilities (i.e., the probability of moving from a state to another), whereas those from states to sensors are emission probabilities (i.e., the probability that in a specific state a sensor has a specific value). (a) HMM model example. Picture inspired by CASAS-HMM [14] and CASAS-HAM [15]. (b) CRF model example. Picture inspired by KROS-CRF [24]. Wash Dishes Wash Dishes Eating Eating Meal Preparation Meal Preparation S01 S01 S02 S02 S03 S03 S04 S04 S05 S05 a11 a12 b12 a21 a22 a23 a31 a32 a33 a13 b21 b23 b32 (a) (a) Wash Dishes Wash Dishes Eating Eating Meal Preparation Meal Preparation S01 S01 S02 S02 S03 S03 S04 S04 S05 S05 a11 a12 a23 a22 a24 a35 b12 b21 b23 b32 (b) (b) Figure 2. Examples of HMM and CRF models. Ellipses represent states (i.e., activities). Rectangles represent sensors. 5. Model Types Cluster hierarchical model portion example obtained by exploiting SVMs. Picture inspired by [35]. Artificial Neural Networks (ANNs) are a sub-symbolic technique, originally inspired by biological neuron networks. They can automatically learn complex mappings and extract a non-linear combination of features. A neural network is composed of many artificial neurons that are linked together according to a specific network architecture. A neural classifier consists of an input layer, a hidden layer, and an output layer. Mappings between input and output features are represented in the composition of activation functions f at a hidden layer, which can be learned through a training process performed using gradient descent optimization methods or resilient backprogagation algorithms. Some techniques stem from data mining methods for market basket analysis (e.g., the Apriori algorithm [36]), which apply a windowing mechanism in order to transform the event/sensor log into what is called a database of transactions. Let I = {i1, . . . , inE} be a set of binary variables corresponding to sensor event types. A transaction is an assignment that binds a value to each of the variables in I, where the values 0 and 1 respectively denote the fact that a certain event happened or not during the considered window. A database of transactions T is an (usually ordered) sequence of transactions each having a, possibly empty, set of properties (e.g., a timestamp). An item is an assignment of the kind ik = {0, 1}. An itemset is an assignment covering a proper subset of the variables in I. An itemset C has support SuppT (C) in the database of transactions T if a fraction of SuppT (C) of transactions in the database contain C. The techniques following this strategy turn the input log into a database of transactions, each of them corresponding to a window. Given two different databases of transactions T1 and T2, the growth rate of an itemset C from T1 to T2 is defined as SuppT2(X) SuppT1(X). Emerging patterns (EP) are those itemsets showing a growth rate greater than a certain threshold ρ. The ratio behind this definition is that an itemset that has high support in its target class (database) and low support in the contrasting class, can be seen as a strong signal, in order to discover the class of a test instance containing it. 5. Model Types Arrows between states are state transition probabilities (i.e., the probability of moving from a state to another), whereas those from states to sensors are emission probabilities (i.e., the probability that in a specific state a sensor has a specific value). (a) HMM model example. Picture inspired by CASAS-HMM [14] and CASAS-HAM [15]. (b) CRF model example. Picture inspired by KROS-CRF [24]. Another statistical tool often employed is represented by Markov Chains (MCs), which are based on the assumption that the probability of an event is only conditional to the previous event. Even if they are very effective for some applications like capacity planning, in the smart spaces context, they are quite limited because they deal with deterministic transactions and modeling an intelligent environment with this formalism results in a very complicated model. Another statistical tool often employed is represented by Markov Chains (MCs), which are based on the assumption that the probability of an event is only conditional to the previous event. Even if they are very effective for some applications like capacity planning, in the smart spaces context, they are quite limited because they deal with deterministic transactions and modeling an intelligent environment with this formalism results in a very complicated model. Support Vector Machines (SVMs) allow to classify both linear and non-linear data. A SVM uses a non-linear mapping to transform the original training data into an higher dimension. Within this new dimension, it searches for the linear optimal separating hyperplane that separates the training data of one class from another. With an appropriate non-linear mapping to a sufficiently high dimension, data from two classes can always be separated. SVMs are good at handling large feature spaces since they employ overfitting protection, which does not necessarily depend on the number of features. Binary Classifiers are built to distinguish activities. Due to their characteristics, SVMs are better in generating other kind of models with a machine learning approach than modeling directly the smart 10 of 23 Future Internet 2019, 11, 23 environment. For instance in [35] authors uses them combined with Naive Bayes Classifiers to learn the activity model built on hierarchical taxonomy formalism shown in Figure 3. Night_wandering Work_computer Eat Dinner Wash_Dishes Wash_Dishes Lunch Wash_Dishes Cook_lunch Snack Meal Preparation Cook Wash_Dishes Cook Figure 3. Cluster hierarchical model portion example obtained by exploiting SVMs. Picture inspired by [35]. Night_wandering Wash_Dishes Cook Figure 3. 5. Model Types Market basket analysis is a special case of affinity analysis that discovers co-occurrence relationships among purchased items inside a single or more transactions. Initial approaches to the development of context-aware systems able to recognize situations were based on predicate logic. Loke [37] introduced a PROLOG extension called LogicCAP; here the 11 of 23 Future Internet 2019, 11, 23 Future Internet 2019, 11, 23 “in-situation” operator captures a common form of reasoning in context-aware applications, which is to ask if an entity E is in a given situation S (denoted as S* > E). In particular, a situation is defined as a set of constraints imposed on output or readings that can be returned by sensors, i.e., if S is the current situation, we expect the sensors to return values satisfying some constraints associated with S. LogicCAP rules use backward chaining like PROLOG, but also utilizes forward chaining in determining situations, i.e., a mix of backward and forward chaining is used in evaluating LogicCAP programs. The work introduces different reasoning techniques with situations including selecting the best action to perform in a certain situation, understanding what situation a certain entity is in (or the most likely) and defining relationships between situations. There are many approaches borrowed from information technology areas, adapted to smart environments. For instance in [38], the authors use temporal logic and model checking to perform activities modeling and recognition. The system proposed is called ARA. A graphical representation of a model example adopted by this approach is shown in Figure 4. It evidences how the activities are composed by the time correlated states between consecutive actions. cpw cpw cpc cpc σt−tce σt−tce+1 σt−tcd σt−1 σt hso hso sop sop ddi ddi σt−tce σt−tce+1 σt−tsop σt−tddi σt Meal Preparation Wash Dishes hso hso sop sop ddi ddi σt−tce hso hso ddi ddi σt−tce σt−tce+1 σt−tmpr σt cmo cmo mpr mpr σt−tce Eating Figure 4. Model formalism used in [38] based on R-pTL formulas. The model is based on correlations between events. Meal Preparation Meal Preparation Wash Dishes Eating Figure 4. Model formalism used in [38] based on R-pTL formulas. The model is based on correlations between events. 5. Model Types Ontologies (denoted as ONTO) represent the last evolution of logic-based approaches and have increasingly gained attention as a generic, formal and explicit way to “capture and specify the domain knowledge with its intrinsic semantics through consensual terminology and formal axioms and constraints” [39]. They provide a formal way to represent sensor data, context, and situations into well-structured terminologies, which make them understandable, shareable, and reusable by both humans and machines. A considerable amount of knowledge engineering effort is expected in constructing the knowledge base, while the inference is well supported by mature algorithms and rule engines. Some examples of using ontologies in identifying situations are given by [40] (later evolved in [20,21]). Instead of using ontologies to infer activities, they use ontologies to validate the result inferred from statistical techniques. 12 of 23 Future Internet 2019, 11, 23 Future Internet 2019, 11, 23 Future Internet 2019, 11, 23 Figure 5 contains three different examples of ontologies employed to model smart spaces. Clearly, each approach adopts the formalism most devised for its application, and many aspects can be modeled under different granularities. “Eating”: { “type”: [ “Personal”] “location”:[“kitchen”, “dining room”] “sensors”:[“S01”,”S02”,”S03”] “action” : [“MoveChair”] “IAM” : [ “hasPrepared”] “attached-to”: [] “duration”: [“700”] } __________________________________ “MealPreparation”: { “type”: [ “Cooking”] “location”:[“kitchen”] “attached-to”: [] “sensors”:[“S01”,”S02”] “IAM” : [ ] action” : [“turnOnCooker”] “duration”: [“300”] } __________________________________ “WashDishes”: { “type”: [ “Cleaning”] “location”:[“kitchen”] “sensors”:[”S02”,”S04”, “S05”] “action” : [“openSink”, “dry”] “IAM” : [ ] “attached-to”: [“dish”,”forks”,”glasses”] “duration”: [“700”] } _____________________________ (a) Container Type Location Device Actuator Sensor S01 Prepare Meal S03 Entities Addings Electric Appliance Toaster Oven Fridge Activity Basic Activity FunctionalActivity BedRoom Activity BathRoom activity Do Housework Kitchen Activity Wash Dishes Eating S02 attachedTo achieve dependOn hasAdding hasActor hasDrinkType hasCookAppl hasHotMealTyp haslocation hashSensor Contextual classes Activity classes Properties (b) Owl:Thing Owl:Thing Artifact Artifact Activity Activity Operation Operation Sensor Sensor Device Device Individual Individual Social Social Personal Personal Physical Physical Meal Prepar. Meal Prepar. Eating Eating Wash Dishes Wash Dishes STOVE⊆COOKING−INSTR∩. (∃USED−FOR.((PREPARE−MEAL∪PREPARE−TEA)∩(∃OCCURS .{APT }))); FREEZER⊆DEVICE∩(∃USED−FOR.(PREPARE−MEAL)∩(∃OCCURS.{APT })); SILVERWARE−DRAWER⊆FOOD−PREP−FORNITURE; FOOD−PREP−FURNITURE⊆FURNITURE∩. (∃USED−FOR.(PREP−TEA∪PREPARE−MEAL)∩(∃OCCURS .{APT })); (c) Figure 5. Examples of Ontology Models. (a) Example of machine-oriented ontology. Picture taken from [41]. (b) Example of taxonomy based ontology. Picture inspired by CHEN-ONT [18]. (c) A portion of the resulting ontology model proposed in RIB-PROB [20,21]. The dashed lines represent the relations super/sub between classes. 5. Model Types (b) (a) Owl:Thing Owl:Thing Artifact Artifact Operation Operation Activity Activity Sensor Sensor Device Device Individual Individual Physical Physical Personal Personal Meal Prepar. Meal Prepar. Eating Eating (c) Figure 5. Examples of Ontology Models. (a) Example of machine-oriented ontology. Picture taken from [41]. (b) Example of taxonomy based ontology. Picture inspired by CHEN-ONT [18]. (c) A portion of the resulting ontology model proposed in RIB-PROB [20,21]. The dashed lines represent the relations super/sub between classes. The way an AmI system makes decisions on the actions can be compared to decision-making in AI agents. As an example, reflex agents with state, as introduced in [42], take as input the current state of the world and a set of Condition-Action rules to choose the action to be performed. Similarly, Augusto [43] introduces the concept of Active DataBase (ADB) composed by Event-Condition-Action (ECA) rules. An ECA rule basically has the form “ON event IF condition THEN action”, where conditions can take into account time. The first attempts to apply techniques taken from the business process management—BPM [44] area were the employment of workflow specifications to anticipate user actions. A workflow is composed by a set of tasks related by qualitative and/or quantitative time relationships. Authors in [45] present a survey of techniques for temporal calculus (i.e., Allen’s Temporal Logic and Point 13 of 23 Future Internet 2019, 11, 23 Algebra) and spatial calculus aiming at decision-making. The SPUBS system [46,47] automatically retrieves these workflows from sensor data. Table 3 shows, for each surveyed paper, information about RQ-B1.1 (Formalism), RQ-B1.2 (Readability Level) and RQ-B1.3 (Granularity). Table 3. Model type (L—Low, M—Medium, H—High)—RQ-B1.1: The type of model—RQ-B1.2: Human Readability Level—RQ-B1.3: Granularity Level. Table 3. Model type (L—Low, M—Medium, H—High)—RQ-B1.1: The type of model—RQ-B1.2: Human Readability Level—RQ-B1.3: Granularity Level. References RQ-B1.1 RQ-B1.2 RQ-B1.3 AUG-ECA [9] ECA H Action AUG-APUBS [10] Action CASAS-HMMNBCRF [16] NB L Activity CASAS-DISCOREC [11–13] HMM M Activity CASAS-HMM [14] Activity CASAS-HMMNBCRF [16] Activity KROS-CRF [24] Activity REIG-SITUATION [25] Situation LES-PHI [31] Activity BUE-WISPS [32] Activity CASAS-HAM [15] MC M Event CASAS-HMMNBCRF [16] CRF L Activity KROS-CRF [24] Activity REIG-SITUATION [25] SVM L Situation STIK-MISVM [27] Activity FLEURY-MCSVM [28] Activity CHEN-ONT [18] ONTO H Activity RIB-PROB [20,21] Action/Activity NUG-EVFUS [22] Action WANG-EP [19] EP L Action/Activity YANG-NN [26] ANN L Activity PALMES-OBJREL [29] Other H Activity 6. Model Construction Modeling formalisms in the literature can be roughly divided into specification-based and learning-based [1]. Research in the field of AmI started when few kinds of sensors were available and the relationships between sensor data and underlying phenomena were easy to establish. Specification-based approaches represent hand-made expert knowledge in logic rules and apply reasoning engines to infer conclusions and to make decisions from sensor data. These techniques evolved in the last years in order to take into account uncertainty. The growing availability of different kinds of sensors made hand-made models impractical to be produced. In order to solve this problem, learning-based methods employ techniques from machine learning and data mining. Specification-based models are usually more human-readable (even though a basic experience with formal logic languages is required), but creating them is very expensive in terms of human resources. Most learning-based models are instead represented using mathematical and statistical formalisms (e.g., HMMs), which make them difficult to be revised by experts and understood by final users. These motivations are at the basis of the research of human-readable automatically inferred formalisms. Learning-based techniques can be divided into supervised and unsupervised techniques. The former expect the input to be previously labeled according to the required output function, hence, they require a big effort for organizing input data in terms of training examples, even though active learning can be employed to ease this task. Unsupervised techniques (or weakly supervised ones, i.e., those where 14 of 23 Future Internet 2019, 11, 23 only a part of the dataset is labeled) can be used to face this challenge but a limited number of works is available in the literature. Unsupervised techniques for AmI knowledge modeling can be useful for other two reasons. Firstly, as stated in the introduction, sometimes knowledge should not be considered as a static resource; instead it should be updated at runtime without a direct intervention of the users [15], hence, updating techniques should rely on labeling of sensor data as little as possible. Moreover, unsupervised techniques may also result useful in supporting passive users, such as guests, that do not participate in the configuration of the system but should benefit from its services as well. Performing learning or mining from sequences of sensor measurements poses the issue of how to group events into aggregates of interests (i.e., actions, activities, situations). 6. Model Construction Even with supervised learning techniques, if labeling is provided at learning time, the same does not hold at runtime where a stream of events is fed into the AmI system. Even though most proposed approaches in the AmI literature (especially supervised learning ones) ignore this aspect, windowing mechanism are needed. As described in [17], the different windowing methods can be classified into three main classes, namely, explicit, time-based and event-based. • Explicit segmentation. In this case, the stream is divided into chunks usually following some kind of classifier previously instructed over a training data set. Unfortunately, as the training data set simply cannot cover all the possible combinations of sensor events, the performance of such a kind of approach usually results in single activities divided into multiple chunks and multiple activities merged. g • Time-based windowing. This approach divides the entire sequence into equal size time intervals. This is a good approach when dealing with data obtained from sources (e.g., sensors like accelerometers and gyroscopes) that operate continuously in time. As can be easily argued, the choice of the window size is fundamental especially in the case of sporadic sensors as a small window size could not contain enough information to be useful whereas a large window size could merge multiple activities when burst of sensors occur. g p • Sensor Event-based windowing. This last approach splits the entire sequence into bins containing an equal number of sensor events. Usually, bins are in this case overlapping with each window containing the last event arrived together with the previous events. Whereas this method usually performs better than the other, it shows drawbacks similar to those introduced for time based windowing. g p • Sensor Event-based windowing. This last approach splits the entire sequence into bins containing an equal number of sensor events. Usually, bins are in this case overlapping with each window containing the last event arrived together with the previous events. Whereas this method usually performs better than the other, it shows drawbacks similar to those introduced for time based windowing. In AUG-ECA [9], the rules composing the model are written by experts, so the model formalism is based on knowledge bases. The rules will be evaluated by exploiting evidential reasoning (ER) and the parameter for the ER are trained in a supervised way. The rules structure follows the Event-Condition-Action (ECA) paradigm, borrowed by database techniques. 6. Model Construction The rules specify how the system has to react to a given event in a specific context. The action performed by the system influences the status of the system itself, potentially generating another event managed with a reaction is other rules. Figure 6 shows some examples of ECA rules used to model some activities. The structure has the typical layout of a IF-THEN construct, in which the conditions represent sensor triggering (or its interpretation), and the conditions are expressed as probabilities regarding the context elements and the auxiliary actions performed. The limits of this approach are related to the big effort needed to model successfully the environment and the possibility of producing conflicting conditions. The evolution in specification-based approaches to overcome the previous limits is represented by ontologies. In CHEN-ONT [18], an activity is modeled as a class: the entities and the properties correlated to them are divided in three groups (see Figure 7). The first one is about the context, the second group represents the causal and/or functional relations, the properties of the third group denote the type and interrelationship between activities. 15 of 23 Future Internet 2019, 11, 23 Future Internet 2019, 11, 23 “Eating”: { IF at_kitchen_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDLater tdRK_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDlater at_reception_off with {(H, 0.9);(M, 0);(L, 0);(N,0)} Confidence THEN <turn on oven> } “Eating”: { IF at_kitchen_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDLater tdRK_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDlater at_reception_off with {(H, 0.9);(M, 0);(L, 0);(N,0)} Confidence THEN <turn on oven> } “WashDishes”: { IF at_kitchen_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDLater at_reception_on with {(H, 1);(M, 0);(L, 0);(N,0)} Confidence ANDlater sink_on with {(H, 0.8);(M, 0.2);(L, 0);(N,0)} confidence THEN <turn on light> AND <turn off oven> } } “MealPreparation”: { IF at_kitchen_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDLater tdRK_on with {(H, 0.9);(M, 0.1);(L, 0);(N,0)} confidence ANDlater at_reception_off with {(H, 0.8);(M, 0.2);(L, 0);(N,0)} Confidence ANDlater fridge_interaction_on with {(H, 1);(M, 0);(L, 0);(N,0)} Confidence THEN <turn on oven> } Figure 6. Example of ECA rules modeling “Eating”, “MealPreparation”, “WashDishes” activities. Pictures inspired by AUG-ECA [9]. Figure 6. Example of ECA rules modeling “Eating”, “MealPreparation”, “WashDishes” activities. Pictures inspired by AUG-ECA [9]. Future Internet 2019, 11, 23 16 of 23 Future Internet 2019, 11, 23 In RIB-PROB [20,21], the multilevel model is obtained combining ontologies and/or grouping elements at previous levels. The Atomic Gestures model is obtained just considering log elements. The Manipulative Gestures are computed considering ontology and axioms. The Simple Activities are obtained grouping Manipulative Gestures. Finally, for Complex Activities, ontologies are involved. Figure 5c represents a portion of the resulting ontology model. The dashed lines represent the relations super/sub between classes. The individual classes have relations that describe dependencies. Moreover Description Logic is employed to support ontological reasoning, which allows also to check the knowledge base consistency. It also infers additional information from registered facts. In NUG-EVFUS [22], the interrelationships between sensors, context and activities are represented as a hierarchical network of ontologies (see Figure 8). A particular activity can be performed or associated with a certain room in the house, this information is modeled with an ontology of the network. Meal Prep S01 Fridge S02 Sink S03 Fridge Cleaning Lunch Figure 8. Hierarchical ontology structure adopted in NUG-EVFUS [22] to model activities in a smart space. S02 S03 S01 Figure 8. Hierarchical ontology structure adopted in NUG-EVFUS [22] to model activities in a smart space. In CASAS-HMM [14], each activity is performed in a protected environment, and the resulting log is recorded and labeled. Then, HMM model is built upon this dataset in a supervised way. The resulting model is shown in Figure 2a. Observations (squares) model the sensor triggering, the states (circles) model the activities that can generate the observations according to certain probabilities. The goal is to infer the activities by processing the observations. This recognition technique is supporting single user data, but the problem of modeling multiple users is introduced. The same team, in CASAS-SVM [17], employs SVM. In this second work, authors propose an interesting analysis of the different windowing strategies to be employed to gather measurements into observation vectors. Finally, in CASAS-HMMNBCRF [16], experiments are performed with the same methodology adding CRF and NB modeling techniques to the analysis. In WANG-EP [19], from log of sequential activities, Emerging Patterns are mined and the resulting set composes the model. In KROS-CRF [24], the model is trained out from a labeled dataset. The log is divided in segments 60-s long and each segment is labeled. 6. Model Construction c es sp e y UG C [9] Container Type Location Device Actuator Sensor S01 Prepare Meal S03 Entities Addings Electric Appliance Toaster Oven Fridge Activity Basic Activity FunctionalActivity BedRoom Activity BathRoom activity Do Housework Kitchen Activity Wash Dishes Eating S02 attachedTo achieve dependOn hasAdding hasActor hasDrinkType hasCookAppl hasHotMealType haslocation hashSensor ntextual classes Activity classes Properties (a) (b) (c) Figure 7. Ontologies used in CHEN-ONT [18] to model the aspects of smart spaces. (a) Ontology example used to model the Smart Environment domain. Picture inspired to CHEN-ONT [18]. (b) Ontology example used to model the activities correlations. Picture taken from CHEN-ONT [18]. (c) Ontology example used to model the sensors properties. Picture taken from CHEN-ONT [18]. (b) Container Type Location Device Actuator Sensor S01 Prepare Meal S03 Entities Addings Electric Appliance Toaster Oven Fridge Activity Basic Activity FunctionalActivity BedRoom Activity BathRoom activity Do Housework Kitchen Activity Wash Dishes Eating S02 attachedTo achieve dependOn hasAdding hasActor hasDrinkType hasCookAppl hasHotMealType haslocation hashSensor Contextual classes Activity classes Properties (a) (b) Contextual classes (a) (b) (c) (c) Figure 7. Ontologies used in CHEN-ONT [18] to model the aspects of smart spaces. (a) Ontology example used to model the Smart Environment domain. Picture inspired to CHEN-ONT [18]. (b) Ontology example used to model the activities correlations. Picture taken from CHEN-ONT [18]. (c) Ontology example used to model the sensors properties. Picture taken from CHEN-ONT [18]. Figure 7. Ontologies used in CHEN-ONT [18] to model the aspects of smart spaces. (a) Ontology example used to model the Smart Environment domain. Picture inspired to CHEN-ONT [18]. (b) Ontology example used to model the activities correlations. Picture taken from CHEN-ONT [18]. (c) Ontology example used to model the sensors properties. Picture taken from CHEN-ONT [18]. Future Internet 2019, 11, 23 The dataset is composed by multi-days logs: one day is used for testing the approach, the remaining for training the models. The resulting model is an undirected graph as in Figure 2b. In REIG-SITUATION [25], a SVM model, built on statistical values extracted from the measurements of a given user, is used for classifying the roles. Then, this information, combined with other statistically extracted ones, is involved into the training of the HMM that models the situations. 17 of 23 Future Internet 2019, 11, 23 In YANG-NN [26], the input vector contains the features to consider, the output vector the classes (activities). The back-propagation learning algorithm is used for training the ANNs. Three neural networks are built on labeled logs: a pre-classifier and two classifiers; static activities and dynamic activities are modeled with separated ANNs. The structure of the neural classifier consists of an input layer, an hidden layer and an output layer. In LES-PHI [31], given the maximum number of features the activity recognition system can use, the system automatically chooses the most discriminative sub-set of features and uses them to learn an ensemble of discriminative static classifiers for the activities that need to be recognized. Then, the class probabilities estimated from the static classifiers are used as inputs into HMMs. In BUE-WISPS [32], the users are asked to perform activities. The resulting log is used for training an HMM. In FLEURY-MCSVM [28], the classes of the classifier model the activities. Binary classifiers are built to distinguish activities: pairwise combinations selection. The number of SVMs for n activities will be n −1. The features used are statistics from measurements. The algorithm proposed in CASAS-DISCOREC [11–13] is to improve the performance of activity recognition algorithms by trying to reduce the part of the dataset that has not been labeled during data acquisition. In particular, for the unlabeled section of the log, the authors employ a pattern mining technique in order to discover, in an unsupervised manner, human activity patterns. A pattern is defined here as a set of events where order is not specified and events can be repeated multiple times. Patterns are mined by iteratively compressing the sensor log. The data mining method used for activity discovery is completely unsupervised without the need of manually segmenting the dataset or choosing windows and allows to discover interwoven activities as well. Future Internet 2019, 11, 23 18 of 23 Future Internet 2019, 11, 23 Events in the event part of the ECA rule always come from sets O and M. Conditions are usually expressed in terms of the values provided by Type C sensors. Finally, the action part contains only Type O sensors. The set of Type O sensor is called mainSeT. The first step of the APUBS method consists of discovering, for each sensor in the mainSeT, the set associatedSeT of potential O and M sensors that can be potentially related to it as triggering events. The method employed is APriori for association rules [36]; the only difference is that possible association rules X ⇒Y are limited to those where cardinality of both X and Y is unitary and Y only contains events contained in mainSeT. Obviously, this step requires a window size value to be specified in order to create transactions. As a second step, the technique discovers the temporal relationships between the events in associatedSeT and those in mainSeT. During this step, non-significant relations are pruned. As a third step, the conditions for the ECA rules are mined with a JRip classifier [48]. In WANG-HIER [23], starting from the raw log, the authors use a K-Medoids clustering method to discover template gestures. This method finds the k representative instances which best represent the clusters. Based on these templates, gestures are identified applying a template matching algorithm: Dynamic Time Warping is a classic dynamic programming based algorithm to match two time series with temporal dynamics. In PALMES-OBJREL [29], the KeyExtract algorithm mines keys from the web that best identify activities. For each activity, the set of most important keys is mined. In the recognition phase, an unsupervised segmentation based on heuristics is performed. Table 4 shows a quick recap of this section and answers questions RQ-B2.1 (Technique Class), RQ-B2.2 (Multi-user Support) and RQ-B2.3 (Additional Labeling Requirement). Table 4. Model construction (Y—Yes, N—No, S—Single, M—Multiple)—RQ-B2.1: Specification vs Learning (Supervised or Unsupervised)—RQ-B2.2: Number of users—RQ-B2.3: Labeling to address multi-user. Table 4. Model construction (Y—Yes, N—No, S—Single, M—Multiple)—RQ-B2.1: Specification vs Learning (Supervised or Unsupervised)—RQ-B2.2: Number of users—RQ-B2.3: Labeling to address multi-user. Future Internet 2019, 11, 23 Starting from singleton patterns, at each step, the proposed technique compresses the logs by exploiting them and iteratively reprocesses the compressed log for recognizing new patterns and further compress the log. When it is difficult to further compress the log, each remaining pattern represents an activity class. Discovered labels are employed to train HMM, BN and SVM models following the same approach as in the supervised works of the same group. In CASAS-HAM [15], the sensor log is considered completely unlabeled. Here temporal patterns (patterns with the addition of temporal information) are discovered similarly as in CASAS-DISCOREC [11–13] and are used for structuring a tree of Markov Chains. Different activations in different timestamps generate new paths in the tree. Depending to temporal constraints, a sub-tree containing Markov Chains at the leafs that model activities is generated. A technique to update the model is also proposed. Here the goal of the model is the actuation of target devices more than recognition. g Authors in STIK-MISVM [27] introduce a weakly supervised approach where two strategies are proposed for assigning labels to unlabeled data. The first strategy is based on the miSVM algorithm. miSVM is a SVM with two levels, the first for assigning labels to unlabeled data, the second one for applying recognition to the activities logs. The second strategy is instead called graph-based label propagation, where the nodes of the graphs are vectors of features. The nodes are connected by weighted edges, the weights represent similarity between nodes. When the entire training set is labeled, an SVM is trained for activity recognition. In AUG-APUBS [10], the system generates ECA rules by considering the typology of the sensors involved in the measurements and the time relations between their activations. APUBS makes clear the difference between different categories of sensors: • Type O sensors installed in objects, thus, providing direct information about the actions of the users. • Type C sensors providing information about the environment (e.g., temperature, day of the we • Type C sensors providing information about the environment (e.g., temperature, day of the week). • Type M sensors providing information about the position of the user inside the house (e.g., in the bedroom). • Type M sensors providing information about the position of the user inside the house (e.g., in the bedroom). 7. Related Work The literature contains several surveys attempting to classify works in the field of smart spaces and ambient intelligence. Papers are presented in chronological orders. None of the reported surveys, clearly states the modality by which papers have been selected. Authors in [7] follow an approach similar to this work, i.e., they separately analyze the different phases of the life-cycle of the models. Differently from our work, for what concerns the model construction phase, they focus on classes of learning algorithms instead of analyzing the specific work. Additionally, specification-based methods are not taken into account. The survey [49] is focused on logical formalisms to represent ambient intelligence contexts and reasoning about them. The analyzed approaches are solely specification-based. Differently from our work, the survey is focused on the reasoning aspect, which is not the focus of our work. The work [50] is an extensive analysis of methods employed in ambient intelligence. This work separately analyzes the different methods without clearly defining a taxonomy. Authors in [1] introduce a clear taxonomy of approaches in the field of context recognition (and more generally, about situation identification). The survey embraces the vast majority of the proposed approaches in the area. Equivalently, paper [51] is a complete work covering not only activity recognition but also fine-grained action recognition. Differently from our work, both surveys are not focusing on the life-cycle of models. Authors in [52] focus on reviewing the possible applications of ambient intelligence in the specific case of health and elderly care. The work is orthogonal to the present paper and all the other reported works, as it is less focused on the pros and cons of each approach, while instead focusing on applications and future perspectives. A manifesto of the applications and principles behind smart spaces and ambient intelligence is presented in [53]. As in [52], authors in [54] take moves from the health care application scenario in order to describe possible applications. Anyway, this work goes more into details of employed techniques with particular focus on classical machine learning methods. Future Internet 2019, 11, 23 Technique RQ-B2.1 RQ-B2.2 RQ-B2.3 AUG-ECA [9] Specification S N CHEN-ONT [18] Specification S N RIB-PROB [20,21] Specification S N NUG-EVFUS [22] Specification S N CASAS-HMM [14] Supervised S N CASAS-SVM [17] Supervised S N CASAS-HMMNBCRF [16] Supervised S N WANG-EP [19] Supervised S N KROS-CRF [24] Supervised S N REIG-SITUATION [25] Supervised M N YANG-NN [26] Supervised S N LES-PHI [31] Supervised S N BUE-WISPS [32] Supervised S N FLEURY-MCSVM [28] Supervised M Y CASAS-DISCOREC [11–13] Weakly Sup. S N STIK-MISVM [27] Weakly Sup. M Y AUG-APUBS [10] Unsupervised S N CASAS-HAM [15] Unsupervised S N WANG-HIER [23] Unsupervised S N PALMES-OBJREL [29] Unsupervised S N Future Internet 2019, 11, 23 19 of 23 19 of 23 Future Internet 2019, 11, 23 Future Internet 2019, 11, 23 20 of 23 As a final point, the validity of the declared results is often difficult to confirm on datasets different from the one used for tests, as in the vast majority of cases source code is not made available. The lack of benchmark data, even though many datasets are made freely available by research groups, makes this situation even harder. Author Contributions: F.L. contributed to methodology, conceptualization, writing and editing; M.M. contributed to methodology, writing and editing, supervision and funding acquisition; D.S. contributed to investigation, conceptualization, writing and editing; T.C. contributed to supervision and funding acquisition. Funding: The work of Daniele Sora has been partly supported by the Lazio regional project SAPERI & Co (FILAS-RU-2014-1113), the work of Francesco Leotta has been partly supported by the Lazio regional project Sapientia (FILAS-RU-2014-1186), all the authors have been also partly supported by the Italian projects Social Museum & Smart Tourism (CTN01-00034-23154), NEPTIS (PON03PE-00214-3) and RoMA—Resilence of Metropolitan Areas (SCN-00064). Conflicts of Interest: The authors declare no conflict of interest. Abbreviations The following abbreviations are used in this manuscript: AAL Ambient Assisted Living ADB Active DataBase AI Artificial Intelligence AmI Ambient Intelligence (A)NN (Artificial) Neural Network BPM Business Process Management CRF Conditional Random Field DST Dampster-Shafer Theory ECA Event-Condition-Action EP Emerging Pattern HCI Human-Computer Interaction HMM Hidden Markov Model IMU Inertial Measurements Unit IoT Internet-of-Things MC Markov Chain NB Naive Bayes RFID Radio-Frequency IDentification RQ Research Question SLR Systematic Literature Review SVM Support Vector Machine UHF Ultra-High Frequency WSN Wireless Sensor Network 1. Ye, J.; Dobson, S.; McKeever, S. Situation identification techniques in pervasive computing: A review. Pervasive Mob. Comput. 2012, 8, 36–66. [CrossRef] 4. Mennicken, S.; Huang, E. Hacking the Natural Habitat: An In-the-Wild Study of Smart Homes, Their Development, and the People Who Live in Them. In Pervasive Computing; Lecture Notes in Computer Science; Kay, J., Lukowicz, P., Tokuda, H., Olivier, P., Krüger, A., Eds.; Springer: Berlin/Heidelberg, Germany, 2012; Volume 7319, pp. 143–160. 1. Ye, J.; Dobson, S.; McKeever, S. Situation identification techniques in pervasive computing: A review. Pervasive Mob. Comput. 2012, 8, 36–66. [CrossRef] 2. Uckelmann, D.; Harrison, M.; Michahelles, F. An architectural approach towards the future Internet-of- Things. In Architecting the Internet-of-Things; Springer: Berlin/Heidelberg, Germany, 2011; pp. 1–24. 3. Augusto, J.; Nugent, C. (Eds.) Smart Homes Can Be Smarter. In Designing Smart Homes; Lecture Notes in Computer Science; Springer: Berlin/Heidelberg, Germany, 2006; Volume 4008, pp. 1–15. 4. Mennicken, S.; Huang, E. Hacking the Natural Habitat: An In-the-Wild Study of Smart Homes, Their Development, and the People Who Live in Them. In Pervasive Computing; Lecture Notes in Computer Science; Kay, J., Lukowicz, P., Tokuda, H., Olivier, P., Krüger, A., Eds.; Springer: Berlin/Heidelberg, Germany, 2012; Volume 7319, pp. 143–160. 8. Discussion In this paper, we applied the SLR approach to classify most prominent approaches in the ambient intelligence area according to how they approach the different stages of a model life cycle. The kind of classification introduced in the paper allows to choose a specific method or approach according to available sensors, amount of labeled data, need for visual analysis, requirements in terms of enactment and decision-making on the environment. The field of ambient intelligence has been very active in the last 10 years, and first products (such as NEST) have hit the market. Unfortunately, as of 2018, no examples of holistic approaches, such as those proposed in the vast majority of academic works, are available outside research labs. The reasons for this are multiple. First of all, the vast majority of approaches either require a big effort in terms of (i) expert time to produce hand-made models in specification based approaches, or (ii) user time to manually label training sets for supervised learning based approaches. Unsupervised methods represent a small percentage on the total amount of academic works. Secondarily, model update is often not taken into account by the proposed approaches. Human habits, the way of performing activities and preferences generally change over time. In the vast majority of cases, proposed solutions do not provide an explicit method to update models, thus, requiring modeling to start from scratch, with the drawbacks identified in the first step. As a third point, support for multiple users performing actions separately and multiple users collaborating in a single activity are often neglected by proposed approaches. Supporting multiple users poses serious limitations to the applicability of available solutions. This problem is even harder to solve when human guests must be taken into account. References 1. Ye, J.; Dobson, S.; McKeever, S. Situation identification techniques in pervasive computing: A review. Pervasive Mob. Comput. 2012, 8, 36–66. [CrossRef] 4. Mennicken, S.; Huang, E. Hacking the Natural Habitat: An In-the-Wild Study of Smart Homes, Their Development, and the People Who Live in Them. In Pervasive Computing; Lecture Notes in Computer Science; Kay, J., Lukowicz, P., Tokuda, H., Olivier, P., Krüger, A., Eds.; Springer: Berlin/Heidelberg, Germany, 2012; Volume 7319, pp. 143–160. 21 of 23 21 of 23 Future Internet 2019, 11, 23 5. Tazari, M.R.; Furfari, F.; Fides-Valero, Á.; Hanke, S.; Höftberger, O.; Kehagias, D.; Mosmondor, M.; Wichert, R.; Wolf, P. The universAAL Reference Model for AAL. In Handbook of Ambient Assisted Living; IOS Press Ebooks: Amsterdam, The Netherlands, 2012; Volume 11, pp. 610–625. 6. Cook, D.; Augusto, J.; Jakkula, V. Ambient intelligence: Technologies, applications, and oppor Pervasive Mob. Comput. 2009, 5, 277–298. [CrossRef] 7. Aztiria, A.; Izaguirre, A.; Augusto, J.C. Learning patterns in ambient intelligence environments: A survey. Artif. Intell. Rev. 2010, 34, 35–51. [CrossRef] 8. Kitchenham, B. Procedures for Performing Systematic Reviews; Keele University: Keele, UK, 2004; Volume 33, pp. 1–26. 9. Augusto, J.C.; Liu, J.; McCullagh, P.; Wang, H.; Yang, J.B. Management of uncertainty and spatio-temporal aspects for monitoring and diagnosis in a smart home. Int. J. Comput. Intell. Syst. 2008, 1, 361–378. [CrossRef] aspects for monitoring and diagnosis in a smart home. Int. J. Comput. Intell. Syst. 2008, 1, 361–378. [CrossRef] 10. Aztiria, A.; Augusto, J.C.; Basagoiti, R.; Izaguirre, A.; Cook, D.J. Discovering frequent user–environment interactions in intelligent environments. Pers. Ubiquitous Comput. 2012, 16, 91–103. [CrossRef] 10. Aztiria, A.; Augusto, J.C.; Basagoiti, R.; Izaguirre, A.; Cook, D.J. Discovering frequent user–environment interactions in intelligent environments. Pers. Ubiquitous Comput. 2012, 16, 91–103. [CrossRef] 11. Cook, D.J.; Krishnan, N.C.; Rashidi, P. Activity discovery and activity recognition: A new partnership. IEEE Trans. Cybern. 2013, 43, 820–828. [CrossRef] [PubMed] 12. Rashidi, P.; Cook, D.J.; Holder, L.B.; Schmitter-Edgecombe, M. Discovering activities to recognize and track in a smart environment. IEEE Trans. Knowl. Data Eng. 2011, 23, 527–539. [CrossRef] 13. Rashidi, P.; Cook, D.J. COM: A method for mining and monitoring human activity patterns in home-based health monitoring systems. ACM Trans. Intell. Syst. Technol. 2013, 4, 64. [CrossRef] 14. Singla, G.; Cook, D.J.; Schmitter-Edgecombe, M. Recognizing independent and joint activities among multiple residents in smart environments. J. Ambient Intell. Human. Comput. 2010, 1, 57–63. [CrossRef] 15. Rashidi, P.; Cook, D.J. References Keeping the resident in the loop: Adapting the smart home to the user. IEEE Trans. Syst. Man Cybern.-Part A Syst. Hum. 2009, 39, 949–959. [CrossRef] 16. Cook, D.J. Learning setting-generalized activity models for smart spaces. IEEE Intell. Syst. 2012, 27, 32–38. [CrossRef] [PubMed] 17. Krishnan, N.C.; Cook, D.J. Activity recognition on streaming sensor data. Pervasive Mob. Comput. 2014, 10, 138–154. [CrossRef] 8. Chen, L.; Nugent, C.D.; Wang, H. A knowledge-driven approach to activity recognition in smart ho IEEE Trans. Knowl. Data Eng. 2012, 24, 961–974. [CrossRef] 9. Gu, T.; Wang, L.; Wu, Z.; Tao, X.; Lu, J. A pattern mining approach to sensor-based human activity recogni IEEE Trans. Knowl. Data Eng. 2011, 23, 1359–1372. [CrossRef] 20. Helaoui, R.; Riboni, D.; Stuckenschmidt, H. A probabilistic ontological framework for the recognition of multilevel human activities. In Proceedings of the 2013 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Zurich, Switzerland, 8–12 September 2013; ACM: New York, NY, USA, 2013; pp. 345–354. 21. Riboni, D.; Sztyler, T.; Civitarese, G.; Stuckenschmidt, H. Unsupervised recognition of interleaved activities of daily living through ontological and probabilistic reasoning. In Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Heidelberg, Germany, 12–16 September 2016; ACM: New York, NY, USA, 2016; pp. 1–12. 22. Hong, X.; Nugent, C.; Mulvenna, M.; McClean, S.; Scotney, B.; Devlin, S. Evidential fusion of sensor data for activity recognition in smart homes. Pervasive Mob. Comput. 2009, 5, 236–252. [CrossRef] 23. Wang, L.; Gu, T.; Tao, X.; Lu, J. A hierarchical approach to real-time activity recognition in body sensor networks. Pervasive Mob. Comput. 2012, 8, 115–130. [CrossRef] 24. Van Kasteren, T.; Noulas, A.; Englebienne, G.; Kröse, B. Accurate activity recognition in a home setting. In Proceedings of the 10th International Conference on Ubiquitous computing, Seoul, Korea, 21–24 September 2008; ACM: New York, NY, USA, 2008; pp. 1–9. 25. Brdiczka, O.; Crowley, J.L.; Reignier, P. Learning situation models in a smart home. IEEE Trans. Syst. Man Cybern. Part B Cybern. 2009, 39, 56–63. [CrossRef] 26. Yang, J.Y.; Wang, J.S.; Chen, Y.P. Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers. Pattern Recognit. Lett. 2008, 29, 2213–2220. [CrossRef] 26. Yang, J.Y.; Wang, J.S.; Chen, Y.P. Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers. Pattern Recognit. Lett. 2008, 29, 2213–2220. [CrossRef] 27. Stikic, M.; Larlus, D.; Ebert, S.; Schiele, B. References Weakly supervised recognition of daily life activities with wearable 26. Yang, J.Y.; Wang, J.S.; Chen, Y.P. Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers. Pattern Recognit. Lett. 2008, 29, 2213–2220. [CrossRef] 27. Stikic, M.; Larlus, D.; Ebert, S.; Schiele, B. Weakly supervised recognition of daily life activities with wearable sensors. IEEE Trans. Pattern Anal. Mach. Intell. 2011, 33, 2521–2537. [CrossRef] 27. Stikic, M.; Larlus, D.; Ebert, S.; Schiele, B. Weakly supervised recognition of daily life activities with wearable sensors. IEEE Trans. Pattern Anal. Mach. Intell. 2011, 33, 2521–2537. [CrossRef] 22 of 23 Future Internet 2019, 11, 23 28. Fleury, A.; Vacher, M.; Noury, N. SVM-based multimodal classification of activities of daily living in health smart homes: Sensors, algorithms, and first experimental results. IEEE Trans. Inf. Technol. Biomed. 2010, 14, 274–283. [CrossRef] 9. Palmes, P.; Pung, H.K.; Gu, T.; Xue, W.; Chen, S. Object relevance weight pattern mining for acti recognition and segmentation. Pervasive Mob. Comput. 2010, 6, 43–57. [CrossRef] 30. Kwapisz, J.R.; Weiss, G.M.; Moore, S.A. Activity recognition using cell phone accelerometers. ACM SigKDD Explor. Newsl. 2011, 12, 74–82. [CrossRef] 31. Lester, J.; Choudhury, T.; Borriello, G. A practical approach to recognizing physical activities. In Proceedings of the International Conference on Pervasive Computing, Dublin, Ireland, 7–10 May 2006; Springer: Berlin/Heidelberg, Germany, 2006; pp. 1–16. 32. Buettner, M.; Prasad, R.; Philipose, M.; Wetherall, D. Recognizing daily activities with RFID-based sensors. In Proceedings of the 11th International Conference on Ubiquitous Computing, Orlando, FL, USA, 30 September–3 October 2009; ACM: New York, NY, USA, 2009; pp. 51–60. 33. Aggarwal, C.C. Managing and Mining Sensor Data; Springer Publishing Company, Incorporated: New York, NY, USA, 2013. 34. Liu, H.; Hussain, F.; Tan, C.L.; Dash, M. Discretization: An enabling technique. Data Min. Knowl. Discov. 2002, 6, 393–423. [CrossRef] 35. Krishnan, N.; Cook, D.J.; Wemlinger, Z. Learning a taxonomy of predefined and discovered activity patterns. J. Ambient Intell. Smart Environ. 2013, 5, 621–637. [PubMed] 36. Agrawal, R.; Srikant, R. Fast algorithms for mining association rules. In Proceedings of the 20th International Conference on Very Large Data Bases, VLDB, Santiago de Chile, Chile, 12–15 September 1994; Volume 1215, pp. 487–499. 37. Loke, S.W. Logic programming for context-aware pervasive computing: Language support, characterizing situations, and integration with the web. In Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence, Beijing, China, 20–24 September 2004; pp. 44–50. 38. References Magherini, T.; Fantechi, A.; Nugent, C.D.; Vicario, E. Using temporal logic and model checking in automated recognition of human activities for ambient-assisted living. IEEE Trans. Hum. Mach. Syst. 2013, 43, 509–521. [CrossRef] 39. Ye, J.; Coyle, L.; Dobson, S.; Nixon, P. Ontology-based models in pervasive computing systems. Knowl. Eng. Rev. 2007, 22, 315–347. [CrossRef] 40. Riboni, D.; Bettini, C. Context-aware activity recognition through a combination of ontological and statistical reasoning. In Proceedings of the Ubiquitous Intelligence Computing, Brisbane, Australia, 7–9 July 2009; pp. 39–53. pp 41. Azkune, G.; Almeida, A.; López-de Ipiña, D.; Chen, L. Extending knowledge-driven activity models through data-driven learning techniques. Expert Syst. Appl. 2015, 42, 3115–3128. [CrossRef] 42. Russell, S.J.; Norvig, P. Artificial Intelligence: A Modern Approach, 3rd ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2010. 43. Augusto, J.C.; Nugent, C.D. The use of temporal reasoning and management of complex events in smart homes. In Proceedings of the 16th European Conference on Artificial Intelligence, Valencia, Spain, 22–27 August 2004; Volume 16, p. 778. 44. Dumas, M.; La Rosa, M.; Mendling, J.; Reijers, H.A. Fundamentals of Business Process Management; S Berlin/Heidelberg, Germany, 2013; Volume 1. 45. Gottfried, B.; Guesgen, H.; Hübner, S. Spatiotemporal Reasoning for Smart Homes. In Designing Smart Homes; Lecture Notes in Computer Science; Augusto, J., Nugent, C., Eds.; Springer: Berlin/Heidelberg, Germany, 2006; Volume 4008, pp. 16–34. 46. Aztiria, A.; Izaguirre, A.; Basagoiti, R.; Augusto, J.C.; Cook, D.J. Discovering Frequent Sets of Actions in Intelligent Environments. In Intelligent Environments; IOS Press: Barcelona, Spain, 2009; pp. 153–160. 47. Aztiria, A.; Izaguirre, A.; Basagoiti, R.; Augusto, J.C.; Cook, D. Automatic modeling of frequent user behaviours in intelligent environments. In Proceedings of the 2010 Sixth International Conference on Intelligent Environments (IE), Kuala Lumpur, Malaysia, 19–21 July 2010; IEEE: Washington, DC, USA, 2010, pp. 7–12. 48. William, C.; others. Fast effective rule induction. In Proceedings of the Twelfth International Conference on Machine Learning, Tahoe City, CA, USA, 9–12 July 1995; pp. 115–123. 23 of 23 Future Internet 2019, 11, 23 49. Bettini, C.; Brdiczka, O.; Henricksen, K.; Indulska, J.; Nicklas, D.; Ranganathan, A.; Riboni, D. A survey of context modelling and reasoning techniques. Pervasive Mob. Comput. 2010, 6, 161–180. [CrossRef] 50. Sadri, F. Ambient Intelligence: A Survey. ACM Comput. Surv. 2011, 43, 36. [CrossRef] 51. Chen, L.; Hoey, J.; Nugent, C.D.; Cook, D.J.; Yu, Z. Sensor-based activity recognition. IEEE Trans. Sy Cybern. Part C Appl. Rev. 2012, 42, 790–808. c⃝2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References [CrossRef] 52. Acampora, G.; Cook, D.J.; Rashidi, P.; Vasilakos, A.V. A Survey on Ambient Intelligence in Healthcare. Proc. IEEE 2013, 101, 2470–2494. [CrossRef] [PubMed] 53. Augusto, J.C.; Callaghan, V.; Cook, D.; Kameas, A.; Satoh, I. Intelligent environments: A manifesto. Hum. Cent. Comput. Inf. Sci. 2013, 3, 12. [CrossRef] 54. Amiribesheli, M.; Benmansour, A.; Bouchachia, A. A review of smart homes in healthcare. J. Ambient Intell. Hum. Comput. 2015, 6, 495–517. [CrossRef] c⃝2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

https://openalex.org/W4281389139

https://www.researchsquare.com/article/rs-1631596/latest.pdf

English

Oral Microbes are a Signature of Disease in the Gut

Research Square (Research Square)

cc-by

Oral Microbes are a Signature of Disease in the Gut Libusha Kelly  (  [email protected] ) Albert Einstein College of Medicine https://orcid.org/0000-0002-7303-1022 Libusha Kelly  (  [email protected] ) Albert Einstein College of Medicine https://orcid.org/0000-0002-7303-1022 Analysis Keywords: oral microbes, dysbiosis, microbiome, healthy microbiome Posted Date: March 20th, 2023 Keywords: oral microbes, dysbiosis, microbiome, healthy microbiome DOI: https://doi.org/10.21203/rs.3.rs-1631596/v2 se:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/15 Abstract Case-control studies have identified human gut microbiome taxa associated with diseases including diabetes, inflammatory bowel, and colorectal cancer. However, many taxa are significantly associated with different diseases, confounding the utility of taxa as disease-specific diagnostics and potentially indicating underlying similarities in physiologically distinct diseases. Using a meta-analysis of 4446 gut metagenomic samples linked to 55 diseases across 21 datasets, we identify taxa consistently associated with health or disease and validate these results on external datasets. We find that taxa associated with disease in the gut are either oral microbiome commensals or are members of the polyphyletic class Clostridiales, whereas healthy taxa are gut specialists. Prevalence of organisms in the oral cavity correlates with gut disease association, revealing a novel, conserved relationship between disease and the gut microbiome. Our work highlights a major limitation of traditional metagenomics analysis methods for identifying disease-specific markers, a key goal of many microbiome studies. 1 Introduction We further observe that taxa associated generally with health tend to be gut specialist of the polyphyletic class Clostridiales which are both general disease markers and prevalent in the gut. By performing a novel meta-analysis on thousands of whole community meta genomes generated by deep sequencing, our study provides novel biological insights into the link between human physiology in health and disease in the gut microbiome. In doing so, we also develop a new analysis framework to disentangle general signatures of disease and health from specific signatures of a disease. 1 Introduction As the cost of genomic sequencing has steadily dropped over the past decades, the scope of sequencing efforts has increased drastically 1–3. Despite terabytes of sequence data on the human microbiome, the community of microorganisms, and their ecosystems in the human body, there remains a gap in understanding how the microbiome, particularly the gut microbiome, interacts with our host physiology 4. Deeper investigation into these links has revealed specific mechanisms by which microorganisms interact with our immune systems and intestinal epithelia 5, 6. There is substantial interest in using the microbiome as a diagnostic tool by identifying taxonomic and functional signatures which correspond to different phenotypes 7. Microbiome disease-signature identification is difficult because of the high-dimensionality of the system and the large number of exogenous variables which can affect a case-control study 8. Batch effects, such as demographic and geographic differences between study populations, are thought to account for a significant portion of the variation in the microbiome and hence the differences in findings between similar studies 9, 10. To address these issues, researchers have begun aggregating studies together to extrapolate common signals for a given phenotype. For example, Thomas et al. and Wirbel et al. performed different meta-analyses on a collection of colorectal cancer microbiome studies and found a similar set of associated bacteria 11, 12. However, many of the taxa associated with specific diseases in these studies were also identified as disease associated in studies of other diseases. For example, the oral commensal Fusobacterium nucleatum has been associated with colorectal cancer 13, periodontitis 14, inflammatory bowel disease 15, and additional conditions 16. Here, we hypothesize that many of the organisms identified as significant in case-control microbiome studies are actually general markers of health and disease. To test this hypothesis, we performed a meta-analysis by aggregating 4446 gut metagenomic samples linked to 55 diseases across 21 case-control studies datasets and ask how specific each taxon is for a given disease (Figure 1). We identify conserved, significant associations between individual taxa across different diseases and conversely identify taxa that are reproducibly associated with health in our case-control studies. We call these taxa ‘general’ biomarkers of health and disease. We validate our health-associated taxa with an external set of studies of healthy microbiomes across the world. Through further investigation of these general disease-associated taxa, we discover a novel, conserved association between oral microbiome taxa found in the gut and disease. A Core Set of Microorganisms are Biomarkers of Health or Disease Across a Variety of Settings. studies which included healthy controls and patients from studies of multiple diseases, including colorectal cancer (CRC), type 2 diabetes (T2D), type 1 diabetes (T1D), metabolic syndromes (IGT), inflammatory bowel disease (IBD), Behcet’s disease (BD), hypertension (HD), liver cirrhosis, and others (Table 1). We hypothesized that many of the differentially abundant microorganisms within these studies are general markers of disease in general, or hallmarks of ’dysbiosis’. To test this hypothesis, we computed and averaged association scores in th Whitney AUC statistics for each taxon in each study. We then generated a core set of taxa whose average association across all studies was statistically significant at a threshold of p<0.01 (Figure 2) and an expanded set of markers at p<0.05 (Figure S1). We identified 261 taxa at a p<.05 (S1) and 41 taxa p < .01 (Figure 2a). Many health associated bacteria included common commensals such as Lachnospiraceae 17 and Ruminococcaceae families 18, 19, while many known pathogen-associated organisms such as Fusobacterium nucleatum 16, Streptococaceae 20–22family, and Solobacterium moorei 23were identified as consistently disease-associated. In particular, many of these amily, and Solobacterium moorei 23were identified as consistently disease-associated. In particular, many of thes Page 2/15 Page 2/15 taxa align with other studies. For example, out of 29 taxa identified as colorectal cancer-associated by Wirbel et al., 19 directly matched with disease- associated markers in our extended setTable 2 12. To demonstrate the consistency of these markers across studies, we performed a two-sided t-test comparing control and experimental groups for each taxon in each study and found that our most significant health markers were health-associated in almost every study independently (and similarly for disease markers) (Figure 2b, Figure S1b). This congruence indicates that our results were not biased by a handful of outlier studies but rather that significant taxa in each of these studies overlap despite involving completely different patient cohorts. Next, we calculated the percent of samples that contained each taxon. We found that healthy markers tend to be present in most samples while some disease markers, such as Anaerostipes 24, are rarely carried and others, such as Clostridium hathewayi 25, are found in most samples (Figure 2d, Figure S1d). Finally, we plotted the relative abundance of the general health and disease markers in microbiomes carrying these taxa. Here, we observe a wide distribution in relative abundance across most taxa (Figure 2d, Figure S1d). A Core Set of Microorganisms are Biomarkers of Health or Disease Across a Variety of Settings. Predominance of Health-Associated Taxa in Microbiomes We next asked if the observation that health-associated markers are more abundant in gut metagenomes generalizes to other studies. We analyzed 15 additional datasets containing only healthy-labeled samples from Western populations and found that, in all cases, healthy taxa have a higher mean relative abundance than disease taxa (Figure 3a). This observation extends to non- Western populations. We analyzed six datasets from geographies such as Tanzania, Fiji Islands, and mountainous village regions in Latin America, and all of them followed the same trend with the exception of Smith et al. 26– 31 (Figure 3b). We also observed that health taxa are more abundant than disease markers even in datasets for which we only had unhealthy patient samples. Taken together, our results suggest that healthy taxa are more predominant in general, regardless of underlying condition 32–35 (Figure 3 c). Oral Microbes are Predominantly Associated With Dysbiosis We next assessed potential biological or functional similarities in our disease-associated taxa. From a review of the literature, we observed that many of our disease-associated taxa are commonly found in oral microbiomes 36. To investigate this relationship in our data, we first quantified the mean abundance of each organism in the oral cavity of healthy individuals using 206 oral microbiome samples from three studies 26, 37, 38. We found that, on average, disease taxa are more abundant in the oral cavity than healthy taxa, whereas healthy taxa are more abundant in the gut (Figure 4 a). No such pattern emerged when performing the same analysis on skin and nasal body sites 39–42. When we compare disease-associated markers with health-associated markers in gut studies, we find that health-associated markers have a higher mean relative abundance in the gut than disease- hen we compare disease-associated markers with health-associated markers in gut studies, we find that health-as ative abundance in the gut than disease- associated markers (Figure 4b). We found a tendency for more abundant gut taxa to be associated with health regardless of whether the mean AUC was statist p=1.26e-8). Most health-associated taxa are close to 0.5 in mean AUC until approximately associated markers (Figure 4b). We found a tendency for more abundant gut taxa to be associated with health regardless of whether the mean AUC was statist p=1.26e-8). Most health-associated taxa are close to 0.5 in mean AUC until approximately p=1.26e-8). A Core Set of Microorganisms are Biomarkers of Health or Disease Across a Variety of Settings. Most health-associated taxa are close to 0.5 in mean AUC until approximately 10^-1 mean abundance, at which point a fanning out occurs, where organisms above this threshold tend to be e associated, although the trend leans towards higher-abundance organisms being more health-associated. (Figu 10^-1 mean abundance, at which point a fanning out occurs, where organisms above this threshold tend to be either strongly health- or disease- associated, although the trend leans towards higher-abundance organisms being more health-associated. (Figure 4c) 10^-1 mean abundance, at which point a fanning out occurs, where organisms above this threshold tend to be either strongly health- or disease- associated, although the trend leans towards higher-abundance organisms being more health-associated. (Figure 4c) By contrast, when we compare the mean relative abundance of bacteria in the oral cavity with their disease vs. health association, we find that the more abundant bacteria are in the oral cavity, the more likely they are to be disease-associated (Figure 4c). We found that, of the 160 taxa that are found both in the oral cavity and are in the core gut group identified by our meta-analysis, 134 are associated with disease and only 26 with health. Furthermore, when we plotted the mean AUC score in gut studies vs. the log mean abundance of each oral taxon, we found that taxa with higher abundances in the oral cavity tend to be more disease-associated (regression slope = -50.4, p=6.84e-22). This relationship also held true if we extended to all oral taxa; higher mean abundance was associated with lower mean AUC, even if that AUC score is not statistically significant. Almost All Disease Markers are Oral Taxa or Class Clostridia When analyzing taxonomic abundance data, we noticed that many taxa were completely absent from most samples, but when they were present, they would follow a consistent distribution (Figure 2c). Hence, the mean relative abundance of a taxa may be biased downward for taxa that are infrequently carried. To Page 3/15 better understand this phenomenon, we visualized our data based not on mean relative abundance but on frequency of carriage. We compared the percentage of samples those different markers are found in across different body sites and discovered a striking pattern: The majority of disease taxa are rarely found in the gut, but rather are common oral commensals (Figure 5a, Figure S2). 3 Discussion Our work demonstrates that microbiome biomarkers identified in case-control studies may not be purely specific to the phenotypes under investigation, which changes the interpretation of these studies’ findings. Our results thus raise an important question for microbiome clinical research: is a taxon identified in a case-control study associated with a specific disease or with general stress pathways that are common across many diseases? This question adds to the list of challenges associated with such studies which includes batch effects 43, sequence annotation errors 44, and sampling strategy 45. t diagnostic tools, developing interventional therapies, Specificity is key to designing robust diagnostic tools, developing interventional therapies, nd understanding mechanistic links between microbes and human pathophysiology. For example, an assay desig etween microbes and human pathophysiology. For example, an assay designed to diagnose CRC using bacterial high sensitivity and is specific enough to not raise false- adouts can only be effective if it has high sensitivity and is specific enough to not raise false- positives when performed on samples from patients with a different disease, such as IBD 46, 47. Ideally, such an assay should be compared against patients with conditions that could have similar signatures to IBD to truly validate its effectiveness48. Historically, microbiome case-control studies are set up to compare a healthy control group against an experimental cohort with some phenotype of interest. Here we show that without also comparing the experimental cohort against ‘similar’ groups (eg. other disease groups), one cannot ascertain specificity. This limitation curtails the effectiveness of follow-up analyses which attempt to elucidate the mechanisms of specific genes, pathways, or taxa identified by a case- control study in the disease phenotype. Our approach identifies general microbiome/disease associations that are robust across multiple disease studies rather than specific to the pathophysiology of an individual disease. From prior studies, mechanisms of microbiome/disease interactions could include short chain fatty acid metabolism, regulatory T cell activity, cytokines interactions, and overall response to bacterial antigens 49. Hence, there is an urgent, unmet need to identify and differentiate general- vs specific- taxa, genes, and pathways associated with disease 50. Our findings also identified a striking pattern in that almost all disease markers are either common oral taxa or belong to the class Clostridia. This could suggest that, in dysbiosis, bacteria which do not normally inhabit the gut may be able to migrate in from the oral cavity 13, 51. A Core Set of Microorganisms are Biomarkers of Health or Disease Across a Variety of Settings. Furthermore, we found that all non-oral commensal disease markers were either from the class Clostridia or belonged to a higher taxonomic order such as genus Peptostreptococcus (Figure 5 b-d). genus Peptostreptococcus (Figure 5 b-d). We next grouped our taxa based on their frequency of carriage and disease score. As Figure 5 c shows, there are no core taxa that are infrequently (<20% of samples) carried in the gut and also health associated. These observations hold true in the extended set of health and disease markers with the exception of two broad categories (family Bacillaceae and Genus Siphoviridae, a phage taxa) and Clostridium sp-L2-50 (Figure S3). Additionally, there are very few core markers commonly carried in the gut and disease associated. Taxa that fit into this category in the core belong to Clostridia, and in the extended group are mostly higher order categories (for example Class Gammaproteobacteria and Order Enterobacteriales). By contrast, there are no core health associated taxa commonly found in oral microbiomes (Figure 5d). This also applies to the extended set again with the exception of a few higher order category taxa (Figure S4). The remaining disease-associated taxa are infrequently found in any body site (Figure 5 b). These are relatively rare organisms like Holdemania sp AP2 (2.37% carriage) and Atopobium sp ICM58 (.08% carriage), while the rarest health-associated taxa is Bacteroides intestinalis (27.08% carriage) (Figure 2 d). There are a few exceptions to this pattern in the extended set, such as the disease marker Lachnospiraceae bacterium 3-1-57FAA-CT1 (29.68% carriage) (Figures 6 and 10). All other core healthy markers are carried by more than half of stool samples and only Clostridia disease markers are ever found at that frequency in the gut. These findings serve as a validation, reinforcing the consistency of the identified taxa as truly general markers of health and disease. 3 Discussion Every step in a study has various risks, from the sample collection protocols to the nucleotide extraction protocols, to the sequencing technologies, primer kits, and base-calling software, to the tools used to annotate metagenomes, and even the statistical analyses performed at the end. Work from other labs has highlighted risks tools used to annotate metagenomes, and even the statistical analyses performed at the end. Work from o due to experimental batch effects and possible solutions for controlling for these issues 9, 10, but to our knowledge, there has been very little discourse about controlling for the specificity of results 66. Our work in this paper highlights both the dangers of ignoring this specificity and the observation that many diseases linked to the microbiome display common signatures. There is a wealth of data now available that can be used to indirectly control for these issues; we expect that future case-control studies will not only take steps to control for experimental noise, but also to compare their results against other (unrelated) studies to quantify which findings are specific and which are nonspecific to their investigation. 3 Discussion Indeed, there is evidence that oral infections such as gingivitis can influence gut disorders such as inflammatory bowel disease 52. One may ask why this does not happen normally; bacteria entering the stomach should usually be destroyed during digestion due to the low pH 53. However, there is evidence that there is always some level of transmission of bacteria between oral to gut environments 54. A disease process could increase this transmission rate, for example by increasing transit time of food and making it easier for oral bacteria to enter the gut 55, 56. We note that the reproducibility of our findings across studies suggests that these taxa are not simply transiting, in which case we would not expect to see the same oral taxa repeatedly, but that they are also taking up residence in the gut. Hence, increased intestinal stress may create an environment that is more conducive to invasion by oral taxa 57. In the ecology literature, there is also evidence that ecosystem invasibility can increase in situations where turnover rises or the overall system becomes less stable 58. Regardless, such an interpretation implies that these markers are not causal but rather are a consequence of intestinal stress. The Clostridia disease markers are an exception to the oral-gut axis observation. Bacteria from class Clostridia are rarely found in the oral cavity and are in some cases gut commensals. These microbes are difficult to classify; the class is polyphyletic and many taxa have been renamed since this dataset was first downloaded in March 2021 59 . However, many Clostridia are spore- forming and, building on the invasibility hypothesis, environmental stress may make it easier for their spores to grow 60–62. Hence, sporulation and altered transit time may represent two parallel routes of entry for microbes into a perturbed gut environment that is more vulnerable to invasion. Page 4/15 Page 4/15 Our study is limited by the number of datasets available for each disease. Hence, we focused on identifying and understanding the general markers. We also note that our observed effect sizes for individual microbes are very small. This, along with the acknowledgement that our results may only be associative rather than causal, suggests that individual markers are poor candidates for probiotics meant to foster a healthy microbiome or for diagnostic assays 63, 64. 4 Methods All figures were generated originally using scripts written in Python and assembled using Adobe Illustrator. 3 Discussion Instead, the combination of many markers should be utilized in order to account for diversity between individuals functional data, in combination with taxonomic data, will also assist in separating true signal from noise. s interest in the microbiome as a research target grows, so too have concerns As interest in the microbiome as a research target grows, so too have concerns emerged about the reproducibility and interpretability of study findings 65. Every step in a study has various risks, from the sample collection protocols to the nucleotide extraction protocols, to the sequencing technologies, primer kits, and base-calling software, to the tools used to annotate metagenomes, and even the statistical analyses performed at the end. Work from other labs has highlighted risks due to experimental batch effects and possible solutions for controlling for these issues 9, 10, but to our knowledge, there has been very little discourse about controlling for the specificity of results 66. Our work in this paper highlights both the dangers of ignoring this specificity and the observation that many diseases linked to the microbiome display common signatures. There is a wealth of data now available that can be used to indirectly control for these issues; we expect that future case-control studies will not only take steps to control for experimental noise, but also to compare their results against other (unrelated) studies to quantify which findings are specific and which are nonspecific to their investigation. emerged about the reproducibility and interpretability of study findings 65. Every step in a study has various risks, from the sample collection protocols to the nucleotide extraction protocols, to the sequencing technologies, primer kits, and base-calling software, to the tools used to annotate metagenomes, and even the statistical analyses performed at the end. Work from other labs has highlighted risks due to experimental batch effects and possible solutions for controlling for these issues 9, 10, but to our knowledge, there has been very little discourse about controlling for the specificity of results 66. Our work in this paper highlights both the dangers of ignoring this specificity and the observation that many diseases linked to the microbiome display common signatures. There is a wealth of data now available merged about the reproducibility and interpretability of study findings 65. Data Curation The data sets for this study were sourced from ‘CuratedMetagenomicData‘ (downloaded Mar 2021) 59, an R package providing uniform metagenomic annotations for all studies included via ‘MetaPhlAn2‘ 67 and ‘HUManN2‘ 68 along with standardized metadata. To prepare this data for analysis, we wrote two Python packages (available on PyPi). The first, ‘curated_metagenomic_data_extractor‘, downloads all availabl annotations and metadata. The metadata is re-uploaded into a (PostgreSQL) database, and the annotations files are uploaded to a Google Cloud Storage bucket. The purpose of this is to make it easier to query samples and studies based on specific criteria (eg. get all samples for a body site, with a specific disease label, etc.). The second package, called ‘curated_metagenomic_data_loader‘, creates datasets for statistical analysis. Here, we write SQL queries to select samples based on their metadata, and the package downloads the annotations for those samples from the storage bucket and consolidates all of this information into a TensorFrame, which is a data structure we created for easily working with this type of data 69 . Dataset Selection Dataset Selection We chose datasets which were available in ‘curatedMetagenomicData’ at the time of analysis g y which contained stool microbiome samples and involved a healthy control group and an unhealthy (labeled disease phenotype) group. These are described in Table 1. Not all studies were classic case-control studies; for example, Vatanen et al. was a longitudinal study following newborn infants across different Eastern European countries 70 . However, every study used had samples labeled with a disease phenotype or as control/healthy. Meta-Analysis Technique Meta-Analysis Technique At a high level, we identified putative health and disease markers by assigning an association At a high level, we identified putative health and disease markers by assigning an association score to each taxon for each dataset. We then looked for taxa that had consistently high (healthy) or low (disease) association scores across all studies. At a high level, we identified putative health and disease markers by assigning an association score to each taxon for each dataset. We then looked for taxa that had consistently high (health At a high level, we identified putative health and disease markers by assigning an association score to each taxon for each dataset. We then looked for taxa that had consistently high (healthy) or low (d on for each dataset. Meta-Analysis Technique Data Curation We then looked for taxa that had consistently high (healthy) or low (disease) association sco e then looked for taxa that had consistently high (healthy) or low (disease) association scores across all studies. score to each taxon for each dataset. We then looked for taxa that had consistently high (healthy) or low (disea For each study and taxa, we performed a Mann-Whitney U test comparing the distribution of relative abundances of that taxa between experimental and control groups 71 . This is a non-parametric test of the null-hypothesis that a random sample chosen from the experimental group is no likely to be greater or lesser than a random sample chosen from the control group. The U- statistic from this test can be converted into an estimate of the probability that a random sample from the control group is greater than a random sample from the experimental group, and this   is called the AUC (area under the curve) 72. The reason for this nomenclature is that in receiver operator characteristic (ROC) analysis for binary classification, the area under a receiver-operator characteristic curve Page 5/15 Page 5/15 (commonly denoted as AUC) is equivalent to the probability that a classifier assigns a greater score to a positive label than a negative label 73. Hence, the U- statistic AUC can be interpreted as a binary classification: if the AUC score is greater than 0.5, the taxa is associated with healthy, and if the score is less than 0.5, it is associated with disease, and 0.5 means no association. This association is quantitative. To assign a final score to a taxon, we computed its mean AUC-score across all studies. We then performed a (two sided) T-test of the null hypothesis that this mean score is 0.5. We used a cutoff of p < 0.01 to generate the core set of taxa that are shown in Figure 1, and a cutoff of p < 0.05 to generate an expanded set of markers. 0.05 to generate an expanded set of markers. Deduplication Deduplication For the purpose of simplifying visualization, we consolidated taxa that were directly related and had the same association scores. This occurs when the taxonomic annotations for a sample give the same relative abundance to a taxon and its parent. For example, s   Clostridium_symbiosum  and s Clostridium_symbiosum|t Clostridium_symbiosum_unclassified had the same score (0.418). Data Curation In such instances, we show only the parent taxon in figures that mention marker names. This was done for Figure 2, Figure 5, Figure S1, and all supplemental figures. Figure 3 and 4 utilized the expanded marker list with no deduplication because including duplicate markers does not change the visualization or interpretation of the figure as duplicate markers sit on top of each other. Comparison to Other Meta-Analyses Our approach can be thought of as a more permissive form of the method used by LefSe algorithm to detect significant markers across studies 74. Briefly, LefSe considers a marker to be significant by requiring that the marker be statistically significantly different between each combination of experimental and control groups across each study being analyzed. LefSe thus performs a Mann-Whitney U test for each pair of experiment / control groups and demands that every one of them be significant. While this approach is effective for small numbers of datasets, the number of tests performed this way (n*(n-1)/2) results in a very strict criterion when considering many studies. To loosen this requirement, we consider those features that are statistically significant in most of the studies considered and on average. To quantify features that are statistically significant in most studies, we compute a Mann- Whitney U test for each study and convert the resulting statistic to an AUC score, where a value greater than 0.5 indicates a positive association, less than 0.5 indicates a negative association, and the closer to 0.5, the weaker the association. To accept a feature, we then require that the mean AUC across all studies for that feature be significantly different from 0.5 (using a two-sided T-test, p <0.01 was used to define ‘core’ healthy and unhealthy marker taxa and p <0.05 was used to define an expanded set of marker taxa). Our approach thus essentially requires that a feature be consistently (but not always) associated with the same side of a study (either positively enriched for the healthy cohort or for the experimental group). While less strict, this method is more flexible, and reflective of the noise in microbiome studies, enabling us to incorporate more studies and account for natural variation that may cause a feature to not always be differentially abundant between study groups. Data Curation Body Site Comparisons Body Site Comparisons To generate the results in Figures 3a, 3b, and 4, we estimated how often each taxon was present in different body sites. To do this, we utilized all of the available datasets in curatedMetagenomicData. For each body site, we took all available samples from that body site and computed the mean relative abundance across all such samples to estimate the mean frequency of a taxon in a site. Organisms with mean relative abundance greater than zero were included. This approach enabled us to statistically characterize oral-associated taxa in our dataset. Funding SK was supported by a National Institutes of Health (NIH) training grant 2T32GM007288-45 (Medical Scientist Training Program) at Albert Einstein College of Medicine and an NIH T32 fellowship on Geographic Medicine and Emerging Infectious Diseases (2T32AI070117-13). LK is supported in part by NIH R01HL069438, a Department of Defense Peer Reviewed Cancer Research Career Development Award (CA171019), the Elsa U Pardee Foundation, and an Irma T. Hirschl Career Scientist award. Computing support for the project was funded by the Google Cloud Research Credits program (GCP19980904) and an Nvidia GPU research grant. Page 6/15 Authors’ contributions SK and LK conceived of the study. SK performed all the of the data processing, statistical analysis, and figure generation. SK and LK co-wrote the text of the paper. Competing interests The authors have no competing interests to disclose. References 1. Gu, W., Miller, S. & Chiu, C. Y. Clinical Metagenomic Next-Generation Sequencing for Pathogen Detection. Annu Rev Pathol 14, 319–338 (2019). 1. Gu, W., Miller, S. & Chiu, C. Y. Clinical Metagenomic Next-Generation Sequencing for Pathogen Detection. Annu Rev Pathol 14, 319–338 (2019). 2. Malla, M. A. et al. Exploring the Human Microbiome: The Potential Future Role of NextGeneration Sequencing in Disease Diagnosis and Treatment. Front Immunol 9, 2868 (2018). 2. Malla, M. A. et al. Exploring the Human Microbiome: The Potential Future Role of NextGeneration Sequencing in Disease Diagnosis and Treatment. Front Immunol 9, 2868 (2018). 3. Levy, S. E. & Boone, B. E. Next-Generation Sequencing Strategies. Cold Spring Harb Perspect Med 9 (2019). 3. Levy, S. E. & Boone, B. E. Next-Generation Sequencing Strategies. Cold Spring Harb Perspect Med 9 (2019). 4. Hatzenpichler, R., Krukenberg, V., Spietz, R. L. & Jay, Z. J. Next-generation physiology approaches to study microbiome function at single cell level. Nat Rev Microbiol 18, 241–256 (2020). 4. Hatzenpichler, R., Krukenberg, V., Spietz, R. L. & Jay, Z. J. Next-generation physiology approaches to study microbiome function at single cell level. Nat Rev Microbiol 18, 241–256 (2020). 4. Hatzenpichler, R., Krukenberg, V., Spietz, R. L. & Jay, Z. J. Next-generation physiology approaches to study microbiome function at single cell level. Nat Rev Microbiol 18, 241–256 (2020). 5. Lambring, C. B. et al. Impact of the Microbiome on the Immune System. Crit Rev Immunol 39, 313–328 (2019). 39, 313–328 (2019). 6. Shi, N., Li, N., Duan, X. & Niu, H. Interaction between the gut microbiome and mucosal immune system. Mil Med Res 4, 14 (2017). 7. Schlaberg, R. Microbiome Diagnostics. Clin Chem 66, 68–76 (2020). 8. Tsilimigras, M. C. & Fodor, A. A. Compositional data analysis of the microbiome: fundamentals, tools, and challenges. Ann Epidemiol 26, 330–5 (2016). Wang, Y. & LêCao, K. A. Managing batch effects in microbiome data. Brief Bioinform 21, 1954–1970 (2020 10. Gibbons, S. M., Duvallet, C. & Alm, E. J. Correcting for batch effects in case-control microbiome studies. PLoS Comput Biol 14, e1006102 (2018). 10. Gibbons, S. M., Duvallet, C. & Alm, E. J. Correcting for batch effects in case-control microbiome studies. PLoS Comput Biol 14, e1006102 (2018). 11. Thomas, A. M. et al. Metagenomic analysis of colorectal cancer datasets identifies crosscohort microbial diagnostic signatures and a link with choline degradation. Nat Med 25, 667–678 (2019). 11. Thomas, A. M. References (2021). 19. Gu,  X. et al. Gut Ruminococcaceae levels at baseline correlate with risk of antibiotic- associated dia 20. Yu, D. et al. Fecal Streptococcus Alteration Is Associated with Gastric Cancer Occurrence and Liver Metastasis. mBio 12, e02994–21 (2021). URL 10.1128/mBio.02994-21. 20. Yu, D. et al. Fecal Streptococcus Alteration Is Associated with Gastric Cancer Occurrence and Liver Metastasis. mBio 12, e02994–21 (2021). URL 10.1128/mBio.02994-21. 21. Vitetta, L., Llewellyn, H. & Oldfield, D. (2019). 21. Vitetta, L., Llewellyn, H. & Oldfield, D. (2019). 21. Vitetta, L., Llewellyn, H. & Oldfield, D. (2019). 22. Garcia-Mantrana, I., Selma-Royo, M., Alcantara, C. & Collado, M. C. (2018). 23. Barrak, I., Stájer, A., Gajdács, M. & Urbán, E. Small, but smelly: the importance of Solobacterium moorei in halitosis and other human infections. Heliyon 6 (2020). 23. Barrak, I., Stájer, A., Gajdács, M. & Urbán, E. Small, but smelly: the importance of Solobacterium moorei in halitosis and other human infections. Heliyon 6 (2020). 24. Identifying Gut Microbiota Associated With Colorectal Cancer Using a Zero-Inflated Lognormal Model. Frontiers in Microbiology 10 (2019). URL 10.3389/fmicb.2019.00826. 24. Identifying Gut Microbiota Associated With Colorectal Cancer Using a Zero-Inflated Lognormal Model. Frontiers in Microbiology 10 (2019). URL 10.3389/fmicb.2019.00826. 25. Hernández-Juárez, L. E. et al. Analyses of publicly available Hungatella hathewayi genomes revealed genetic distances indicating they belong to more than one species. Virulence 12 (1950). 25. Hernández-Juárez, L. E. et al. Analyses of publicly available Hungatella hathewayi genomes revealed genetic distances indicating they belong to mor than one species. Virulence 12 (1950). 26. Brito, I. L. et al. Mobile genes in the human microbiome are structured from global to individual scales. 27. Dhakan, D. B. et al. The unique composition of Indian gut microbiome, gene catalogue, and associated fecal metabolome deciphered using multi- omics approaches. Gigascience 8 (2019). 27. Dhakan, D. B. et al. The unique composition of Indian gut microbiome, gene catalogue, and associated fecal metabolome deciphered using multi- omics approaches. Gigascience 8 (2019). 28. Liu, W. et al. Unique Features of Ethnic Mongolian Gut Microbiome revealed by metagenomic analysis. 29. Pehrsson, E. C. et al. Interconnected microbiomes and resistomes in low-income human habitats. Nature 533, 212–6 (2016). 29. Pehrsson, E. C. et al. Interconnected microbiomes and resistomes in low-income human habitats. Nature 533, 212–6 (2016). 9. Pehrsson, E. C. et al. Interconnected microbiomes and resistomes in low-income human habitats. Nature 53 29. Pehrsson, E. C. et al. References Interconnected microbiomes and resistomes in low-income human habitats. Nature 533, 212–6 (2016). 30. Rampelli, S. et al. Metagenome Sequencing of the Hadza Hunter-Gatherer Gut Microbiota. 30. Rampelli, S. et al. Metagenome Sequencing of the Hadza Hunter-Gatherer Gut Microbiota. 30. Rampelli, S. et al. Metagenome Sequencing of the Hadza Hunter-Gatherer Gut Microbiota. 30. Rampelli, S. et al. Metagenome Sequencing of the Hadza Hunter-Gatherer Gut Microbiota. Curr Biol 25, 1682–93 (2015). 31. Smits, S. A. et al. Seasonal cycling in the gut microbiome of the Hadza hunter-gatherers of Tanzania. Science 357, 802–806 (2017). 32. Wen, C. et al. Quantitative metagenomics reveals unique gut microbiome biomarkers in ankylosing spondylitis. Genome Biol 18, 142 (2017). 32. Wen, C. et al. Quantitative metagenomics reveals unique gut microbiome biomarkers in ankylosing spondylitis. Genome Biol 18, 142 (2017). 32. Wen, C. et al. Quantitative metagenomics reveals unique gut microbiome biomarkers in ankylosing spondylitis. Genome Biol 18, 142 (2017). 33. Gopalakrishnan, V., Helmink, B. A., Spencer, C. N., Reuben, A. & Wargo, J. A. The Influence of the Gut Microbiome on Cancer, Immunity, and Cancer Immunotherapy. Cancer Cell 33, 570–580 (2018). 33. Gopalakrishnan, V., Helmink, B. A., Spencer, C. N., Reuben, A. & Wargo, J. A. The Influence of the Gut Microbiome on Cancer, Immunity, and Cancer Immunotherapy. Cancer Cell 33, 570–580 (2018). 33. Gopalakrishnan, V., Helmink, B. A., Spencer, C. N., Reuben, A. & Wargo, J. A. The Influence of the Gut Microbiome on Cancer, Immunity, and Cancer Immunotherapy. Cancer Cell 33, 570–580 (2018). 34. Loomba, R. et al. Gut Microbiome-Based Metagenomic Signature for Non-invasive Detection of Advanced Fibrosis in Human Nonalcoholic Fatty Liver Disease. Cell Metab 25 (2017). 34. Loomba, R. et al. Gut Microbiome-Based Metagenomic Signature for Non-invasive Detection of Advanced Fibrosis in Human Nonalcoholic Fatty Liver Disease. Cell Metab 25 (2017). 34. Loomba, R. et al. Gut Microbiome-Based Metagenomic Signature for Non-invasive Detection of Advanced Fibrosis in Human Nonalcoholic Fatty Liver Disease. Cell Metab 25 (2017). 35. Matson, V. et al. The commensal microbiome is associated with anti-PD-1 efficacy in metastatic melanoma patients. Science 359, 104–108 (2018). 5. Matson, V. et al. The commensal microbiome is associated with anti-PD-1 efficacy in metastatic melanoma 35. Matson, V. et al. The commensal microbiome is associated with anti-PD-1 efficacy in metastatic melanoma patients. Science 359, 104–108 (2018). 36. Willis, J. R. & Gabaldón, T. The Human Oral Microbiome in Health and Disease: From Sequences to. Ecosystems. References et al. Metagenomic analysis of colorectal cancer datasets identifies crosscohort microbial diagnostic signatures and a link with choline degradation. Nat Med 25, 667–678 (2019). 11. Thomas, A. M. et al. Metagenomic analysis of colorectal cancer datasets identifies crosscohort microbial diagnostic signatures and a link with choline degradation. Nat Med 25, 667–678 (2019). 12. Wirbel, J. et al. Meta-analysis of fecal metagenomes reveals global microbial signatures that are specific for colorectal cancer. Nat Med 25, 679–689 (2019). 12. Wirbel, J. et al. Meta-analysis of fecal metagenomes reveals global microbial signatures that are specific for colorectal cancer. Nat Med 25, 679–689 (2019). 13. Park, S. Y. et al. Oral-Gut Microbiome Axis in Gastrointestinal Disease and Cancer. Cancers (Basel) 13 (2 14. More Than Just a Periodontal Pathogen –the Research Progress on Fusobacterium nucleatum. Frontiers in Cellular and Infection Microbiology 12 (2022). URL 10.3389/fcimb.2022. 815318. 14. More Than Just a Periodontal Pathogen –the Research Progress on Fusobacterium nucleatum. Frontiers in Cellular and Infection Microbiology 12 (2022). URL 10.3389/fcimb.2022. 815318. 14. More Than Just a Periodontal Pathogen –the Research Progress on Fusobacterium nucleatum. Frontiers in Cellular and Infection Microbiology 12 (2022). URL 10.3389/fcimb.2022. 815318. 15. Huh, J. W. & Roh, T. Y. Opportunistic detection of Fusobacterium nucleatum as a marker for the early gut microbial dysbiosis. BMC Microbiol 20, 208– 208 (2020). 15. Huh, J. W. & Roh, T. Y. Opportunistic detection of Fusobacterium nucleatum as a marker for the early gut microbial dysbiosis. BMC Microbiol 20, 208– 208 (2020). 16. Han, Y. W. Fusobacterium nucleatum: a commensal-turned pathogen. Curr Opin Microbiol 16. Han, Y. W. Fusobacterium nucleatum: a commensal-turned pathogen. Curr Opin Microbiol 23, 141–147 (2015). Page 7/15 17. Vacca, M. et al. The Controversial Role of Human Gut Lachnospiraceae. Microorganisms 8, 573–573 (2020). 17. Vacca, M. et al. The Controversial Role of Human Gut Lachnospiraceae. Microorganisms 8, 573–573 (2 18. Lee, G., You, H. J. & Bajaj, J. S. Distinct signatures of gut microbiome and metabolites associated with significant fibrosis in non-obese NAFLD. Nat Commun 11, 4982–4982 (2020). 18. Lee, G., You, H. J. & Bajaj, J. S. Distinct signatures of gut microbiome and metabolites associated with significant fibrosis in non-obese NAFLD. Nat Commun 11, 4982–4982 (2020). 19. Gu,  X. et al. Gut Ruminococcaceae levels at baseline correlate with risk of antibiotic- associated diarrhea. (2021). 19. Gu,  X. et al. Gut Ruminococcaceae levels at baseline correlate with risk of antibiotic- associated diarrhea. References Microorganisms 8, 308–308 (2020). 37. Ferretti, P. et al. Mother-to-Infant Microbial Transmission from Different Body Sites Shapes the Developing Infant Gut Microbiome. Cell Host Microbe 24 (2018). , y p , ( ) 36. Willis, J. R. & Gabaldón, T. The Human Oral Microbiome in Health and Disease: From Sequences to. Ecosystems. Microorganisms 8, 308–308 (2020). 37. Ferretti, P. et al. Mother-to-Infant Microbial Transmission from Different Body Sites Shapes the Developing Infant Gut Microbiome. Cell Host Microbe 24 (2018). 36. Willis, J. R. & Gabaldón, T. The Human Oral Microbiome in Health and Disease: From Sequences to. Eco 37. Ferretti, P. et al. Mother-to-Infant Microbial Transmission from Different Body Sites Shapes the Develop 24 (2018). 37. Ferretti, P. et al. Mother-to-Infant Microbial Transmission from Different Body Sites Shapes the Developing Infant Gut Microbiome. Cell Host Microbe 24 (2018). 38. Olm, M. R. et al. Identical bacterial populations colonize premature infant gut, skin, and oral microbiomes and exhibit different in situ growth rates. Genome Res 27, 601–612 (2017). Page 8/15 39. Tett, A. et al. Unexplored diversity and strain-level structure of the skin microbiome associated with psoriasis. NPJ Biofilms Microbiomes 3, 14 (2017). Page 8/15 39. Tett, A. et al. Unexplored diversity and strain-level structure of the skin microbiome associated with psoriasis. NPJ Biofilms Microbiomes 3, 14 (2017). Page 8/15 39. Tett, A. et al. Unexplored diversity and strain-level structure of the skin microbiome associated with psoriasis. NPJ Biofilms Microbiomes 3, 14 (2017). Page 8/15 40. Chng, K., Tay, A. & Li, C. Whole metagenome profiling reveals skin microbiome-dependent susceptibility to atopic dermatitis flare. Nat Microbiol 1, 16106–16106 (2016). 40. Chng, K., Tay, A. & Li, C. Whole metagenome profiling reveals skin microbiome-dependent susceptibility to atopic dermatitis flare. Nat Microbiol 1, 16106–16106 (2016). 41. Oh, J. et al. Biogeography and individuality shape function in the human skin metagenome. 41. Oh, J. et al. Biogeography and individuality shape function in the human skin metagenome. Nature 514, 59–64 (2014). 42. Structure, function and diversity of the healthy human microbiome. Nature 486, 207–14 (2012). 43. Tierney, B. T. et al. Systematically assessing microbiome-disease associations identifies drivers of inconsistency in metagenomic research. PLoS Biol 20, e3001556 (2022). 43. Tierney, B. T. et al. Systematically assessing microbiome-disease associations identifies drivers of inconsistency in metagenomic research. PLoS Biol 20, e3001556 (2022). 44. Sun, Z., Huang, S. & Zhang, M. Challenges in benchmarking metagenomic profilers. References Nat Methods 18, 618–626 (2021). 44. Sun, Z., Huang, S. & Zhang, M. Challenges in benchmarking metagenomic profilers. Nat Methods 18, 618–626 (2021). 45. Current Sampling Methods for Gut Microbiota: A Call for More Precise Devices. Frontiers in Cellular and Infection Microbiology 10 (2020). URL 10.3389/fcimb.2020.00151. 45. Current Sampling Methods for Gut Microbiota: A Call for More Precise Devices. Frontiers in Cellular and Infection Microbiology 10 (2020). URL 10.3389/fcimb.2020.00151. 45. Current Sampling Methods for Gut Microbiota: A Call for More Precise Devices. Frontiers in Cellular and Infection Microbiology 10 (2020). URL 10.3389/fcimb.2020.00151. 46. Vogtmann, E. et al. Colorectal Cancer and the Human Gut Microbiome: Reproducibility with Whole-Genome Shotgun Sequencing. PLoS One 11, e0155362 (2016). 46. Vogtmann, E. et al. Colorectal Cancer and the Human Gut Microbiome: Reproducibility with Whole-Genome Shotgun Sequencing. PLoS One 11, e0155362 (2016). 46. Vogtmann, E. et al. Colorectal Cancer and the Human Gut Microbiome: Reproducibility with Whole-Genome Shotgun Sequencing. PLoS One 11, e0155362 (2016). 47. Zeller, G. et al. Potential of fecal microbiota for early-stage detection of colorectal cancer. Mol Syst Biol 10, 766 (2014). 8. Olesen, S., Alm, E. Dysbiosis is not an answer. Nat Microbiol 1, 16228 (2016). https://doi.org/10.1038/nmic 49. Zheng, D., Liwinski, T. & Elinav, E. Interaction between microbiota and immunity in health and disease. Cell Res 30, 492–506 (2020). 49. Zheng, D., Liwinski, T. & Elinav, E. Interaction between microbiota and immunity in health and disease. Cell Res 30, 492–506 (2020). 49. Zheng, D., Liwinski, T. & Elinav, E. Interaction between microbiota and immunity in health and disease. Cell Res 30, 492–506 (2020). 50. Minot, S. S. & Willis, A. D. Clustering co-abundant genes identifies components of the gut microbiome that are reproducibly associated with colorectal cancer and inflammatory bowel disease. Microbiome 7, 110 (2019). 50. Minot, S. S. & Willis, A. D. Clustering co-abundant genes identifies components of the gut microbiome that are reproducibly associated with colorectal cancer and inflammatory bowel disease. Microbiome 7, 110 (2019). 50. Minot, S. S. & Willis, A. D. Clustering co-abundant genes identifies components of the gut microbiome that are reproducibly associated with colorectal cancer and inflammatory bowel disease. Microbiome 7, 110 (2019). 51. Olsen, I. & Yamazaki, K. Can oral bacteria affect the microbiome of the gut? J Oral Microbiol 11, 1586422 (2019). 52. Byrd,  K. M. &  Gulati, A. S. The Gum-Gut Axis in Inflammatory Bowel Diseases: A Hypothesis-Driven Review of Associations and Advances. Front Immunol 12, 620124 (2021). 52. Byrd,  K. M. &  Gulati, A. S. The Gum-Gut Axis in Inflammatory Bowel Diseases: A Hypothesis-Driven Review of Associations and Advances. Fron Immunol 12, 620124 (2021). 53. Tennant, S. M. et al. Influence of gastric acid on susceptibility to infection with ingested bacterial pathogens. Infect Immun 76, 639–45 (2008). 53. Tennant, S. M. et al. Influence of gastric acid on susceptibility to infection with ingested bacterial pathogens. Infect Immun 76, 639–45 (2008). 53. Tennant, S. M. et al. Influence of gastric acid on susceptibility to infection with ingested bacterial pathogens. Infect Immun 76, 639–45 (2008). 54. Schmidt, T. S. et al. Extensive transmission of microbes along the gastrointestinal tract. {eLife} 54. Schmidt, T. S. et al. Extensive transmission of microbes along the gastrointestinal tract. {eLife} 8 (2019). URL 10.7554/elife.42693. 8 (2019). URL 10.7554/elife.42693. 55. Kojima, M. et al. Bowel movement frequency and risk of colorectal cancer in a large cohort study of Japanese men and women. Br J Cancer 90, 1397–401 (2004). 55. Kojima, M. et al. Bowel movement frequency and risk of colorectal cancer in a large cohort study of Japanese men and women. Br J Cancer 90, 1397–401 (2004). 56. Haase, A. M. et al. Regional gastrointestinal transit times in severe ulcerative colitis. Neurogastroenterol Motil 28, 217–24 (2016). 56. Haase, A. M. et al. Regional gastrointestinal transit times in severe ulcerative colitis. Neurogastroenterol Motil 28, 217–24 (2016). 57. Albright, M. B. N. et al. Solutions in microbiome engineering: prioritizing barriers to organism establishment. ISME J 16, 331–338 (2022). 58. Hui, C., Richardson, D. M. & Landi, P. Defining invasiveness and invasibility in ecological networks. Biol Invasions 18, 971–983 (2016). 57. Albright, M. B. N. et al. Solutions in microbiome engineering: prioritizing barriers to organism establishment. ISME J 16, 331–338 (2022). 58. Hui, C., Richardson, D. M. & Landi, P. Defining invasiveness and invasibility in ecological networks. Biol Invasions 18, 971–983 (2016). D. M. & Landi, P. Defining invasiveness and invasibility in ecological networks. Biol Invasions 18, 971–983 (2016) Page 9/15 59. Pasolli, E., Schiffer, L. & Manghi, P. Accessible, curated metagenomic data through ExperimentHub. Nat Methods 14, 1023–1024 (2017). 60. Clostridia: Sporeforming Anaerobic Bacilli - Medical Microbiology - NCBI Bookshelf. 59. Pasolli, E., Schiffer, L. & Manghi, P. Accessible, curated metagenomic data through ExperimentHub. Nat Methods 14, 1023–1024 (2017). 60 Clostridia: Sporeforming Anaerobic Bacilli Medical Microbiology NCBI Bookshelf 59. Pasolli, E., Schiffer, L. & Manghi, P. Accessible, curated metagenomic data through ExperimentHub. Nat Methods 14, 1023–1024 (2017). 60. Clostridia: Sporeforming Anaerobic Bacilli - Medical Microbiology - NCBI Bookshelf. 61. Al-Hinai, M. A., Jones, S. W. & Papoutsakis, E. T. The Clostridium sporulation programs: diversity and preservation of endospore differentiation. Microbiol Mol Biol Rev 79, 19–37 (2015). wanathan, V. K. & Vedantam, G. Spore-forming Bacilli and Clostridia in human disease. Future Microbiol 5, 1109– 62. Mallozzi, M., Viswanathan, V. K. & Vedantam, G. Spore-forming Bacilli and Clostridia in human disease. 63. Cunningham, M. et al. Shaping the Future of Probiotics and Prebiotics. Trends in Microbiology 29, 667–685 (2021). URL https://doi org/10 1016/j tim 2021 01 003 63. Cunningham, M. et al. Shaping the Future of Probiotics and Prebiotics. Trends in Microbiology 29, 667–685 (2021). URL https://doi.org/10.1016/j.tim.2021.01.003. 64. 8 (2019). URL 10.7554/elife.42693. Damhorst, M. W. G. L., Adelman, Michael, H., Woodworth, C. S. & Kraft. Current Capabilities of Gut Microbiome-Based Diagnostics and the Promise of Clinical Application. Pages S270-S275 223 (2021). 65. Schloss, P. D. Identifying and Overcoming Threats to Reproducibility, Replicability, Robustness, and Generalizability in Microbiome Research. (2018). 66. Wirbel, J., Zych, K. & Essex, M. Microbiome meta-analysis and cross-disease comparison enabled by the SIAMCAT machine learning toolbox. Genome Biol 22, 93–93 (2021). 67. Truong, D., Franzosa, E. & Tickle, T. MetaPhlAn2 for enhanced metagenomic taxonomic profiling. Nat Methods 12, 902–903 (2015). 68. Franzosa, E. A. et al. Species-level functional profiling of metagenomes and metatranscriptomes. Nature methods 15, 962–968 (2018). 69. S, K. & L, K. Fireworks: Reproducible Machine Learning and Preprocessing with PyTorch. 69. S, K. & L, K. Fireworks: Reproducible Machine Learning and Preprocessing with PyTorch. The Journal of Open Source Software (2019). 70. Vatanen, T. et al. Variation in Microbiome LPS Immunogenicity Contributes to Autoimmunity in Humans. Cell 165, 842–53 (2016). 70. Vatanen, T. et al. Variation in Microbiome LPS Immunogenicity Contributes to Autoimmunity in Humans 71. Mann, H. B. & Whitney, D. R. On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other. The Annals of Mathematical Statistics 18, 50–60 (1947). 71. Mann, H. B. & Whitney, D. R. On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other. The Annals of Mathematical Statistics 18, 50–60 (1947). 71. Mann, H. B. & Whitney, D. R. On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other. The Annals of Mathematical Statistics 18, 50–60 (1947). 72. Horton, B. AUC Meets the Wilcoxon-Mann-Whitney U-Statistic (2017). URL https://blog. revolutionanalytics.com/2017/03/auc-meets-u-stat.html. 72. Horton, B. AUC Meets the Wilcoxon-Mann-Whitney U-Statistic (2017). URL https://blog. revolutionanalyt 72. Horton, B. AUC Meets the Wilcoxon-Mann-Whitney U-Statistic (2017). URL https://blog. revolutionanalytics.com/2017/03/auc-meets-u-stat.html. 72. Horton, B. AUC Meets the Wilcoxon-Mann-Whitney U-Statistic (2017). URL https://blog. revolutionanalytics.com/2017/03/auc-meets-u-stat.html. 73. Drury, M. The Probabilistic Interpretation of AUC (2017). URL http://madrury.github.io/ jekyll/update/statistics/2017/06/21/auc-proof.html. 74. Segata,  N.,  Izard,  J. & Waldron, L. Metagenomic biomarker discovery and explanation. Genome Biol 12, 60–60 (2011). 75. David, L. A. et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature 505, 559–63 (2014). l. Gut microbiome development along the colorectal adenoma-carcinoma sequence. Nat Commun 6, 6528 (2015) 76. Feng, Q. et al. Gut microbiome development along the colorectal adenoma-carcinoma sequence. 8 (2019). URL 10.7554/elife.42693. Identification and assembly of genomes and genetic elements in complex metagenomic samples without using reference genomes. Nat Biotechnol 32, 822–8 (2014). 84. Nielsen, H. B. et al. Identification and assembly of genomes and genetic elements in complex metagenomic samples without using reference genomes. Nat Biotechnol 32, 822–8 (2014). 85. Qin, J. et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. 85. Qin, J. et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. 85. Qin, J. et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. 86. Qin, N. et al. Alterations of the human gut microbiome in liver cirrhosis. Nature 513, 59–64 (2014). 87. Vincent, C. et al. Bloom and bust: intestinal microbiota dynamics in response to hospital exposures and Clostridium difficile colonization or infection. Microbiome 4, 12 (2016). 87. Vincent, C. et al. Bloom and bust: intestinal microbiota dynamics in response to hospital exposures and Clostridium difficile colonization or infection. Microbiome 4, 12 (2016). 87. Vincent, C. et al. Bloom and bust: intestinal microbiota dynamics in response to hospital exposures and Clostridium difficile colonization or infection. Microbiome 4, 12 (2016). 88. Ye, Z. et al. A metagenomic study of the gut microbiome in Behcet’s disease. Microbiome 6, 135 (2018) 88. Ye, Z. et al. A metagenomic study of the gut microbiome in Behcet’s disease. Microbiome 6, 135 (2018). 88. Ye, Z. et al. A metagenomic study of the gut microbiome in Behcet’s disease. Microbiome 6, 135 (2018). 89. Yu, J. et al. Metagenomic analysis of faecal microbiome as a tool towards targeted non- invasive biomarkers for colorectal cancer. Gut 66, 70–78 (2017). 89. Yu, J. et al. Metagenomic analysis of faecal microbiome as a tool towards targeted non- invasive biomarkers for colorectal cancer. Gut 66, 70–78 (2017). 90. Milanese, A., Mende, D. R. & Paoli, L. Microbial abundance, activity and population genomic profiling with mOTUs2. Nat Commun 10, 1014–1014 (2019). 90. Milanese, A., Mende, D. R. & Paoli, L. Microbial abundance, activity and population genomic profiling with mOTUs2. Nat Commun 10, 1014–1014 (2019). 8 (2019). URL 10.7554/elife.42693. Nat Commun 6, 6528 (2015). 77. Hannigan, G. D., Duhaime, M. B., 4th Ruffin, M., Koumpouras, C. C. & Schloss, P. D. Diagnostic Potential and Interactive Dynamics of the Colorectal Cancer Virome. mBio 9 (2018). 77. Hannigan, G. D., Duhaime, M. B., 4th Ruffin, M., Koumpouras, C. C. & Schloss, P. D. Diagnostic Potential and Interactive Dynamics of the Colorectal Cancer Virome. mBio 9 (2018). 77. Hannigan, G. D., Duhaime, M. B., 4th Ruffin, M., Koumpouras, C. C. & Schloss, P. D. Diagnostic Potential and Interactive Dynamics of the Colorectal Cancer Virome. mBio 9 (2018). 78. Jie, Z. et al. The gut microbiome in atherosclerotic cardiovascular disease. Nat Commun 8, 845 (2017). Page 10/15 79. Karlsson, F. H. et al. Gut metagenome in European women with normal, impaired and diabetic glucose control. Nature 498, 99–103 (2013). Page 10/15 79. Karlsson, F. H. et al. Gut metagenome in European women with normal, impaired and diabetic glucose control. Nature 498, 99–103 (2013). Page 10/15 79. Karlsson, F. H. et al. Gut metagenome in European women with normal, impaired and diabetic glucose control. Nature 498, 99–103 (2013). 80. Kieser, S. et al. Bangladeshi children with acute diarrhoea show faecal microbiomes with increased Streptococcus abundance, irrespective of diarrhoea aetiology. Environ Microbiol 20, 2256–2269 (2018). 80. Kieser, S. et al. Bangladeshi children with acute diarrhoea show faecal microbiomes with increased Streptococcus abundance, irrespective of diarrhoea aetiology. Environ Microbiol 20, 2256–2269 (2018). 81. Kostic, A. D. et al. The dynamics of the human infant gut microbiome in development and in progression toward type 1 diabetes. Cell Host Microbe 17, 260–73 (2015). 81. Kostic, A. D. et al. The dynamics of the human infant gut microbiome in development and in progression toward type 1 diabetes. Cell Host Microbe 1 260–73 (2015). 2. Li, J. et al. Gut microbiota dysbiosis contributes to the development of hypertension. Microbiome 5, 14 (201 83. Li, S. S. et al. Durable coexistence of donor and recipient strains after fecal microbiota transplantation. Science 352, 586–9 (2016). 83. Li, S. S. et al. Durable coexistence of donor and recipient strains after fecal microbiota transplantation. Science 352, 586–9 (2016). 84. Nielsen, H. B. et al. Identification and assembly of genomes and genetic elements in complex metagenomic samples without using reference genomes. Nat Biotechnol 32, 822–8 (2014). 84. Nielsen, H. B. et al. 80.         Kieser, S. et al. Bangladeshi children with acute diarrhoea show faecal microbiomes with increased Streptococcus abundance, irrespective of diarrhoea aetiology. Environ Microbiol 20, 2256–2269 (2018). Tables Tables 1-2 are available in the supplementary files section. Tables 1-2 are available in the supplementary files section. Figures Page 11/15 Page 11/15 Figure 1 Protocol overview. a) We assign significance values to organisms on a per-study basis using a Mann-Whitney U test. b) We assign overall scores to organism by averaging the AUC score across all studies (see methods). When the mean AUC is significantly different from 0.5 (two-sided T-test), we consider the mark health-associated (>0.5) or disease-associated (<0.5), indicating that the marker is consistently associated with that category. Figure 1 Protocol overview. a) We assign significance values to organisms on a per-study basis using a Mann-Whitney U test. b) We assign overall scores to organisms by averaging the AUC score across all studies (see methods). When the mean AUC is significantly different from 0.5 (two-sided T-test), we consider the marker health-associated (>0.5) or disease-associated (<0.5), indicating that the marker is consistently associated with that category. Protocol overview. a) We assign significance values to organisms on a per-study basis using a Mann-Whitney U test. b) We assign overall scores to organisms by averaging the AUC score across all studies (see methods). When the mean AUC is significantly different from 0.5 (two-sided T-test), we consider the marker health-associated (>0.5) or disease-associated (<0.5), indicating that the marker is consistently associated with that category. Page 12/15 Figure 2 Most significant biomarkers associated with health or disease (Core set - p<0.01). a) Boxplots show distribution by study of disease score (AUC-score from Mann-Whitney U test) for each of the taxa highlighted. Above 0.5 indicates health association and below 0.5 indicates disease association. Markers are identified whenever mean association score is statistically away from 0.5, via two-sided t-test. b) Heatmap showing direction of association of taxa by study. Green indicates health association, magenta indicates disease association,  and white indicates none (as determined by Mann-Whitney AUC score). Most biomarker taxa are consistently associated with one or the other. c) Log-distribution of relative abundances of Figure 2 Figure 2 Most significant biomarkers associated with health or disease (Core set - p<0.01). a) Boxplots show distribution by study of disease score (AUC-score from Mann-Whitney U test) for each of the taxa highlighted. Above 0.5 indicates health association and below 0.5 indicates disease association. Markers are identified whenever mean association score is statistically away from 0.5, via two-sided t-test. b) Heatmap showing direction of association of taxa by study. Green indicates health association, magenta indicates disease association,  and white indicates none (as determined by Mann-Whitney AUC score). Most biomarker taxa are consistently associated with one or the other. c) Log-distribution of relative abundances of each taxa when they are present in a microbiome. d) Box-plots show the percentage of samples in our dataset containing the taxa in question at all. determined by Mann-Whitney AUC score). Most biomarker taxa are consistently associated with one or the other. c) Log-distribution of relative abundances of each taxa when they are present in a microbiome. d) Box-plots show the percentage of samples in our dataset containing the taxa in question at all. Page 13/15 Figure 3 Figure 3 Distribution of extended set healthy and disease marker mean relative abundances in external v samples consistently show that healthy markers have higher mean relative abundance than dis non-western populations and c) cohorts with only samples from diseased individuals. Figure 3 Distribution of extended set healthy and disease marker mean relative abundances in external validation studies. a) Datasets having only healthy patient samples consistently show that healthy markers have higher mean relative abundance than disease markers. This observation ex- tends to patients from b) non-western populations and c) cohorts with only samples from diseased individuals. Page 14/15 Figure 4 Figure 4 Page 14/15 Page 14/15 Page 14/15 a) Boxplots show the (log) distributions of relative abundance of expanded healthy and disease markers across different body sites. Healthy markers are statistically higher abundance in the gut, whereas disease markers are statistically more abundant in the oral microbiome, and no significant pattern emerges for nares or skin sites. Scatter plots demonstrate a relationship between a taxon’s disease score and its mean relative abundance in b) stool samples and c) oral samples. This relationship holds regardless of whether the taxon is statistically significant. Extended set marker taxa are color-coded as red (disease) and green (healthy), while all other taxa are colored in grey. Figure 5 Comparison of marker taxa association score vs frequency in the gut environment for core taxa (p<0.01). a) Leftmost column shows the disease score for each marker taxon, and the subsequent columns indicate the percentage of samples in a given body site that contain that taxon. The scatterplots on the right compares the association score of a taxon vs its mean relative abundance b) in the gut and c) in the oral microbiome, respectively. Horizontal line delineates disease markers from healthy markers and vertical dotted line separates taxa that are present in more or less than 20% of samples. d) Scatterplot compares abundance in the gut vs abundance in the oral environment color-coded by association score. Figure 2 p p g y g ( ) green (healthy), while all other taxa are colored in grey. Figure 5 Comparison of marker taxa association score vs frequency in the gut environment for core taxa (p<0.01). a) Leftmost column shows the disease score for each marker taxon, and the subsequent columns indicate the percentage of samples in a given body site that contain that taxon. The scatterplots on the right compares the association score of a taxon vs its mean relative abundance b) in the gut and c) in the oral microbiome, respectively. Horizontal line delineates disease markers from healthy markers and vertical dotted line separates taxa that are present in more or less than 20% of samples. d) Scatterplot compares abundance in the gut vs abundance in the oral environment color-coded by association score. Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. table1.pdf table2.pdf figures1v1.pdf figures2v1.pdf figures3v1.pdf figures4v1.pdf figures5v1.pdf table1.pdf table2.pdf figures1v1.pdf figures2v1.pdf figures3v1.pdf figures4v1.pdf figures5v1.pdf table1.pdf table2.pdf figures1v1.pdf figures2v1.pdf figures3v1.pdf figures4v1.pdf figures5v1.pdf Page 15/15

https://openalex.org/W3177483626

https://engrxiv.org/preprint/download/1577/3242

English

Statistical Analysis on Shear Strength Parameter from Index Properties of Fine-grained Soils

Social Science Research Network

cc-by

Tariku Tafari Bakala1*, Emer Tucay Quezon2 and Mohammed Yasin3 1Civil Engineering Department, Assosa University, Assosa, Benishangul- Gumuz Region, Ethiopia. 2Civil Engineering and Construction Engineering and Management Streams, Ambo Institute of Technology, Ambo University, Ambo, Oromia Region, Ethiopia. 3Civil Engineering, Jimma Institute of Technology, Jimma University, Jimma, Oromia, Ethiopia. Article Information DOI: 10.9734/JERR/2021/v20i417291 Editor(s): (1) Dr. Guang Yih Sheu, Chang-Jung Christian University, Taiwan. Reviewers: (1) Jian-Hong Wu, National Cheng Kung University, Taiwan. (2) Lim Hwee San, Universiti Sains Malaysia (USM), Malaysia. Complete Peer review History: http://www.sdiarticle4.com/review-history/66257 DOI: 10.9734/JERR/2021/v20i417291 Editor(s): (1) Dr. Guang Yih Sheu, Chang-Jung Christian University, Taiwan. Reviewers: (1) Jian-Hong Wu, National Cheng Kung University, Taiwan. (2) Lim Hwee San, Universiti Sains Malaysia (USM), Malaysia. Complete Peer review History: http://www.sdiarticle4.com/review-history/66257 Received 05 January 2021 Accepted 08 March 2021 Published 18 March 2021 Original Research Article Authors’ contributions This work was carried out in collaboration among all authors. Author TTB designed the study, performed the statistical analysis, wrote the protocol and wrote the first draft of the manuscript. Author ETQ managed the analyses of the study. Author MY managed the literature searches. All authors read and approved the final manuscript. Journal of Engineering Research and Reports 20(4): 15-28, 2021; Article no.JERR.66257 ISSN: 2582-2926 Statistical Analysis on Shear Strength Parameter from Index Properties of Fine-grained Soils Tariku Tafari Bakala1*, Emer Tucay Quezon2 and Mohammed Yasin3 1Civil Engineering Department, Assosa University, Assosa, Benishangul- Gumuz Region, Ethiopia. 2Civil Engineering and Construction Engineering and Management Streams, Ambo Institute of Technology, Ambo University, Ambo, Oromia Region, Ethiopia. 3Civil Engineering, Jimma Institute of Technology, Jimma University, Jimma, Oromia, Ethiopia. Authors’ contributions This work was carried out in collaboration among all authors. Author TTB designed the study, performed the statistical analysis, wrote the protocol and wrote the first draft of the manuscript. Author ETQ managed the analyses of the study. Author MY managed the literature searches. All authors read and approved the final manuscript. Article Information DOI: 10.9734/JERR/2021/v20i417291 Editor(s): (1) Dr. Guang Yih Sheu, Chang-Jung Christian University, Taiwan. Reviewers: (1) Jian-Hong Wu, National Cheng Kung University, Taiwan. (2) Lim Hwee San, Universiti Sains Malaysia (USM), Malaysia. Complete Peer review History: http://www.sdiarticle4.com/review-history/66257 Received 05 January 2021 Accepted 08 March 2021 Published 18 March 2021 ABSTRACT Original Research Article 20(4): 15-28, 2021; Article no.JERR.6625 ISSN: 2582-2926 1. INTRODUCTION Statistical modeling for a prediction in geotechnical engineering has been used to model different engineering properties of soils. This is pointed out that the prominence of statistical modeling for prediction in geotechnical practice is greatly vigorous. One of the most significant engineering properties of soil is its capacity to respond sliding along internal surfaces within a mass. The stability of structures built on soil depends on the shearing resistance obtainable by the soil along probable surfaces of slippage [1]. It is quite essential that an engineer has to safeguard the structure is safe against shear catastrophe in the soil that supports it and does not undergo excessive settlement [2]. It is often necessary for the geotechnical engineer to quickly characterize the soil and determine its engineering properties to estimate the soil's appropriateness for any industrial practices. Consequently, statistical analysis is a crucial method to predict soils' engineering properties, especially for developing countries like Ethiopia, where there is a financial limitation, lack of test equipment, and limited time, which is used for design purposes. Undrained shear strength is one of the parameters to the bearing capacity of soil that could convey any structure rest on it. Some laboratory tests needed to obtain these parameters are costly and laborious, while other soil properties like index properties can be achieved quicker and cheaper. Index properties are the basis for distinguishing soils [3]. However, index properties can be obtained simply with low-cost equipment when compared to strength properties equipment. This study was done in Agaro town, one of the rapidly growing southwestern parts of Ethiopia. A few soil tests were carried out in this town before, although it is known as a coffee production center wherein construction undertakings have been enormously creating. The undrained shear strength is used to estimate the short-term bearing capacity of fine-grained soils for foundations and estimate slopes' short- term stability. Besides, it compares the shear strength of soils from a site to create soil strength variability quickly and cost-effectively and use for stress-strain characteristics under fast (undrained) loading conditions [4-6]. In this research, undrained shear strength was performed by correlating with index properties of soils, which can be used to minimize cost, effort, and time for any geotechnical practice to analyze and design conditions of the study area. ABSTRACT Shear strength is the essential engineering property of soil required to analyze and design foundations, retaining walls, bridges, embankment, and related infrastructure. The laboratory equipment and field instruments are not sufficient in developing countries to obtain soil engineering properties, especially strength properties. Thus, Geotechnical engineers usually endeavor to develop statistical models that best fit a particular area and soil type, especially for analysis and design purposes. In this research, a Statistical Analysis on the Shear Strength parameter from the Index Properties of Fine-Grained Soils was studied. For predicting the undrained shear strength parameter, single linear regression (SLR) and multiple linear regressions (MLR) analyses were developed. To develop the intended statistical models for a study, SAS JMP Pro 13, SPSS v22, and Microsoft Excel-2013 software were introduced. The results of a study indicated that undrained shear strength(Cu) was significantly correlated with liquid limit(LL), plastic limit(PL), bulk density (ρbulk), dry density(ρdry), natural moisture content(NMC), and plasticity index(PI). While it was not significantly correlated with a specific gravity (Gs) and liquidity index (LI) of study area soil. Corresponding author: Email: [email protected], [email protected]; *Corresponding author: Email: [email protected], [email protected]; Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Finally, a strong Model of Cu with a coefficient of determination (R2 = 0.806), good significance level, and less Std. error was obtained from multiple linear regression (MLR) analysis. The developed model can figure undrained shear strength parameter and wide application in the construction industry to minimize the cost, effort, and time for laboratory tests of shear strength parameter of a study area. Finally, a strong Model of Cu with a coefficient of determination (R2 = 0.806), good significance level, and less Std. error was obtained from multiple linear regression (MLR) analysis. The developed model can figure undrained shear strength parameter and wide application in the construction industry to minimize the cost, effort, and time for laboratory tests of shear strength parameter of a study area. Keywords: Index properties; undrained shear strength; linear; statistical analysis. 2.2.1 Natural moisture content (NMC) This study was accompanied in Agaro town, as indicated in Fig. 1, found in the Jimma Zone of the Oromia Region, Ethiopia. It sits at estimated latitude and longitude of 7°51′N 36°35′E and has an altitude of 1670 meters above sea level. It is 393 km from the capital, Addis Ababa, and 46 km from Jimma town. Moisture contents of the soil samples were determined in the laboratory as per ASTM D 2216 standard. A set of samples were dried to a constant weight using oven-dry at a temperature of 105ºC. 2.2 Methods The test was performed as per ASTM D854-58, Standard Test for Specific Gravity of Soil Solids by density bottle, procedure. For this work, various field visits (reconnaissance) were done to differentiate the study area's real situations. Furthermore, undrained shear strength and index properties of study area soil were performed in a Jimma Institute of Technology laboratory. Based on obtained test results, linear statistical analyses were done between undrained shear strength and index properties of study area soil. 2. MATERIALS AND METHODS tests were performed to determine the index properties and undrained shear strength of study area soils. 2.2.3 Dry density (ρdry) and bulk density (ρbulk) The density of soil was determined according to ASTM D 2937 (a standard test for the density of soil in place by the drive cylinder method). This technique is achieved to determine the in-place density of undisturbed soil found by pushing or drilling a thin-walled cylinder. The location of test pits was selected to characterize the soil types found in the study area. It was conducted the excavations of test pits. From excavated test pits, thirty representative undisturbed and disturbed samples were taken by Shelby tube & plastic bags, respectively. 1. INTRODUCTION In this study area, no laboratory equipment is available to test engineering properties of soils, especially strength parameters, for analysis and design purposes. From the related literature review, one author mentioned that not always possible to conduct the tests on every new situation. To cope with such problems, numerical solutions have been developed to estimate the shear strength parameters [7]. Most of the time, for the need to analyze and design foundations, slope stability and other infrastructure engineers follow some techniques apart from doing an applicable investigation of subsoil incident, which may reason for the destruction of structures on it. This study developed a statistical model that helped forecast the undrained shear strength from soils' index properties. For restraint of time, costs, and lack of testing equipment of shear strength, it is accessible for an engineer or consultants, contractors, clients, municipality, and researchers to get whatever they want for analysis and design purposes area from simple index properties test results. Experimental determination of the strength parameters used for design purposes is widespread, challenging to perform, and costly compared to index properties of soils. 16 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 2.2.4 Grain size analysis This test was performed as per ASTM D422 standard to determine the percentage of different grain sizes contained within a soil. The mechanical or sieve analysis is used to decide the distribution of the coarser, larger-sized particles, and the hydrometer analysis method was used to determine the distribution of the finer particles, respectively. For this study, both wet sieve analysis and hydrometer analysis was done. For determination of natural moisture content, undrained shear strength, and density tests, undisturbed samples were used. In contrast, disturbed samples were used to determine index properties tests such as specific gravity, liquid limit, plastic limit, plasticity index, and grain size analysis. In this study, the following laboratory Fig. 1. Location of the study area Fig. 1. Location of the study area 17 17 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 2.2.5 Atterberg limit’s test is a term used to denote the association or relationship between two (or more) quantitative variables. This analysis is fundamentally based on the assumption of a straight–line with the construction of a scatter plot or scatter diagram [a graphical of the data] with one variable on the X-axis and the other on the Y-axis [9]. In this study, the undrained shear strength (Cu) was taken as the predicted variable (dependent) while the predictors (independent) variables represented by the specific gravity, liquid limit, plastic limit, plasticity index, liquidity index, bulk density, dry density, and natural moisture content. Before the execution of the regression analysis using the test results, a scatter plot matrix was produced by applying the SAS JMP Pro.13 to study the relations developed between the dependent variable and the predictor variables by visualizing to determine the model that best outfits the test results as indicated on Fig. 2 and Fig. 3. This test was performed as per ASTM D-4318 standard for Liquid Limit (LL), Plastic Limit (PL), and Plasticity Index (PI) of soils. The air-dried samples were organized by drying the specimen in the air. The quotas of the samples passing the No. 40 (0.425mm) sieve were used for the preparation of the sample for the test. 2.2.6 Undrained shear strength (Cu) Undrained shear strength (Cu) was determined according to ASTM D 2166 standard. The test was conducted on undisturbed samples collected by Shelby tube sampler. These sample properties were used during the statistical analysis between the undrained shear strength and index properties of soils using the SAS JMP Pro 13, SPPS v22 software, and excel 2013 Microsoft. Moreover, statistical regression analyses of single and multiple models of test results were carried out. Statistical models by regression analysis were analyzed and developed to fit the test results. Beneath the discussions of the obtained results, the developed models' fitness was inspected in different ways. From scatter plots offered in Fig. 2 and Fig. 3, a visual method of displaying a relationship between variables is plotted in a two-dimensional coordinate system. The scatter plots' assessment indicated that a real indication of the points lie scattered arbitrarily as a straight or looks like a straight line, mainly for the liquid limit, plastic limit, natural moisture content, bulk density, dry density, and plasticity index. However, the remaining independent variables, such as specific gravity and liquidity index, by some extent, outliers away from the possible visual straight. The above scatter plots are indicated a linear response, and hence, a linear regression model expressed the association between the focus parameters. 3.1.1 Scatter plot strategy To develop correlations, the first step is creating a scatter plot of the data [8]. Correlation analysis Fig. 2. Scatter plot diagram of undrained shear strength versus Gs, NMC, ρbulk, ρdry Fig. 2. Scatter plot diagram of undrained shear strength versus Gs, NMC, ρbulk, ρdry 18 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 3.1.2 Normality test linear regression (MLR) models to describe the strength of (as correlation indicated in Table 2.) cohesive soil from soil index properties using a statistical approach. Multiple linear regression is a method of analysis for assessing the strength of the relationship between each set of independent variables and a single response variable. In contrast, when only a single explanatory variable is involved, it is generally referred to as simple linear regression [12]. It is essential to check normality before proceeding with any applicable statistical procedures. If the assumption of normality is violated, interpretation and inference may not be reliable or valid. For N (number of samples, in this case, thirty) is less than 2,000 (for a small number of the sample), it is recommended to read the Shapiro-Wilk statistic that does not reject the null hypothesis of normality for p>0.05 [10]. Shapiro-Wilk test has a proper performance with a sample size of 7-2000. It is not possible to apply all the available tests to evaluate normality in any of the statistical software programs. However, it is possible to run the two commonly used Kolmogorov-Smirnov and Shapiro-Wilk tests with SPSS [11]. Based on Table 1, both predicted, and predictors data were normally distributed (i.e.p-value more than 0.05). To do this modeling, statistical Software SAS JMP Pro.13, SPSS V22 and Microsoft excel 2013 soft wares were used to study the significance of individual predictor variables as well as to get the best model. In view of that, the thirty-laboratory test results of the independent and dependent variables were used in the regression analysis to get the intended statistical model. To detect the influence of one variable on the other, a stepwise linear regression has been analyzed, and as a result, the respective correlation coefficients and level of significance were determined. 3.2 Regression Analysis Significance level (α) and Pearson correlation coefficient (R) in correlations Gs NMC ρbulk ρdry LL PL PI LI Cu Gs R 1 -0.111 0.108 0.103 -0.076 -0.124 0.063 -0.036 0.187 Sig.(2tail) 0.561 0.571 0.587 0.688 0.515 0.739 0.849 0.322 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 NMC R - 0.111 1.000 -0.777 -0.879 0.832 0.756 0.547 0.500 -0.813 Sig.(2tail) 0.561 0.000 0.000 0.000 0.000 0.002 0.001 0.000 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 ρbulk R 0.108 -0.777 1.000 0.981 -0.805 -0.713 -0.566 -0.260 0.835 Sig.(2tail) 0.571 0.000 0.000 0.000 0.000 0.001 0.166 0.000 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 ρdry R 0.103 -0.879 0.981 1.000 -0.855 -0.762 -0.591 -0.356 0.866 Sig.(2tail) 0.587 0.000 0.000 0.000 0.000 0.001 0.053 0.000 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 LL R - 0.076 0.832 -0.805 -0.855 1.000 0.927 0.618 0.062 -0.897 Sig.(2tail) 0.688 0.000 0.000 0.000 0.000 0.000 0.746 0.000 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 PL R - 0.124 0.756 -0.713 -0.762 0.927 1.000 0.278 -0.113 -0.822 Sig.(2tail) 0.515 0.000 0.000 0.000 0.000 0.137 0.554 0.000 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 PI R 0.063 0.547 -0.566 -0.591 0.618 0.278 1.000 0.393 -0.575 Sig.(2tail) 0.739 0.002 0.001 0.001 0.000 0.137 0.031 0.001 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 LI R - 0.036 0.562 -0.260 -0.356 0.062 -0.113 0.393 1.000 -0.171 Sig.(2tail) 0.849 0.001 0.166 0.053 0.746 0.554 0.031 0.367 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 Cu R 0.187 -0.813 0.835 0.866 -0.897 -0.822 -0.575 -0.171 1.000 Sig.(2tail) 0.322 0.000 0.000 0.000 0.000 0.000 0.001 0.367 N 30 30.000 30.000 30.000 30.000 30.000 30.000 30.000 30.000 As cited by Roy et al. [7], the procedure is commonly used to produce a parsimonious model that maximizes accuracy with an optionally reduced number of predictor variables. Table 2. linear relationships shows that the 3.2.1 Simple linear regression (SLR) analysis 3.2.1.1 Model-A Model between undrained shear strength (Cu) and specific gravity (Gs). Table 2. Significance level (α) and Pearson correlation coefficient (R) in correlations As cited by Roy et al. [7], the procedure is commonly used to produce a parsimonious model that maximizes accuracy with an optionally reduced number of predictor variables. Model between undrained shear strength (Cu) and specific gravity (Gs). Table 2. 3.2 Regression Analysis In this study, an effort was made to apply single linear regression model (SLR) and multiple Table 1. Normality tests Table 1. Normality tests Table 1. Normality tests Kolmogorov-Smirnov Shapiro-wilk Statistic df Sig. Statistic Df Sig.(p-value) Gs 0.134 30 0.177 0.947 30 0.136 NMC 0.102 30 0.2 0.966 30 0.441 ρbulk 0.161 30 0.046 0.946 30 0.129 ρdry 0.173 30 0.022 0.959 30 0.296 LL 0.103 30 0.2 0.979 30 0.802 PL 0.137 30 0.156 0.937 30 0.074 PI 0.134 30 0.18 0.964 30 0.386 LI 0.131 30 0.2 0.966 30 0.438 Cu 0.095 30 0.2 0.962 30 0.349 Fig. 3. Scatter plot diagram of undrained shear strength with LL, PL, PI, LI Table 1. Normality tests Kolmogorov-Smirnov Shapiro-wilk Statistic df Sig. Statistic Df Sig.(p-value) Gs 0.134 30 0.177 0.947 30 0.136 NMC 0.102 30 0.2 0.966 30 0.441 ρbulk 0.161 30 0.046 0.946 30 0.129 ρdry 0.173 30 0.022 0.959 30 0.296 LL 0.103 30 0.2 0.979 30 0.802 PL 0.137 30 0.156 0.937 30 0.074 PI 0.134 30 0.18 0.964 30 0.386 LI 0.131 30 0.2 0.966 30 0.438 Cu 0.095 30 0.2 0.962 30 0.349 Fig. 3. Scatter plot diagram of undrained shear strength with LL, PL, PI, LI Fig. 3. Scatter plot diagram of undrained shear strength with LL, PL, PI, LI 19 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Table 2. Model between Cu and dry density (ρdry) Cu = -29.670 + 73.031*ρdry, with R2=.750, α = .000 <0.05 The statistical output details specified that the relationship developed between ρdry and Cu is significant (α <0.05) and good determinant factor. 3.2.1.2 Model-B The statistical output details indicated that the relationship developed between Cu and ρbulk is significant (α<0.05), and a good relationship exists between the correlation variables. Model between Cu and Natural moisture content (NMC) Cu = 148.515 - 2.078*NMC, with R2=0.662,α = 0.00 <0.05, N = 30 3.2 Regression Analysis linear relationships shows that the correlation between Cu with liquid limit (LL), plastic limit (PL), natural moisture content (NMC), density (ρbulk,ρdry) & plasticity index (PI) relatively stronger. However, Cu has a weak correlation with specific gravity (Gs) & Liquidity index (LI). In fact, the strength of fine- grained soil has a more significant association with soil consistency. Consequently, those parameters have resulted in a relatively strong correlation with the strength parameter (Cu). This was due to the presence of more clay and silty in that soils. In this study, many alternative linear regression analyses that best fit the obtained test results were carried out. The resulting Model, after correlating Cu with Gs is obtained from SAS JMP pro 13 & SPSS v22 outputs with its corresponding statistical parameters as on Fig. 4. Cu = -75.786 + 52.014*Gs, with R2= 0.035, α = 0.322 > 0.05, N = 30 Cu = -75.786 + 52.014*Gs, with R2= 0.035, α = 0.322 > 0.05, N = 30 The statistical output details showed that the relationship developed between Gs and Cu is insignificant (i.e.α>0.05). Furthermore, the relationship between correlation variables is weak (R2 <0.5). 20 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 3.2.1.4 Model-D The statistical output details indicated that the relationship developed between NMC and Cu is significant (α<0.05), and a good correlation happened concerning the correlating variables. Fig. 5 shows the graphical output of the model between Cu and Natural moisture content (NMC). Model between Cu and dry density (ρdry) Model between Cu and dry density (ρdry) Model of Cu with NMC and LI The statistical output details indicated that the relationship developed between LI and Cu is insignificant (α>0.05), and a weak relationship exists between the correlation variables. Table 3.illustrates that the developed single linear regression (SLR) models based on the level of the significance (α) and coefficient of determination (R2), Cu value has a strong relationship with LL, ρdry, ρbulk, PL & NMC (i.e., from order 1 to 5).On the other hand, PI, Gs & LI (i.e., Orders from 6 to 8) indicated a weak relationship (R2 <0.5) and insignificant level for Gs & LI (i.e.α>0.05) with Cu. Those predictors were also good indicators to form better multiple linear regression analysis that could provide better models for the prediction of dependent variables of intensive area. The resulting Model after correlating Cu with NMC and LI is obtained from SAS JMP pro 13 & SPSS v22 outputs with its corresponding statistical parameters: Cu= 163.309+.559* LI – 2.677*NMC, R2 =.781, α = 0.000 < 0.05, Tolerance= .685 > 0.2 & VIF =1.461 < 10, Durbin-Watson= 2.629 ~ 2 The resulting Model after correlating Cu with NMC and LI is obtained from SAS JMP pro 13 & SPSS v22 outputs with its corresponding statistical parameters: Cu= 163.309+.559* LI – 2.677*NMC, R2 =.781, α = 0.000 < 0.05, Tolerance= .685 > 0.2 & VIF =1.461 < 10, Durbin-Watson= 2.629 ~ 2 The details of the statistical output of Model-1 indicated that the relationship developed between predictors with no concern for the value of 1 to 3.Hence, after going through several alternative groupings of predictors. This model contains plastic limit (PL) and plasticity index (PI) with a good significance level (i.e.α=0), and the strongest determination coefficient (R2=0.806) is modeled and taken as the best model. 3.2.1.6 Model-F applied. For this section, the significance level and correlation coefficient of predictors on each other obtained from the single linear regression analysis and the scatter plot were used. For independent variables highly correlated (interdependent) to each other (i.e., Correlated at 0.50 or 0.60 and above), then one might decide to combine (aggregate) them into a composite variable or eliminate one or more of the highly correlated variables [13]. Spotting multicollinearity among a set of explanatory variables might not be easy. A useful approach is examining the variance inflation factors (VIFs) or the tolerances of the explanatory variables. Accordingly, VIFs above 10 or tolerances below 0.1 are seen as a cause of concern [12]. Moreover, Durbin-Watson used to examine multicollinearity of predictors with no concern for the value of 1 to 3.Hence, after going through several alternative groupings of predictors, a model which contains plastic limit (PL) and plasticity index (PI) with a good significance level (i.e.α=0). The strongest determination coefficient (R2=0.806) is modeled and taken as the best model. Model between Cu and Plastic Limit (PL) The statistical output details indicated that the relationship developed between Cu and PL is significant (α < 0.05), and a strong correlation exists between the correlation variables. 3.2.1.8 Model-H Model between Cu and liquidity Index (LI) 3.2.1.7 Model-G The statistical output details indicated that the relationship developed between PI and Cu is significant (α<0.05), but a weak relationship exists between the correlation variables. 3.2.1.5 Model-E Model between Cu and bulk density (ρbulk) as indicated on Fig. 6. Model between Cu and Liquid limit (LL) Model between Cu and Liquid limit (LL) Cu = 220.604 - 2.323*LL, R2= 0.805, α =0 .000 < 0.05 Cu = -54.278 + 65.452*ρbulk, with R2=.698, α = 0.00 <0.05, N = 30 21 Fig. 4. Linear fit of Gs-Cu Fig. 5. Linear fit of NMC-Cu Fig. 4. Linear fit of Gs-Cu Fig. 4. Linear fit of Gs-Cu Fig. 4. Linear fit of Gs-Cu Fig. 5. Linear fit of NMC-Cu 21 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.662 Fig. 6. Linear fit of ρbulk-Cu Fig. 7. Linear fit of ρdry-Cu Fig. 8. Linear fit of LL-Cu etails of the statistical output showed that ationship developed between LL and Cu is significant (α<0.05) and has a stron relationship. Fig. 6. Linear fit of ρbulk-Cu Fig. 7. Linear fit of ρdry-Cu Fig. 7. Linear fit of ρdry-Cu Fig. 8. Linear fit of LL-Cu The details of the statistical output showed that the relationship developed between LL and Cu is significant (α<0.05) and has a strong relationship. The details of the statistical output showed that the relationship developed between LL and Cu is significant (α<0.05) and has a strong relationship. significant (α<0.05) and has a strong relationship. The details of the statistical output showed that the relationship developed between LL and Cu is 22 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 3.2.1.6 Model-F Model of Cu with PL and LI Model of Cu with PL and LI Multiple Linear Regression (MLR) analysis is tried to model the relationship between two or more explanatory variables and a predicted variable by fitting an equation to experimental data. For this study, the stepwise regression analysis method of variable selection was The resulting Model after correlating Cu with PL and LI is obtained from SAS JMP pro 13 & SPSSv22 outputs-:-Cu= 164.670-2.697* PL- .357*LI, R2 =.746, α =0 .000 < 0.05, 23 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Tolerance=.987 > 0.2 & VIF=1.013 < 10, Durbin- Watson=2.566~2 .The details of the statistical out-put of Model-2 indicate that the relationship between Cu and PL and LI is significant (α<0.05). Besides, the R2 value of the multiple regression analysis is improved than the R2 value of the individual parameters, i.e., PL and LI. the "best model" for the prediction of undrained shear strength (Cu) of the study area based on the relative correlation coefficient(R), determinant factor (R2) & significance level (α) of all developed models. 3.2.3 Validation of predicted value from the actual (measured) value of Cu To check the validity of developed models, experimental values obtained from test samples (actual) should be compared with the predicted value [14]. Considering the acceptability of Model-3 as the best model, it can e used to approximate the study area's undrained shear strength parameter. Moreover, it is possible to understand from Fig.15. the relationship between the predicted and the measured (actual) value was strong. 3.2.2.3 Model-3 Model of Cu with PL and PI Model of Cu with PL and PI Model of Cu with PL and PI The details of the statistical output of Model 3 indicate that the relationship developed between Cu with PL and PI is significant (α<0.05). The R2 value of the multiple regression analysis is improved than the R2 value of the individual parameters, i.e., PL, PI. Cu= 224.032-2.272* PL- 2.485*PI, R2 =.806, α =0.000 < 0.05, Tolerance=.923> 0.2 & VIF=1.084< 10, Durbin- Watson=2.791~ 2 In general, the scatter plot on Fig. 15 illustrated that the predicted Cu value scatters near the straight line, through which the actual and predicted Cu value is equal. However, there is a little bit of variation between the actual and the measured Cu. Based on Table 4, all models are good since all models are significant and the coefficient of determinations is strong. However, Model-3 is 24 Fig. 9. Linear fit of LI-Cu Fig. 10. Linear fit of PL-Cu Fig. 9. Linear fit of LI-Cu Fig. 9. Linear fit of LI-Cu Fig. 10. Linear fit of PL-Cu 24 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Fig. 11. Linear fit of PI-Cu Fig. 12. Model-1 3D scatter plot Fig. 11. Linear fit of PI-Cu Fig. 12. Model-1 3D scatter plot Fig. 13. Model-2 3D scatter plot Fig. 11. Linear fit of PI-Cu Fig. 12. Model-1 3D scatter plot Fig. 12. Model-1 3D scatter plot Fig. 12. Model-1 3D scatter plot Fig. 13. Model-2 3D scatter plot Fig. 13. Model-2 3D scatter plot 25 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JER 26 Fig. 14. Model-3 3D scatter plot Fig. 15. Plots of predicted and actual values of undrained shear strength (Cu) Fig. 16. Plots of predicted and actual values of undrained shear strength (Cu) R² = 0.88 55 65 75 50 55 60 65 70 Predicted Cu,kPa Actual Cu, kPa Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Fig. 14. Model-3 3D scatter plot Fig. 15. Plots of predicted and actual values of undrained shear strength (Cu) R² = 0.88 75 Cu,kPa Fig. 14. Model-3 3D scatter plot Fig. 14. Model-3 3D scatter plot Fig. 14. Model-3 3D scatter plot Fig. 15. Plots of predicted and actual values of undrained shear strength (Cu) g p g ( ) R² = 0.88 55 65 75 50 55 60 65 70 Predicted Cu,kPa Actual Cu, kPa Fig. 16. CONSENT It is not applicable. Model of Cu with PL and PI In contrast, it was not significantly correlated with this study area soil's specific gravity and liquidity index. The validation of the developed model is led by using these test results of the study area. Subject to the relative correlation coefficient(R), determinant factor (R2) & significance level(α), Model-3 (i.e. Cu= 224.032-2.272* PL-2.485*PI) is preferred among the different alternative models discussed & developed above. Consequently, from Fig.16 relation of measured (actual) and the predicted value of Cu is exhibited a little variation (R2=0.88). The research result would be beneficial for individuals, researchers, consultants, contractors, municipal and other government agencies involved in building construction and other structures in the study area. It is also suggested to do such statistical modeling on other areas of the country by different agencies and researchers. 4. CONCLUSION Furthermore, it is desirable to conduct comparative statistical modeling between undisturbed and remolded soils to get Cu value from soil index properties. It was studied the statistical analysis of undrained shear strength from index properties of Fine-Grained Soil. For this purpose, different models were developed to predict Cu value from Gs, NMC, ρbulk, ρdry, LL, PL, and PI & LI. Consequently, the best Model from all with a better coefficient of determination (R2 = 0.806), good significance level, and less Std. error were obtained from multiple linear regression (MLR) analysis as below: It is not applicable. It is not applicable. ETHICAL APPROVAL Cu=224.032-2.272*PL-2.485*PI, R2=.806, p- value =0.000 < 0.05, Tolerance =.923> 0.2 & VIF=1.084< 10, Durbin-Watson=2.791~2. Model of Cu with PL and PI Plots of predicted and actual values of undrained shear strength (Cu) 26 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 Table 3. Summary of Single Linear Regression (SLR) models No. Model name SLR models from different variables R2 Significance level,α Rank based on α and R2 1 Model-A Cu = -75.786 + 52.014*Gs 0.035 0.322 7 2 Model-B Cu = 148.515 - 2.078*NMC 0.662 0.00 5 3 Model-C Cu=-54.278+ 65.452*ρbulk 0.698 0.00 3 4 Model-D Cu = -29.670 + 73.031*ρdry 0.75 0.00 2 5 Model-E Cu = 220.604 - 2.323*LL 0.805 0.00 1 6 Model-F Cu = 154.661- 2.602*PL 0.676 0.00 4 7 Model-G Cu = 185.778 - 3.807*PI 0.330 0.001 6 8 Model-H Cu = 65.606 - 0.22914*LI 0.029 0.367 8 Table 4. Summary of Multiple Linear Regression (MLR) models No. Model name MLR models from different variables R R2 α Rank based on R,α and R2 1 Model-1 Cu= 163.309+.559* LI -2.677*NMC .884 0.781 0.0 2 2 Model-2 Cu= 164.670-2.697* PL-.357*LI .864 0.746 0.0 3 3 Model-3 Cu= 224.032-2.272* PL-2.485*PI .898 0.806 0.0 1 3.2.4 Validation of Predicted Value with additional test results The validation of the developed model is led by using these test results of the study area. Subject to the relative correlation coefficient(R), determinant factor (R2) & significance level(α), Model-3 (i.e. Cu= 224.032-2.272* PL-2.485*PI) is preferred among the different alternative models discussed & developed above. Consequently, from Fig.16 relation of measured (actual) and the predicted value of Cu is exhibited a little variation (R2=0.88). 4. CONCLUSION It was studied the statistical analysis of liquid limit, plastic limit, bulk density, dry density, natural moisture content, and plasticity index. In contrast, it was not significantly correlated with this study area soil's specific gravity and liquidity index. The research result would be beneficial for individuals, researchers, consultants, contractors, municipal and other government agencies involved in building construction and other structures in the study area. It is also suggested to do such statistical modeling on other areas of the country by different agencies and researchers. Furthermore, it is desirable to conduct comparative statistical modeling between Table 3. Summary of Single Linear Regression (SLR) models Table 4. Summary of Multiple Linear Regression (MLR) models 3.2.4 Validation of Predicted Value with additional test results liquid limit, plastic limit, bulk density, dry density, natural moisture content, and plasticity index. ACKNOWLEDGEMENT From the study result, undrained shear strength parameters were significantly correlated with The authors would like to thank Assosa University, Jimma Institute of Technology, Jimma 27 Bakala et al.; JERR, 20(4): 15-28, 2021; Article no.JERR.66257 7. Roy, Surendra and Das, Gurcharan.Statistical models for the prediction of shear strength parameters at Sirsa, India. International Journal of Civil and Structural Engineering. 2014;4(4):483- 498. University, and the Ethiopian Road Authority (ERA) for allowing this work to progress. This work may not be achieved without supportive sponsoring. COMPETING INTERESTS 8. Andualem, Gedeon. Developing Correlation between Dynamic Cone Penetration Index and Undrained Shear Strength of Soils Found in Debre Markos Town. Addis Ababa University, Addis Ababa Institute of Technology. Addis Ababa: A master’s thesis presented to School of graduate studies, Addis Ababa University; 2015. 8. Authors have declared that no conflicts of interest exist. © 2021 Bakala et al.; This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. REFERENCES 1. Rajeev, Jain, Pramod, Sharma and Sanja, Bhandari. Unconsolidated Undrained Shear Strength of Remolded Clays by Anns Technique.International Journal of Engineering Research and Technology (IJERT).2013;2(6):2827-2832. 9. Adunoye, G.O. Fines Content and Angle of Internal Friction of a Lateritic Soil: An Experimental Study. American Journal of Engineering Research. 2014;3(3):16-21. gy ( ) ( ) 2. Zumrawi, Magdi M. E. and Mohammed, Lina A. D.Correlation of Placement Conditions and Soil Intrinsic Properties with Shear Strength of Cohesive Soils.7th Annual Conference for Postgraduate Studies and Scientific Research - Basic Sciences and Engineering Studies. Khartoum; 2016. 10. Park, Hun Myoung. Univariate Analysis and Normality Test Using SAS, STATA, and SPSS.The Trustees of Indiana University; 2006. 11. Molavi, H Vardanjani, et al..Cleansing and preparation of data.J Oral Health Oral Epidemiol.2016; 5(4):171-185. 12. Sabine Landau and Everitt, Brian S. A Handbook of Statistical Analyses using SPSS. New York: A CRC Press Company; 2004. 3. Teferra, Alemayehu and Leikun, Mesfin. Principle of Foundation Engineering. Addis Ababa: Addis Ababa university press; 1999. 3. Teferra, Alemayehu and Leikun, Mesfin. Principle of Foundation Engineering. Addis Ababa: Addis Ababa university press; 1999. 13. Leech, L. Nancy, Barrett, Karen C. and Morgan, George A. SPSS for Intermediate Statistics: Use and Interpretation. 2nd ed. London: Lawrence Erlbaum Associates, Inc; 2005. 4. Budhu, Muni. Soil Mechanics and Foundations. 3rd Ed. New York: John Wiley & Sons, Inc; 2011. 5. Gunaratne, Manjriker. The Foundation Engineering Handbook. New York: CRC Press; 2006. 14. Khalid, Usama, et al.. Prediction of Unconfined Compressive Strength from Index Properties of Soils. Sci. Int. (Lahore). 2015;27(5):4071-4075. 6. Ethiopian Building Code Standard. Addis Ababa: Ministry of Urban Development and Construction; 2013. 28

https://openalex.org/W4392251727

https://sciendo.com/pdf/10.2478/amns-2024-0433

English

Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory

Applied mathematics and nonlinear sciences

cc-by

Applied Mathematics and Nonlinear Sciences https://www.sciendo.com Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory Yuting Liao1,† 1. School of Mechanical Engineering, Hunan Mechanical and Electrical Vocational Technical, Changsha, Hunan, 430151, China. Communicated by Z. Sabir Received January 8, 2024 Accepted January 13, 2024 Available online February 26, 2024 Communicated by Z. Sabir Received January 8, 2024 Accepted January 13, 2024 Available online February 26, 2024 Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 1. School of Mechanical Engineering, Hunan Mechanical and Electrical Vocational Technical, Changsha, Hunan, 430151, China. Applied Mathematics and Nonlinear Sciences https://www.sciendo.com Abstract With the continuous development of innovative education and the growth of the demand for personalized learning, this paper explores the evaluation of students’ learning ability and resource recommendation methods in Civic Education under the guidance of OBE (student-centered, outcome-oriented, continuous improvement) educational philosophy to better achieve the educational goals. By constructing a learning ability evaluation model for Civic Education, this paper establishes a learning ability evaluation and Civic Education resource recommendation model based on Bayesian decision making by using Bayesian Decision Theory’s a posteriori probability for uncertainty reasoning and combining the personalized recommendation function of collaborative filtering algorithm. By analyzing the data of a university’s ideological and political education course in the fall of 2020, this paper not only realizes the accurate evaluation of students’ learning ability, but also tests the model’s effectiveness in learning ability evaluation and resource recommendation. The findings show that students perform well in effective communication, collaboration and independent learning ability, whose evaluation mean values are 0.62, 0.58 and 0.55, respectively. In addition, some students’ learning ability significantly improves after receiving the recommended test questions, with the increase ranging from 34.36% to 36.67%. These findings demonstrate the model’s excellent learning ability evaluation and resource recommendation performance. The research in this paper provides valuable references for evaluating students’ learning ability and resource recommendation in Civics and Political Science education, which can help further improve students’ learning effectiveness and achieve educational goals. : Bayesian decision theory; Collaborative filtering recommendation; Learning ability evaluation; Civic 1 Introduction The daily ideological and political education of students is the main position of ideological and political education in colleges and universities, which plays an important role in the formation of the soul of college students, cultural enrichment, ability to expand, professionalism, physical and mental health development and other aspects [1-3]. Xi Jinping pointed out that “for a value to be truly effective, it must be integrated into social life so that people can perceive it and understand it in practice. We should pay attention to linking what we advocate with people’s daily lives and make efforts to put them into practice [4-6].” Ideological and political education in colleges and universities, as the main channel and the main position of ideological work, should actively implement the requirements of “changing in response to events, advancing in response to the times, and new in response to the situation,” and under the premise of keeping the overall goal unchanged, extend it vertically and expand it horizontally, and creatively give the education goal new connotations of the times, and should be synergistic and sequential, and be tightly linked from the individual to the society, from the theory to the society, from the theory to the society, from the individual to the society, from the theory to the society, and from the theory to the society. From individual to society, from theory to practice, we should implement the ideological and political education objectives of colleges and universities in the new period in order to cultivate college students to become moral and talented, all-round development of socialist modernization talents [7-9]. Under the leadership of the Party and the country’s goal, and under the guidance of the fundamental task of making people moral, the goal of ideological and political education in colleges and universities needs to be adjusted in a timely manner, focusing on the cultivation and guidance of students’ ideals and beliefs, political awareness, moral character, cultural literacy, and other aspects. Realizing the goal of ideological and political education in colleges and universities in the new period is both a theoretical and practical issue, which requires adherence to the principle of directionality and subjectivity as a prerequisite, grasping the basic laws of ideological and political education as a foundation, and realizing the unity of theoretical teaching and practical guidance [10]. Zhang, A. et al. ISSN 2444-8656 https://doi.org/10.2478/amns-2024-0433 ISSN 2444-8656 https://doi.org/10.2478/amns-2024-0433 ISSN 2444-8656 https://doi.org/10.2478/amns-2024-0433 Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 2 1 Introduction discussed the current situation of Civic and Political teaching in the Internet era and proposed a multidimensional ideological and political quality model and evaluation index for college students based on a computer platform. The multidimensional reform path of ideological and political education was analyzed to realize the comprehensive development of students’ comprehensive quality [11]. Qiao, S. et al. analyzed the ideological and political quality of members of sports clubs in colleges and universities based on the fuzzy evaluation theory and, at the same time, put forward the educational path of ideological and political education integrating sports clubs [12]. Ying, X. et al. examined and analyzed the relationship between education and trust in government and trust in religion based on data from China’s Comprehensive Social 2010 Survey and found that higher levels of education were associated with higher preferences for religious trust than trust in government [13]. Serra, G. et al. studied the politics of socialist education in Ghana, noting that the study of ideology facilitates official ideology creation and dissemination, but it also questions the application of social science education and provides a window for political engagement by foreign agents [14]. Shi, Y. analyzed the effectiveness of ideological and political education for students in engineering higher vocational colleges and universities under the background of intelligent manufacturing, put forward the basic principles of ideological and political education, combined with theory and practice, reformed and optimized ideological and political education, and adhered to the pioneering nature of values education while promoting the diversification of the individual’s value orientation [15]. Luo, P. In order to cope with the challenges of globalization and informatization to ideological and political education, based on the AMOS structural model, exploring the method and content optimization of ideological education, pointing out that college counselors need to combine different contents and Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 3 3 methods of ideological education to achieve reasonable and scientific ideological education, and to provide reasonable and effective guidance at different stages of students’ thinking [16]. Mu, Y. et al. analyzed and managed the physical and mental qualities of college students based on multidimensional data mining. The aim is to shape the healthy psychology of college students and achieve the goal of moral education work [17]. Liu, G. 1 Introduction et al., based on the existing research, elaborated on the three characteristics of the output-oriented education concept, pointed out that the existing problems are outdated teaching methods, teaching materials are out of practice, etc., and put forward relevant optimization of the teaching suggestions such as strengthening the construction of the teaching team, optimizing the teaching content and teaching materials, etc. [18]. Valor, C. et al. examined the association between students’ ethical competencies and sustainable consumption, pointing out that among them, ethical competencies are related to sustainable consumption and arguing that the development of these competencies can contribute to the development of students’ sustainable consumption attitudes and behaviors [19]. Teng, Y. et al. explored what elements of multimedia technology can play an important role in the teaching of moral education and studied the interactive approach to multimedia moral education, pointing out that this multimedia teaching approach improves the effectiveness of teaching in the classroom [20]. In this paper, we first take the three dimensions of cognitive ability, interpersonal ability, and self- competence as the entry point to construct the evaluation model of learning ability in Civic and Political Education. Secondly, the a priori probability and a posteriori probability are calculated through Bayesian decision theory, and uncertainty reasoning is carried out so as to estimate the probability of the learning ability of students in Civic and Political Education. The similarity is calculated by a collaborative filtering recommendation algorithm, the similarity ranking is selected, and specific test question resources are recommended for students with high similarity to the target students. Then, the Z ideological and political education course of universities in the fall of 2020 was selected as a research sample, and relevant data were collected and substituted into the Bayesian decision theory-based learning ability evaluation and resource recommendation model to evaluate the learning ability of students in ideological and political education. Finally, the resource recommendation effect of the constructed model is explored by testing the learning ability evaluation and the difficulty of the test questions, as well as comparing the students’ learning ability before and after answering the recommended test questions. 2.1 Objectives of Civic Education Based on OBE Concepts Civic education is an education for students’ ideals, beliefs, and values, which can help students integrate their individual values with the collective values of the society and realize the collective values of the society by realizing the individual values. Civic education, in the process of realizing the goals of changing the world, serving society, and so on, must gather the landing point in the human body and must cultivate students into excellent talents with character, blood, ability, and soul through the way of ideological guidance. The main content of the OBE education concept includes “student-centered, result-oriented, continuous improvement,” which can be applied to the exploration of the realization of ideological and political education goals in colleges and universities. Under the cultivation goal of ideological and political education, we take students as the center, study the learning ability of students in ideological education courses, and recommend resources. Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 2.2 Evaluation Model of Learning Competence in Civic Education To understand the learning ability of students in the Civic Education program, it is especially important to construct a scientific and reasonable evaluation model of learning ability in Civic Education. Starting from the three dimensions of cognitive ability, interpersonal ability, and self- competence, the Civic and Political Education Learning Competence Evaluation Model is shown in Table 1, which includes nine indicators, including critical thinking ability, innovation and creativity, problem-solving ability, information literacy, effective communication ability, collaboration ability, leadership ability, independent learning ability, and the development and maintenance of academic intention. Table 1. Evaluation model of learning ability of ideological and political education Ideological and political education learning ability Table 1. Evaluation model of learning ability of ideological and political education Ideological and political education learning ability Index Content Critical thinking ability A1 Evaluate, integrate, and critically analyze the information collected. Innovation and creativity A2 Creativity and innovation Problem solving ability A3 Identify existing problems, find the original context of the problem, develop a solution and put it into practice. Information literacy A4 Ability to understand and acquire information. Ability to communicate effectively A5 Organize and express your ideas clearly, listen to and accept feedback, suggestions and ideas from others, and provide appropriate advice and feedback to peers. Collaborative ability A6 Form a team with others and set team goals together, collaborate with others to complete tasks. Leadership A7 Guidance and command ability, and the effect on others. Autonomous learning ability A8 Grasp of learning objectives, and self-monitoring, evaluation, reflection and adjustment. Develop and maintain academic ideas A9 Positive and optimistic academic attitude, willing to invest time and effort, and enhance learning and research perseverance, overcome academic difficulties 3 Evaluation of Learning Ability and Resource Recommendation Model for Civic Table 1. Evaluation model of learning ability of ideological and political education Ideological and political education learning ability 3 Evaluation of Learning Ability and Resource Recommendation Model for Civic Education in Colleges and Universities 3.1 Bayesian decision theory Bayesian decision-making describes uncertainty in terms of a posteriori probabilities in the presence of incomplete information and performs uncertainty inference based on a posteriori probabilities. In Bayesian decision-making, a posteriori probabilities are calculated using Bayes’ theorem. Bayesian decision theory provides a method for optimal decision-making given currently known information. 1) Bayes’ Theorem 1) Bayes’ Theorem 1) Bayes’ Theorem 1) Bayes’ Theorem The a posteriori probability is also exhaustive in nature, i.e.: ( ) 0 1 1 P | m i i c = =  x (2) (2) Define conditional probability: knowing that an unknown sample belongs to class c , the probability that the sample is 0 x is the conditional probability that 0 x is in class c , i.e. ( ) 0 | P c =  x x x , abbreviated as ( ) 0 | P c x . Conditional probability gives the probability distribution of a class, i.e., the probability distribution of a class in the sample space. If the sample space is a continuous sample space consisting of continuous attributes, the conditional probability can be expressed as a probability density function. Bayes’ theorem describes the relationship between prior probability, posterior probability, and conditional probability as follows: ( ) ( ) ( ) ( ) ( ) ( ) 0 0 0 0 0 1 | ( ) | ( ) | | m i i i P c P c P c P c P c P P c P c = = =  x x x x x (3) (3) 2) Bayesian decision criterion 2) Bayesian decision criterion 1) Bayes’ Theorem Bayesian decision theory is the basic theory of statistical pattern recognition. Suppose the classification problem to be studied has m category and the category space is   1, m c c =  Y . Define a priori probability: the a priori probability of a class c is the probability that an unknown sample in the sample space X belongs to that class before any observation, i.e., P( | ) c   x x X , abbreviated as ( ) P c . Define a priori probability: the a priori probability of a class c is the probability that an unknown sample in the sample space X belongs to that class before any observation, i.e., P( | ) c   x x X , abbreviated as ( ) P c . The a priori probability is the probability of an event occurring based on historical information or subjective judgment. The a priori probability has not been experimentally confirmed and is a pre-test Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 5 Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 5 5 probability. Since for classification problems, a sample can and can only belong to one class, the a priori probability of all classes is exhaustive, i.e.: probability. Since for classification problems, a sample can and can only belong to one class, the a priori probability of all classes is exhaustive, i.e.: ( ) 1 1 m i i P c = =  (1) (1) Define a posteriori probability: given a definite sample 0x in the sample space, the probability that the sample belongs to class c is the a posteriori probability that class c is in the given sample of evidence 0 x , i.e., ( ) 0 | P c  = x x x , abbreviated as ( ) 0 | P c x . The a posteriori probability is a more realistic probability obtained by combining the new additional information obtained from the survey evidence and correcting the a priori probability. 2) Bayesian decision criterion 2) Bayesian decision criterion In classification problems, there are usually objectives to be considered for classification, and a suitable decision rule can be selected according to the purpose of classification. Commonly used decision rules are the minimum error rate criterion and the minimum risk criterion. (1) Minimum error rate criterion Minimum error rate criterion (1) Minimum error rate criterion In classification problems, people often want to minimize the error of classification, and from such a need, the minimum error rate can be derived by maximizing the a posteriori probability. This is the minimum error rate criterion and can be expressed as: ˆ arg max ( | ) l y P l  = Y x (4) (4) Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 6 It can be shown that maximizing the posterior probability yields the minimum expected error rate. The expected error rate can be expressed as follows, with ( | ) P e x denoting the probability of dividing x incorrectly: It can be shown that maximizing the posterior probability yields the minimum expected error rate. (1) Minimum error rate criterion Since the a posteriori probability can be calculated by Bayes theorem, the decision rule can be further expressed as: ˆ arg max ( | ) ( | ) ( ) arg max ( ) arg max ( | ) ( ) l l l y P l P l P l P P l P l    = = = Y Y Y x x x x (8) (8) (2) Minimum Risk Criterion (1) Minimum error rate criterion The expected error rate can be expressed as follows, with ( | ) P e x denoting the probability of dividing x incorrectly: ( ) ( | ) ( ) P e P e P d  =  x X x x x (5) ( ) ( | ) ( ) P e P e P d  =  x X x x x (5) (5) For any point x , if it is assigned to class l , the error rate is: For any point x , if it is assigned to class l , the error rate is: For any point x , if it is assigned to class l , the error rate is: ( )  | ( ) ( | ) ( ) ( )(1 ( | )) l c l P e P P c P P P l − = = −  Y x x x x x x (6) (6) Thus, minimizing the expected error rate ( ) P e is equivalent to minimizing ( ) | ( ) l P e P x x for each x , i.e: x , i.e: ( ) ˆ arg min | ( ) arg min ( )(1 ( | )) arg min ( | ) l l l l y P e P P P l P l    = = − = Y Y Y x x x x x (7) ( ) ˆ arg min | ( ) arg min ( )(1 ( | )) arg min ( | ) l l l l y P e P P P l P l    = = − = Y Y Y x x x x x (7) (7) That is, it maximizes the a posteriori probability. Therefore, the minimum error rate criterion is also known as the maximum a posteriori criterion. That is, it maximizes the a posteriori probability. Therefore, the minimum error rate criterion is also known as the maximum a posteriori criterion. (2) Minimum Risk Criterion The Minimum Risk Bayesian Criterion is the Bayesian decision-making in consideration of various errors that may cause different losses. The minimum risk criterion needs to provide a table of decisions representing the losses caused by different decisions. In a classification problem with a total of m class, the decision space A has m different decisions corresponding to assignments to different classes, assuming that the loss of taking the sample in class i , ic , to take decision j , j a , to assign it to class j is ( ) | j i a c  . The conditional risk ( ) | i R a x of taking sample x to decision ia to assign it to class i is: ( ) ( ) ( ) 1 | | | m i i j j j R a a c P c  = = x x (9) (9) Minimizing the conditional risk for each x given minimizes the expected risk. Thus the minimum risk criterion can be expressed as: Minimizing the conditional risk for each x given minimizes the expected risk. Thus the minimum risk criterion can be expressed as: Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 7 Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 7 ( ) ( ) ( ) 1 ˆ arg min ( | ) arg min | | l m j j j l j y R l l c P c P c    = = =  Y Y x x (10) (10) The 0-1 loss function is a commonly used loss function that has a loss of 0 for all correct decisions and a loss of 1 for all incorrect decisions, i.e.: The 0-1 loss function is a commonly used loss function that has a loss of 0 for all correct decisions and a loss of 1 for all incorrect decisions, i.e.: ( ) 0 | 1, , 1 i j i j a c i m i j  =  = =    (11) (11) When using a 0-1 loss function, the minimum risk criterion is equivalent to the minimum error rate criterion. 3) Bayesian classification methods 3) Bayesian classification methods 3) Bayesian classification methods 3) Bayesian classification methods Bayesian classification decisions can be made using the minimum error and minimum risk criteria, respectively. However, the probability structure of each category in practice is not clear, so the Bayesian classification method requires an estimation of the probability structure needed for Bayesian decision-making based on the training samples. Depending on the probability estimation method, different Bayesian classifiers can be designed. Suppose the training set with n sample is ( ) ( )   1 1 , , , , n n y y =  T x x and the category space is   1, m c c =  Y . Usually the class prior probability can be estimated by dividing the number of samples in the class by the total number of samples. That is, ( ) / , i i i P c n n n = is the number of samples in the i th class ic in the training set, i.e., ( ) 1 n i j i j n I c c = = =  . The key to Bayesian classification methods lies in the estimation of the class conditional probability P( | ) c x . The more common Bayesian classification methods are plain Bayesian classification and Bayesian networks. (2) Minimum Risk Criterion It is as follows: When using a 0-1 loss function, the minimum risk criterion is equivalent to the minimum error rate criterion. It is as follows: ( ) ( ) ( ) ( ) ( ) ( ) ( ) arg min | arg min | | arg min | arg min 1 | arg max | i i j j j m j j i i i R a a c P c P c P c P c  = = = = − =   x x x x x (12) (12) (1) Plain Bayesian Classification (NBC) The Plain Bayesian algorithm is a classic algorithm for data mining. Plain Bayesian classification makes a class conditional independence assumption that simplifies the calculation of class conditional probabilities. The conditional independence assumption assumes that all attributes in each class are independent of each other. That is, it is assumed that all attributes in each class are independent of each other. Assuming that sample ( ) 1 2 , , , d x x x =  x is such that the estimation of the conditional probability can be simplified as: Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 8 ( ) ( ) ( ) 1 1 ( | ) , , | | d d i i P c P x x c P x c = = = x (13) (13) (2) Bayesian Networks (BBN) A Bayesian network consists of two parts: the network structure and the conditional probability table. Its network structure is a directed acyclic graph. Each node represents a random variable or attribute, and each arc represents a probabilistic dependency. For each variable (each node of the network), the Bayesian network maintains a conditional probability table (CPT), which maintains the conditional probabilities of each possible combination of values of the current node variable for its parent variable. In a network graph, given all the parent nodes of a variable, the variable is conditionally independent of its non-descendants in the network graph so that the estimate for the joint probability can be expressed as: ( ) ( ) 1 ( | ) | Parents , d i i i P c P x x c = = x (14) (14) Where ( ) Parents ix denotes the parent node of the i nd attribute. Where ( ) Parents ix denotes the parent node of the i nd attribute. 3.2 Collaborative Filtering Recommendation Algorithm To further improve the user similarity accuracy, a penalty factor is introduced as shown in Equation (17): Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 9 9 ( ) ( ) , 1/ log(1 | ( ) |) | ( ) | | ( ) | i N u N v u v N i sim N u N v   + =   (17) (17) ( ) N i denotes the set of behavioral feedbacks that user u and user v have had the same behavioral feedbacks for item i , which penalizes the effect of popular items in the user’s interest preference list on the user’s similarity by 1/ log(1 | ( ) |) N i + , i.e., the same behavioral feedbacks from users for the cold items are more reflective of the similarity of interest preferences between them. On the basis of the calculated user similarity, the User-based recommendation algorithm can be calculated by Equation (18) to recommend to the user the items preferred by the k users whose interest preferences are closest to theirs: ( ) N i denotes the set of behavioral feedbacks that user u and user v have had the same behavioral feedbacks for item i , which penalizes the effect of popular items in the user’s interest preference list on the user’s similarity by 1/ log(1 | ( ) |) N i + , i.e., the same behavioral feedbacks from users for the cold items are more reflective of the similarity of interest preferences between them. 3.2 Collaborative Filtering Recommendation Algorithm On the basis of the calculated user similarity, the User-based recommendation algorithm can be calculated by Equation (18) to recommend to the user the items preferred by the k users whose interest preferences are closest to theirs: , , , ( , ) ( ) u v u v v i v sim u k N i p sim r  =   (18) 2) Item-based collaborative filtering recommendation algorithm , , , ( , ) ( ) u v u v v i v sim u k N i p sim r  =   (18) (18) 2) Item-based collaborative filtering recommendation algorithm 2) Item-based collaborative filtering recommendation algorithm 2) Item-based collaborative filtering recommendation algorithm The item-based collaborative filtering recommendation algorithm will continue to recommend items with higher similarity to the items previously preferred by the user to the user. The key to the recommendation lies in the calculation of item similarity, given that ( ) N i indicates the number of users who prefer item i , and ( ) N j indicates the number of users who prefer item j , the similarity of the above items can be derived from Equation (19): , | ( ) ( ) | | ( ) | i j N i N j sim N i = (19) (19) However, Eq. (19) does not take into account the effect on item similarity caused by popular items, i.e., when an item is popular, the value of ,i j sim may be close to 1. Therefore, Eq. (20) adds a factor for penalizing items that are popular: However, Eq. (19) does not take into account the effect on item similarity caused by popular items, i.e., when an item is popular, the value of ,i j sim may be close to 1. Therefore, Eq. 3.2 Collaborative Filtering Recommendation Algorithm Collaborative filtering recommendation uses data processing technology to mine and analyze users’ historical behavioral information, discover users or projects with similar characteristics, and then use relevant algorithms to provide personalized recommendation services for users. Neighborhood-based recommendation algorithms can be subdivided into user-based collaborative filtering recommendation algorithms and project-based collaborative filtering recommendation algorithms. 1) User-based collaborative filtering recommendation algorithm User-based collaborative filtering recommendation algorithm will be recommended to the user to be recommended to other users with similar interests, preferred favorite items recommended to the user. The key to this recommendation lies in the calculation of user similarity, given user u and user v , it is stipulated that ( ) N u denotes the set of items for which user u has positive feedback on the item, and ( ) N v denotes the set of items for which user v has positive feedback on the item. The preference similarity of user u and user v can be simply calculated using equation (15) or equation (16): (16): , | ( ) ( ) | | ( ) ( ) | u v N u N v sim N u N v = (15) , | ( ) ( ) | | ( ) | | ( ) | u v N u N v sim N u N v  =  (16) , | ( ) ( ) | | ( ) ( ) | u v N u N v sim N u N v = (15) , | ( ) ( ) | | ( ) | | ( ) | u v N u N v sim N u N v  =  (16) (15) (16) It should be noted that there are times when users do not generate the same behavioral feedback for the same item, i.e., | ( ) ( ) | 0 N u N v  = . Therefore, to further reduce the computational complexity, an inverted table of users and items can be created, i.e., an ensemble of pairs of | ( ) ( ) | 0 N u N v   can be computed before proceeding to the next step. 4 Evaluation of students’ learning ability and recommendation of resources for civic education in higher education institutions The ideological and political education course of College Z in the fall of 2020 was selected as the object of study, and the data of students’ answers were collected after the data were sorted and cleaned; the total number of valid subjects was 200, and the number of valid test questions was 120. Among them, the valid test questions were given corresponding ability labels according to the evaluation model of learning ability of ideological and political education. 3.2 Collaborative Filtering Recommendation Algorithm (20) adds a factor for penalizing items that are popular: , | ( ) ( ) | | ( ) | | ( ) | i j N i N j sim N i N j =  (20) (20) In order to further improve the accuracy of the recommendation results and increase the coverage and diversity of the recommendation results, it normalizes the item similarity by the maximum value of the matrix, as shown in Equation (21): In order to further improve the accuracy of the recommendation results and increase the coverage and diversity of the recommendation results, it normalizes the item similarity by the maximum value of the matrix, as shown in Equation (21): , , , _ max i j i j j i j sim new sim sim = (21) (21) On the basis of the calculated item similarity, the k item recommended by the User-Based recommendation algorithm to the user with higher similarity to his/her previously preferred items can be calculated by Eq. (22): On the basis of the calculated item similarity, the k item recommended by the User-Based recommendation algorithm to the user with higher similarity to his/her previously preferred items can be calculated by Eq. (22): , , , ( ) ( , ) u j j i u i i N u sim j k p sim r  =   (22) (22) ( ) ( , ) i N u sim j k  where ( , ) sim j k denotes the set of k items with high similarity to item j , ,j i sim is the similarity between item i and item j , and , u i r denotes the user u preference for item i , i.e., the Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 10 higher the similarity compared to the user’s historical preference, the higher the item will be ranked in the recommendation list. 4.1 Evaluation of Civic Education Students’ Learning Ability Substituting the statistical data into the model teaching calculation, the results of the learning ability evaluation of the four students of the Civic and Political Education course are shown in Fig. 1, (a) to (d) is the learning ability evaluation of students A, B, C and D, respectively. The ability value greater than zero in the figure indicates that the student’s development of that ability is normal, and the higher the ability value, the higher the ability level. Ability values less than zero indicate that the student’s development of the ability is deficient, and they cannot accurately apply the ability to solve problems, where the mean value represents the average score of all the tested students. Student C’s learning abilities were well-balanced, with no significant shortcomings. His collaborative and creative skills were the best, with assessment values of 0.88 and 0.71, respectively, while Student A’s critical thinking skills, information literacy, and effective communication skills were better developed, with assessment values of 0.75, 0.63, and 0.85, respectively, but there were shortcomings in problem-solving skills and leadership skills, with values of -0.57 and -0.11 respectively. Student B had better developed critical thinking skills, information literacy, collaboration skills, and self- directed learning skills, all with assessment values greater than 0.6, but had shortfalls in innovative and creative skills, problem-solving skills, and development of sustaining academic intent. Student D had shortcomings in innovation, creativity, and information literacy, both with assessment values less than 0. All other learning competencies were well developed. Overall, the students of the selected Civic Education program developed better in effective communication skills, collaboration ratio, and independent learning skills, with mean values of 0.62, 0.58, and 0.55, respectively, and poorly in innovative and creative skills, development of sustaining academic intention and critical thinking, with evaluation mean values of 0.13, 0.15 and 0.21. (a) Assessment of student A’s learning ability (b) Assessment of student B’s learning ability (b) Assessment of student B’s learning ability (b) Assessment of student B’s learning ability (a) Assessment of student A’s learning ability Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 11 Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 11 (c) Assessment of student C’s learning ability (d) Assessment of student D’s learning ability Figure 1. 4.1 Evaluation of Civic Education Students’ Learning Ability Evaluation results of students’ learning ability (c) Assessment of student C’s learning ability (d) Assessment of student D’s learning ability (d) Figure 1. Evaluation results of students’ learning ability Figure 1. Evaluation results of students’ learning ability 4.2 Analysis of the Effect of Recommending Civic and Political Education Resources Randomly selected students and Civic Education topics as the model experimental object, through the ability evaluation results derived from the real data to draw an ability evaluation trend and the trend of the difficulty of the test questions, Civic Education student’s ability evaluation trend shown in Figure 2. It can be found that when the number of test questions is 1, the estimated value of the ability parameter will be around 0, not the extreme value, which effectively shows that the model based on Bayesian decision theory in this paper will achieve better estimation results even with a very small sample size. In addition, as the number of questions answered by students increases, their ability rating values will slowly converge to the boundary values of the pre-set range of values of the a priori probability of -3 and 3. With the increase in the number of questions, the learning ability of student 1 and student 2 shows an upward tendency and the maximum values of their learning ability are 2.67 and 2.62, respectively. The learning ability of student 3 and student 4 shows a downward tendency with the increase in the number of questions, and the minimum values are -2.52 and -2.62, respectively. Figure 2. Trend of students’ ability evaluation in ideological and political education Figure 2. Trend of students’ ability evaluation in ideological and political education The trend of test question difficulty in Civic Education is shown in Figure 3. With the increase in the number of respondents, the difficulty value of the test questions tends to converge to the boundary values of -3 and 3 of the range of values of the a priori probability set in advance, and it can also be found that the estimated value of the difficulty of the test questions does not easily take the extreme value when the number of respondents is 1. As the number of respondents increases, the difficulty of Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 12 questions 1, 2, and 5 gradually increases, and when the number of respondents reaches 15, their difficulty is 1.43, 1.66, and 1.77, respectively. Questions 3 and 4 decrease in difficulty as the number of respondents increases, and when the number of respondents reaches 15, their difficulty is -0.45 and -1.25, respectively. Figure 3. Difficulty trend of ideological and political education questions Figure 3. 4.2 Analysis of the Effect of Recommending Civic and Political Education Resources Although it is not excluded that there are relevant disturbing factors affecting the accuracy of the data, on the whole, the test question recommendation model based on Bayesian decision theory has a certain effect on the improvement of the student’s ability level, and it can help the students to carry out good training for the weak ability points. 4.2 Analysis of the Effect of Recommending Civic and Political Education Resources Difficulty trend of ideological and political education questions On the premise that the scientific nature of the learning ability evaluation and resource recommendation model based on Bayesian decision theory was verified above, in order to verify the validity and accuracy of the model further, the test questions were firstly recommended to the students of the Civic and Political Education course, and then the students were allowed to answer and collect the results of the answers, and then the Civic and Political Education Learning Ability Evaluation Model was utilized to calculate the learning ability of the students, and finally, the results of the calculation were compared with the ability level of the students before the recommended test questions were answered. Finally, the calculated results are compared with the ability level before answering the recommended test questions. The learning ability level of the students in the Civic and Political Education course is divided into five levels, which are recorded as Level 1, Level 2, Level 3, Level 4, and Level 5. The valid student data of this experiment is 100 items after the students answered the test questions recommended by the test question recommendation algorithm. The number of students who corresponded to the improvement of the ability was 82 students, which accounted for 82% of the students, and the number of students who had the improvement of the ability level was 15 students, which accounted for 15% of the students. There are 15 students, accounting for 15%. Among them, the learning ability levels of five random students before and after the model recommendation and their corresponding probabilities are shown in Table 2, where the probabilities represent the learning ability levels that the students may be at. Students S1, S3, S4, and S5 all improved their corresponding learning ability after answering the recommended test questions, with an improvement of 34.36%, 36.11%, 35.48, and 36.67%, respectively, and there was not much difference in the corresponding learning ability improvement among the four students. Among them, student S5 had the largest improvement in learning ability, while student S1 had the smallest improvement. Student S2’s learning ability level increased from Level 2 to Level 3 after answering the recommended test questions, and the probability increased from 27% of Level 2 to 44% of Level 3, and student S2 made more progress. 3) After answering the recommended test questions of the model, the corresponding learning ability of students S1, S3, S4, and S5 is improved by 34.36%, 36.11%, 35.48, and 36.67%, 4.2 Analysis of the Effect of Recommending Civic and Political Education Resources Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 13 Table 2. 5 students’ learning ability levels and their corresponding probabilities Student Recommendation Level 1 Level 2 Level 3 Level 4 Level 5 S1 Before recommendation 0.15 0.32 0.14 0.17 0.22 After recommendation 0.11 0.43 0.27 0.13 0.06 S2 Before recommendation 0.21 0.27 0.16 0.17 0.19 After recommendation 0.22 0.12 0.44 0.11 0.11 S3 Before recommendation 0.15 0.21 0.36 0.17 0.11 After recommendation 0.12 0.14 0.49 0.13 0.12 S4 Before recommendation 0.12 0.11 0.28 0.31 0.18 After recommendation 0.15 0.13 0.18 0.42 0.12 S5 Before recommendation 0.14 0.3 0.22 0.17 0.17 After recommendation 0.17 0.41 0.16 0.14 0.12 Through the validation experiments and students’ feedback, it can be concluded that the test question recommendation based on Bayesian decision theory is effective, which not only improves the probability of Civic Education students’ learning ability level but also enables the ability level to be improved. Although it is not excluded that there are relevant disturbing factors affecting the accuracy of the data, on the whole, the test question recommendation model based on Bayesian decision theory has a certain effect on the improvement of the student’s ability level, and it can help the students to carry out good training for the weak ability points. Table 2. 5 students’ learning ability levels and their corresponding probabilities Student Recommendation Level 1 Level 2 Level 3 Level 4 Level 5 S1 Before recommendation 0.15 0.32 0.14 0.17 0.22 After recommendation 0.11 0.43 0.27 0.13 0.06 S2 Before recommendation 0.21 0.27 0.16 0.17 0.19 After recommendation 0.22 0.12 0.44 0.11 0.11 S3 Before recommendation 0.15 0.21 0.36 0.17 0.11 After recommendation 0.12 0.14 0.49 0.13 0.12 S4 Before recommendation 0.12 0.11 0.28 0.31 0.18 After recommendation 0.15 0.13 0.18 0.42 0.12 S5 Before recommendation 0.14 0.3 0.22 0.17 0.17 After recommendation 0.17 0.41 0.16 0.14 0.12 Through the validation experiments and students’ feedback, it can be concluded that the test question recommendation based on Bayesian decision theory is effective, which not only improves the probability of Civic Education students’ learning ability level but also enables the ability level to be improved. 5 Conclusion This paper constructs a learning ability evaluation model for Civic Education under the guidance of the OBE education concept. It also combines Bayesian decision theory and collaborative filtering recommendation algorithm to establish a learning ability evaluation and resource recommendation model based on Bayesian decision theory. Taking the ideological and political education course of College Z in the fall of 2020 as the research object, the model is used to evaluate students’ learning ability, and the scientificity and effectiveness of the model are proved through experiments. The final research results are as follows: 1) Among the selected students of the ideological and political education course, the evaluation mean values of effective communication and exchangeability, collaboration ratio, and independent learning ability are 0.62, 0.58, and 0.55, respectively, and these three learning abilities of students are well developed. While the evaluation mean values of innovation and creativity, development of maintaining academic intention, and critical thinking were 0.13, 0.15, and 0.21, respectively, students need to strengthen the exercise and enhancement of these three abilities. 2) With the increase in the number of questions answered, the student’s learning ability or the difficulty of the test questions tends to converge to the boundary values of -3 and 3 of the range of values of the a priori probability set beforehand, and when the number of questions answered is 1, the estimated values of the learning ability and the difficulty of the test questions will not be taken to the extreme value, which verifies the scientific validity of the decision-making theory based on Bayesian decision-making. 3) After answering the recommended test questions of the model, the corresponding learning ability of students S1, S3, S4, and S5 is improved by 34.36%, 36.11%, 35.48, and 36.67%, Yuting Liao. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-16 14 respectively, and the learning ability level of student S2 is improved from Level 2 to Level 3, which indicates that the test question recommendation based on the Bayesian Decision Theory has good effect, and verifies the effectiveness of the model. respectively, and the learning ability level of student S2 is improved from Level 2 to Level 3, which indicates that the test question recommendation based on the Bayesian Decision Theory has good effect, and verifies the effectiveness of the model. References [1] Zhao, X. (2017). Promotion of education path of college students’ marxist values in new media environment. Boletin Tecnico/Technical Bulletin, 55(19), 319-323. [1] Zhao, X. (2017). Promotion of education path of college students’ marxist values in new media environment. Boletin Tecnico/Technical Bulletin, 55(19), 319-323. [2] An, Z., Ding, S., Lv, T., & Zhang, P. (2017). Study on the education model of socialist core values of college students in the micro era. Boletin Tecnico/Technical Bulletin, 55(20), 159-165. [2] An, Z., Ding, S., Lv, T., & Zhang, P. (2017). Study on the education model of socialist core values of college students in the micro era. Boletin Tecnico/Technical Bulletin, 55(20), 159-165. [2] An, Z., Ding, S., Lv, T., & Zhang, P. (2017). Study on the education model of socialist core values of college students in the micro era. Boletin Tecnico/Technical Bulletin, 55(20), 159-165. [3] Dong, D. (2021). Emotional rationality: a study on the adolescents’ socialization in education based on sang language. International Journal of Electrical Engineering Education, 002072092110020. [4] School, of, Information, Studies, University, & of, et al. (2017). Life values of library and information studies students and faculty. Education for Information, 10(1), 3-15. [5] Ye, S. (2017). Study on the multidimensional education method of college students’ political teaching with computer platform. Boletin Tecnico/Technical Bulletin, 55(16), 450-456. [6] Tian, G., & Tsai, W. H. (2020). Ideological education and practical training at a county party school: shaping local governance in contemporary china. The China Journal, 85, 000-000. [7] Straubhaar, R. (2019). Teaching for america across two hemispheres: comparing the ideological appeal of the teach for all teacher education model in the united states and brazil. Journal of Teacher Education. [8] Baltodano, M. P. (2019). 2017 cae presidential address: intergenerational disruption or ideological incongruency?. Anthropology & Education Quarterly, 50(4). [9] Jung, H., & Gil, J. A. (2019). Does college education make people politically liberal?: evidence from a natural experiment in south korea. Social Science Research. [10] Russell, S. G., Sirota, S. L., & Ahmed, A. K. (2019). Human rights education in south africa: ideological shifts and curricular reforms. Comparative Education Review, 63(1), 000-000. [11] Zhang, A., & Liu, F. (2017). Research on the computer-based multidimensional ideological and political education of college students. Revista de la Facultad de Ingenieria, 32(8), 303-310. [12] Qiao, S., & Zhang, Y. (2017). Research on the interaction path between ideological and political education and sports associations based on fuzzy evaluation theory. 5 Conclusion 4) By constructing a model based on Bayesian decision theory, this paper can be used in the evaluation of learning ability and resource recommendation of students in the Civic Education course, which provides the idea of optimizing the path to achieve the goal of Civic Education. However, the research in this paper has not fully considered the influence of other interfering factors, which still needs further improvement and refinement. 4) By constructing a model based on Bayesian decision theory, this paper can be used in the evaluation of learning ability and resource recommendation of students in the Civic Education course, which provides the idea of optimizing the path to achieve the goal of Civic Education. However, the research in this paper has not fully considered the influence of other interfering factors, which still needs further improvement and refinement. [16] Luo, P. (2017). Influence of university counselor’s ideological and political education on students based on amos structure model. Revista de la Facultad de Ingenieria, 32(11), 698-704. References Revista de la Facultad de Ingenieria, 32(11), 752-758. [13] Ying, X., Yunping, T., & Fenggang, Y. (2017). Does ideological education in china suppress trust in religion and foster trust in government?. Religions, 8(5), 94. [14] Serra, G., & Gerits, F. (2019). The politics of socialist education in ghana: the kwame nkrumah ideological institute, 1961–6. The Journal of African History, 60(3), 407-428. [15] Shi, Y. (2017). Research on the effectiveness of student political education in engineering higher vocational colleges based on intelligent manufacturing background. Boletin Tecnico/Technical Bulletin, 55(11), 65-70. [16] Luo, P. (2017). Influence of university counselor’s ideological and political education on students based on amos structure model. Revista de la Facultad de Ingenieria, 32(11), 698-704. Research on the realization path of ideological and political education goals in colleges and universities based on Bayesian decision theory 15 15 [17] Mu, Y. (2021). Analyzing the physical and mental quality of college students using multidimensional data mining. Mobile information systems. [17] Mu, Y. (2021). Analyzing the physical and mental quality of college students using multidimensional data mining. Mobile information systems. [18] Liu, G., & Zhang, G. (2021). Teaching reform of moral education based on the concept of output oriented education. International Journal of Electrical Engineering Education, 002072092110052. [18] Liu, G., & Zhang, G. (2021). Teaching reform of moral education based on the concept of output oriented education. International Journal of Electrical Engineering Education, 002072092110052. [18] Liu, G., & Zhang, G. (2021). Teaching reform of moral education based on the concept of output oriented education. International Journal of Electrical Engineering Education, 002072092110052. [19] Valor, C., Antonetti, P., & Merino, A. (2020). The relationship between moral competences and sustainable consumption among higher education students. Journal of Cleaner Production, 248, 119161. [19] Valor, C., Antonetti, P., & Merino, A. (2020). The relationship between moral competences and sustainable consumption among higher education students. Journal of Cleaner Production, 248, 119161. [19] Valor, C., Antonetti, P., & Merino, A. (2020). The relationship between moral competences and sustainable consumption among higher education students. Journal of Cleaner Production, 248, 119161. [20] Teng, Y., Jiang, P. X., & Wang, K. (2020). Moral education teaching in colleges and universities based on the application of multimedia technology. International Journal of Electrical Engineering Education, 002072092094059. [20] Teng, Y., Jiang, P. X., & Wang, K. (2020). Moral education teaching in colleges and universities based on the application of multimedia technology. [20] Teng, Y., Jiang, P. X., & Wang, K. (2020). Moral education teaching in colleges and universities based on the application of multimedia technology. International Journal of Electrical Engineering Education, 002072092094059. References International Journal of Electrical Engineering Education, 002072092094059.

https://openalex.org/W2804981611

https://bmcmusculoskeletdisord.biomedcentral.com/track/pdf/10.1186/s12891-018-2016-8

English

Intra- and interobserver reliability of glenoid fracture classifications by Ideberg, Euler and AO

BMC musculoskeletal disorders

cc-by

Gilbert et al. BMC Musculoskeletal Disorders (2018) 19:89 https://doi.org/10.1186/s12891-018-2016-8 Gilbert et al. BMC Musculoskeletal Disorders (2018) 19:89 https://doi.org/10.1186/s12891-018-2016-8 Open Access * Correspondence: [email protected] †Equal contributors 1Department of Trauma, Hand, Plastic and Reconstructive Surgery, Julius-Maximilians-University of Würzburg, Würzburg, Germany Full list of author information is available at the end of the article Intra- and interobserver reliability of glenoid fracture classifications by Ideberg, Euler and AO F. Gilbert1*† , L. Eden1†, R. Meffert1, F. Konietschke2, J. Lotz3, L. Bauer3 and W. Staab3 Abstract Background: Representing 3%–5% of shoulder girdle injuries scapula fractures are rare. Furthermore, approximately 1% of scapula fractures are intraarticularfractures of the glenoid fossa. Because of uncertain fracture morphology and limited experience, the treatment of glenoid fossa fractures is difficult. The glenoid fracture classification by Ideberg (1984) and Euler (1996) is still commonly used in literature. In 2013 a new glenoid fracture classification was introduced by the AO. The purpose of this study was to examine the new AO classification in clinical practice in comparison with the classifications by Ideberg and Euler. Methods: In total CT images of 84 patients with glenoid fossa fractures from 2005 to 2018 were included. Parasagittal, paracoronary and axial reconstructions were examined according to the classifications of Ideberg, Euler and the AO by 3 investigators (orthopedic surgeon, radiologist, student of medicine) at three individual time settings. Inter- and intraobserver reliability of the three classification systems were ascertained by computing Inter- and Intraclass (ICCs) correlation coefficients using Spearman’s rank correlation coefficient, 95%-confidence intervals as well as F-tests for correlation coefficients. Results: Inter- and intraobserver reliability for the AO classification showed a perspicuous coherence (R = 0.74 and R = 0.79). Low to moderate intraobserver reliability for Ideberg (R = 0.46) and Euler classification (R = 0.41) was found. Furthermore, data show a low Interobserver reliability for both Ideberg and Euler classification (R < 0.2). Both the Inter- and Intraclass reliability using AO is significantly higher than those using Ideberg and Euler (p < 0.05). Using the new AO classification, it was possible to find a proper class for every glenoid fossa fracture. On average, according to Euler classification 10 of 84 fractures were not classifiable whereas to Ideberg classification 21 of 84 fractures were not classifiable. Conclusion: The new AO classification system introduced 2013 facilitates reliable grading of glenoid fossa fractures with high inter- and intraobserver reliability in 84 patients using CT images. It should possibly be applied in order to enable a valid, reliable and consistent academic description of glenoid fossa fractures. The established classifications by Euler and Ideberg are not capable of providing a similar reliability. Keywords: Scapula, Glenoid, Fracture, Classification, Diagnosis, Reliability, Comparison © The Author(s). © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Background classification are found in the current literature. For ex- ample, Bartoníček et al. question its practical relevance and claim it includes a hypothetical fracture [14]. By contrary, ter Meulen et al. conclude that the new AO classification allows adequate characterization and dis- crimination of glenoid fracture patterns focusing on the number of fragments, fragmented articular surface area and their relation to the type of injury [15]. Comprising 3%–5% of shoulder girdle injuries scapula fractures are uncommon. Scapula fractures are uncom- mon but make up 3–5% of all shoulder girdle injuries. Moreover, intraarticular glenoid fossa fractures represent only 1% of scapula fractures [1]. However, fractures of the scapula may not only lead to serious pain but also affect regular function of the shoulder girdle leading to scapulothoracic dyskinesis, malunion, nonunion, rotator cuff dysfunction or impingement, respectively [1–3]. Hence, the purpose of this study was to examine the new AO classification in clinical practice in comparison to the classifications by Ideberg and Euler. We hypothe- sized that applying the new AO classification system leads to superior inter- and intraobserver correlation than using the classifications by Ideberg or Euler. Due to different injury mechanisms, glenoid avulsions, rim fractures and fossa fractures should be distinguished from each other. Anterior dislocations of the shoulder caused by low energy- or sports trauma are generally re- sponsible for glenoid avulsions and rim fractures [3, 4]. By contrary, high energy blunt-force mechanisms are ac- countable for glenoid fossa fractures [2, 5]. Thus, glenoid fossa fractures are frequently found with concomitant injuries to chest, head, brachial plexus and humerus. Ac- cording to Voleti 80% to 95% of glenoid fossa fractures are associated with additional injuries, whereas to van Oostveen up to 60% [1, 6]. Abstract 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Gilbert et al. BMC Musculoskeletal Disorders (2018) 19:89 Page 2 of 5 Page 2 of 5 Methods A retrospective study was performed in patients who were diagnosed with fractures of the glenoid fossa from 2007 to 2018. A total of 84 cases were found. 73 males and 11 females with a median age of 34 years (range 16– 71). The dominant hand was injured in 50 patients. Every patient had posteroanterior(PA), 30° oblique pro- nation and lateral radiographs following injury and then had a CT scan with 3D reconstruction if further evalu- ation was required for the confirmation of diagnoses. The 64-channel volume CT (Light speed® VCT XT, GE Healthcare, Milwaukee, USA) was used with the follow- ing protocol; slice thickness of 2 mm, tube voltage of 120 kV, tube current of 1.10 mAs, helical type scan, 0.5 s of rotation, and reconstruction with bone kernel. We classified glenoid fractures according to the classifica- tions by Ideberg, Euler and AO based (Tables 1, 2 and 3) on 3D CT images (Fig. 1). The treatment of scapula fractures has changed from preferably conservative to operative. Recent reviews of Zlowodzki and Lantry have shown that operative treat- ment was used for 80% of all glenoid fractures with good to excellent results in 82% of the cases [7, 8]. In general, indication for surgery depends on instability, degree of dislocation and articular surface fragment size [6]. On the one hand, there is a high number of publications about glenoid fracture treatment. On the other hand, comparing these publications is challenging because many different classification systems for description of scapula fractures were used. An orthopaedic surgeon with 10 years of experience, a consultant radiologist with 9 years of experience, and a 4th-year medical student who was trained for two months prior to the study, evaluated 84 cases of glenoid fractures three times during a 9-week interval using the classifica- tion systems by Ideberg, Euler and the AO based on para- sagittal, paracoronary and axial CT scans of the shoulder. All observers were blinded to the treatment. To determine In summary, plenty of classification systems for glen- oid fractures exist. One of the most commonly used classification is the one by Ideberg [9, 10]. Based on standard radiographs solely, Ideberg originally classified intraarticular fractures into 5 main types of fracture pat- terns. Later his classification was altered by Goss and Mayo [2, 11]. Additionally Goss showed that Ideberg’s classification has got no prognostic value [2]. Methods Table 1 Ideberg classification of glenoid fossa fractures Type Ia Anterior rim fracture Type Ib Posterior rim fracture Type II Fracture line through glenoid fossa exiting scapula laterally Type III Fracture line through glenoid fossa exiting scapula superiorly Type IV Fracture line through glenoid fossa exiting scapula medially Type Va Combination of types II and IV Type Vb Combination of types III and IV Type Vc Combination of types II, III, and IV Type VI Severe comminution Table 1 Ideberg classification of glenoid fossa fractures Type Ia Anterior rim fracture Another popular scapular fracture classification espe- cially in German literature is Euler’s classification [12]. In the subdivision for glenoid fractures Euler distin- guishes between 6 different types of fracture patterns. Moreover, types of fracture patterns can be combined. In 2013, the Orthopaedic Trauma Association (OTA) and the Arbeitsgemeinschaft für Osteosynthesefragen (AO) came up with their new OTA/AO classification of scapular fractures based on the analysis of 45 CT scans [13]. This classification divides fractures of the glenoid in 11 possible fracture patterns dividing the glenoid fossa in four quadrants. Diverse opinions about the new AO Page 3 of 5 Gilbert et al. BMC Musculoskeletal Disorders (2018) 19:89 Table 2 Euler and Rüedi classification for scapular fractures A Fractures of the body of scapula Isolated or multifragmentary B Fractures of the process B1 spine B2 coracoid B3 acromion C Fractures of scapular neck C1 anatomical neck C2 surgical neck C3 surgical neck with a. fracture clavicle and acromion b. torn CC and CA ligaments D Articular fractures D1 glenoid rim D2 glenoid fossa with D3 scapula neck and body fracture a. inferior glenoid fragment b. horizontal split of scapula c. coracoglenoid block formation d. comminuted fractures E Fracture combination with humeral head fractures the intra- and interobserver reliability of the classifica- tions, we conducted Spearman’s rank correlation and 95% confidence intervals. intraobserver reliability for both Ideberg (R = 0.46; 95% CI = −0.16–0.95) and Euler classification (R = 0.41; 95% CI = 0.02–0.87). Interobserver reliability for Ideberg and Euler classification showed no correlation (R < 0.2). The AO classification has a significantly higher reliability compared to the two competitors (p < 0.05). Furthermore, the confi- dence intervals do not overlap. Inter- and intraobserver reliability of the AO classification was higher than those of the classifications by Ideberg and Euler. Using new AO classification, every glenoid fossa fracture was classifiable. Methods On average, according to Euler classification 10 of 84 frac- tures were not classifiable whereas in the Ideberg classifi- cation 21 of 84 fractures were not classifiable. Discussion The findings of our study demonstrates that the new AO classification obtains a higher inter- and intraobserver reli- ability (R = 0.74 and R = 0.79) compared to Ideberg (R = 0.46) and Euler classifications (R = 0.41). This result is supported by the discovery that making use of the new AO classifica- tion, the observers had no problems finding a suitable class for every fracture of the glenoid fossa. By contrast, the ob- servers were not able to classify every glenoid fracture ac- cording to the classifications of Ideberg (21 of 84) and Euler (10 of 84). Therefore, the new AO classification from 2013 Statistics Data was analyzed by computing inter-and intraclass ob- server reliability measures (Spearman’s rank correlation coefficient) as well as 95% confidence intervals. Reliabil- ity is targeted with an ICC ≥0.8. Hypotheses in the dif- ferences among the reliability measures were tested using approximate F-tests for correlation coefficients as well as Fisher’s test. Results were interpreted as signifi- cant if p < 0.05 [16]. .All computations were conducted using the freely available statistical computing environ- ment R version 3.0.1 software (R Foundation for Statis- tical Computing, Vienna, Austria; http://www.r-project. org) using the psych package [17]. This retrospective study was approved by the institutional review board at our hospital. Results The interobserver and intraobserver reliability of the new AO classification showed a notable coherence (R = 0.74; 95% CI = 0.64–0.91 and R = 0.79; 95% CI = 0.68–0.84, re- spectively). We found a low to moderate correlation for Table 3 AO/OTA classification (incomplete) Scapula, extra-articular (not glenoid) (14-A) A1: Acromion A2. Coracoid A3. Body Partial articular (glenoid) (14-B) B1 Anterior rim B2 Posterior rim B3 Inferior Rim 1.1 Anterior rim, noncomminuted 1.2 Anterior rim, comminuted 2.1 Posterior rim, noncomminuted 2.2 Posterior rim, comminuted 3.1 Inferior rim, noncomminuted 3.2 Inferior rim, comminuted Total articular (glenoid) (14-C) C1 Extra-articular glenoid neck C2 Intra-articular with neck C3 Intra-articular with body 1.1 noncomminuted 1.2 comminuted 2.1 Intra-articular with neck, articular noncomminuted, neck noncomminuted 2.2 Intra-articular with neck 2.3 comminuted, articular noncomminuted Intra-articular with glenoid neck, articular comminuted Page 4 of 5 Gilbert et al. BMC Musculoskeletal Disorders (2018) 19:89 Fig. 1 Examples of glenoid fractures and in combination with the scapular body and neck Fig. 1 Examples of glenoid fractures and in combination with the scapular body and neck Fig. 1 Examples of glenoid fractures and in combination with the scapular body and neck Fig. 1 Examples of glenoid fractures and in combination with the scapular body and neck classification. These findings were supported by Neuhaus et al. found a proportion of rater agreement of 81% between 135 orthopedic surgeons in a web-based survey with 35 scapula fractures for the new AO/OTA classification, which is conclusive to our findings [18]. may facilitate the classification of glenoid fractures compared to Ideberg and Euler classification. Different opinions about the new AO classification exist in recent literature. Bartoníček et al. not only doubted the practical relevance of the AO classification, and presented a new classification themselves [14]. Our study shows that in contrast to other established classifi- cations the AO classification provides satisfying results regarding inter- and intraobserver reliability. The prac- tical relevance of a classification should be developed based on high reliability from the view point of further- ing communication among orthopedic surgeons and re- searchers, and developing appropriate treatment plans by using a common classification system. Our study includes several limitations. Having only a small sample size of 84 cases might be too limited in order to highly recommend the superiority of the new AO classification compared with Ideberg and Euler clas- sification. Our study lacks a clinical and radiological follow-up. Conclusions Received: 3 January 2018 Accepted: 21 March 2018 Competing interests Competing interests The authors declare that they have no competing interests. • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. 12. Wiedemann E. Frakturen der Skapula. Unfallchirurg. 2004;107:1124–33. 12. Wiedemann E. Frakturen der Skapula. Unfallchirurg. 2004;107:1124–33. 13. Jaeger M, et al. The AO Foundation and Orthopaedic trauma association (AO/OTA) scapula fracture classification system: focus on glenoid fossa involvement. J Shoulder Elb Surg. 2013;22:512–20. Funding Thi bli Funding This publication was funded by the German Research Foundation (DFG) and the University of Wuerzburg in the funding programme Open Access Publishing. 9. Ideberg, R. Fractures of the scapula involving glenoid fossa. in Surgery of the shoulder 63–66. 9. Ideberg, R. Fractures of the scapula involving glenoid fossa. in Surgery of the shoulder 63–66. 10. Ideberg R, Grevsten S, Larsson S. Epidemiology of scapular fractures incidence and classification of 338 fractures. Acta Orthop Scand. 1995;66:395–7. 10. Ideberg R, Grevsten S, Larsson S. Epidemiology of scapular fractures incidence and classification of 338 fractures. Acta Orthop Scand. 1995;66:395–7. 11. Mayo, K. A., Benirschke, S. K. & Mast, J. W. Displaced fractures of the glenoid fossa. Results of open reduction and internal fixation. Clin. Orthop. Relat Res 122–130 (1998). Results Therefore, the clinical impact or importance is difficult to assess. Furthermore, there might be a bias of the reliability test results because every hospital uses a different protocol of CT. Finally, not evaluating the time spent during the classification of injury, information about the immediacy of the classifications cannot be presented. However, this is the first study examining in- ter- and intraobserver reliability of 3 different classifica- tion systems (Ideberg, Euler, AO) with the aim of comparing them in one study. Using three observers working in different disciplines and with different states of knowledge, this study may allow an assumption about the convenience of the mentioned classification systems. Further studies are needed not only to prove the reliabil- ity of the AO classification but also to examine its clin- ical relevance. Ter Meulen et al. came to the result that the new AO classification enables appropriate distinction of glenoid frac- tures examining 3D CT models of 53 fractures [15]. They found a significant variation of articular surface area and number of fracture fragments among the different classes of the AO classification. This finding may be a possible ex- planation for the high inter- and intraobserver reliability of the AO classification found in our study. Additionally, ter Meulen et al. observed a significant relation between high- vs low-energy trauma and the fragmented surface area and high- vs low-energy trauma and the number of fragments, which supports a probable clinical relevance of the AO Page 5 of 5 Page 5 of 5 Gilbert et al. BMC Musculoskeletal Disorders (2018) 19:89 Authors’ contributions FG: made substantial contribution in design and conception of the study, wrote the manuscript LE: made substantial contribution in design and conception of the study, wrote the manuscript RM: made substantial contribution in design and conception of the study, revisited the manuscript critically FK: performed statistical analysis of the data and interpretation of data. Revisited the manuscript critically JL: performed acquisition of data and interpretation of data, revisited the manuscript critically LB: performed acquisition of data and interpretation of data, revisited the manuscript critically WS: made substantial contribution in design and conception of the study, revisited the manuscript critically. All authors approved the final version of the manuscript. 14. Bartoníček J, Tuček M, Klika D, Chochola A. Pathoanatomy and computed tomography classification of glenoid fossa fractures based on ninety patients. Int Orthop. 2016;40:2383–92. 14. Bartoníček J, Tuček M, Klika D, Chochola A. Pathoanatomy and computed tomography classification of glenoid fossa fractures based on ninety patients. Int Orthop. 2016;40:2383–92. 15. ter Meulen DP, Janssen SJ, Hageman MGJS, Ring DC. Quantitative three- dimensional computed tomography analysis of glenoid fracture patterns according to the AO/OTA classification. J Shoulder Elb Surg. 2016;25:269–75. 15. ter Meulen DP, Janssen SJ, Hageman MGJS, Ring DC. Quantitative three- dimensional computed tomography analysis of glenoid fracture patterns according to the AO/OTA classification. J Shoulder Elb Surg. 2016;25:269–75 16. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420–8. 16. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420–8. 17. Revelle W. Procedures for psychological, psychometric, and personality. R- package psych. 2017; https://cran.r-project.org/web/packages/psych/psych.pdf. 17. Revelle W. Procedures for psychological, psychometric, and personality. R- package psych. 2017; https://cran.r-project.org/web/packages/psych/psych.pdf. 18. Neuhaus V, Bot AGJ, Guitton TG, Ring DC. Scapula Fractures. J Orthop Trauma. 2014;28:124–9. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Acknowledgments None. 7. Zlowodzki M, Bhandari M, Zelle BA, Kregor PJ, Cole PA. Treatment of scapula fractures: systematic review of 520 fractures in 22 case series. J Orthop Trauma. 2006;20:230–3. 7. Zlowodzki M, Bhandari M, Zelle BA, Kregor PJ, Cole PA. Treatment of scapula fractures: systematic review of 520 fractures in 22 case series. J Orthop Trauma. 2006;20:230–3. 8. Lantry JM, Roberts CS, Giannoudis PV. Operative treatment of scapular fractures: a systematic review. Injury. 2008;39:271–83. 8. Lantry JM, Roberts CS, Giannoudis PV. Operative treatment of scapular fractures: a systematic review. Injury. 2008;39:271–83. References 1. Voleti PB, Namdari S, Mehta S. Fractures of the scapula. Adv Orthop. 2012; 2012:1–7. 2. Goss TP. Fractures of the glenoid cavity. J Bone Joint Surg Am. 1992;74:299 305. 3. Maquieira GJ, Espinosa N, Gerber C, Eid K. Non-operative treatment of large anterior glenoid rim fractures after traumatic anterior dislocation of the shoulder. J Bone Joint Surg Br. 2007;89:1347–51. 4. Bigliani LU, Newton PM, Steinmann SP, Connor PM, McIlveen SJ. Glenoid rim lesions associated with recurrent anterior dislocation of the shoulder. Am J Sports Med. 1998;26:41–5. 4. Bigliani LU, Newton PM, Steinmann SP, Connor PM, McIlveen SJ. Glenoid rim lesions associated with recurrent anterior dislocation of the shoulder. Am J Sports Med. 1998;26:41–5. Abbreviations O “ b 5. Guttentag IJ, Rechtine GR. Fractures of the scapula. A review of the literature Orthop Rev. 1988;17:147–58. AO: “Arbeitgsgemeinschaft Osteosynthesefragen”; CT: Computed tomography; ICC: Inter- and Intraclass correlation; kV: Kilovolt; mAs: Miliampere-seconds; mm: Millimeter; OTA: Orthopaedic Trauma Association 5. Guttentag IJ, Rechtine GR. Fractures of literature Orthop Rev. 1988;17:147–58. 6. van Oostveen, D. P. H., Temmerman, O. P. P., Burger, B. J., van Noort, A. & Robinson, M. Glenoid fractures: a review of pathology, classification, treatment and results. Acta Orthop Belg 80, 88–98 (2014). 6. van Oostveen, D. P. H., Temmerman, O. P. P., Burger, B. J., van Noort, A. & Robinson, M. Glenoid fractures: a review of pathology, classification, treatment and results. Acta Orthop Belg 80, 88–98 (2014). Received: 3 January 2018 Accepted: 21 March 2018 The 2013 newly introduced AO classification system allows reliable grading of glenoid fossa fractures with high inter- and intraobserver reliability in 84 patients using CT images. It should possibly be applied in order to enable a valid, reli- able and consistent academic description of glenoid fossa fractures. The established classifications by Euler and Ide- berg are not capable of providing a similar reliability. Consent for publication Not applicable. • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: Ethics approval and consent to participate We declare that the study has been performed in accordance with the Declaration of Helsinki and has been approved by the ethical Committee of the University of Würzburg (Ethical Committee Approval: Nr: 55/15 Date 1st Februar 2015). Written informed consent to participate in the study was obtained from each participant. 18. Neuhaus V, Bot AGJ, Guitton TG, Ring DC. Scapula Fractures. J Orthop Trauma. 2014;28:124–9. Author details 1 1Department of Trauma, Hand, Plastic and Reconstructive Surgery, Julius-Maximilians-University of Würzburg, Würzburg, Germany. 2Department of Mathematical Sciences, The University of Texas at Dallas, Richardson, TX, USA. 3Institute for Diagnostic and Interventional Radiology, University Medical Center Göttingen, Göttingen, Germany.

https://openalex.org/W4239633502

https://epress.lib.uts.edu.au/journals/index.php/sydney_journal/article/download/1505/1677

English

Warriewood

Sydney journal

cc-by

Virginia Macleod Warriewood is on Sydney's northern beaches, between Mona Vale and North Narrabeen, in the Pittwater local government area. This was once a 'wet' part of the coast. Lagoons and swamps were typical of the northern beaches and east coast of New South Wales. Narrabeen Creek flows through the middle of Warriewood, and Mullet Creek marks its southern boundary. Early nineteenth-century maps mark most of the land between the south-east corner of Pittwater across to Mona Vale Beach and south, including Warriewood Valley, as swamp. The local Guringai Aboriginal people would have found these swamps rich in food supplies – fish, birds, plants and naturally fresh water. Land grants and agriculture The first land grant in Warriewood was given to James Jenkins. In 1824 he received 1,000 acres (405 hectares) and the following year a further 250 acres (101 hectares), all of which he called Cabbage Tree Hill Farm. Cabbage tree palms were common along the peninsula. The palm fronds were used for thatch and hats, the trunks for fencing and building, and the pith was fed to livestock. By 1829 Jenkins had a weatherboard cottage, 14 horses and 440 horned cattle, and he had cleared 17 acres (7 hectares) and put up 300 rods (150 metres) of fencing. Local timbers – ironbark, stringybark, grey gum, turpentine, mahogany and forest oak – were used for building, fencing, firewood and other purposes. Later in the nineteenth century, the Macpherson family farmed Jenkins's land, which became known as Warriewood. Goods were taken to Sydney via Lane Cove Road, or overland to Manly and then by ferry to Sydney. 2(2) June 2010 2(2) June 2010 2(2) June 2010 ISSN 1835-0151 Sydney Journal http://epress.lib.uts.edu.au/ojs/index.php/sydney_journal/index http://epress.lib.uts.edu.au/ojs/index.php/sydney_journal/index Market gardens He collected tomatoes in crates from local farms in the evening and took them to market, leaving at 3 am, every day of the week. George Dunn moved there as a child in the 1920s and he remembers helping on the family farm with his eight siblings. As an adult, Dunn worked as a carrier. He collected tomatoes in crates from local farms in the evening and took them to market, leaving at 3 am, every day of the week. From 1951, the state government restricted development in this area, as it was unable to provide the necessary infrastructure – water and sewerage. No building was allowed on any block smaller than two hectares (five acres) and rural zoning was established and maintained for this area. From 1951, the state government restricted development in this area, as it was unable to provide the necessary infrastructure – water and sewerage. No building was allowed on any block smaller than two hectares (five acres) and rural zoning was established and maintained for this area. Farm to suburb By the 1960s, market gardening in the area had declined because produce shipped from other states was cheaper. As a result, some farmers redeveloped their land into nursery gardens. At the same time there was pressure from the community to allow subdivision. Some areas along Pittwater and Mona Vale Roads were rezoned to allow for commercial development. However, residential subdivision was only achieved in 1991 under the state government urban development plans. Gradually the rural character of Warriewood began to change, as the valley became a suburb. There was considerable planning for infrastructure (water, roads and other services) to accommodate more than 1,500 residential blocks and the approximately 5,000 or so people who would soon live there. Today street names in the new subdivisions – Apollo, Daydream, Jubilee and Ponderosa – are tributes to the varieties of tomatoes which once grew on the same land. Virginia Macleod was local studies librarian at Mona Vale Library, Sydney, and works as a professional historian Market gardens In 1906 the land was subdivided and advertised as suitable for both residential and farm blocks. A brochure by Henry Halloran promoting the sale referred to the 'salubrious' climate, the 'sheltered vale' protected from adverse winds, and the quality of the soil. It cited a Mr Duffy, [who] with no labor but his own, obtained 43 pounds of peas from an acre and a quarter which less than a year before was bush-land. Mr Duffy, [who] with no labor but his own, obtained 43 pounds of peas from an acre and a quarter which less than a year before was bush-land. Despite Halloran's sales pitch, market gardening seems to have only begun in the 1920s. Perhaps this had something to do with transport. In 1925 Roseville Bridge opened, and then in 1932 the Harbour Bridge, making it much easier to get goods to market in Sydney. Market gardening is typically practised on the edges of cities, where land, water supply and labour are cheaper. Warriewood became known as 'Glass City', because at one time there were 3,500 glasshouses there. Production was at its peak in the period between 1947 and 1954, and 100 pounds (45 kilograms) of tomatoes per glass house each year was a profitable return. Field crops such as beans, potatoes, lettuce, carrots, peas, cabbage, marrow, beetroot and capsicum were also grown, but not under glass. Sydney Journal is part of the Dictionary of Sydney project www.dictionaryofsydney.org 66 2(2) June 2010 2(2) June 2010 Sydney Journal Warriewood Virginia Macleod pp 66–67 Typically, market gardeners are recent migrants. This was certainly the case for Warriewood, which became home to Italians and people from the country then called Yugoslavia. Some Yugoslav families would offer work to newly arrived fellow-countrymen, and it is no surprise that the first Yugoslav club in Australia was started at Warriewood in 1939, at the corner of Vineyard Street and Warriewood Road. Several migrant families came from Broken Hill after the depression in the mining industry in 1932. Typically, market gardeners are recent migrants. This was certainly the case for Warriewood, which became home to Italians and people from the country then called Yugoslavia. Some George Dunn moved there as a child in the 1920s and he remembers helping on the family farm with his eight siblings. As an adult, Dunn worked as a carrier. Sydney Journal is part of the Dictionary of Sydney project www.dictionaryofsydney.org References Margrit Koettig, 'Ingleside Warriewood Urban Land Release Assessment of Aboriginal Sites', 1993 Tropman and Tropman Architects, 'Ingleside Warriewood Urban Land Release Heritage Study', 1993 Tropman and Tropman Architects, 'Ingleside Warriewood Urban Land Release Heritage Study', 1993 Sydney Journal is part of the Dictionary of Sydney project www dictionaryofsydney org 67 Sydney Journal is part of the Dictionary of Sydney project www.dictionaryofsydney.org 67

https://openalex.org/W3121985235

https://journals.plos.org/plosgenetics/article/file?id=10.1371/journal.pgen.1009325&type=printable

English

Ablation of DNA-methyltransferase 3A in skeletal muscle does not affect energy metabolism or exercise capacity

PLOS genetics

cc-by

Editor: Yvonne Bo¨ttcher, University of Oslo, NORWAY Received: July 29, 2020 Accepted: January 4, 2021 Published: January 29, 2021 Received: July 29, 2020 Accepted: January 4, 2021 Published: January 29, 2021 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pgen.1009325 RESEARCH ARTICLE Ablation of DNA-methyltransferase 3A in skeletal muscle does not affect energy metabolism or exercise capacity Lewin SmallID1, Lars R. IngerslevID1, Eleonora ManittaID1, Rhianna C. LakerID1, Ann N. HansenID1, Brendan DeeneyID1, Alain Carrie´ 2, Philippe Couvert2, Romain BarrèsID1* 1 Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark, 2 Sorbonne Universite´-INSERM UMR_S 1166 ICAN, Pitie´-Salpêtrière Hospital, Paris, France Lewin SmallID1, Lars R. IngerslevID1, Eleonora ManittaID1, Rhianna C. LakerID1, Ann N. HansenID1, Brendan DeeneyID1, Alain Carrie´ 2, Philippe Couvert2, Romain BarrèsID1* 1 Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark, 2 Sorbonne Universite´-INSERM UMR_S 1166 ICAN, Pitie´-Salpêtrière Hospital, Paris, France a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * [email protected] * [email protected] PLOS GENETICS PLOS GENETICS Abstract In response to physical exercise and diet, skeletal muscle adapts to energetic demands through large transcriptional changes. This remodelling is associated with changes in skele- tal muscle DNA methylation which may participate in the metabolic adaptation to extracellu- lar stimuli. Yet, the mechanisms by which muscle-borne DNA methylation machinery responds to diet and exercise and impacts muscle function are unknown. Here, we investi- gated the function of de novo DNA methylation in fully differentiated skeletal muscle. We generated muscle-specific DNA methyltransferase 3A (DNMT3A) knockout mice (mD3AKO) and investigated the impact of DNMT3A ablation on skeletal muscle DNA meth- ylation, exercise capacity and energy metabolism. Loss of DNMT3A reduced DNA methyla- tion in skeletal muscle over multiple genomic contexts and altered the transcription of genes known to be influenced by DNA methylation, but did not affect exercise capacity and whole- body energy metabolism compared to wild type mice. Loss of DNMT3A did not alter skeletal muscle mitochondrial function or the transcriptional response to exercise however did influ- ence the expression of genes involved in muscle development. These data suggest that DNMT3A does not have a large role in the function of mature skeletal muscle although a role in muscle development and differentiation is likely. OPEN ACCESS Citation: Small L, Ingerslev LR, Manitta E, Laker RC, Hansen AN, Deeney B, et al. (2021) Ablation of DNA-methyltransferase 3A in skeletal muscle does not affect energy metabolism or exercise capacity. PLoS Genet 17(1): e1009325. https://doi.org/ 10.1371/journal.pgen.1009325 Editor: Yvonne Bo¨ttcher, University of Oslo, NORWAY Editor: Yvonne Bo¨ttcher, University of Oslo, NORWAY Introduction Regulation of gene expression is a critical process in all cells and plays a defining role in achiev- ing tissue- and cell-specificity, in the face of an identical genetic code. The control of gene expression is dependent upon transcription factors which, upon activation, bind to transcrip- tion factor-specific DNA motifs in promoter and/or enhancer regions, recruit co-factors and initiate transcription. Epigenetic modifications to DNA and histones add an additional level of control to gene expression by altering the structure of DNA and subsequent accessibility of transcriptional machinery to gene regulatory elements. DNA methylation is a major epigenetic modification that occurs at the 5’ position of cytosine residues within a cytosine-guanine (CpG) dinucleotide resulting in 5-methylcytosine. The presence or absence of 5-methylcyto- sine modulates the recruitment of methyl binding domain (MBD) proteins to the DNA and depending on the genomic context, may regulate transcription factor binding [1]. Competing interests: The authors have declared that no competing interests exist. Skeletal muscle is a highly plastic tissue that displays a robust adaptive response to environ- mental stimuli such as physical exercise and diet. Muscle contraction results in large-scale transcriptional remodelling of metabolic, antioxidant and contractile genes which participate in the adaptation to the increased demands placed on the tissue [2]. This transcriptional remodelling is associated with changes in DNA methylation after both acute exercise [3–5] and exercise training [5–7]. For example, intense exercise in humans (80% maximal aerobic capacity) resulted in an immediate reduction in promoter methylation of PPARGC1A, TFAM, MEF2A and PDK4 and expression levels of these genes were upregulated 3 hours later in an intensity-dependent manner [3]. However, exercise-mediated demethylation of promoter regions is transient and re-methylation quickly occurs, within 1–3 hours [3], suggesting possi- ble directed methylation. Changes in nutritional status, either through diabetes status [8], fast- ing [9], weight loss [10] or diet [7,11] have also been shown to alter DNA methylation in skeletal muscle. Despite an abundance of evidence that environmental stimuli such as exercise and diet can affect DNA methylation in skeletal muscle [12,13], the mechanisms by which DNA methyla- tion machinery responds to these stimuli are still unclear. The regulation of DNA methylation is complex and involves multiple enzymes that actively methylate (DNA methyltransferases, DNMTs) or oxidize for potential demethylation (Ten-eleven translocation enzymes, TETs) of cytosine residues. DNA may also be demethylated passively (non-enzymatically) during cell division. Author summary Copyright: © 2021 Small et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Skeletal muscle is a plastic tissue able to adapt to environmental stimuli such as exercise and diet in order to respond to energetic demand. One of the ways in which skeletal mus- cle can rapidly react to these stimuli is DNA methylation. This is when chemical groups are attached to DNA, potentially influencing the transcription of genes. We investigated the function of DNA methylation in skeletal muscle by generating mice that lacked one of the main enzymes responsible for de novo DNA methylation, DNA methyltransferase 3A (DNMT3A), specifically in muscle. We found that loss of DNMT3A reduced DNA meth- ylation in muscle however this did not lead to differences in exercise capacity or energy Data Availability Statement: The authors confirm that all data underlying the findings are fully available without restriction. Transcriptomic and DNA methylation data have been deposited in a 1 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function publicly available database (https://www.ncbi.nlm. nih.gov/geo/query/acc.cgi?acc=GSE152349) and numerical data underlying all figures is provided in a spreadsheet in the supporting information (S2 Spreadsheet). metabolism. This suggests that DNMT3a is not involved in the adaptation of skeletal mus- cle to diet or exercise. metabolism. This suggests that DNMT3a is not involved in the adaptation of skeletal mus- cle to diet or exercise. Funding: The Novo Nordisk Foundation Center for Basic Metabolic Research is an independent research center at the University of Copenhagen, partially funded by an unrestricted donation from the Novo Nordisk Foundation (NNF18CC0034900). L.S. is supported by a research grant from the Danish Diabetes Academy, which is funded by the Novo Nordisk Foundation (NNF17SA0031406). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. mD3AKO mice have reduced skeletal muscle DNA methylation In order to delete Dnmt3a in muscle, we utilized the Cre-lox system with a muscle creatine kinase promoter driving Cre recombinase expression. mD3AKO mice had substantially reduced DNMT3A at the protein level compared to WT mice, specifically in skeletal muscle and heart and not in other tissues such as liver, kidney or brain (Fig 1A and 1B). DNA methyl- ation analysis and RNA sequencing were performed on the soleus muscle as this muscle dis- plays dynamic de-methylation and then re-methylation of metabolically important genes following contraction [3]. mRNA levels of Dnmt3a showed a ~60% reduction in the soleus, EDL and quadriceps muscles of mD3AKO mice compared to wild type (WT) mice (Figs 1C and S1A). As mCK-Cre is expressed quite late during skeletal muscle differentiation [16], the residual Dnmt3a expression likely comes from satellite cells and non-muscle cells (immune cells, fibroblasts, adipocytes). mRNA levels of Dnmt3b, the other major de novo DNA methyl- transferase isoform, showed no compensatory upregulation in mD3AKO muscle and was expressed at a much lower level than Dnmt3a (Fig 1C). To quantify DNA methylation, we per- formed reduced representation bisulfite sequencing (RRBS) on the soleus muscle of WT and mD3AKO mice. There was a small but significant reduction in global DNA methylation (per- centage of quantified CpG sites that were methylated) in the muscle of mD3AKO animals (Fig 1D). This corresponded to reduced methylation in mD3AKO muscle compared to WT in all genomic contexts examined (Fig 1E) however, the largest fold changes were in CpG islands (CGI, Fig 1F). After correcting for multiple testing (False Discovery Rate, FDR 5%), we found 1422 differentially methylated clusters between WT and mD3AKO muscle. 1386 were hypo- methylated and 36 were hypermethylated. Methylation at 9655 promoter regions was quanti- fied, 184 promoter regions were significantly hypomethylated and 13 promoter regions were significantly hypermethylated in mD3AKO muscle compared to WT (Fig 1G). Gene ontology analysis (gene ontology—biological process) of hypomethylated clusters included significantly enriched terms related to muscle development (S1B Fig), while there was no significant enrich- ment of terms from hypermethylated regions. Gene ontology analysis of hypomethylated pro- moter regions revealed only two terms: “inhibition of cysteine-type endopeptidase activity” and “zymogen inhibition” (S1C Fig). Introduction The classical model of DNA methylation proposes that DNMT1 is responsible for maintaining DNA methylation marks through cell divisions, while DNMT3A and DNMT3B are thought to be primarily responsible for de novo methylation events [14]. Early work in the field observed that knockdown of DNMT3B in muscle cells in vitro could impede fatty acid- induced methylation of the PPARGC1A promoter [8]. However, as Dnmt3a is the most highly expressed DNMT isoform in skeletal muscle [15] we decided to focus on the role of DNMT3A in differentiated skeletal muscle to investigate the importance of de novo DNA methylation in the response of skeletal muscle to environmental stimuli (exercise and diet). We generated muscle-specific DNMT3A knockout mice (mD3AKO) and investigated the impact of the loss of muscle DNMT3A on skeletal muscle DNA methylation, exercise capacity, energy metabolism and mitochondrial function. Additionally, by integrating DNA PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 2 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function methylation and transcriptomics data, we report novel DNMT3A targets in differentiated skel- etal muscle, regulated by promoter/gene body methylation. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 DNMT3A knockout in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise (C) Dnmt3 isoform mRNA expression (quantified by RNA-seq) in the soleus muscle analysed by individual t-tests, n = 12. (D) Percentage of total methylated cytosine residues quantified by RRBS in soleus muscle, analysed by t-test and (E) Quantified CpG methylation in different genomic contexts, analysed by 2-way ANOVA. (F) Fold-change of methylated cytosine residues comparing mD3AKO to WT muscle in different genomic contexts. (G) Volcano plot of differentially methylated promoter regions, red points are significant, n = 23–24 mice.  P < 0.05,  P < 0.005,  P < 0.0005, † P < 0.0005 main effect of genotype. Data show the mean (bar) with individual data points or violin plots. https://doi.org/10.1371/journal.pgen.1009325.g001 Fig 1. Ablation of DNMT3A in muscle reduces DNA methylation over multiple genomic contexts. (A) Western blot of DNMT3A protein in multiple tissues from WT and mD3AKO mice and (B) quantitation of DNMT3A protein in skeletal muscle, analysed by 2-way ANOVA, n = 3. (C) Dnmt3 isoform mRNA expression (quantified by RNA-seq) in the soleus muscle analysed by individual t-tests, n = 12. (D) Percentage of total methylated cytosine residues quantified by RRBS in soleus muscle, analysed by t-test and (E) Quantified CpG methylation in different genomic contexts, analysed by 2-way ANOVA. (F) Fold-change of methylated cytosine residues comparing mD3AKO to WT muscle in different genomic contexts. (G) Volcano plot of differentially methylated promoter regions, red points are significant, n = 23–24 mice.  P < 0.05,  P < 0.005,  P < 0.0005, † P < 0.0005 main effect of genotype. Data show the mean (bar) with individual data points or violin plots. https://doi.org/10.1371/journal.pgen.1009325.g001 Fig 2. Ablation of DNMT3A in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise. In a cohort of 32-36-week-old chow-fed mice, (A) voluntary wheel running distance and (B) speed in mice averaged for a 24-hour period from a period of 2 weeks, n = 8–10. (C) Distance run during a graded exercise tolerance test and (D) blood lactate before and directly after test, n = 6–8. Analysed by 2-way ANOVA for a main effect of genotype and a main effect of sex (A-C) or a main effect of genotype and a main effect of exercise, D. (E) Simplified schematic of acute running treadmill experiment in a separate cohort of 12-week old mice. DNMT3A knockout in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise (F) Number of significantly differentially expressed genes in the soleus muscle when comparing genotype, exercise and a genotype/ exercise interaction. Expression of exercise responsive genes (G) Nr4a3, (H) Hk2, (I) Ppargc1a and (J) Atf3 in the soleus muscle of combined male and female mice, n = 5–7. ♀♂P < 0.05 main effect of sex, # P < 0.05 main effect of exercise. Data show the mean (bar) with individual data points. https://doi.org/10.1371/journal.pgen.1009325.g002 Fig 2. Ablation of DNMT3A in skeletal muscle does not affect exercise capacity or the transcriptional response to acute Fig 2. Ablation of DNMT3A in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise. In a cohort of 32-36-week-old chow-fed mice, (A) voluntary wheel running distance and (B) speed in mice averaged for a 24-hour period from a period of 2 weeks, n = 8–10. (C) Distance run during a graded exercise tolerance test and (D) blood lactate before and directly after test, n = 6–8. Analysed by 2-way ANOVA for a main effect of genotype and a main effect of sex (A-C) or a main effect of genotype and a main effect of exercise, D. (E) Simplified schematic of acute running treadmill experiment in a separate cohort of 12-week old mice. (F) Number of significantly differentially expressed genes in the soleus muscle when comparing genotype, exercise and a genotype/ exercise interaction. Expression of exercise responsive genes (G) Nr4a3, (H) Hk2, (I) Ppargc1a and (J) Atf3 in the soleus muscle of combined male and female mice, n = 5–7. ♀♂P < 0.05 main effect of sex, # P < 0.05 main effect of exercise. Data show the mean (bar) with individual data points. https://doi.org/10.1371/journal.pgen.1009325.g002 Fig 2. Ablation of DNMT3A in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise. In a cohort of 32-36-week-old chow-fed mice, (A) voluntary wheel running distance and (B) speed in mice averaged for a 24-hour period from a period of 2 weeks, n = 8–10. (C) Distance run during a graded exercise tolerance test and (D) blood lactate before and directly after test, n = 6–8. Analysed by 2-way ANOVA for a main effect of genotype and a main effect of sex (A-C) or a main effect of genotype and a main effect of exercise, D. (E) Simplified schematic of acute running treadmill experiment in a separate cohort of 12-week old mice. DNMT3A knockout in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise In order to assess exercise capacity in mD3AKO mice, we investigated both the capacity for these mice to run voluntarily, with free access to running wheels, as well as during a forced treadmill exercise tolerance test. For both exercise regimens, we observed that males ran a shorter distance compared to females. mD3AKO mice with free access to a running wheel ran a similar distance and speed compared to WT mice (Fig 2A and 2B). Similarly, there was no genotype-specific differences in the distance run during forced exercise (Fig 2C). Blood lactate, a measure of glucose utilization during exercise, showed an expected increase after exercise however there was no effect of genotype (Fig 2D). In a separate cohort of 12-week-old mice, we assessed the transcriptional response to exercise in WT and mD3AKO mice by harvesting soleus muscle 60 minutes following a 30-minute tread- mill exercise bout comparing it to muscle from non-exercised mice (Fig 2E). We found 1615 3 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 4 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function Fig 1. Ablation of DNMT3A in muscle reduces DNA methylation over multiple genomic contexts. (A) Western blot of DNMT3A protein in multiple tissues from WT and mD3AKO mice and (B) quantitation of DNMT3A protein in skeletal muscle, analysed by 2-way ANOVA, n = 3. (C) Dnmt3 isoform mRNA expression (quantified by RNA-seq) in the soleus muscle analysed by individual t-tests, n = 12. (D) Percentage of total methylated cytosine residues quantified by RRBS in soleus muscle, analysed by t-test and (E) Quantified CpG methylation in different genomic contexts, analysed by 2-way ANOVA. (F) Fold-change of methylated cytosine residues comparing mD3AKO to WT muscle in different genomic contexts. (G) Volcano plot of differentially methylated promoter regions, red points are significant, n = 23–24 mice.  P < 0.05,  P < 0.005,  P < 0.0005, † P < 0.0005 main effect of genotype. Data show the mean (bar) with individual data points or violin plots. https://doi.org/10.1371/journal.pgen.1009325.g001 Fig 1. Ablation of DNMT3A in muscle reduces DNA methylation over multiple genomic contexts. (A) Western blot of DNMT3A protein in multiple tissues from WT and mD3AKO mice and (B) quantitation of DNMT3A protein in skeletal muscle, analysed by 2-way ANOVA, n = 3. DNMT3A knockout in skeletal muscle does not affect exercise capacity or the transcriptional response to acute exercise (F) Number of significantly differentially expressed genes in the soleus muscle when comparing genotype, exercise and a genotype/ exercise interaction. Expression of exercise responsive genes (G) Nr4a3, (H) Hk2, (I) Ppargc1a and (J) Atf3 in the soleus muscle of combined male and female mice, n = 5–7. ♀♂P < 0.05 main effect of sex, # P < 0.05 main effect of exercise. Data show the mean (bar) with individual data points. https://doi.org/10.1371/journal.pgen.1009325.g002 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 5 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function genes with a significant change in expression between the exercised and sedentary muscles (FDR 5%), while 56 were altered by genotype. However, there were no genes that had a significant exer- cise/genotype interaction (Fig 2F). Well-characterized exercise-responsive genes such as Nr4a3, Ppargc1a and Hk2 (Fig 2G–2I) showed robust increases in expression after exercise but here again, we found no interaction between exercise and genotype. Activating transcription factor 3 (Atf3) displayed the largest increase in expression after exercise (~15 fold increased, Fig 2J). Because few differences were found in exercise capacity or the transcriptional response to exer- cise between WT and mD3AKO mice, we used the methylation and RNA-seq data generated from soleus muscle to investigate the interaction between methylation and mRNA changes in response to genotype and exercise. When the changes in muscle DNA methylation were visual- ized by principal component analysis (PCA), a technique used to separate multi-dimensional data based on variance, there was a clear separation of the muscle methylation data by genotype (Fig 3A), however no clear separation by exercise (Fig 3B), suggesting that ablation of DNMT3A has a far greater effect on methylation in muscle than exercise. This was also evident when look- ing at the number of differentially methylated regions when comparing the effect of genotype to the effect of time after exercise (S2A Fig). Interestingly, the RNA-seq data showed the opposite result, with multidimensional scaling (MDS) plots highlighting a far greater separation by exercise (Fig 3D) than genotype (Fig 3C). A separation by sex from the RNA-seq data was also apparent (Fig 3C). Global DNA methylation after exercise was not altered by time after exercise however, there was a clear reduction in DNA methylation in the muscle of mD3AKO mice (S2B Fig). mD3AKO mice have normal body composition, whole-body energy expenditure, substrate utilization and glucose tolerance To investigate if DNMT3A in skeletal muscle contributes to whole-body energy metabolism, we measured body composition, whole-body energy expenditure (EE) and glucose tolerance in 30-week-old WT and mD3AKO mice fed a standard chow diet. mD3AKO had a similar body weight, lean mass and fat mass compared to WT mice (Fig 4A–4D). We observed no dif- ference in substrate utilization (RER, Fig 4E and 4F) or EE (Fig 4G and 4H) between WT and mD3AKO mice, although these measurements displayed clear diurnal variations, as expected. Male mice had a significantly higher rate of EE than female mice (Fig 4G and 4H) and there was a positive correlation between EE and lean mass regardless of sex and diet however this was not altered by genotype (Fig 4I). Glucose tolerance, fasting blood glucose, plasma insulin and NEFA levels were not affected by genotype (Fig 4J–4N) however, fasting plasma insulin levels were substantially higher in male mice (Fig 4M). Muscle mass, myofiber size and fiber composition (determined by analysis of gene transcription of slow and fast twitch muscle markers) were all similar between WT and mD3AKO mice (S3 Fig). mD3AKO mice have a normal response to high-fat feeding We next wanted to determine if skeletal muscle DNMT3A had a role in high fat diet (HFD)- induced glucose intolerance or adiposity as a previous investigation found that DNMT3A abla- tion in adipose tissue improved glucose tolerance in mice fed a HFD but not in chow-fed mice [17]. Male WT or mD3AKO mice were fed a HFD (45% calories from fat) for a period of 12 weeks after which body composition was analysed and a glucose tolerance test was performed. Indirect calorimetry was performed 2 weeks later (14 weeks of diet, Fig 5A). HFD-fed mice had substantially increased body and fat mass compared to chow-fed mice however there was no effect of genotype (Fig 5B and 5C). Similarly, there was no difference in glucose tolerance between HFD-fed mD3AKO and WT mice, although HFD-fed mice had a higher glucose excursion than chow-fed mice indicating relative glucose intolerance (Fig 5D and 5E). The PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 6 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function Fig 3. Genotype has a stronger effect on skeletal muscle DNA methylation than exercise while the opposite is true of skeletal muscle mRNA abundance. PCA of DNA methylation data from RRBS labelled for (A) genotype and (B) time after exercise. MDS plots of RNA-seq data labelled for (C) genotype and (D) time after exercise. From soleus muscle of 12-week old male and female WT and mD3AKO mice sacrificed 10, 30 or 60 minutes after 30 minutes of treadmill running, or not exercised (rest). Fig 3. Genotype has a stronger effect on skeletal muscle DNA methylation than exercise while the opposite is true of skeletal muscle mRNA abundance. PCA of DNA methylation data from RRBS labelled for (A) genotype and (B) time after exercise. MDS plots of RNA-seq data labelled for (C) genotype and (D) time after exercise. From soleus muscle of 12-week old male and female WT and mD3AKO mice sacrificed 10, 30 or 60 minutes after 30 minutes of treadmill running, or not exercised (rest). Fig 3. Genotype has a stronger effect on skeletal muscle DNA methylation than exercise while the opposite is true of skeletal muscle mRNA abundance. PCA of DNA methylation data from RRBS labelled for (A) genotype and (B) time after exercise. MDS plots of RNA-seq data labelled for (C) genotype and (D) time after exercise. mD3AKO mice have a normal response to high-fat feeding From soleus muscle of 12-week old male and female WT and mD3AKO mice sacrificed 10, 30 or 60 minutes after 30 minutes of treadmill running, or not exercised (rest) https://doi.org/10.1371/journal.pgen.1009325.g003 whole-body EE and RER of HFD-fed mD3AKO mice were similar to WT mice (Fig 5F and 5G). Independent of genotype, HFD-feeding caused an expected decrease in 24-hour RER compared to chow-fed mice indicating a greater reliance on fat oxidation (Figs 4F and 5G). Additionally, we analysed Dnmt3a expression in the muscle of WT mice fed either chow or a HFD. 16–18 weeks of high-fat feeding did not alter the expression of Dnmt3a in soleus or quadriceps muscle compared to age-matched chow-fed mice however, Dnmt3a expression was increased ~1.4 fold in the EDL of HFD-fed mice (Fig 5H). Loss of DNMT3A in muscle does not affect mitochondrial content or function As skeletal muscle is responsible for a substantial proportion of whole-body energy utili- zation [18] and DNMT3A has been reported to be present in the mitochondria [19] we 7 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 8 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 8 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function Fig 4. mD3AKO mice have normal body composition, glucose tolerance and energy expenditure. (A) Body weight, (B) lean mass, (C) fat mass and (D) percentage of body mass in chow-fed 30-week-old male and female WT and mD3AKO mice, n = 12– 20. (E) Diurnal respiratory exchange ratio and (F) average during the light and dark phase. (G) Diurnal energy expenditure and (H) total during the light and dark phase, measured in 32-36-week-old chow-fed male and female WT and mD3AKO mice, n = 4. (I) ANCOVA comparing the relationship between lean mass and energy expenditure in WT and mD3AKO mice combining sex and diet, n = 14. (J) Oral glucose tolerance test curves, (K) the area under the curve of the oGTT curves and (L) blood glucose after a 6 hour fast in chow-fed 30-week-old male and female WT and mD3AKO mice. (M) Plasma insulin and (N) plasma NEFA from blood taken after sacrifice in 6-hour fasted 36-40-week-old mice. n = 9–15. Analysed by 2-way ANOVA for a main effect of genotype and a main effect of sex. ♀♂P < 0.05 main effect of sex. Data show the mean (bar) with individual data points or mean +SD. https://doi.org/10.1371/journal.pgen.1009325.g004 https://doi.org/10.1371/journal.pgen.1009325.g004 investigated if DNMT3A is required for proper mitochondrial function. The soleus and EDL from mD3AKO mice had similar abundances of OXPHOS complexes to WT mus- cle, although EDL from male mice had a significant increase in complex III abundance compared to female mice (Fig 6A–6C). Similarly, soleus mtDNA copy number, a measure of mitochondrial content, was not different between genotypes although male mice dis- played an increase in soleus mtDNA copy number compared to females (Fig 6D). Next, we isolated muscle satellite cells from male mD3AKO and WT mice to investigate mito- chondrial function in vitro. Loss of DNMT3A in muscle does not affect mitochondrial content or function mD3AKO-derived primary myotubes had a significant (70%) reduction in mRNA levels of Dnmt3a only at day 5 of differentiation (harvested for RNA in parallel with the Seahorse experiments) (Fig 6E). Compared to myotubes derived from WT mice, 5-day differentiated primary myotubes from mD3AKO mice showed no differ- ences in basal or FCCP-uncoupled oxygen consumption (Fig 6F) or the extracellular acidification rate (Fig 6G) during Seahorse experiments. Fig 5. Ablation of DNMT3A in skeletal muscle does not affect the whole-body response to high-fat feeding. (A) 18-week-old male WT and mD3AKO mice were fed a HFD for a period of 12 weeks after which body composition and glucose tolerance was determined. Indirect calorimetry was performed at 14 weeks of HFD. (B) Body weight, (C) fat mass, (D) oral glucose tolerance test curves and (E) the area under the curve of the oGTT curves. The mean of the chow values is depicted by the dotted line, n = 9–12. (F) Respiratory exchange ratio and (G) energy expenditure, n = 6. (H) Dnmt3a expression in the soleus, EDL and quadriceps muscles of WT chow and HFD-fed mice compared to the housekeeping gene 18S, fold-change compared to WT, n = 9–10. Analysed by individual t-tests.  P < 0.05. Data show the mean (bar) with individual data points or mean +SD. https://doi org/10 1371/journal pgen 1009325 g005 Fig 5. Ablation of DNMT3A in skeletal muscle does not affect the whole-body response to high-fat feeding. (A) 18-week-old male WT and mD3AKO mice were fed a HFD for a period of 12 weeks after which body composition and glucose tolerance was determined. Indirect calorimetry was performed at 14 weeks of HFD. (B) Body weight, (C) fat mass, (D) oral glucose tolerance test curves and (E) the area under the curve of the oGTT curves. The mean of the chow values is depicted by the dotted line, n = 9–12. (F) Respiratory exchange ratio and (G) energy expenditure, n = 6. (H) Dnmt3a expression in the soleus, EDL and quadriceps muscles of WT chow and HFD-fed mice compared to the housekeeping gene 18S, fold-change compared to WT, n = 9–10. Analysed by individual t-tests.  P < 0.05. Data show the mean (bar) with individual data points or mean +SD. https://doi.org/10.1371/journal.pgen.1009325.g005 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 9 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function Fig 6. Loss of DNMT3A in muscle does not affect mitochondrial content or function Loss of DNMT3A does not affect skeletal muscle mitochondrial function. (A) Representative blot and densitometry of OXPHOS complexes in the (B) soleus and (C) EDL muscle and (D) mtDNA copy number in the soleus muscle of 12-week old mice, analysed by individual 2-way ANOVAs for a main effect of genotype and a main effect of sex, n = 11–13. (E) Dnmt3a expression in primary myotubes isolated from male WT and mD3AKO mice at differentiation day 0, 1, 3 and 5 compared to the housekeeping gene 18S, fold-change compared to WT, n = 3, analysed by individual t-tests, corrected for multiple comparisons. (F) Oxygen consumption rate and (G) extracellular acidification rate determined using the Seahorse XF Analyser Mito Stress Test in the same myotubes at differentiation day 5. Drug treatments were oligomycin, FCCP and antimycin A + rotenone. n = 3, analysed by individual t-tests of the average of each state. ♀♂P < 0.05 main effect of sex,  P < 0.005. Data show the mean (bar) with individual data points or mean +SD. Loss of DNMT3A in muscle does not affect mitochondrial content or function Loss of DNMT3A does not affect skeletal muscle mitochondrial function. (A) Representative blot and densitometry of OXPHOS complexes in the (B) soleus and (C) EDL muscle and (D) mtDNA copy number in the soleus muscle of 12-week old mice, analysed by individual 2-way ANOVAs for a main effect of genotype and a main effect of sex, n = 11–13. (E) Dnmt3a expression in primary myotubes isolated from male WT and mD3AKO mice at differentiation day 0, 1, 3 and 5 compared to the housekeeping gene 18S, fold-change compared to WT, n = 3, analysed by individual t-tests, corrected for multiple comparisons. (F) Oxygen consumption rate and (G) extracellular acidification rate determined using the Seahorse XF Analyser Mito Stress Test in the same myotubes at differentiation day 5. Drug treatments were oligomycin, FCCP and antimycin A + rotenone. n = 3, analysed by individual t-tests of the average of each state. ♀♂P < 0.05 main effect of sex,  P < 0.005. Data show the mean (bar) with individual data points or mean +SD. h //d i /10 1371/j l 1009325 006 skeletal muscle mitochondrial function. (A) Representative blot and densitometry of OXPHOS complexes in the (B) soleus Fig 6. Loss of DNMT3A does not affect skeletal muscle mitochondrial function. (A) Representative blot and densit Fig 6. Loss of DNMT3A does not affect skeletal muscle mitochondrial function. (A) Representative blot and densitometry of OXPHOS complexes in the (B) soleus and (C) EDL muscle and (D) mtDNA copy number in the soleus muscle of 12-week old mice, analysed by individual 2-way ANOVAs for a main effect of genotype and a main effect of sex, n = 11–13. (E) Dnmt3a expression in primary myotubes isolated from male WT and mD3AKO mice at differentiation day 0, 1, 3 and 5 compared to the housekeeping gene 18S, fold-change compared to WT, n = 3, analysed by individual t-tests, corrected for multiple comparisons. (F) Oxygen consumption rate and (G) extracellular acidification rate determined using the Seahorse XF Analyser Mito Stress Test in the same myotubes at differentiation day 5. Drug treatments were oligomycin, FCCP and antimycin A + rotenone. n = 3, analysed by individual t-tests of the average of each state. ♀♂P < 0.05 main effect of sex,  P < 0.005. Data show the mean (bar) with individual data points or mean +SD. Fig 6. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 Loss of DNMT3A alters methylation and transcription of genes involved in skeletal muscle development (F) Significant GO-BP (gene ontology—biological processes) terms from the differentially expressed transcripts. (G) mRNA abundance of significantly differentially expressed genes involved in muscle development from RNA-seq data, expressed as fold over WT. From soleus muscle of 12-week old male and female WT and mD3AKO mice, n = 12,  Padj < 0.05,  Padj < 0.0005. Data show the mean (bar) with individual data points. https://doi.org/10.1371/journal.pgen.1009325.g007 https://doi.org/10.1371/journal.pgen.1009325.g007 relating to embryonic development (Fig 7F). Additionally, genes related to “metabolism of xenobiotics and glutathione” were significantly downregulated while genes related to the term “muscle filament sliding” were significantly upregulated. Several genes involved in muscle cell development including Myh8, a perinatally expressed myosin heavy chain isoform and Pax7 and Myog, two transcription factors crucial for normal muscle cell development, were signifi- cantly upregulated in the skeletal muscle of mD3AKO mice (Fig 7G). Genes involved in the “development” gene ontology term have a star next to their gene symbol (Fig 7B). relating to embryonic development (Fig 7F). Additionally, genes related to “metabolism of xenobiotics and glutathione” were significantly downregulated while genes related to the term “muscle filament sliding” were significantly upregulated. Several genes involved in muscle cell development including Myh8, a perinatally expressed myosin heavy chain isoform and Pax7 and Myog, two transcription factors crucial for normal muscle cell development, were signifi- cantly upregulated in the skeletal muscle of mD3AKO mice (Fig 7G). Genes involved in the “development” gene ontology term have a star next to their gene symbol (Fig 7B). In order to validate some of the differentially methylated/expressed genes found in soleus muscle (Fig 7E) in other muscle types, we performed targeted bisulfite pyrosequencing and RT-qPCR on the EDL (white, mainly type II fibers) and quadriceps (mixed, type I and II fibers) muscles from a separate cohort of mice. We included H19 in this analysis, due to presence of a differentially methylated CpG site in the gene body and excluded Rhoh and Rab17 due to low expression levels (CPM < 1), Sim2 due to technical issues with pyrosequencing as well as the uncharacterized gene, 1600002K03Rik. Loss of DNMT3A alters methylation and transcription of genes involved in skeletal muscle development We found 56 genes differentially expressed between WT and mD3AKO soleus muscle (Fig 7A). Of those, 43 genes were significantly upregulated while 13 were significantly downregu- lated (Fig 7B). The most upregulated gene was Myh8 (~11-fold increase compared to WT) while the most downregulated was Dnmt3a. When percent methylation of promoter regions was compared to log fold-changes (log FC) of the corresponding transcripts, comparing mD3AKO muscle to WT muscle, there was a small but significant negative correlation (R = -0.026) suggesting that transcriptional changes between WT and mD3AKO muscle are on the whole not caused by altered promoter methylation (Fig 7C). Altered gene body methylation (comparing muscle between the different genotypes) had a stronger negative correlation with RNA abundance than promoter methylation (R = 0.046, Fig 7D). Promoter and gene body methylation of differentially expressed genes after exercise displayed a similar lack of correla- tion to gene expression (S4 Fig). We found 8 genes which displayed both significantly altered DNA methylation (either at the promoter region or within the gene body) and altered tran- script level between WT and mD3AKO soleus muscle, all of these had a reduction in DNA methylation and an increase in expression when comparing between the muscle of mD3AKO and WT mice (Fig 7E). Gene ontology analysis of differentially expressed genes revealed enriched terms mostly related to a downregulation of DNA methylation and DNA methylation 10 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function 11 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function Fig 7. Ablation of DNMT3A in skeletal muscle alters transcription of differentially methylated genes and genes involved in embryonic development. (A) Volcano plot of all transcripts between the muscle of WT and mD3AKO mice, significant differential expression is indicated by red dots. (B) Heatmap of differentially expressed transcripts between WT and mD3AKO muscle, displayed as LogFC compared to the mean of the WT male rest group,  represent developmental genes. Change in % methylation in (C) promoter regions and (D) gene bodies vs LogFC of RNA abundance of mD3AKO muscle compared to WT muscle, showing significant inverse correlation (simple linear regression). (E) Genes that are both significantly differentially expressed and have significantly altered DNA methylation within the promoter region or gene body between WT and mD3AKO muscle. Loss of DNMT3A alters methylation and transcription of genes involved in skeletal muscle development Similarly to the RRBS data in soleus muscle (Fig 8A), all the regions studied (H19, Aldh1l1, Agtrap, Pax7, and Eef1a1) showed a clear and significant reduction in methylation when comparing mD3AKO to WT in both the EDL and quadriceps muscles in all of the CpG sites studied, apart from the first CpG measured in the H19 cluster (Fig 8B). However, when examining levels of the corresponding transcripts, only 3 of the 5 tar- gets (H19, Aldh1l1 and Agtrap) displayed a similar increase in expression in the EDL and quadriceps muscles (Fig 8D) compared to the soleus muscle (Fig 8C) when comparing mD3AKO to WT mice. Together, these data suggest that DNMT3A targets regions in multiple muscle types that are associated with muscle type-dependent increases in gene transcription. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 Discussion There is substantial evidence that exercise and diet can alter DNA methylation in skeletal mus- cle tissue. However, it is currently unclear both by which mechanisms DNA methylation is altered and if these changes influence skeletal muscle physiology. In order to investigate the role of de novo DNA methylation in these processes, we utilized a mouse model with DNMT3A ablated specifically in mature muscle (mD3AKO mice). Due to the major role of DNMT3A in embryonic and early-life development (most whole-body DNMT3A-null mice die before 4-weeks of age [20]), we wanted to use a model in which DNMT3A was knocked out only in fully differentiated skeletal muscle fibers (Fig 6E) in order to remove the possibility that exercise or metabolic phenotypes were driven by changes in muscle differentiation. DNMT3A protein levels were also reduced in the heart however, as few differences were found in whole-body energy metabolism or the response to exercise, we did not further examine the impact of DNMT3A ablation in heart. Knockout of DNMT3A substantially altered the DNA methylome of skeletal muscle, evi- denced by robust separation of genotypes after principal component analysis. This separation was far greater than the separation following exercise, which suggests that ablation of DNMT3A has a greater impact than exercise on DNA methylation in skeletal muscle. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 12 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 13 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function Fig 8. Hypomethylation of specific regions in mD3AKO muscle is consistent over multiple skeletal muscle types. Analysis of gene body/ promoter methylation and transcription of putative DNMT3A target genes, H19, Aldh1l1, Agtrap, Pax7 and Eef1a1. (A) CpG methylation of significantly differentially methylated gene body/promoter regions from soleus RRBS data, FDR 5%, n = 3–24 (some CpG sites were only detected in some samples). (B) CpG methylation of the same regions from targeted bisulfite pyrosequencing in the EDL and quadriceps muscles of a separate cohort of male WT and mD3AKO mice, n = 10, analysed by individual t-tests between WT and mD3AKO for each muscle, corrected for multiple comparisons (total number of CpG sites analysed). (C) Significantly differentially expressed transcripts from the corresponding genes from soleus RNA-seq data, FC compared to WT, n = 12, FDR 5%. Discussion (D) Gene expression of the same genes by qPCR in the EDL and quadriceps muscles compared to the housekeeping gene 18S of a separate cohort of male WT and mD3AKO mice, FC compared to WT, n = 5–10 (due to low expression of some genes, samples with CT values over 40 were excluded), analysed by 2-way ANOVA for a main effect of genotype.  AdjP < 0.0001 for each CpG site (both muscles). † P < 0.05 main effect of genotype. Data show the mean with individual data points. https://doi.org/10.1371/journal.pgen.1009325.g008 https://doi.org/10.1371/journal.pgen.1009325.g008 However, global DNA methylation measured as a percentage of quantified CpG sites was reduced by an average of only ~1% (WT mice ~25%, mD3AKO mice ~24%). This suggests that a large proportion of DNA methylation in skeletal muscle is static after development and this methylation is likely maintained by DNMT1 rather than the de novo methyltransferases. Interestingly, this loss of DNA methylation occurs in all genomic contexts studied which sug- gests that de novo methylation catalysed by DNMT3A is not specific to promoters or CGIs. Potentially, the minor global hypomethylation of DNA caused by DNMT3A ablation may indicate a loss of the maintenance function of DNMT3A [21]. Supporting this idea is the dif- ferential methylation and transcription of the imprinted gene H19, which is normally main- tained in a semi-methylated state (paternal allele is methylated). Despite this, if we plot change in promoter DNA methylation, there is a clear bias for a loss of methylation if we compare mD3AKO to WT muscle (Fig 1F). However, global DNA methylation measured as a percentage of quantified CpG sites was reduced by an average of only ~1% (WT mice ~25%, mD3AKO mice ~24%). This suggests that a large proportion of DNA methylation in skeletal muscle is static after development and this methylation is likely maintained by DNMT1 rather than the de novo methyltransferases. Interestingly, this loss of DNA methylation occurs in all genomic contexts studied which sug- gests that de novo methylation catalysed by DNMT3A is not specific to promoters or CGIs. Potentially, the minor global hypomethylation of DNA caused by DNMT3A ablation may indicate a loss of the maintenance function of DNMT3A [21]. Supporting this idea is the dif- ferential methylation and transcription of the imprinted gene H19, which is normally main- tained in a semi-methylated state (paternal allele is methylated). Discussion Despite this, if we plot change in promoter DNA methylation, there is a clear bias for a loss of methylation if we compare mD3AKO to WT muscle (Fig 1F). We found no difference in either voluntary or forced exercise capacity between WT and mD3AKO mice. Similarly, the transcriptional response to acute exercise in skeletal muscle was the same between genotypes suggesting that de novo methylation by DNMT3A does not medi- ate exercise-responsive transcription. Comparably, we found that acute exercise causes a much more robust effect on the transcriptome than on the methylome. This would suggest that much of the transcriptomic response to acute exercise may not be driven by de novo CpG methylation/demethylation. A similarly small overlap between significantly altered promoter methylation and corresponding transcripts after exercise was found in a previous investigation performed by our lab in human skeletal muscle [7]. As acute exercise is associated with demethylation of exercise-responsive genes [3], ablation of enzymes participating in DNA demethylation such as the TET enzymes, instead of the de novo DNA methyltransferases, may have caused more effects on the transcriptome after exercise. Yet, it was speculated that de novo DNA methyltransferase activity was needed to remethylate exercise-demethylated gene promoters and shut down transcription [3]. If such a mechanism is at play, our data showing that transcription after exercise is not altered in skeletal muscle lacking DNMT3A suggest that DNMT3A does not participate in de novo methylation of exercise-demethylated genes. As skeletal muscle is responsible for a large proportion of postprandial glucose disposal [22] we examined glucose metabolism in mD3AKO mice. Similar to a previous report, ablation of DNMT3A in the muscle of chow-fed mice did not affect glucose tolerance or fasting blood glu- cose levels [23]. Additionally, there was no difference in glucose tolerance between genotypes in male mice fed a HFD for a period of 12 weeks despite an expected significant increase in body weight and fat mass and a worsening of glucose tolerance in HFD-fed mice compared to chow-fed mice. This suggests that DNMT3A does not impact on the adaptation of muscle to a high-fat diet. This was not the case in adipose tissue where ablation of DNMT3A has been reported to improve glucose tolerance in mice fed a HFD, but not in chow-fed mice [17]. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 Discussion Unlike the adipose tissue which displays a clear upregulation of Dnmt3a expression in obese PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 14 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function mice or mice fed a HFD [17,24], we found little difference in Dnmt3a expression in skeletal muscle between age matched chow and 16–18 week HFD-fed mice. This is opposed to a study reporting a ~2-fold increase in Dnmt3a expression in the skeletal muscle of 3-month HFD-fed mice compared to chow-fed controls [17]. There are several reports that suggest DNMT3A may have a role in influencing mitochon- drial DNA (mtDNA) methylation [19,25,26], potentially impacting mitochondrial function. However, the presence of cytosine methylation in mtDNA has been questioned and may be largely overestimated due to incomplete bisulfite conversion of coiled mtDNA [27]. The other de novo DNA methyltransferase isoform, DNMT3B, plays a role in the regulation of mitochon- drial density in skeletal muscle cells by fatty acid-induced methylation of the PPPARC1A pro- moter [8]. In the current study, no differences were found in skeletal muscle mitochondrial content, mitochondrial function in isolated muscle cells or energy expenditure at a whole- body level between WT and mD3AKO mice suggesting that DNMT3A is not important for the regulation of mitochondrial function or density in mature skeletal muscle. One advantage of the current study is that DNA methylation and mRNA abundance were determined from the same soleus muscle and therefore the relationship between altered DNA methylation and transcription can be examined. DNA methylation, both at promoter regions and in gene bodies, displayed a slight inverse correlation with transcript levels of the corre- sponding genes, with a stronger correlation with gene bodies than promoter regions. However, after multiple correction, only 8 genes displayed both a significant reduction in methylation (in the promoter region or gene body) and a corresponding increase in gene expression (Fig 7E). A similar lack of correlation between changed promoter methylation and gene transcription was observed in genes that were differentially expressed during exercise. This is comparable to reports that describe a low overlap between differentially methylated promoter regions and dif- ferentially expressed transcripts after exercise training [7,28] or exposure to a short term HFD in human skeletal muscle [11]. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 Discussion It is possible that differential methylation of genomic enhancer regions upstream or downstream of promoter regions may have a larger role in regulation of transcription which were not captured with this analysis, due to the bias of RRBS towards CpG islands. Genes that did change with genotype were part of gene ontology sets related to DNA methylation, glutathione metabolism and muscle filament sliding. Altered glutathione metabo- lism may be a consequence of reduced DNA methylation as the substrate for cytosine methyla- tion, s-adenosyl methionine shares the same amino acid precursors as glutathione: methionine and homocysteine [29]. Together, this dataset greatly expands the known DNMT3A targets in skeletal muscle [23,30] (evidenced by reduced promoter/gene-body methylation and a corre- sponding increase in the transcript level) and indicates possible enhancer interactions govern- ing the other differentially expressed genes. Additionally, 5 putative DNMT3A targets (H19, Aldh1l1, Agtrap, Pax7 and Eef1a1) were validated in multiple skeletal muscles (soleus, EDL and quadriceps) displaying significant promoter/gene body hypomethylation in mD3AKO muscle. In 3 of the targets (H19, Aldh1l1 and Agtrap), this was accompanied by an increase in the tran- script level of the corresponding gene in all of the muscles studied. There have now been several investigations utilizing muscle-specific DNMT3A knockout mice. When DNMT3A is knocked out employing Cre recombinase driven by a muscle satellite cell marker (Pax3), mice are smaller due to substantially reduced skeletal muscle mass and muscle regeneration is compromised [30]. When the Cre recombinase is driven by a muscle marker that is intermediate in muscle cell development (alpha-actin), mice have normal body and skeletal muscle mass however muscle regeneration is still compromised [23]. In the cur- rent study, Cre recombinase was driven by the muscle creatine kinase promoter which is expressed late in skeletal muscle cell development [16], and no effect of DNMT3A knockout was seen on body weight, skeletal muscle size or morphology, presumably as DNMT3A is still PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 15 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function present in satellite cells. However, loss of DNMT3A in skeletal muscle altered the transcription of genes involved in embryonic development and morphogenesis (Sim2, Eya4, Hoxc9, Cux1, Hoxa10, Wnt16, Elf3) as well as upregulating genes involved in muscle cell development such as Myh8, Pax7 and Myog. Animals To excise Dnmt3a, LoxP sites flanked exon 19 of Dnmt3a, which codes for a PC motif in the catalytic domain. Cre-mediated recombination would thereby lead to a frameshift for all Dnmt3a variants as previously described [33]. Muscle specific DNMT3A knockout mice were generated by crossing mice heterozygous for muscle creatine kinase promoter-driven Cre recombinase (mCK-Cre-/+) with mice homozygous for floxed Dnmt3a (Dnmt3aflox/flox, gift from Prof En Li from the Cardiovascular Research Center of Massachusetts General Hospital, USA) on a C57BL/6 background. This cross produced the experimental animals, genotyped as either WT (Dnmt3aflox/flox) or mCK-DNMT3A (Dnmt3aflox/flox mCK-Cre Tg). Mice were group-housed with littermates of the same sex (2–7 in a cage), maintained at 22˚C±1˚C, and were fed either chow diet (Altromin diet #1310) or a high-fat diet (HFD, 45% calories as fat, Research Diets D12451). Mice had ad libitum access to food and water and were kept at a 12:12h light:dark cycle. Apart from the mice in the acute exercise protocol detailed below, mice were killed between 36–40 weeks of age at 14:00 after a 6 hour fast by intraperitoneal pen- tobarbital overdose (100 mg/kg of body weight) and cervical dislocation, blood was taken by cardiac puncture (prior to cervical dislocation) and muscle tissues were rapidly dissected and snap-frozen. Ethics statement Experiments involving mice were approved by the Danish Animal Experiments Expectorate and performed according to local committee guidelines (license numbers 2011/561-92 and 2019-15-0201-01638). Discussion Overexpression of the myogenic transcription factors Myog [31] and Pax7 [32] has been reported to inhibit skeletal muscle growth and differentiation. This sug- gests that earlier deletion of DNMT3A in the process of skeletal muscle differentiation may have a more substantial effect on muscle function. In conclusion, ablation of DNMT3A in differentiated skeletal muscle does not affect skeletal muscle size, exercise capacity, glucose metabolism or mitochondrial content or function despite having a substantial effect on muscle DNA methylation. Additionally, this investigation revealed that changes in DNA methylation and gene transcription in skeletal muscle in response to acute exercise are not driven by DNMT3A and identified multiple novel DNMT3A targets over different skeletal muscle types validated by RRBS, bisulfite pyrosequen- cing and gene expression. Together with previous investigations highlighted above, our data suggest that the major role of DNMT3A in skeletal muscle is likely to be in the development and differentiation of muscle tissue either during embryonic development or muscle regenera- tion from satellite cells, rather than a role in mature skeletal muscle function. Exercise tolerance test and voluntary wheel running 32–36 week-old chow-fed mice were familiarized to the treadmill for three consecutive days with 10 min of treadmill running with 0% incline at 10 m/min finishing 2 days before the exer- cise tolerance test. On the day of the exercise tolerance test, at 14:00, mice underwent a graded treadmill test, at a 5% incline, in which they were warmed up with a speed of 10 m/min for 10 minutes and then every 4 minutes speed was increased by 2 m/min to a maximum of 30 m/ min. Mice that lagged at the charged coil for a period of 10 seconds were taken off the treadmill and the running time and distance were recorded. Blood lactate was determined by hand-held lactate meter (Lactate Pro 2, Axonlab) from tail blood, prior to and immediately following treadmill running. A separate cohort of 32–36 week-old chow-fed mice were used for volun- tary wheel running experiments. Mice were singly housed and acclimatized to wheel running cages for 1 week. After acclimatization, mice were kept in running wheel cages for an addi- tional 2 weeks and running distance and speed were monitored using bicycle computers. Acute exercise protocol for transcriptomics and methylation analysis 12-week-old male and female mice were used in these experiments. Mice were familiarized to a treadmill (Exer3/6, Columbus Instruments) for three consecutive days with 10 min of tread- mill running with 0% incline at 10 m/min. On the day of acute exercise mice performed 30 min of treadmill running with 5% incline at 16m/min. Mice were sacrificed by cervical 16 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function dislocation at 10, 30 or 60 min immediately after completion of treadmill running. A separate group of mice were sacrificed at the same time after 30 minutes on a sham treadmill arena ded- icated as “rest”. Soleus muscles were dissected and snap frozen in liquid nitrogen for later anal- ysis. RRBS was performed on the soleus muscle from all timepoints while RNA-seq was performed on the “rest” and 60 minutes after exercise timepoints. Assessment of body composition, indirect calorimetry and oral glucose tolerance test Body composition and glucose tolerance were determined in mice at 30 weeks of age (12 weeks of high-fat feeding for HFD mice). Lean and fat mass were measured using the EchoMRI-500 (EchoMRI LLC, Houston, USA) according to the manufacturers’ instructions, excluding body water. For the oral glucose tolerance test, mice were fasted for 6 hours (food removed at 8:00, oGTT started at 14:00) and glucose was administered by gavage with a laryn- geal cannula corresponding to 2.5g of glucose/kg of lean mass. Blood glucose levels were moni- tored from the tail tip using a hand-held glucometer (Contour XT, Bayer) before, and for 90 minutes following glucose administration. Indirect calorimetry was performed in 31–32 week old mice. Mice were acclimatized to the Home Cage System Phenomaster (TSE systems) cages (without running wheels) for 1 week prior to recording measurements. Gas exchange was monitored every 20 minutes for a period of 5 days and was averaged to hourly intervals for a 24-hour period. Plasma analysis Plasma insulin was analysed by ELISA (Ultra Sensitive Mouse Insulin ELISA kit, Crystal Chem, #90080). Plasma NEFA was analysed by NEFA HR(2) kit (Wako). RNA sequencing RNA was checked for quality using the Bioanalyser instrument (Agilent Technologies) and subject to the Illumina TruSeq Stranded Total RNA with Ribo-Zero Gold protocol (Illumina) and performed as described. Libraries were subjected to 75-bp paired-end sequencing on NextSeq 550 (Illumina). Approximately 32 million reads/sample were assigned to genes with 17389 genes surviving the expression threshold. Immunoblotting Tissue was homogenized in RIPA buffer containing protease (SIGMAFAST Protease Inhibitor Cocktail Tablet, Sigma) and phosphatase inhibitors (10 mM NaF, 1 mM Na3VO4) by steel bead disruption (3 x 90 sec at 30 Hz using the TissueLyser II, Qiagen). Protein amounts were normalized after determination of protein concentration using BCA protein assay kit (Pierce) in a standard laemmli buffer and subjected to SDS-PAGE, transferred to a PVDF membrane and blotted for DNMT3A (Abcam, #ab188470) or OXPHOS rodent cocktail (Abcam, #ab110413). Protein loading was determined using Bio-Rad stain-free technology. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 17 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function RNA and DNA isolation Muscle tissue was homogenized in 350–600 μl of RLT plus buffer (Qiagen) with β-mercap- toethanol using steel beads and subjected to 3 x 90 sec at 30 Hz of disruption using the Tissue- Lyser II (Qiagen). RNA and DNA were simultaneously isolated using the RNA/DNA/miRNA AllPrep kit (Qiagen) as per instructions. Primary muscle cells were lysed directly in RLT buffer and RNA extracted using the RNeasy kit (Qiagen). RNA was eluted in 45 μl of RNase free water and DNA in 60 μl of elution buffer (Qiagen). Genome-wide DNA methylation analysis Reduced Representation Bisulfite Sequencing (RRBS) was performed as described [7]. Briefly, DNA samples were incubated with MspI enzyme (NEB) in order to fragment DNA at CCGG positions to enrich for CpG regions. Bisulfite conversion was performed using the EZ DNA methylation Kit (Zymo Research) according to the manufacturer’s instructions. DNA was then PCR amplified and ligated to TruSeq (Illumina) sequencing adapters. Libraries were sub- jected to 75-bp single-end sequencing on a NextSeq 550 (Illumina). On average, 18.3 million reads/sample were aligned and 205304 CpGs had sufficient coverage in every sample. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 Bioinformatic analysis RNA sequencing reads were processed as previously described [34], with the difference that the gene model used was mm10 GENCODE Comprehensive gene annotations version 15 pri- mary assembly (https://www.gencodegenes.org/mouse/release_M15.html). RRBS reads were preprocessed with Trim Galore v. 0.4.0 (https://github.com/FelixKrueger/TrimGalore), a wrapper for cutadapt v. 1.9.1 [35], with the—rrbs flag set and aligned to the mm10 genome assembly using Bismark v. 0.18.1 [36], with default parameters and finally CpG methylation calls were summarized using the bismark2bedgraph script which is part of the Bismark soft- ware suite. Samples with less than 10 million aligned reads were observed to have different global levels of methylation and were excluded from further analysis. For each CpG, observa- tions from both strands were added together to form an overall methylation level per CpG, and only CpGs covered by eight or more reads in every sample was used for testing for differ- ential methylation. Chromosomes M, X and Y were excluded, the former due to the lack of methylation on chromosome M and the latter two since the two sexes were not equally repre- sented in all groups and the sex differences were not of interest. Testing for differential methyl- ation was performed using the procedure described by Chen et al. version 2 [37], briefly the ratio between methylated and unmethylated counts is modelled by a negative binomial linear model as implemented in the edgeR framework [38]. The linear model consisted of the main effects of genotype, time post exercise and an interaction between the two. In addition to dif- ferential methylation analysis at the single CpG level, differential methylation was also tested using counts aggregated within promoters, defined as 3000 bp up stream and 1000 bp down- stream of TSS, and gene bodies both defined by the same gene model used for RNA-seq analy- sis. Finally, CpGs were also aggregated and tested by CpG clusters where each cluster was PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 18 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function defined by scanning along the genome and grouping CpGs no more than 100bp apart. The fraction of methylation was calculated for each CpG included in the analysis and used both to annotate the differential methylation results and for performing a principal component analy- sis of the sample methylation levels. Bioinformatic analysis Gene ontology of differentially methylated clusters was performed using the Genomic Regions Enrichment of Annotations Tool (GREAT) [39] using the standard settings. Soleus mtDNA copy number was calculated from RRBS data by deter- mining the ratio of reads that aligned to chromosome M compared to total aligned reads, as previously described [40]. Primary muscle cells Muscle satellite cells were isolated from the complete hindlimbs of 30-week-old male mice (3 WT and 3 mD3AKO) as described [41]. Satellite cells were seeded at 20,000 cells/well (96-well plate format) on Seahorse XFe96 Cell Culture Microplates coated with 1% Matrigel basement membrane (Corning) and differentiated into myotubes in high glucose DMEM with 2% horse serum for 5 days. Real-time measurements of oxygen consumption rate (OCR) and extracellu- lar acidification rate (ECAR) were performed using a Seahorse XFe96 Extracellular Flux Ana- lyzer (Agilent Technologies) utilising the Seahorse XF Cell Mito Stress Test kit (Agilent Technologies). OCR and ECAR were measured under basal conditions and after injection of final concentrations of 1 μM oligomycin, 2.3 μM FCCP, or 2.55 μM antimycin A combined with 1 μM rotenone. The measured values were normalized to protein levels by lysing the cells in 50 mM NaOH (70˚C for 1 hour) and performing a BCA protein assay. 14–32 technical rep- licates (wells) were measured for each of the 3 primary muscle cell lines per genotype and an average was taken. The same satellite cells were grown and differentiated in parallel in 12-well plates for RNA extraction. Targeted bisulfite pyrosequencing Bisulfite-pyrosequencing was performed on EDL and quadriceps samples from 36–40 week- old male WT and mD3AKO mice. 1 μg of genomic DNA was bisulfite-converted using the EZ DNA Methylation-Lightning kit (Zymo Research) and 30 ng of bisulfite-converted DNA per assay was amplified with the Qiagen PyroMark PCR kit following the manufacturer’s instruc- tions. Amplicons were sequenced on a PyroMark Q48 Autoprep instrument (Qiagen). Pyrose- quencing primers were designed using the Qiagen PyroMark Assay Design Software 2.0 and are displayed in S1 Table. Targets were chosen from differentially methylated clusters from the soleus RRBS data. Data were analysed with the provided PyroMark Q48 Autoprep software. cDNA synthesis and qPCR cDNA was synthesized using the Bio-Rad iScript cDNA synthesis kit from 1 μg of RNA. qPCR was performed using conventional Sybr Green chemistry utilizing primers displayed in S1 Table. All primers were used at a final concentration of 200 nM. Relative quantification was determined by comparing samples to a standard curve of pooled cDNA and normalized to the housekeeping gene 18S. Statistical analysis Data are expressed as means, error bars depict standard deviation. For mouse experiments uti- lising both female and male mice, results were analysed by 2-way ANOVA for a main effect of genotype and a main effect of sex. When exercise was included as a factor (Fig 2), both sexes were analysed individually by 2-way ANOVA for a main effect of genotype and a main effect of exercise. When only genotype was tested, results were analysed by a student’s t-test. When comparing between different muscle types, results were analysed by 2-way ANOVA for a main effect of genotype and a main effect of muscle. N number represents individual animals or cell lines derived from individual animals. Statistical analysis was performed in GraphPad Prism software (Prism 8). A p-value of less than 0.05 was considered statistically significant. Differen- tial expression and differential methylation data were analysed for a main effect of genotype, a main effect of exercise and a genotype/exercise interaction. As there were no significant exer- cise/genotype interactions in either the RNA-seq or RRBS datasets, exercised and non-exer- cised mice were grouped together for the analysis performed in Fig 7. A false discovery rate (FDR) of less than 5% was considered significant. Data availability RNA-seq and RRBS data are archived for public access at the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE152349) under accession number GSE152349. A summarised spreadsheet of the RNA-seq data is included in the supplemental information (S1 Spreadsheet). All of the numerical data underlying the figures is available in the supplemental information (S2 Spreadsheet). Analysis of myofiber size Plantaris muscles from 12-week-old mice were frozen in isopentane and embedded in paraffin blocks. 2 μm muscle sections were stained using Picrosirius red (Direct Red 80, Sigma) for 15 minutes and dehydrated in 99% ethanol. Slides were imaged on a Zeiss Primovert light micro- scope with the objective lens at 20x magnification. Myofiber cross sectional area (CSA) was 19 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function calculated using ImageJ software (National Institutes of Health, Bethesda, MD) from at least 4 fields of view per slide and the average CSA per mouse was calculated. Acknowledgments The authors would like to thank the staff of the Department of Experimental Medicine, Uni- versity of Copenhagen for assistance with animal care. We acknowledge the Single-Cell Omics platform at the Center for Basic Metabolic Research (CBMR) for technical and computational expertise and support. S2 Spreadsheet. Raw numerical data from all figures. Raw data is displayed in an Excel spreadsheet with one sheet for each figure and supplemental figure. (XLSX) S2 Spreadsheet. Raw numerical data from all figures. Raw data is displayed in an Excel spreadsheet with one sheet for each figure and supplemental figure. (XLSX) Supporting information Change in % methylation in pro- moter regions and gene bodies at (A, D) 10, (B, E) 30 and (C, F) 60 minutes after 30 minutes of treadmill running compared to rested muscle vs LogFC of RNA abundance in muscle 60 min- utes after 30 minutes of treadmill running compared to rested muscle. A significant correlation is demonstrated by a blue line. (TIF) S4 Fig. Correlations between the expression of differentially regulated transcripts by exer- cise and the DNA methylation of corresponding regions. Change in % methylation in pro- moter regions and gene bodies at (A, D) 10, (B, E) 30 and (C, F) 60 minutes after 30 minutes of treadmill running compared to rested muscle vs LogFC of RNA abundance in muscle 60 min- utes after 30 minutes of treadmill running compared to rested muscle. A significant correlation is demonstrated by a blue line. (TIF) S1 Spreadsheet. Summarized RNA-seq data of the soleus muscle. Displaying main effect of genotype, main effect of exercise and a genotype/exercise interaction. (XLSX) S1 Spreadsheet. Summarized RNA-seq data of the soleus muscle. Displaying main effect of genotype, main effect of exercise and a genotype/exercise interaction. (XLSX) Supporting information S1 Fig. Skeletal muscle expression of Dnmt3a and gene ontology of differentially methyl- ated regions between WT and mD3AKO muscle. (A) Dnmt3a expression in the EDL and quadriceps muscles of 36-40-week-old male WT and mD3AKO mice compared to the house- keeping gene 18S, fold-change compared to WT, n = 10. Analysed by individual t-tests. Signifi- cant GO-BP (gene ontology—biological processes) terms from hypomethylated (B) clusters and (C) promoter regions when comparing mD3AKO to WT soleus muscle. There were no significant terms for hypermethylated regions. FDR 5%.  P < 0.0005. Data are the mean (bar) with individual data points. S2 Fig. Modest changes in DNA methylation in muscle after exercise. (A) Number of signif- icantly differentially methylated regions in the soleus muscle when comparing genotype, time after exercise and a genotype/exercise interaction. (B) Percentage of total methylated cytosine residues quantified by RRBS in the soleus muscle of non-exercised mice (rest) or taken 10, 30 or 60 minutes after 30 minutes of treadmill running, combined male and female. Analysed by 2-way ANOVA, n = 6–7. Data are the mean (bar) with individual data points. (TIF) S3 Fig. No difference in skeletal muscle size or morphology between mD3AKO and WT mice. (A) Soleus and tibialis muscle weights of 36–40 week-old male and female WT and mD3AKO mice, n = 15–21. (B) Representative sirius red staining of the plantaris muscle of WT and mD3AKO mice from 12-week-old mice and (C) myofiber cross-sectional area, PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 20 / 24 PLOS GENETICS The role of DNMT3A in skeletal muscle function n = 7–9. Analysed by 2-way ANOVA for a main effect of genotype and a main effect of sex. (D) Gene expression of markers of slow and fast twitch fibers from the soleus muscle of 12-week old mice (male and female combined) normalised to WT, n = 12, analysed by individ- ual t-tests. ♀♂P < 0.05 main effect of sex. Data are box plots or the mean (bar) with individual data points. (TIF) S4 Fig. Correlations between the expression of differentially regulated transcripts by exer- cise and the DNA methylation of corresponding regions. References 1. Robertson KD, Wolffe AP. DNA methylation in health and disease. Nat Rev Genet. 2000; 1: 11–19. https://doi.org/10.1038/35049533 PMID: 11262868 2. Pillon NJ, Gabriel BM, Dollet L, Smith JAB, Puig LS, Botella J, et al. Transcriptomic profiling of skeletal muscle adaptations to exercise and inactivity. Nat Commun. 2020; 11: 470. https://doi.org/10.1038/ s41467-019-13869-w PMID: 31980607 3. Barrès R, Yan J, Egan B, Treebak JT, Rasmussen M, Fritz T, et al. Acute exercise remodels promoter methylation in human skeletal muscle. Cell metabolism. 2012; 15: 405–411. https://doi.org/10.1016/j. cmet.2012.01.001 PMID: 22405075 4. Bajpeyi S, Covington JD, Taylor EM, Stewart LK, Galgani JE, Henagan TM. Skeletal Muscle PGC1α -1 Nucleosome Position and -260nt DNA Methylation Determine Exercise Response and Prevent Ectopic Lipid Accumulation in Men. Endocrinology. 2017; 158: 2190–2199. https://doi.org/10.1210/en.2017- 00051 PMID: 28398573 5. Seaborne RA, Strauss J, Cocks M, Shepherd S, O’Brien TD, Someren KA van, et al. Human Skeletal Muscle Possesses an Epigenetic Memory of Hypertrophy. Sci Rep-uk. 2018; 8: 1898. https://doi.org/ 10.1038/s41598-018-20287-3 PMID: 29382913 6. Kanzleiter T, Ja¨hnert M, Schulze G, Selbig J, Hallahan N, Schwenk RW, et al. Exercise training alters DNA methylation patterns in genes related to muscle growth and differentiation in mice. Am J Physiol- ogy Endocrinol Metabolism. 2015; 308: E912–20. https://doi.org/10.1152/ajpendo.00289.2014 PMID: 25805191 7. Laker RC, Garde C, Camera DM, Smiles WJ, Zierath JR, Hawley JA, et al. Transcriptomic and epige- netic responses to short-term nutrient-exercise stress in humans. Scientific reports. 2017; 7: 15134. https://doi.org/10.1038/s41598-017-15420-7 PMID: 29123172 8. Barrès R, Osler ME, Yan J, Rune A, Fritz T, Caidahl K, et al. Non-CpG methylation of the PGC-1alpha promoter through DNMT3B controls mitochondrial density. Cell metabolism. 2009; 10: 189–198. https:// doi.org/10.1016/j.cmet.2009.07.011 PMID: 19723495 9. Gillberg L, Ro¨nn T, Jørgensen SW, Perfilyev A, Hjort L, Nilsson E, et al. Fasting unmasks differential fat and muscle transcriptional regulation of metabolic gene sets in low versus normal birth weight men. Ebiomedicine. 2019; 47: 341–351. https://doi.org/10.1016/j.ebiom.2019.08.017 PMID: 31439477 10. Barres R, Kirchner H, Rasmussen M, Yan J, Kantor FR, Krook A, et al. Weight loss after gastric bypass surgery in human obesity remodels promoter methylation. Cell Reports. 2013; 3: 1020–7. https://doi. org/10.1016/j.celrep.2013.03.018 PMID: 23583180 11. Jacobsen SC, Brøns C, Bork-Jensen J, Ribel-Madsen R, Yang B, Lara E, et al. Effects of short-term high-fat overfeeding on genome-wide DNA methylation in the skeletal muscle of healthy young men. Diabetologia. 2012; 55: 3341–9. https://doi.org/10.1007/s00125-012-2717-8 PMID: 22961225 12. Writing – original draft: Lewin Small, Romain Barrès. Writing – review & editing: Lewin Small, Lars R. Ingerslev, Eleonora Manitta, Rhianna C. Laker, Ann N. Hansen, Philippe Couvert, Romain Barrès. Author Contributions Conceptualization: Lewin Small, Rhianna C. Laker, Romain Barrès. Conceptualization: Lewin Small, Rhianna C. Laker, Romain Barrès. Data curation: Lewin Small, Lars R. Ingerslev. Data curation: Lewin Small, Lars R. Ingerslev. Data curation: Lewin Small, Lars R. Ingerslev. Formal analysis: Lewin Small, Lars R. Ingerslev. Formal analysis: Lewin Small, Lars R. Ingerslev. Funding acquisition: Lewin Small, Romain Barrès. Funding acquisition: Lewin Small, Romain Barrès. Funding acquisition: Lewin Small, Romain Barrès. Investigation: Lewin Small, Eleonora Manitta, Rhianna C. Laker, Ann N. Hansen, Brendan Deeney. Methodology: Lewin Small, Lars R. Ingerslev, Eleonora Manitta, Rhianna C. Laker, Romain Barrès. Project administration: Romain Barrès. Project administration: Romain Barrès. Resources: Alain Carrie´, Philippe Couvert, Romain Barrès. Resources: Alain Carrie´, Philippe Couvert, Romain Barrès. Software: Lars R. Ingerslev. Software: Lars R. Ingerslev. Supervision: Romain Barrès. Validation: Lewin Small, Lars R. Ingerslev. Validation: Lewin Small, Lars R. Ingerslev. 21 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function Visualization: Lewin Small. Visualization: Lewin Small. References Jacques M, Hiam D, Craig J, Barrès R, Eynon N, Voisin S. Epigenetic changes in healthy human skele- tal muscle following exercise- a systematic review. Epigenetics. 2019; 14: 633–648. https://doi.org/10. 1080/15592294.2019.1614416 PMID: 31046576 13. Sharples AP, Stewart CE, Seaborne RA. Does skeletal muscle have an ‘epi’-memory? The role of epi- genetics in nutritional programming, metabolic disease, aging and exercise. Aging Cell. 2016; 15: 603– 16. https://doi.org/10.1111/acel.12486 PMID: 27102569 14. Jeltsch A, Jurkowska RZ. New concepts in DNA methylation. Trends Biochem Sci. 2014; 39: 310–318. https://doi.org/10.1016/j.tibs.2014.05.002 PMID: 24947342 15. Robertson KD, Uzvolgyi E, Liang G, Talmadge C, Sumegi J, Gonzales FA, et al. The human DNA methyltransferases (DNMTs) 1, 3a and 3b: coordinate mRNA expression in normal tissues and overex- pression in tumors. Nucleic Acids Res. 1999; 27: 2291–2298. https://doi.org/10.1093/nar/27.11.2291 PMID: 10325416 16. Bi P, Yue F, Sato Y, Wirbisky S, Liu W, Shan T, et al. Stage-specific effects of Notch activation during skeletal myogenesis. eLife. 2016; 5. https://doi.org/10.7554/eLife.17355 PMID: 27644105 22 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function 17. You D, Nilsson E, Tenen DE, Lyubetskaya A, Lo JC, Jiang R, et al. Dnmt3a is an epigenetic mediator of adipose insulin resistance. eLife. 2017; 6: 205. https://doi.org/10.7554/eLife.30766 PMID: 29091029 18. Zurlo F, Larson K, Bogardus C, Ravussin E. Skeletal muscle metabolism is a major determinant of rest- ing energy expenditure. Journal of Clinical Investigation. 1990; 86: 1423–1427. https://doi.org/10.1172/ JCI114857 PMID: 2243122 19. Dou X, Boyd-Kirkup JD, McDermott J, Zhang X, Li F, Rong B, et al. The strand-biased mitochondrial DNA methylome and its regulation by DNMT3A. Genome Res. 2019; 29: 1622–1634. https://doi.org/10. 1101/gr.234021.117 PMID: 31537639 20. Okano M, Bell DW, Haber DA, Li E. DNA Methyltransferases Dnmt3a and Dnmt3b Are Essential for De Novo Methylation and Mammalian Development. Cell. 1999; 99: 247–257. https://doi.org/10.1016/ s0092-8674(00)81656-6 PMID: 10555141 21. Jones PA, Liang G. Rethinking how DNA methylation patterns are maintained. Nat Rev Genet. 2009; 10: 805–811. https://doi.org/10.1038/nrg2651 PMID: 19789556 22. Ferrannini E, Bjorkman O, Reichard GA, Pilo A, Olsson M, Wahren J, et al. The disposal of an oral glu- cose load in healthy subjects. A quantitative study. Diabetes. 1985; 34: 580–588. https://doi.org/10. 2337/diab.34.6.580 PMID: 3891471 23. Hatazawa Y, Ono Y, Hirose Y, Kanai S, Fujii NL, Machida S, et al. Reduced Dnmt3a increases Gdf5 expression with suppressed satellite cell differentiation and impaired skeletal muscle regeneration. The FASEB Journal. References 2018; 32: 1452–1467. https://doi.org/10.1096/fj.201700573R PMID: 29146735 24. Kamei Y, Suganami T, Ehara T, Kanai S, Hayashi K, Yamamoto Y, et al. Increased expression of DNA methyltransferase 3a in obese adipose tissue: studies with transgenic mice. Obes Silver Spring Md. 2009; 18: 314–21. https://doi.org/10.1038/oby.2009.246 PMID: 19680236 25. Wong M, Gertz B, Chestnut BA, Martin LJ. Mitochondrial DNMT3A and DNA methylation in skeletal muscle and CNS of transgenic mouse models of ALS. Front Cell Neurosci. 2013; 7: 279. https://doi.org/ 10.3389/fncel.2013.00279 PMID: 24399935 26. Patil V, Cuenin C, Chung F, Aguilera JRR, Fernandez-Jimenez N, Romero-Garmendia I, et al. Human mitochondrial DNA is extensively methylated in a non-CpG context. Nucleic Acids Res. 2019; 47: 10072–10085. https://doi.org/10.1093/nar/gkz762 PMID: 31665742 27. Mechta M, Ingerslev LR, Fabre O, Picard M, Barrès R. Evidence Suggesting Absence of Mitochondrial DNA Methylation. Frontiers Genetics. 2017; 8: 166. https://doi.org/10.3389/fgene.2017.00166 PMID: 29163634 28. Lindholm ME, Marabita F, Gomez-Cabrero D, Rundqvist H, Ekstro¨m TJ, Tegne´r J, et al. An integrative analysis reveals coordinated reprogramming of the epigenome and the transcriptome in human skeletal muscle after training. Epigenetics. 2015; 9: 1557–1569. https://doi.org/10.4161/15592294.2014. 982445 PMID: 25484259 29. Garcı´a-Gime´nez JL, Roma´-Mateo C, Pe´rez-Machado G, Peiro´-Chova L, Pallardo´ FV. Role of glutathi- one in the regulation of epigenetic mechanisms in disease. Free Radic Biology Medicine. 2017; 112: 36–48. https://doi.org/10.1016/j.freeradbiomed.2017.07.008 PMID: 28705657 30. Naito M, Mori M, Inagawa M, Miyata K, Hashimoto N, Tanaka S, et al. Dnmt3a Regulates Proliferation of Muscle Satellite Cells via p57Kip2. Tajbakhsh S, editor. PLoS genetics. 2016; 12: e1006167. https:// doi.org/10.1371/journal.pgen.1006167 PMID: 27415617 31. Gundersen K, Rabben I, Klocke BJ, Merlie JP. Overexpression of myogenin in muscles of transgenic mice: interaction with Id-1, negative crossregulation of myogenic factors, and induction of extrasynaptic acetylcholine receptor expression. Mol Cell Biol. 1995; 15: 7127–7134. https://doi.org/10.1128/mcb.15. 12.7127 PMID: 8524280 32. Olguin HC, Olwin BB. Pax-7 up-regulation inhibits myogenesis and cell cycle progression in satellite cells: a potential mechanism for self-renewal. Dev Biol. 2004; 275: 375–388. https://doi.org/10.1016/j. ydbio.2004.08.015 PMID: 15501225 33. Kaneda M, Okano M, Hata K, Sado T, Tsujimoto N, Li E, et al. Essential role for de novo DNA methyl- transferase Dnmt3a in paternal and maternal imprinting. Nature. 2004; 429: 900–903. https://doi.org/ 10.1038/nature02633 PMID: 15215868 34. Williams K, Ingerslev LR, Bork-Jensen J, Wohlwend M, Hansen AN, Small L, et al. Skeletal muscle enhancer interactions identify genes controlling whole-body metabolism. Nat Commun. 2020; 11: 2695. https://doi.org/10.1038/s41467-020-16537-6 PMID: 32483258 35. Martin M. PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 References Cutadapt removes adapter sequences from high-throughput sequencing reads. Embnet J. 2011; 17: 10–12. https://doi.org/10.14806/ej.17.1.200 36. Krueger F, Andrews SR. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applica- tions. Bioinform Oxf Engl. 2011; 27: 1571–2. https://doi.org/10.1093/bioinformatics/btr167 PMID: 21493656 23 / 24 PLOS Genetics | https://doi.org/10.1371/journal.pgen.1009325 January 29, 2021 PLOS GENETICS The role of DNMT3A in skeletal muscle function 37. Chen Y, Pal B, Visvader JE, Smyth GK. Differential methylation analysis of reduced representation bisulfite sequencing experiments using edgeR. F1000research. 2017; 6: 2055. https://doi.org/10. 12688/f1000research.13196.2 PMID: 29333247 38. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinform Oxf Engl. 2009; 26: 139–40. https://doi.org/10.1093/ bioinformatics/btp616 PMID: 19910308 39. McLean CY, Bristor D, Hiller M, Clarke SL, Schaar BT, Lowe CB, et al. GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol. 2010; 28: 495–501. https://doi.org/10.1038/nbt. 1630 PMID: 20436461 40. Longchamps RJ, Castellani CA, Yang SY, Newcomb CE, Sumpter JA, Lane J, et al. Evaluation of mito- chondrial DNA copy number estimation techniques. Plos One. 2020; 15: e0228166. https://doi.org/10. 1371/journal.pone.0228166 PMID: 32004343 41. Shahini A, Vydiam K, Choudhury D, Rajabian N, Nguyen T, Lei P, et al. Efficient and high yield isolation of myoblasts from skeletal muscle. Stem cell research. 2018; 30: 122–129. https://doi.org/10.1016/j.scr. 2018.05.017 PMID: 29879622 24 / 24

https://openalex.org/W2011129003

https://zenodo.org/records/1447327/files/article.pdf

German

Über Friedrich Kottlers Abhandlung “Über Einsteins Äquivalenzhypothese und die Gravitation”

Annalen der Physik

public-domain

1) Ann. d. Phys. 60. p. 955. 1916. 639 639 5. Uber P r 4 e d r i c h E o t t l e r s Abhamdlurng ,, &er Eirnsteirns Aquivalemdqipothese urnd die Gravitation(6 1); 5. Uber P r 4 e d r i c h E o t t l e r s Abhamdlurng ,, &er Eirnsteirns Aquivalemdqipothese urnd die Gravitation(6 1); worn A. EirnsteCm. Unter den Arbeiten, welche sich kritisch mit der all- gemeinen Relativitatstheorie beschiiftigen, sind besonders die- jenigen Ko t tlers bemerkenswert, denn dieser Fachgenosse ist wirklich in den Geist der Theorie eingedrungen. Mit der letzlen dieser -4rbeiten will ich mich hier auseinandersetzen. Ko t tler behauptet, ich hatte das von mir aufgestellte ,,Aquivalenzprinzip", durch welches ich die Begriife der ,,triigen Masse" und der ,,schweren Masse" zu einem ein- heitlichen Begriffe zu vereinigen strebte, in meinen spateren Arbeiten wieder anfgegeben. Diese Meinung muB darauf be- ruhen, daI3 wir beide nicht dasselbe als ,,;4qnivalenzprinzip" bewichnen; denn nach meiner Auffassung ruht meine Theorie ausschlieBlich auf diesem Prinzip. Deshalb sei folgendes wiederholt : 1. Der Grenzfell der speziellen Relaticitiitsthsorie. Ein raumzeitlich endliches Gebiet sei frei von einem Gravitations- felde, d. h. es sei moqlich, ein Bezugssystem K (.,Galileisches System") aufzustellen, relativ zu welcheni fur das genamite Gebiet folgendes gilt. Koordinaten seien in bekannter Weise mit dem EinheitsmaBstab, Zeiten mit der Einheitsuhr un- mittelbar meBbar, wie dies in der speziellen Relativitiits- theorie vorausgesetzt zu werden pflegt. In bezug auf dieses System bewege sich ein isolierter ma terieller Punkt gerad- linig und gleichformig, wie es von Galilei vorausgesetzt wurde. 1. Der Grenzfell der speziellen Relaticitiitsthsorie. Ein raumzeitlich endliches Gebiet sei frei von einem Gravitations- felde, d. h. es sei moqlich, ein Bezugssystem K (.,Galileisches System") aufzustellen, relativ zu welcheni fur das genamite Gebiet folgendes gilt. Koordinaten seien in bekannter Weise mit dem EinheitsmaBstab, Zeiten mit der Einheitsuhr un- mittelbar meBbar, wie dies in der speziellen Relativitiits- theorie vorausgesetzt zu werden pflegt. In bezug auf dieses System bewege sich ein isolierter ma terieller Punkt gerad- linig und gleichformig, wie es von Galilei vorausgesetzt wurde. 2. dquivalenzprinzip. Ausgehend von diesem Grenzfall der speziellen Relativitatstheorie kann man sich fragen, ob 1) Ann. d. Phys. 60. p. 955. 1916. 640 A . Einstein. ein in dem betrachteten Gebiete relativ zu X gleicliforniig beschleunigter Beobachter seinen Zustand als beschleunigt auffassen muD, oder ob ihm nach den (angenahert) bekannten Naturgesetzen eine Auffassung ubrig bleibt, vermoge derer er seinen Zustand als ,,Ruhe" deuten kann. Prgziser aus- gedruckt : Erlauben uns die in gewisser Annaherung bekannten Naturgesetze ein in bezug auf I< gleichformig heschleunigtes Bezugssystem K' als ruhend zu betrachten? Oder etwas all- gemeiner : Lal3t sich das Relativitatsprinzip auch auf relativ zneinaiider (glsichformig) beschleunigte Bezugssys tenie auv- dehnen? 5. Uber P r 4 e d r i c h E o t t l e r s Abhamdlurng ,, &er Eirnsteirns Aquivalemdqipothese urnd die Gravitation(6 1); Die Antwort lautet : Soweit wir die Xaturgesetze wirklich kennen, hjndert uns nichts daran, das System K' als ruhend zu betraohten, wenn wir relativ zu K' ein (in erster Anniiherung homogenes) Schwerefeld als vorhsnden annehmen ; denn wie in einem homogenen Schrvereield, so auch in bezug auf unser System IC' fallen alle Kiirper unabhangig vorl ihrer physikalischen Natur mit derselben Beschleunigung. Die Voraussetzung, daB man in aller Strenge I<' als ruhend behandeln durfe, ohne daD irgendein Naturgesetz in bezug auf k" nicht erfullt ware, nenne ich ,,hpivalenzprinzip". L 3. Das Schwerefeld nicht nur lcinemu2isch bedingt. Man kann die vorige Betrachtung auch umkehren. Sei das init dem oben betrachteten Schwerefelde ausgestaltete System K' das ursprungliche. Dann kann man ein neues, gegen K' be- schleunigtes Bezugssystem K einfuhren, mit Bezug suf welches sich (isolierte) Massen (in dem betrachteten Gebiete) gerad- linig gleichformig bewegen. Aber man darf nun nicht weiter- gehen und sagen: 1st K' ein mit einem beliebigen Gravitations- feld versehenes Bezugssystem, so jst stets ein Bezugssystern 1; auffindbar, in bezug auf welches sich isolierte Massen gerad- linig gleichformig bewegen, d. h. in bezug auf welches kein Gravitationsfeld existiert. Die Absurditat einer solchen T70r- aussetzung liegt auf der Hand. 1st das Gravitationsfdd in bezug auf K' zum Beispiel das eines ruhenden Massenpunlites, so laBt sich dieses Feld fur die ganze Umgebung des Massenpunktes gewiB durch kein noch SO feines Transformationskunststuck hinwegtrans€ormieren. Man darf also keineswegs behaupten, das Gravitationsfeld sei gewissermaaen rein kinematisch zu erklaren; eine ,,kinematkche, nicht dynamischr buffassung der Gravitation" ist nicht moglich. Purch blol3e Trans- L ..._ Uber Eimteim Aquivaknzh ypothese und die Gravitation. 641 formation aus einem Galileischen System in ein anderes durch Beschleunigungstransformationen lernen wir also nicht be- liebige Gravitationsfelder kennen, sondern solche ganz speziellbr Art, welche aber doch denselben Gesetzen genugen mussen nie alle anderen Gravitationsfelder. Dies ist nur wieder eine andere Formulierung des Aquivalenzpinzips (speziell in seiner Anwendung auf die Gravitation). Eine Gravitationstheorie verletzt also das Aquivalenz- prinzip in dem Sinne, wie ich es verstehe, nur dann, wenn die Gleichungen der Gravitation in keinem Bezugssystem I<' erfullt sind, welches relativ zu einem galileischen Bezugs- system ungleichformig bewegt ist. DaB dieser Vorwurf gegm meine Theorie mit allgemein kovarisnten Gleichungen nicht erhobm werden kann, ist evident; denn hier sind die Glei- chungcii bezuglich sines jeden Bezugssystems erfullt. Die Forderung der allgeminen Kouarianz der Gleichungen umfapt die des Bquivalenzprinzips ah ganz speziellen Fall. 4. 5. Uber P r 4 e d r i c h E o t t l e r s Abhamdlurng ,, &er Eirnsteirns Aquivalemdqipothese urnd die Gravitation(6 1); Sind die Kraftfte des Grawitatwnsfeldes , ,reale" Krajte ? Bot,tler rugt es, daB ich in den Bewegungsgleichungen das zweite Glied als den Ausdruck des Einflusses des Schwere- feldes auf den Massenpunkt, das erste Glied gewissermaBen als den Ausdruck der Galileischen Tragheit interpretiere. Dadurch wiirden ,,wirkliche Krafte des Schwerefeldes" ein- gefuhrt, was dem Geiste des Aquivalenzprinzipes nicht ent- spreche. Hierauf antworte ich, daB jene Gleichung als Ganzes allgemein kovariant, also jedenfalls der Aquivalenz- hypothese gemaB ist. Die von mir eingefuhrte Benennung der Teile ist prinzipiell bedeutungslos und einzig dazu be- stimmt, unseren physikalischen Denkgewohnheiten entgegen- zukommen. Dies gilt auch insbesondere von den Begriffen (Komponenten des Gravitationsfeldes) und to' (Energie- komponenten des Gravitationsfeldes). Die Einfuhrung dieser Benennungen ist prinzipiell unnotig, erscheint mir aber Mr g p p g Annalen der Phyaik. IV. Folge. 61. 42 42 A. Einstein. Uber Einsteins A'quiualcnzhypothse usw. 642 die Aufrechterhaltung der Kontinuitat der Gedanken wenig- stens einstweilen nicht wertlos ; deshalb habe ich diese GrolSeu eingefiihrt, trotzdem ihnen kein Tensorcharakter zukomnit . I)em Aquivalenzprinzip aber ist stets Genuge geleistet, 11 e i n ~ die Gleichungen kovsriant sind. 5. Es ist wahr, daB ich die allgemeine Kovarisnz dt.1 Gleichungen durch das Aufgeben der gewcihnlichen Zeit - messung und der Euklidischen Raummessung habe crkaufrn miissen. Kottler glaubt ohne dies Opfer auskommen zu konnen. Aber bereits im Falle des von ihm betrachteten iin Bornschen Sinne relativ zu einem Galileischen System be- schleunigten Systems K mu13 man auf die gewohnliche Zeit- messung verxichten. Da ist vom Standpunkt der Relativitats- theorie der Gedanke schon naheliegeiid, daB aucli die ge- wohnliche Raummessung aufgegeben wtrden miisse. Von dieser Notwendigkeit wird sich Hr. Ko t tler sicherlich selbst iiberzeugen, wenii er die ihm vorschwebenden theoretischx PIBne allgemein durchzufiihren suchen wird. 5. Es ist wahr, daB ich die allgemeine Kovarisnz dt.1 Gleichungen durch das Aufgeben der gewcihnlichen Zeit - messung und der Euklidischen Raummessung habe crkaufrn miissen. Kottler glaubt ohne dies Opfer auskommen zu konnen. Aber bereits im Falle des von ihm betrachteten iin Bornschen Sinne relativ zu einem Galileischen System be- schleunigten Systems K mu13 man auf die gewohnliche Zeit- messung verxichten. Da ist vom Standpunkt der Relativitats- theorie der Gedanke schon naheliegeiid, daB aucli die ge- wohnliche Raummessung aufgegeben wtrden miisse. Von dieser Notwendigkeit wird sich Hr. Ko t tler sicherlich selbst iiberzeugen, wenii er die ihm vorschwebenden theoretischx PIBne allgemein durchzufiihren suchen wird. Oktober 1916. (Eingegangen 19. Oktober 1916 )

https://openalex.org/W2977587960

http://umu.diva-portal.org/smash/get/diva2:1391865/FULLTEXT01

English

Aggregatibacter Actinomycetemcomitans: Clinical Significance of a Pathobiont Subjected to Ample Changes in Classification and Nomenclature

Pathogens

cc-by

http://www.diva-portal.org http://www.diva-portal.org http://www.diva-portal.org This is the published version of a paper published in .



 Abstract: Aggregatibacter actinomycetemcomitans is a Gram-negative bacterium that is part of the oral microbiota. The aggregative nature of this pathogen or pathobiont is crucial to its involvement in human disease. It has been cultured from non-oral infections for more than a century, while its portrayal as an aetiological agent in periodontitis has emerged more recently. A. actinomycetemcomitans is one species among a plethora of microorganisms that constitute the oral microbiota. Although A. actinomycetemcomitans encodes several putative toxins, the complex interplay with other partners of the oral microbiota and the suppression of host response may be central for inflammation and infection in the oral cavity. The aim of this review is to provide a comprehensive update on the clinical significance, classification, and characterisation of A. actinomycetemcomitans, which has exclusive or predominant host specificity for humans. Keywords: adherence; endocarditis; fimbriae; JP2; leukotoxin; periodontitis pathogens pathogens pathogens Pathogens 2019, 8, 243; doi:10.3390/pathogens8040243 Niels Nørskov-Lauritsen 1 , Rolf Claesson 2, Anne Birkeholm Jensen 3, Carola Höglund Åberg 4 and Dorte Haubek 3,* 1 Department of Clinical Microbiology, Aarhus University Hospital, DK-8200 Aarhus N, Denmark; [email protected] 2 Department of Odontology, Division of Oral Microbiology, Umeå University, S-901 87 Umeå, Sweden; [email protected] 3 Department of Dentistry and Oral Health Aarhus University DK 8000 Aarhus C Denmark; abj@dent au dk g g * Correspondence: [email protected] This is the published version of a paper published in . Access to the published version may require subscription. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-167619 1. Introduction Aggregatibacter actinomycetemcomitans is the type species of genus Aggregatibacter, which is part of bacterial family Pasteurellaceae. [Bacterium actinomycetem comitans] was cultured from actinomycotic lesions of humans in the early 20th century. The absence of related microorganisms rendered it difficult to classify this Gram-negative, fastidious rod, and isolates cultured from invasive infections were referred to national reference institutions. The expanding field of oral microbiology with a focus on periodontitis, particularly the localized, severe form that affects adolescents, caused a renewed interest in the bacterial species. In 2006, the current species name was adopted, and A. actionomycetemcomtians became type species of a new bacterial genus, Aggregatibacter. Influential events in the narrative of A. actinomycetemcomitans are listed in Table 1. Pathogens 2019, 8, 243; doi:10.3390/pathogens8040243 www.mdpi.com/journal/pathogens 2 of 18 Pathogens 2019, 8, 243 Table 1. Seminal events in the history of Aggregatibacter actinomycetemcomitans. Year Event Reference 1912 Klinger describes [Bacterium actinomycetem comitans] [1] 1929 Topley and Wilson relocate the species to genus Actinobacillus [2] 1962 King and Tatum describe the close phenotypic similarity of [Actinobacillus actinomycetemcomitans] with [Haemophilus aphrophilus] [3] 1976 [Actinobacillus actinomycetemcomitans] is associated with periodontitis in adolescents [4,5] 1979 Extraction and partial characterisation of Ltx, a leukotoxin capable of specific lyse of human polymorphonuclear leukocytes [6] 1982 The Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, and Kingella (HACEK) group of fastidious, Gram-negative bacteria causing infective endocarditis, is conceived [7] 1982 Serum antibody levels link [Actinobacillus actinomycetemcomitans] with localized juvenile periodontitis [8] 1983 Three distinct surface antigens are identified and a particularly high periodontopathogenic potential of serotype b is indicated [9] 1994 The 530-bp deletion in the ltx promoter region is associated with enhanced expression of Ltx and becomes a marker for the so-called JP2 genotype of [Actinobacillus actinomycetemcomitans] [10] 2006 A new bacterial genus, Aggregatibacter is created with Aggregatibacter actinomycetemcomitans being the type species [11] 2008 Clinical follow-up studies unequivocally demonstrate that carriage of the JP2 clone of Aggregatibacter actinomycetemcomitans is linked with aggressive periodontitis [12] A. actinomycetemcomitans is one species among a plethora of microorganisms that constitute the oral microbiota. It has been estimated that at least 500 different bacterial species colonise the oral cavity [13–15], and half of these may have been cultivated and validly named because of vigorous efforts directed to the cultivation of oral bacteria. 2. Taxonomy, Classification, Serotype (St) and Population Structure 2. Taxonomy, Classification, Serotype (St) and Population Structur More than 100 years ago, [Bacterium actinomycetem comitans] was co-isolated with Actinomyces from actinomycotic lesions of humans [1] (Actinomyces, ray fungus, referring to the radial arrangement of filaments in Actinomyces bovis sulfur granules; actinomycosis, a chronic disease characterised by hard granulomatous masses). Ample changes have occurred in the classification and nomenclature of this species. Despite the limited similarity with Actinobacillus lignieresii, it was reclassified as [Actinobacillus actinomycetemcomitans] in a seminal textbook from 1929 [2]. According to Cowan [21], the bacterium was placed in this genus because ‘neither Topley nor Wilson could think where to put it’. In 1962 the phenotypic resemblance of [Actinobacillus actinomycetemcomitans] with [Haemophilus aphrophilus] was described [3], and a subsequent relocation of [Actinobacillus actinomycetemcomitans] to genus Haemophilus occurred [22]. Nomenclatural classification as [Haemophilus actinomycetemcomitans] within the genus Haemophilus permitted antimicrobial susceptibility testing according to standards outlined by the US Clinical and Laboratory Standards Institute. Disk diffusion could be performed and interpreted on Haemophilus test medium (HTM) in 5% CO2, and HTM broth microdilution testing was carried out in ambient air [23]. However, the nomenclatural relocation did not result in a satisfying classification, because neither [Actinobacillus actinomycetemcomitans] nor [Haemophilus aphrophilus] are adequately related to Haemophilus influenzae, the type species of the genus Haemophilus. Finally, in 2006 the new genus Aggregatibacter was created to accomodate Aggregatibacter actinomycetemcomitans, Aggregatibacter aphrophilus and Aggregatibacter segnis [11]. A fourth Aggregatibacter species, Aggregatibacter kilianii, has recently been named (Figure 1) [24]. More than 100 years ago, [Bacterium actinomycetem comitans] was co-isolated with Actinomyce from actinomycotic lesions of humans [1] (Actinomyces, ray fungus, referring to the radia arrangement of filaments in Actinomyces bovis sulfur granules; actinomycosis, a chronic diseas characterised by hard granulomatous masses). Ample changes have occurred in the classification and nomenclature of this species. Despite the limited similarity with Actinobacillus lignieresii, it wa reclassified as [Actinobacillus actinomycetemcomitans] in a seminal textbook from 1929 [2]. Accordin to Cowan [21], the bacterium was placed in this genus because ‘neither Topley nor Wilson could thin where to put it’. In 1962 the phenotypic resemblance of [Actinobacillus actinomycetemcomitans] with [Haemophilus aphrophilus] was described [3], and a subsequent relocation of [Actinobacillu actinomycetemcomitans] to genus Haemophilus occurred [22]. Nomenclatural classification a [Haemophilus actinomycetemcomitans] within the genus Haemophilus permitted antimicrobia susceptibility testing according to standards outlined by the US Clinical and Laboratory Standard Institute. 1. Introduction Analysis of a large number of 16S rRNA gene clones from studies of the oral microbiota increased the number of taxa to 619 [16], and the number is steadily increasing (www.homd.org). Bacterial species cannot be validly named in the absence of a cultured type strain [17]. Although “taxa”, “phylotypes” or “operative taxonomic units” revealed by deep sequencing of polymerase chain reaction (PCR)-amplified 16S rRNA genes have relevance for recognition of microbial fluctuations in health and disease, only cultivable microbiota can be made subject to extensive characterisation, including adherence, animal experiments, antimicrobial susceptibility, co-culture, generation of mutants, and growth characteristics. p y g g Carriage of A. actinomycetemcomitans appears to be highly host-specific. Although the spread and dissemination of bacterial clones occur, these are not frequent events; hosts tend to carry their strain from teething to edentulous old age [18]. Yet, the species encompasses properties that sometimes reveal its significance in human disease. Particularly, a single serotype b clonal lineage designated the JP2 clone is associated with a severe form of localised periodontitis and tooth loss in adolescents [12]. But rather than being the causative agent of aggressive periodontitis, A. actinomycetemcomitans may be necessary for the action of a consortium of bacterial partners by suppressing host defences [19]. It may be classified as a low abundance oral pathobiont, defined as a member of the microbiota that exerts specific effects on the host’s mucosal immune system associated with the development of disease [20]. Although A. actinomycetemcomitans may accompany (comitans) Actinomyces, the narrative of a pathobiont is not valid for other invasive infections such as infectious endocarditis, where A. actinomycetemcomitans—when identified—is detected as the sole pathogen by culture and/or PCR on removed heart valves. Severe periodontitis and infective endocarditis are two prominent diseases of very different prevalence, symptoms, and outcome. Although they may share a causative microorganism, a number of conditions is still unknown, and host factors, oral hygiene, and incidental circumstances may be instrumental. The aim of the present review is to provide a comprehensive update on the characterisation, classification and clinical significance of A. actinomycetemcomitans with a particular focus on selected clinical entities. 1. Introduction Adhesion, persistence, and inactivation of immune cells are probably essential for the understanding of the intimate association with the host, and these factors are detailed for the purpose Pathogens 2019, 8, 243 Pathogens 2019, 8, x 3 of 18 3 o d f h of the elucidation of pathogenicity. A number of relevant publications and reviews of other important biochemical mechanisms of this bacterial species are listed in the relevant sections. purpose of the elucidation of pathogenicity. A number of relevant publications and reviews of othe important biochemical mechanisms of this bacterial species are listed in the relevant sections. 2. Taxonomy, Classification, Serotype (St) and Population Structure 2. Taxonomy, Classification, Serotype (St) and Population Structur Disk diffusion could be performed and interpreted on Haemophilus test medium (HTM) in 5% CO2, and HTM broth microdilution testing was carried out in ambient air [23]. However, th nomenclatural relocation did not result in a satisfying classification, because neither [Actinobacillu actinomycetemcomitans] nor [Haemophilus aphrophilus] are adequately related to Haemophilus influenzae the type species of the genus Haemophilus. Finally, in 2006 the new genus Aggregatibacter was created to accomodate Aggregatibacter actinomycetemcomitans, Aggregatibacter aphrophilus and Aggregatibacte segnis [11]. A fourth Aggregatibacter species, Aggregatibacter kilianii, has recently been named (Figur 1) [24]. Figure 1. Comparison of Aggregatibacter strains by whole genome sequences; distinct species are separated by dotted lines (strain PN_491 is unclustered). A total of 3261 positions with single nucleotide polymorphism (SNP) are included in the dataset. Values at nodes are percentages of bootstrap replications supporting the node (500 replicates). Bar represents 200 SNPs. Reprinted from Journal of Clinical Microbiology [24] with permission. Figure 1. Comparison of Aggregatibacter strains by whole genome sequences; distinct species are separated by dotted lines (strain PN_491 is unclustered). A total of 3261 positions with single nucleotide polymorphism (SNP) are included in the dataset. Values at nodes are percentages of bootstrap replications supporting the node (500 replicates). Bar represents 200 SNPs. Reprinted from Journal of Clinical Microbiology [24] with permission. I th l 1980 th di ti t f ti f A ti t it id tifi d [9] hil Figure 1. Comparison of Aggregatibacter strains by whole genome sequences; distinct species are separated by dotted lines (strain PN_491 is unclustered). A total of 3261 positions with single nucleotide polymorphism (SNP) are included in the dataset. Values at nodes are percentages o bootstrap replications supporting the node (500 replicates). Bar represents 200 SNPs. Reprinted from J l f Cli i l Mi bi l [24] ith i i Figure 1. Comparison of Aggregatibacter strains by whole genome sequences; distinct species are separated by dotted lines (strain PN_491 is unclustered). A total of 3261 positions with single nucleotide polymorphism (SNP) are included in the dataset. Values at nodes are percentages of bootstrap replications supporting the node (500 replicates). Bar represents 200 SNPs. Reprinted from Journal of Clinical Microbiology [24] with permission. Figure 1. Comparison of Aggregatibacter strains by whole genome sequences; distinct species ar separated by dotted lines (strain PN_491 is unclustered). A total of 3261 positions with singl nucleotide polymorphism (SNP) are included in the dataset. 2. Taxonomy, Classification, Serotype (St) and Population Structure 2. Taxonomy, Classification, Serotype (St) and Population Structur Values at nodes are percentages o bootstrap replications supporting the node (500 replicates). Bar represents 200 SNPs. Reprinted from Journal of Clinical Microbiology [24] with permission Figure 1. Comparison of Aggregatibacter strains by whole genome sequences; distinct species are separated by dotted lines (strain PN_491 is unclustered). A total of 3261 positions with single nucleotide polymorphism (SNP) are included in the dataset. Values at nodes are percentages of bootstrap replications supporting the node (500 replicates). Bar represents 200 SNPs. Reprinted from Journal of Clinical Microbiology [24] with permission. f gy p In the early 1980s, three distinct surface antigens of A. actinomycetemcomitans were identified [9], while six serotypes (a through f) were recognised by 2001. The serological specificity is defined by structurally In the early 1980s, three distinct surface antigens of A. actinomycetemcomitans were identified [9], while six serotypes (a through f) were recognised by 2001. The serological specificity is defined by structurally 4 of 18 Pathogens 2019, 8, 243 and antigenically distinct O-polysaccharide components of their respective lipopolysaccharide molecules. AseventhSt, designatedSt(g), witha1:1ratioof2,4-di-O-methyl-rhamnoseand2,3,6-tri-O-methyl-glucose, was recently added [25]. St(a), St(b), and St(c) are globally dominant [26], but the distribution may vary according to ethnicity and geography. In Scandinavia, the three dominant serotypes are equally prevalent, while predominance of St(c) is observed in Chinese, Japanese, Korean, Thai and Vietnamese populations [27–31]; a noticeable high prevalence of St(e) has been reported among Japanese periodontitis patients [32]. Assessment of serotype-specific antibodies supports these findings, as all early-onset periodontitis patients from Turkey and Brazil had elevated antibody levels to St(c) and St(a), while St(b) levels were higher in the US [33,34]. g An initial characterisation of the population structure of the species was published in 1994, using multi-locus enzyme electrophoresis [35]. Two large and four small divisions were identified, with division I (St(a) and St(d)) and III (St(b) and St(c)) encompassing 34% and 58% of the 97 strains analysed, respectively. Two St(e) strains occupied separate divisions (II and VI), one St(c) strain constituted electrophoretic division IV, while division V was composed of two St(a) and one non-serotypable strain. Sequencing of a 16S rRNA gene fragment from 35 strains suggested a different structure with three major clusters [36]. RNA cluster I included 12 strains of four serotypes (a, d, e, and f), all 10 St(b) strains belonged to RNA cluster II, while RNA cluster III only included St(c) strains (N = 10). 2. Taxonomy, Classification, Serotype (St) and Population Structure 2. Taxonomy, Classification, Serotype (St) and Population Structur Strains of particular serotypes were not exclusively confined to specific RNA clusters; one St(a) strain belonged to the St(b) cluster (II), and two divergent RNA clusters were composed of single strains, namely a St(c) (RNA cluster IV) and a St(e) strain (RNA cluster V), respectively [36]. One study attempted to establish a multi-locus sequence typing (MLST) scheme for A. actinomycetemcomitans [37]. Six gene fragments from the Haemophilus influenzae MLST scheme were used. The investigation focused on the JP2 clone, which contributed 66 of 82 strains. MLST has insufficient power to reveal dissemination patterns of clonally related strains, and point mutations of two pseudogenes present in the JP2 clone were more versatile in this respect [37]. Moreover, a MLST web site was not organised and, therefore, the benefits of a portable typing scheme were not corroborated. But MLST of 16 non-JP2 strains carefully selected from the enzyme electrophoresis study [35] suggested the existence of four phylogenetic clusters, rooted by an outgroup consisting of an uncommon St(e) strain. Two related clusters were composed of St(b) and St(c) strains, respectively, while a more distinct cluster encompassed strains of St(a), St(d) and St(e) [37]. Restriction fragment length polymorphism using various restriction enzymes and arbitrarily-primed PCR has been used to differentiate types of A. actinomycetemcomitans cultured from patients with severe periodontitis and healthy controls [30,37–42]. The method is versatile and discriminative, but lacks portability and a common nomenclature; thus, it is of value for individual studies of specific strains, but lacks general applicability and descriptive significance. Finally, whole genome sequencing has been introduced for characterisation of the species [43–45]. In the largest study, sequences from two human strains of Aggregatibacter aphrophilus, 30 human A. actinomycetemcomitans strains, and one St(b) strain isolated from a rhesus macaque Old World monkey were used for selection of 397 core genes which were concatenated and trimmed to produce a single alignment of 335,400 bp [45]. Five clades were recognised, designated clade b, clade c, clade e/f, clade a/d and clade e’. Although the analysis clearly separated six strains of serotype b from six strains of serotype c, a close similarity was observed between these two clades, as well as between clade a/d and e/f. In contrast, the clade designated e’, encompassing four St(e) strains, was phylogenetically distinct. 2. Taxonomy, Classification, Serotype (St) and Population Structure 2. Taxonomy, Classification, Serotype (St) and Population Structur The open reading frames necessary for expression of St(e) antigen were highly conserved between clade e and clade e’ strains, but e’ strains were found to be missing the genomic island that carries genes encoding the cytolethal distending toxin. Moreover, the clade e’ strains were more related to an Old World primate strain and carried the unusual 16S rRNA type V sequence (RNA types as defined by Kaplan et al. [36]). Although bacterial species are not defined by DNA sequence, average nucleotide identity (ANI) values locate whole genome sequences from this group/clade outside the species boundary [44]. Thus, strains belonging to the so-called clade e’ (as well as the rhesus macaque 5 of 18 Pathogens 2019, 8, 243 monkey strain) may possibly be transferred to new species, and A. actinomycetemcomitans may be restricted to strains with exclusive host specificity for humans. well as the rhesus macaque monkey strain) may possibly be transferred to new species, and A actinomycetemcomitans may be restricted to strains with exclusive host specificity for humans. monkey strain) may possibly be transferred to new species, and A. actinomycetemcomitans may be restricted to strains with exclusive host specificity for humans. well as the rhesus macaque monkey strain) may possibly be transferred to new species, and A actinomycetemcomitans may be restricted to strains with exclusive host specificity for humans. A recent study compared whole genome sequences of strains from blood stream infections supplemented with oral reference strains [46]. Exclusion of so-called clade e’ strains increased the number of core genes present in all strains from 1146 to 1357. Strains of A. actinomycetemcomitans are basically divided into three lineages (numbering of lineages differs from reference [44]). Lineage I encompasses the type strain and consists of two groups (St(b) and St(c), respectively). Lineage II consists of St(a) plus St(d)-(g). In contrast to lineage I, many strains of different serotypes from this lineage are competent for natural transformation, and the average size of genomes is approximately 10% larger than in lineage I. Lineage III also expresses St(a) membrane O polysaccharide, and the genome size is comparable to lineage II. However, all six investigated strains were incompetent for transformation due to inactivation of multiple competence genes [46]. y y p y A recent study compared whole genome sequences of strains from blood stream infection supplemented with oral reference strains [46]. 2. Taxonomy, Classification, Serotype (St) and Population Structure 2. Taxonomy, Classification, Serotype (St) and Population Structur Exclusion of so-called clade e’ strains increased the number of core genes present in all strains from 1146 to 1357. Strains of A. actinomycetemcomitans are basically divided into three lineages (numbering of lineages differs from reference [44]). Lineage encompasses the type strain and consists of two groups (St(b) and St(c), respectively). Lineage I consists of St(a) plus St(d)-(g). In contrast to lineage I, many strains of different serotypes from thi lineage are competent for natural transformation, and the average size of genomes is approximately 10% larger than in lineage I. Lineage III also expresses St(a) membrane O polysaccharide, and the genome size is comparable to lineage II. However, all six investigated strains were incompetent fo transformation due to inactivation of multiple competence genes [46]. In conclusion, St designations are valuable for initial typing of clinical strains, but insufficient for characterisation. Recognition of a general MLST scheme could be helpful, and whole genome sequences could be used for MLST and in silico serotyping, as well as further characterisation and epidemiologic investigations. The species description consisting of three separate lineages is figurative, but more knowledge on the new lineage III is needed to disclose the relevance for phenotype, host specificity and pathogenicity. p p g [ ] In conclusion, St designations are valuable for initial typing of clinical strains, but insufficien for characterisation. Recognition of a general MLST scheme could be helpful, and whole genome sequences could be used for MLST and in silico serotyping, as well as further characterisation and epidemiologic investigations. The species description consisting of three separate lineages is figurative, but more knowledge on the new lineage III is needed to disclose the relevance for phenotype, host specificity and pathogenicity. 3. General Characteristics 3. General Characteristic Diameter of colonies did not reach 2 mm after 3 days incubation in 5% CO2. (B) Clinical isolate incubated on TSBV (tryptic soy-serum-bacitracin-vancomycin) agar for 4 days in 5% CO2. Expression of the distinctive “star-shaped” colony is facilitated by growth on TSBV agar. Pictures by courtesy of Jan Berg Gertsen and Rolf Claesson. 3. General Characteristics 3. General Characteristic A. actinomycetemcomitans is a fastidious, facultatively anaerobic, non-motile, small Gram-negative rod, 0.4–0.5 µm × 1.0–1.5 µm in size. Microscopically, the cells may appear as cocci in broth and in clinical samples. It grows poorly in ambient air, but well in 5% CO2 [47]. Colonies on chocolate agar are small, with a diameter of ≤0.5 mm after 24 h, but may exceed 1–2 mm after 48 h [48]. Primary colonies are rough-textured and adhere strongly to the surface of agar plates (Figure 2). A. actinomycetemcomitans is a fastidious, facultatively anaerobic, non-motile, small Gram negative rod, 0.4–0.5 µm x 1.0–1.5 µm in size. Microscopically, the cells may appear as cocci in broth and in clinical samples. It grows poorly in ambient air, but well in 5% CO2 [47]. Colonies on chocolat agar are small, with a diameter of ≤0.5 mm after 24 h, but may exceed 1–2 mm after 48 h [48]. Primary colonies are rough-textured and adhere strongly to the surface of agar plates (Figure 2). (B) (A) (A) (B) Figure 2. (A) Tenacious, rough-textured colonies of A. actinomycetemcomitans strain HK1651 on chocolate agar. Diameter of colonies did not reach 2 mm after 3 days incubation in 5% CO2. (B) Clinical isolate incubated on TSBV (tryptic soy-serum-bacitracin-vancomycin) agar for 4 days in 5% CO2. Expression of the distinctive “star-shaped” colony is facilitated by growth on TSBV agar. Pictures by courtesy of Jan Berg Gertsen and Rolf Claesson. Figure 2. (A) Tenacious, rough-textured colonies of A. actinomycetemcomitans strain HK1651 on chocolate agar. Diameter of colonies did not reach 2 mm after 3 days incubation in 5% CO2. (B) Clinical isolate incubated on TSBV (tryptic soy-serum-bacitracin-vancomycin) agar for 4 days in 5% CO2. Expression of the distinctive “star-shaped” colony is facilitated by growth on TSBV agar. Pictures by courtesy of Jan Berg Gertsen and Rolf Claesson. (A) (B) Figure 2. (A) Tenacious, rough-textured colonies of A. actinomycetemcomitans strain HK1651 on chocolate agar. Diameter of colonies did not reach 2 mm after 3 days incubation in 5% CO2. (B) Clinica isolate incubated on TSBV (tryptic soy-serum-bacitracin-vancomycin) agar for 4 days in 5% CO2 Expression of the distinctive “star-shaped” colony is facilitated by growth on TSBV agar. Pictures by courtesy of Jan Berg Gertsen and Rolf Claesson. Figure 2. (A) Tenacious, rough-textured colonies of A. actinomycetemcomitans strain HK1651 on chocolate agar. Pathogens 2019, 8, 243 Pathogens 2019, 8, 243 6 of 18 The selective medium TSBV (tryptic soy-serum-bacitracin-vancomycin) agar [53] is commonly used for culture. If Enterobacterales are present in significant amounts in the samples, a modified version of TSBV is recommended [54]. Detection of A. actinomycetemcomitans in clinical samples renders limited information on prediction, progression, and treatment planning of periodontal disease. For these purposes, the proportion of the bacterium at diseased sites is more relevant. This is in line with the ecological plaque hypothesis [55]. The detection level of A. actinomycetemcomitans is around 100 viable bacteria (colony-forming units) per mL. Fusobacterium nucleatum and other strict anaerobes will grow on TSBV in the absence of oxygen. The total concentration of viable bacteria is estimated by parallel cultivation on 5% blood agar plates, and the proportion of A. actinomycetemcomitans in the sample can be calculated. A. actinomycetemcomitans is suspected when rough-textured, tenacious colonies appear on selective agar after one or two days (Figure 2). The species is distinguished from closely related bacteria by a positive catalase reaction and negative β-galactosidase reaction. Salient biochemical characters of A. actinomycetemcomitans have been published [56]. In addition, the bacterium is readily identified by MALDI-TOF mass spectrometry [57]; however, the current version of the Bruker database (v3.1) only includes mass spectra from a limited number of strains, and modest log-scores are not unusual when clinical strains are examined. Leukotoxicity, i.e., the capacity of the bacterium to kill or inactivate immune cells, is properly determined in biological assays involving human cell lines [58], but a semi-quantitative method based on hemolysis on blood agar plates has been reported [42,59]. Quantification of the leukotoxin by enzyme-linked immunosorbent assay (ELISA) is also used; most studies have assessed the leukotoxin released from the surface of the bacterium, either during growth in broth [60], or by treatment of bacteria cultured in media that inhibit leukotoxin release with a hypertonic salt solution [42]. Leukotoxicity may also be estimated by determination of the total amount of leukotoxin produced by the strain. Bacterial suspensions are solubilized by SDS, and the leukotoxin is subsequently quantitated by Western blot–based methodology [60]. It is anticipated that the amount of leukotoxin released from the bacterial cell surface reflects the total amount of leukotoxin produced, but this relationship remains to be corroborated. Polymerase chain reaction (PCR) is frequently used for identification and characterisation of A. actinomycetemcomitans in clinical samples. Pathogens 2019, 8, 243 The leukotoxin promoter was an early focal point [59]. PCR amplification of the ltx promoter region and visualization on gel can discriminate the JP2 genotype from other strains of the species [61], but preferential amplification of smaller products characterised by a 530-bp deletion will overestimate the prevalence of the JP2 genotype. Recent improvements in PCR offer more precise quantification of periodontal pathogens in a complex plaque biofilm [62]. By real-time or quantitative PCR (qPCR), the instrument reports the cycling threshold (CT)-value, which can estimate the concentration of the target in the sample. qPCR has been used to separately quantitate JP2 and non-JP2 genotypes [63]. To approximate the total number of bacteria by qPCR, the 16S rRNA gene is generally targeted. The method can only provide a rough estimate, as primers and probes may preferentially bind to certain bacterial phyla, and because the number of copies of the gene varies substantially between different bacterial species [64]. Serotypes a through f can be identified by PCR as described [65,66]. A method for detecting St(g) has not yet been described. 3.1. Recovery, Phenotype, and Molecular Detection 3.1. Recovery, Phenotype, and Molecular Detection Relevant sites in the oral cavity for sampling of A. actinomycetemcomitans are periodontal pockets, the mucosa, and saliva. Sampling techniques include use of sterile paper points to be inserted in periodontal pockets, cotton swab for the mucosa, and chewing on a piece of paraffin for the collection of stimulated saliva. For transport of paper points, the VMGAIII-medium is recommended [49]; samples collected with cotton swap can be transported in a salt buffer or in TE-buffer [50]. Saliva can be transported in tubes without additives. For short-time transportation, saliva can be transported in tubes without additives. Otherwise, it can be frozen or stored at room temperature in a Saliva DNA Preservation Buffer. Proteomic analysis of gingival crevicular fluid and saliva is an expanding diagnostic field that may require improvements in standardised collection techniques and devices Relevant sites in the oral cavity for sampling of A. actinomycetemcomitans are periodontal pockets, the mucosa, and saliva. Sampling techniques include use of sterile paper points to be inserted in periodontal pockets, cotton swab for the mucosa, and chewing on a piece of paraffin for the collection of stimulated saliva. For transport of paper points, the VMGAIII-medium is recommended [49]; samples collected with cotton swap can be transported in a salt buffer or in TE-buffer [50]. Saliva can be transported in tubes without additives. For short-time transportation, saliva can be transported in tubes without additives. Otherwise, it can be frozen or stored at room temperature in a Saliva DNA Preservation Buffer. Proteomic analysis of gingival crevicular fluid and saliva is an expanding diagnostic field that may require improvements in standardised collection techniques and devices [51,52]. 3.2. Aggregative Properties and the Leukotoxin Gene Operon A. actinomycetemcomitans expresses three potential toxins, fimbriae and a number of adhesins, plus a number of other gene products that may have significance for microbial interplay, persistence, transformation to planktonic state, and pathogenicity (Table 2). 7 of 18 Pathogens 2019, 8, 243 Table 2. Genomic characteristics and putative virulence determinants of A. actinomycetemcomitans. nomic Characteristics and Putative Virulence Determinants of A. actinomycetemcomitans References Widespread colonization island or the tad locus [67] Autotransporter adhesins Aae, EmaA and Omp100/ApiA [68–70] A leukotoxin of the repeats in toxin (RTX) family with specificity for leukocytes [71] Growth-inhibitory factor cytolethal distending toxin [72,73] CagE, capable of inducing apoptosis [74] Dispersin B, a biofilm-releasing beta-hexosaminidase [75] Vesicle-independent extracellular release of proinflammatory lipoprotein [76] Widespread colonization island or the tad locus [67] Autotransporter adhesins Aae, EmaA and Omp100/ApiA [68–7 A leukotoxin of the repeats in toxin (RTX) family with specificity for leukocytes [71] Growth-inhibitory factor cytolethal distending toxin [72,73 CagE, capable of inducing apoptosis [74] Dispersin B, a biofilm-releasing beta-hexosaminidase [75] Vesicle-independent extracellular release of proinflammatory lipoprotein [76] The distinct growth in broth as small granules adhering to the walls of the test tube was included in the initial description of [Bacterium actinomycetem comitans] [1]. Fresh isolates of A. actinomycetemcomitans invariably form colonies that are rough-textured with an opaque, star-shaped internal structure (Figure 2B). Subculture in broth yields clumps of autoaggregated cells that attach tightly to the glass, leaving a clear broth. A. actinomycetemcomitans possesses fimbriae, and these appendages can be irreversibly lost after prolonged subculture in the laboratory [77]. Antibodies to synthetic fimbrial peptide significantly reduce the binding of A. actinomycetemcomitans to saliva-coated hydroxyapatite beads, buccal epithelial cells and a fibroblast cell line, indicating a decisive role of these structures for adherence to multiple surfaces [78]. Moreover, autoaggregation (spontaneous formation of aggregates with rapid settling in un-agitated suspensions) was completely lost by a smooth-colony, isogenic variant [79]. Fimbriae are assembled as bundles of 5- to 7-nm-diameter pili composed of a 6.5 kDa protein designated Flp (fimbrial low-molecular-weight protein) [80,81]. The RcpA/B (rough colony proteins) were the first outer membrane proteins identified that were associated with rough colony variants [82], and they are encoded by a 14-gene locus designated the tad locus. The Tad (tight adherence) macromolecular transport system is a subtype of type II secretion. The tad locus is composed of nine tad, three rcp and two flp genes [67]. 3.2. Aggregative Properties and the Leukotoxin Gene Operon Mutation analysis of the naturally competent strain D7S revealed flp-1, rcpA, rcpB, tadB, tadD, tadE and tadF to be indispensable for expression of fimbriae, while mutants of five other genes expressed reduced levels of fimbriae, or fimbriae that had different gross appearance [83,84]. In a rat model, the tad locus was critical for colonizing the oral cavity and for pathogenesis, measured as maxillae bone loss and A. actinomycetemcomitans-specific IgG levels [85]. Many pathogenic bacteria can undergo phase variation, but smooth-to-rough conversion has not been substantiated for A. actinomycetemcomitans. Rather, the rough-to-smooth conversion is typically caused by mutations in the flp promoter region, and replacement with wild-type promoter can restore the rough phenotype [86]. However, one study indicated that smooth strains could re-express the fimbriae in low humidity environments [87]. In addition to expression of fimbriae decisive for autoaggregation and adherence to a wide range of solid surfaces (biofilm formation), A. actinomycetemcomitans encodes a spectrum of autotransporter adhesins, proteins that promote their own transport from the periplasm to the exterior surface, where they may be decisive for adhesion to specific human cellular epitopes. A homologue with similarity to the monomeric H. influenzae autotransporter, Hap, was designated Aae. Inactivation of aae in two naturally transformable strains caused a 70% reduction in adhesion to cultured epithelial cells [68]. Aae exhibits specificity for buccal epithelial cells from humans and Old World primates, and does not bind to human pharyngeal or cervical epithelial cells [88]. Two trimeric autotransporters with homology to the YadA adhesin/invasin family were identified. Omp100 has also been designated Api (Aggregatibacter putative invasin). Escherichia coli expressing ApiA bound to various types of human collagen plus fibronectin. Adhesion to human cells was specific to buccal epithelial cells from humans and Old-World primates, although the specificity was not as prominent as observed for AaE [70,89]. Screening of a large number of insertion transposon mutants identified the extracellular matrix adhesin A encoded by emaA, which is involved in collagen adhesion [90]. Collagen prevail in the supporting tissue of cardiac valves, and EmaA (extracellular matrix adhesin) may play a role in the pathogenesis of infective endocarditis [91]. Pathogens 2019, 8, 243 Pathogens 2019, 8, 243 8 of 18 Iron is an essential transition metal for nearly all forms of life. The host limits the availability of iron through a process termed nutritional immunity [92]. 3.3. Geographic Dissemination of Specific Genotypes The JP2 clone of A. actinomycetemcomitans is suggested to have arisen 2400 years ago in the northern Mediterranean part of Africa [37]. The bacterial clone is endemically present in Moroccan and Ghanaian populations [12,101] and almost exclusively detected among individuals of African origin [37,102]. However, among 17 JP2 clone carriers, living in Sweden and identified during 2000–2014, ten were of Scandinavian heritage [31]. Among six of the identified JP2 clone carriers, three were of Swedish origin. Detection of the JP2 clone of A. actinomycetemcomitans has not been reported in Asian populations [30,100,103,104]. The occurrence of the JP2 clone of A. actinomycetemcomitans in Caucasians may be caused by horizontal transmission, and may weaken the theory of racial tropism of the clone [59]. More data and research are needed to explain the dissemination of the leukotoxic JP2 clone of A. actinomycetemcomitans. Other genotypes characterised by an increased leukotoxic potential comprise a 640-bp deletion cultured from a host of Ethiopian origin [99], an 886-bp insertion sequence from a host of Japanese origin [100], and two strains of serotype c, originating from Thailand with a JP2-like deletion in the promoter region of ltx, and with virulence of similar magnitude to the JP2 genotype strains [105]. All these genotypes were collected from individuals with severe periodontitis. 3.2. Aggregative Properties and the Leukotoxin Gene Operon Haemolysis can be an initial step for release of iron from heme by Gram-negative bacteria. The RTX (repeats in toxin) family is an important group of toxins, whose name refers to glycine- and apartate-rich, calcium-binding repeats in the carboxy terminus of the toxin proteins [93]. RTX toxins are produced by many Gram-negative bacteria including members of family Pasteurellaceae – it has, indeed, been proposed that these toxins may originate in Pasteurellaceae [94]. In 1977, it was shown that polymorphonuclear leukocytes exposed to gingival bacterial plaque in vitro released lysosomal constituents [95], and the leukotoxin (Ltx) of A. actinomycetemcomitans was extracted and partially characterised in 1979 [6]. Ltx is a RTX cytolysin. By 1989, the gene was cloned and analysed [96,97], and the 530-bp deletion in the ltx promoter associated with enhanced expression of Ltx characterising the JP2 genotype was subsequently described [10]. The difference between minimally toxic and highly toxic strains were convincingly illustrated in clinical studies from Northern Africa [12]. The significance of the 530-bp deletion may reside in a potential transcriptional terminator spanning 100 bp [60]. The leukotoxin of A. actinomycetemcomitans is highly specific for human and primate white blood cells and is capable of neutralising local mucosal immune responses. However, purified leukotoxin can lyse sheep and human erythrocytes in vitro, and beta-haemolysis can be demonstrated on certain media [98]. In addition to the JP2 genotype characterised by the 530-bp promoter deletion, two other leukotoxin promoter variants have been reported. One genotype is characterised by a slightly enlarged (640-bp) deletion [99], while the other promoter variant carries an 886-bp insertion sequence [100]. Both these variants produce levels of leukotoxin similar to the JP2 genotype of A. actinomycetemcomitans. 3.3. Geographic Dissemination of Specific Genotypes 4. Prevalence and Clinical Significance Cultivable A. actinomycetemcomitans is present in at least 10% of periodontally healthy children with primary dentition [106]. An influential publication found carrier rates of 20% for normal juveniles, 36% for normal adults, 50% for adult periodontitis patients, and 90% for young periodontitis patients [107]. Early studies failed to culture the species from edentulous infants [108,109], but molecular studies using PCR on unstimulated saliva samples have challenged this association: 37 of 59 completely edentulous infants were positive for A. actinomycetemcomitans, reaching 100% at 12 months of age [110]. Vertical transmission is common. Two studies reported detection rates by culture of 30–60% in children of adult periodontitis patient, and the genotypes of the strains were always identical [111,112]. A smaller study from Brazil of women with severe chronic periodontitis did not corroborate this finding, as the 9 of 18 9 of 1 Pathogens 2019, 8, 243 Pathogens 2019, 8, x two culture-positive children carried genotypes that were different from those of their mothers [113]. Horizontal transmission of A. actinomycetemcomitans can occur, and transmission rates between 14% and 60% between spouses have been estimated [18,114]. However, members of most families with aggressive periodontitis also harbour additional clonal types of A. actinomycetemcomitans [115]. chronic periodontitis did not corroborate this finding, as the two culture-positive children carried genotypes that were different from those of their mothers [113]. Horizontal transmission of A actinomycetemcomitans can occur, and transmission rates between 14% and 60% between spouses hav been estimated [18,114]. However, members of most families with aggressive periodontitis also harbour additional clonal types of A actinomycetemcomitans [115] Once colonized, A. actinomycetemcomitans remains detectable in patients with periodontitis. Irrespective of periodontal treatment, colonisation by the same strain is remarkably stable within subjects for periods of 5 to 12 years, as revealed by restriction fragment length polymorphism [40], serotyping combined with arbitrarily primed PCR [116], or JP2 clone-specific PCR [117]. Genomic stability during persistent oral infections has been demonstrated by genome sequencing of strains cultured from the same individual 10 years later [118]. harbour additional clonal types of A. actinomycetemcomitans [115]. Once colonized, A. actinomycetemcomitans remains detectable in patients with periodontitis Irrespective of periodontal treatment, colonisation by the same strain is remarkably stable within subjects for periods of 5 to 12 years, as revealed by restriction fragment length polymorphism [40] serotyping combined with arbitrarily primed PCR [116], or JP2 clone-specific PCR [117]. 4. Prevalence and Clinical Significance Genomi stability during persistent oral infections has been demonstrated by genome sequencing of strain cultured from the same individual 10 years later [118] The natural habitat of A. actinomycetemcomitans is the oral cavity, but A. actinomycetemcomitans can be isolated from a variety of oral as well as non-oral infectious diseases, including arthritis, bacteraemia, endocarditis, osteomyelitis, skin infections, urinary tract infections and various types of abscesses [119]. Characterisation of 52 non-oral strains showed similarity to oral strains [120], and the portal of entry for systemic infections is usually the oral cavity [121]. cultured from the same individual 10 years later [118]. The natural habitat of A. actinomycetemcomitans is the oral cavity, but A. actinomycetemcomitan can be isolated from a variety of oral as well as non-oral infectious diseases, including arthritis bacteraemia, endocarditis, osteomyelitis, skin infections, urinary tract infections and various types o abscesses [119]. Characterisation of 52 non-oral strains showed similarity to oral strains [120], and th portal of entry for systemic infections is usually the oral cavity [121]. 4.1. Infective Endocarditis 4.1. Infective Endocarditi The oral cavity is the only known habitat of A. actinomycetemcomitans, but only a few layers of crevicular epithelial cells separate the gingival location from the parenteral space of the host. Entry into the blood stream has not been quantitated, but incidental introductions may occur during tooth brushing, injuries, chewing of granular matters etc., and this may be accelerated by the presence of periodontitis. A. actinomycetemcomitans was originally co-isolated with Actinomyces from actinomycotic lesions [1], and the association with Actinomyces has been confirmed by case reports of infections in a variety of anatomical localizations. Among Actinomyces species, co-isolation of A. actinomycetemcomitans appears restricted to Actinomyces israelii [122,123]. f The oral cavity is the only known habitat of A. actinomycetemcomitans, but only a few layers o crevicular epithelial cells separate the gingival location from the parenteral space of the host. Entr into the blood stream has not been quantitated, but incidental introductions may occur during toot brushing, injuries, chewing of granular matters etc., and this may be accelerated by the presence o periodontitis. A. actinomycetemcomitans was originally co-isolated with Actinomyces from actinomycotic lesions [1], and the association with Actinomyces has been confirmed by case reports o infections in a variety of anatomical localizations. Among Actinomyces species, co-isolation of A actinomycetemcomitans appears restricted to Actinomyces israelii [122 123] Infective endocarditis is an infection of the endocardium, the lining of the interior surfaces of the chambers of the heart. It usually affects the heart valves (Figure 3A), where corrosion and incidental exposure of sub-endothelium tissue during the extensive motion of the valves may serve as a starting point for bacterial adhesion. actinomycetemcomitans appears restricted to Actinomyces israelii [122,123]. Infective endocarditis is an infection of the endocardium, the lining of the interior surfaces of th chambers of the heart. It usually affects the heart valves (Figure 3A), where corrosion and incidenta exposure of sub-endothelium tissue during the extensive motion of the valves may serve as a startin point for bacterial adhesion (A) (B) Figure 3. Imaging signs of infections and inflammation that may be associated with A. actinomycetemcomitans. (A) Transesophageal echocardiography of a heart with mitral valve infective endocarditis. The arrow marks a large vegetation on the posterior leaflet between left atrium (LA) and left ventricle (LV); usually, vegetations caused by A. actinomycetemcomitans are of smaller size. (B) 14-year old girl of African ethnicity. 4.1. Infective Endocarditis 4.1. Infective Endocarditi The radiograph shows an extensive and apparently rapid loss of the periodontal support of the lower incisor 31. Pictures by courtesy of close clinical collaborators of th th Figure 3. Imaging signs of infections and inflammation that may be associated with A. actinomycetemcomitans. (A) Transesophageal echocardiography of a heart with mitral valve infective endocarditis. The arrow marks a large vegetation on the posterior leaflet between left atrium (LA) and left ventricle (LV); usually, vegetations caused by A. actinomycetemcomitans are of smaller size. (B) 14-year old girl of African ethnicity. The radiograph shows an extensive and apparently rapid loss of the periodontal support of the lower incisor 31. Pictures by courtesy of close clinical collaborators of the authors. (B) (A) (B) (A) Figure 3. Imaging signs of infections and inflammation that may be associated with A actinomycetemcomitans. (A) Transesophageal echocardiography of a heart with mitral valve infective endocarditis. The arrow marks a large vegetation on the posterior leaflet between left atrium (LA and left ventricle (LV); usually, vegetations caused by A. actinomycetemcomitans are of smaller size. (B 14-year old girl of African ethnicity. The radiograph shows an extensive and apparently rapid loss o the periodontal support of the lower incisor 31. Pictures by courtesy of close clinical collaborators o Figure 3. Imaging signs of infections and inflammation that may be associated with A. actinomycetemcomitans. (A) Transesophageal echocardiography of a heart with mitral valve infective endocarditis. The arrow marks a large vegetation on the posterior leaflet between left atrium (LA) and left ventricle (LV); usually, vegetations caused by A. actinomycetemcomitans are of smaller size. (B) 14-year old girl of African ethnicity. The radiograph shows an extensive and apparently rapid loss of the periodontal support of the lower incisor 31. Pictures by courtesy of close clinical collaborators of the authors. the authors. A. actinomycetemcomitans is part of the Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella and Kingella (HACEK) group of fastidious Gram-negative bacteria that is responsible for 1.4–3% o i f ti d diti [124 125] Th i i ll i l d d H hil i A. actinomycetemcomitans is part of the Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, and Kingella (HACEK) group of fastidious Gram-negative bacteria that is responsible for 1.4–3% of cases infective endocarditis [124,125]. The group originally included Haemophilus species, Actinobacillus the authors. A. 4.1. Infective Endocarditis 4.1. Infective Endocarditi actinomycetemcomitans is part of the Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella and Kingella (HACEK) group of fastidious Gram-negative bacteria that is responsible for 1.4–3% o cases infective endocarditis [124 125] The group originally included Haemophilus species A. actinomycetemcomitans is part of the Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, and Kingella (HACEK) group of fastidious Gram-negative bacteria that is responsible for 1.4–3% of cases infective endocarditis [124,125]. The group originally included Haemophilus species, Actinobacillus 10 of 18 10 of 18 Pathogens 2019, 8, 243 actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, and Kingella kingae [7]. The HACEK acronym is still valid, but currently denotes non-influenzae Haemophilus sp., Aggregatibacter sp., Cardiobacterium sp., E. corrodens, and Kingella sp. [126]. A. actinomycetemcomitans is the HACEK organism most strongly associated with infective endocarditis [121,125], and bacteraemia with A. actinomycetemcomitans necessitates clarification of this putative focus of infection. In a retrospective study of 87 cases of HACEK bacteraemia from New Zealand, the association between HACEK bacteraemia and infective endocarditis varied among bacterial species ranging from 0% (E. corrodens) to 100% (A. actinomycetemcomitans) [127]. Specific features of infective endocarditis caused by A. actinomycetemcomitans have been reviewed [121]. 4.2. The Complex Interplay with Periodontitis Periodontitis is an inflammatory disease associated with loss of connective tissue and bone around teeth (Figure 3B). The bacterial tooth biofilm initiates the gingival inflammation, and further progression of the periodontal lesion depends on dysbiotic ecological changes within the gingival sulcus area. Unfavourable lifestyles and hygiene contribute to the development and progression of periodontitis, which has been designated as one of mankind’s most common chronic inflammatory diseases [48]. ch has been designated as one of mankind’s most common chronic inflammatory diseases [48]. The complexity of the periodontal microbiota and the variety of clinical symptoms delayed the identification of specific microbial aetiological agents. In 1996, A. actinomycetemcomitans, Porphyromonas gingivalis, and Tannerella forsythia were officially designated as aetiological agents of periodontitis [128]. A. actinomycetemcomitans was targeted based on prevalence studies in health and disease, serum antibody levels, and presence of virulence determinants (Tables 1 and 2). More recently, attention has been directed to the complex interplay between other cultivable and other non-cultivable bacteria in the oral microbiota, as well as to the interplay with the host [16,48,129,130]. Indeed, it has been suggested that A. actinomycetemcomitans conducts its business by concealing itself from the scrutiny of the host immune system, or even being a community activist that suppresses host responses to allow overgrowth of its collaborators [19]. The earlier classification of aggressive periodontitis was based mainly on the clinical presentation and the rapid loss of periodontal tissue [131]. A new classification scheme has been adopted, in which chronic and aggressive forms of the disease are now merged into a single category, which is characterised by a multi-dimensional staging and grading system [132,133]. Staging assesses severity and extent of disease at presentation, and attempts to include the complexity of disease management. The grading provides an evaluation of the risk of progression, and attempts to predict response to standard periodontal therapy [132]. 5. Therapy Treatment of periodontitis aims to stop the progression of the periodontal lesion and to maximise periodontal health [134]. Mechanical debridement of biofilm is considered the most effective therapy, but must be combined with a detailed oral hygiene. If periodontal lesions persist after 3–6 months, a second phase of therapy is planned. A favourable healing potential has been documented for lesions associated with the rapidly progressive, localised periodontitis that affects adolescents [135]. Systemic antibiotics should only be administered as adjunctive therapy in selected cases. Access surgery with regenerative techniques have been used for periodontitis stages III–IV [132,134]. Notable risk factors are non-compliance, smoking, elevated gingival bleeding index, and inadequate plaque control [136]. Very different amoxicillin resistance rates have been reported, ranging from 0% in Switzerland [137], over 33% in Spain [138] to 84% in the United Kingdom [139]. The mechanisms of resistance were not reported. Production of β-lactamase is the most common cause of β-lactam resistance in Gram-negative bacteria, but these enzymes have not been detected in A. actinomycetemcomitans. The fastidious nature of the bacterium is a challenge for antimicrobial susceptibility testing, and methodology as well as interpretative criteria must be addressed when reports of resistance are evaluated. A recent 11 of 18 Pathogens 2019, 8, 243 investigation using different methods could not confirm the emergence of resistance to β-lactams in A. actinomycetemcomitans; the study included strains that had previously been reported as resistant [140]. Thus, there is currently no convincing evidence for replacement of oral amoxicillin when antimicrobial agents are indicated for treatment of A. actinomycetemcomitans-associated periodontitis. Gram-negative bacteria are generally more susceptible to the cephalosporin-class than the penicillin-class of β-lactams. For infective endocarditis, an intravenous course of at least four weeks with a third-generation cephalosporin, or a combination of ampicillin and an aminoglycoside, is recommended [121]. Recently, a well-designed randomised study reported favourable outcomes for oral antimicrobial follow up regimens given to patients with infective endocarditis deemed clinically stable and without complications [141]. A. actinomycetemcomitans could be a candidate microorganism for use of partial oral antimicrobial treatment of infective endocarditis, but the relative rare occurrence of HACEK bacteraemia poses difficulties for additional clinical studies. Abbreviations aae, Aggregatibacter autotransporter adhesin; Api, Aggregatibacter putative invasion; CT, cycling threshold; emaA, extracellular matrix adhesin A; HACEK, Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, and Kingella; JP2 clone, a specific juvenile periodontitis-related bacterial clone; Ltx, leukotoxin; MLST, multilocus sequence type; Omp, outer membrane protein; PCR, polymerase chain reaction; RTX, repeats in toxin; St, serotype; SNP, single nucleotide polymorphism; YadA, Yersinia adhesin A. 6. Conclusions A. actinomycetemcomitans is part of the human microbiota. It can be cultured from one-third of healthy adults, while PCR-based methods suggest a more ubiquitous presence. The bacterium’s tenacious, aggregative character is instrumental for the remarkable genotype stability in colonised hosts, and for progression to persistent, distant infections after incidental entry into the parental space. A. actinomycetemcomitans are commonly detected if adolescents present with periodontitis. Chronically inflamed gingival crevices may spark the repeated, intermittent entry into the blood stream. Its participation in the disease process of periodontitis is beyond reasonable doubt, but its orchestration of severe periodontitis continues to be fascinating and disputed. Adhesion and leukotoxic features are well-described, but interplay with other members of the oral microbiota is more difficult to elucidate, as is the interchangeable position of eliciting antibody response and “staying under the radar”. The recent division into three subspecies or lineages has not been investigated by clinical studies linking disease and lineage. Disease-specific treatment options are currently widely accepted. Author Contributions: All authors made a substantial, direct, and intellectual contribution to the work. N.N.-L. compiled the contributions and made the first draft of the manuscript. All authors approved it for publication. Funding: This research received no external funding. Acknowledgments: Mogens Kilian is thanked for profound inspiration. Furthermore, we thank the members of the European Network for Aggregatibacter actinomycetemcomitans Research (ENAaR; https://projects.au.dk/ aggregatibacter/) for valuable discussions. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. References Applications of molecular ecology in the characterisatio uncultured microorganisms associated with human disease. Rev. Med. Microbiol. 1997, 8, 91–101. [Cross 15. Wilson, M.J.; Weightman, A.J.; Wade, W.G. Applications of molecular ecology in the characterisation of uncultured microorganisms associated with human disease. Rev. Med. Microbiol. 1997, 8, 91–101. [CrossRef] 16. Dewhirst, F.E.; Chen, T.; Izard, J.; Paster, B.J.; Tanner, A.C.; Yu, W.H.; Lakshmanan, A.; Wade, W.G. The human oral microbiome. J. Bacteriol. 2010, 192, 5002–5017. uncultured microorganisms associated with human disease. Rev. Med. Microbiol. 1997, 8, 91–101. [CrossRef] 16. Dewhirst, F.E.; Chen, T.; Izard, J.; Paster, B.J.; Tanner, A.C.; Yu, W.H.; Lakshmanan, A.; Wade, W.G. The human oral microbiome. J. Bacteriol. 2010, 192, 5002–5017. 6. Dewhirst, F.E.; Chen, T.; Izard, J.; Paster, B.J.; Tanner, A.C.; Yu, W.H.; Lakshmanan, A.; Wade, W.G. The hu oral microbiome. J. Bacteriol. 2010, 192, 5002–5017. 17. Bisgaard, M.; Christensen, H.; Clermont, D.; Dijkshoorn, L.; Janda, J.M.; Moore, E.R.B.; Nemec, A.; Nørskov-Lauritsen, N.; Overmann, J.; Reubsaet, F.A.G. The use of genomic DNA sequences as type material for valid publication of bacterial species names will have severe implications for clinical microbiology and related disciplines. Diagn. Microbiol. Infect. Dis. 2019, 95, 102–103. [CrossRef] [PubMed] 18. Asikainen, S.; Chen, C. Oral ecology and person-to-person transmission of Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis. Periodontol 2000 1999, 20, 65–81. 19. Fine, D.H.; Pati, A.G.; Velusamy, S.K. Aggregatibacter actinomycetemcomitans (Aa) Under the Radar: Myths and Misunderstandings of Aa and Its Role in Aggressive Periodontitis. Front. Immunol. 2019, 10, 728. [CrossRef] 20. Hornef, M. Pathogens, commensal symbionts, and pathobionts: Discovery and functional effects on the host. 19. Fine, D.H.; Pati, A.G.; Velusamy, S.K. Aggregatibacter actinomycetemcomitans (Aa) Under the Radar: Myths and Misunderstandings of Aa and Its Role in Aggressive Periodontitis. Front. Immunol. 2019, 10, 728. [CrossRef] 20. Hornef, M. Pathogens, commensal symbionts, and pathobionts: Discovery and functional effects on the host. ILAR J. 2015, 56, 159–162. [CrossRef] 21. Cowan, S.T. Cowan and Steel’s Manual for the Identification of Medical Bacteria, 2nd ed.; Cambridge University Press: Cambridge, UK, 1974; p. 95. 22. Potts, T.V.; Mitra, T.; O’Keefe, T.; Zambon, J.J.; Genco, R.J. Relationships among isolates of oral haemophili as determined by DNA-DNA hybridization. Arch. Microbiol. 1986, 145, 136–141. [CrossRef] 23. Munson, E.; Carroll, K.C. What’s in a name? New bacterial species and changes to taxonomic status from 2012 through 2015. J. Clin. Microbiol. 2017, 55, 24–42. [PubMed] 24. References 1. Klinger, R. Untersuchungen über menschliche Aktinomykose. Zentralbl. Bakteriol. 1912, 62, 191–200. (In German) 1. Klinger, R. Untersuchungen über menschliche Aktinomykose. Zentralbl. Bakteriol. 1912, 62, 191–200. (In German) 2. Topley, W.W.C.; Wilson, G.S. The Principles of Bacteriology and Immunity; Edward Arnold: London, UK, 1929; pp. 1–587. 3. King, E.O.; Tatum, H.W. Actinobacillus actinomycetemcomitans and Hemophilus aphrophilus. J. Infect. Dis. 1962, 111, 85–94. [CrossRef] [PubMed] 4. Slots, J. The predominant cultivable organisms in juvenile periodontitis. Scand. J. Dent. Res. 1976, 84, 1–10. [CrossRef] [PubMed] 12 of 18 Pathogens 2019, 8, 243 5. Newman, M.G.; Socransky, S.S.; Savitt, E.D.; Propas, D.A.; Crawford, A. Studies of the microbiology of periodontosis. J. Periodontol. 1976, 47, 373–379. [CrossRef] [PubMed] 6. Tsai, C.C.; McArthur, W.P.; Baehni, P.C.; Hammond, B.F.; Taichman, N.S. Extraction and partial characterization of a leukotoxin from a plaque-derived Gram-negative microorganism. Infect. Immun. 1979, 25, 427–439. [PubMed] 7. Geraci, J.E.; Wilson, W.R. Symposium on infective endocarditis. III. Endocarditis due to gram-negative bacteria. Report of 56 cases. Mayo Clin. Proc. 1982, 57, 145–148. 8. Ebersole, J.L.; Taubman, M.A.; Smith, D.J.; Genco, R.J.; Frey, D.E. Human immune responses to oral micro-organisms. I. Association of localized juvenile periodontitis (LJP) with serum antibody responses to Actinobacillus actinomycetemcomitans. Clin. Exp. Immunol. 1982, 47, 43–52. 9. Zambon, J.J.; Slots, J.; Genco, R.J. Serology of oral Actinobacillus actinomycetemcomitans and serotype distribution in human periodontal disease. Infect. Immun. 1983, 41, 19–27. 10. Brogan, J.M.; Lally, E.T.; Poulsen, K.; Kilian, M.; Demuth, D.R. Regulation of Actinobacillus actinomycetemcomitans leukotoxin expression: Analysis of the promoter regions of leukotoxic and minimally leukotoxic strains. Infect. Immun. 1994, 62, 501–508. 11. Nørskov-Lauritsen, N.; Kilian, M. Reclassification of Actinobacillus actinomycetemcomitans, Haemophilus aphrophilus, Haemophilus paraphrophilus and Haemophilus segnis as Aggregatibacter actinomycetemcomitans gen. nov., comb. nov., Aggregatibacter aphrophilus comb. nov. and Aggregatibacter segnis comb. nov., and emended description of Aggregatibacter aphrophilus to include V factor-dependent and V factor-independent isolates. Int. J. Syst. Evol. Microbiol. 2006, 56, 2135–2146. 12. Haubek, D.; Ennibi, O.K.; Poulsen, K.; Vaeth, M.; Poulsen, S.; Kilian, M. Risk of aggressive periodontitis in adolescent carriers of the JP2 clone of Aggregatibacter (Actinobacillus) actinomycetemcomitans in Morocco: A prospective longitudinal cohort study. Lancet 2008, 371, 237–242. [CrossRef] 13. Moore, W.E.; Moore, L.V. The bacteria of periodontal diseases. Periodontology 2000 1994, 5, 66–77. [CrossRef] [PubMed] 14. Socransky, S.S.; Haffajee, A.D. Evidence of bacterial etiology: A historical perspective. Periodontology 2000 1994, 5, 7–25. [CrossRef] [PubMed] 5. Wilson, M.J.; Weightman, A.J.; Wade, W.G. References Murra, M.; Lützen, L.; Barut, A.; Zbinden, R.; Lund, M.; Villesen, P.; Nørskov-Lauritsen, N. Whole-genome sequencing of aggregatibacter species isolated from human clinical specimens and description of aggregatibacter kilianii sp. nov. J. Clin. Microbiol. 2018, 56, e00053-18. [PubMed] 25. Takada, K.; Saito, M.; Tsuzukibashi, O.; Kawashima, Y.; Ishida, S.; Hirasawa, M. Characterization of a new serotype g isolate of Aggregatibacter actinomycetemcomitans. Mol. Oral Microbiol. 2010, 25, 200–206. [CrossRef] [PubMed] 13 of 18 13 of 18 Pathogens 2019, 8, 243 26. Brígido, J.A.; da Silveira, V.R.; Rego, R.O.; Nogueira, N.A. Serotypes of Aggregatibacter actinomycetemcomitans in relation to periodontal status and geographic origin of individuals—A review of the literature. Med. Oral Patol. Oral Cir. Bucal 2014, 19, e184–e191. [CrossRef] 27. Saarela, M.; Asikainen, S.; Alaluusua, S.; Pyhälä, L.; Lai, C.H.; Jousimies-Somer, H. Frequency and stability of mono-or poly-infection by Actinobacillus actinomycetemcomitans serotypes a, b, c, d or e. Oral Microbiol. Immunol. 1992, 7, 277–279. [CrossRef] 28. Thiha, K.; Takeuchi, Y.; Umeda, M.; Huang, Y.; Ohnishi, M.; Ishikawa, I. Identification of periodontopathic bacteria in gingival tissue of Japanese periodontitis patients. Oral Microbiol. Immunol. 2007, 22, 201–207. [CrossRef] 29. Rylev, M.; Kilian, M. Prevalence and distribution of principal periodontal pathogens worldwide. J. Clin. Periodontol. 2011, 3, 346–361. [CrossRef] 30. Bandhaya, P.; Saraithong, P.; Likittanasombat, K.; Hengprasith, B.; Torrungruang, K. Aggregatibacter actinomycetemcomitans serotypes, the JP2 clone and cytolethal distending toxin genes in a Thai population. J. Clin. Periodontol. 2012, 39, 519–525. 31. Claesson, R.; Höglund Åberg, C.; Haubek, D.; Johansson, A. Age-related prevalence and characteristics of Aggregatibacter actinomycetemcomitans in periodontitis patients living in Sweden. J. Oral Microbiol. 2017, 9, 1334504. [CrossRef] 32. Yamamoto, M.; Nishihara, T.; Koseki, T.; He, T.; Yamato, K.; Zhang, Y.J.; Nakashima, K.; Oda, S.; Ishikawa, I. Prevalence of Actinobacillus actinomycetemcomitans serotypes in Japanese patients with periodontitis. J. Periodontal Res. 1997, 32, 676–681. [CrossRef] 33. Celenligil, H.; Ebersole, J.L. Analysis of serum antibody responses to periodontopathogens in early-onset periodontitis patients from different geographical locations. J. Clin. Periodontol. 1998, 25, 994–1002. [CrossRef] [PubMed] 34. Saraiva, L.; Rebeis, E.S.; Martins Ede, S.; Sekiguchi, R.T.; Ando-Suguimoto, E.S.; Mafra, C.E.; Holzhausen, M.; Romito, G.A.; Mayer, M.P. IgG sera levels against a subset of periodontopathogens and severity of disease in aggressive periodontitis patients: A cross-sectional study of selected pocket sites. J. Clin. Periodontol. 2014, 41, 943–951. [CrossRef] [PubMed] 35. Poulsen, K.; Theilade, E.; Lally, E.T.; Demuth, D.R.; Kilian, M. References Population structure of Actinobacillus actinomycetemcomitans: A framework for studies of disease-associated properties. Microbiology 1994, 140, 2049–2060. [CrossRef] [PubMed] 36. Kaplan, J.B.; Schreiner, H.C.; Furgang, D.; Fine, D.H. Population structure and genetic diversity of Actinobacillus actinomycetemcomitans strains isolated from localized juvenile periodontitis patients. J. Clin. Microbiol. 2002, 40, 1181–1187. [CrossRef] 37. Haubek, D.; Poulsen, K.; Kilian, M. Microevolution and patterns of dissemination of the JP2 clone of Aggregatibacter (Actinobacillus) Actinomycetemcomitans. Infect. Immun. 2007, 75, 3080–3088. [CrossRef] p Aggregatibacter (Actinobacillus) Actinomycetemcomitans. Infect. Immun. 2007, 75, 3080–3088. [CrossRef] 38. DiRienzo, J.M.; Slots, J. Genetic approach to the study of epidemiology and pathogenesis of Actinobacillus actinomycetemcomitans in localized juvenile periodontitis. Arch. Oral Biol. 1990, 35, S79–S84. [CrossRef] 38. DiRienzo, J.M.; Slots, J. Genetic approach to the study of epidemiology and pathogenesis of Actinobacillus actinomycetemcomitans in localized juvenile periodontitis. Arch. Oral Biol. 1990, 35, S79–S84. [CrossRef] 39. Zambon, J.J.; Sunday, G.J.; Smutko, J.S. Molecular genetic analysis of Actinobacillus actinomycetemcomitans epidemiology. J. Periodontol. 1990, 61, 75–80. [CrossRef] 40. DiRienzo, J.M.; Slots, J.; Sixou, M.; Sol, M.A.; Harmon, R.; McKay, T.L. Specific genetic variants of Actinobacillus actinomycetemcomitans correlate with disease and health in a regional population of families with localized juvenile periodontitis. Infect. Immun. 1994, 62, 3058–3065. 41. Eriksen, K.T.; Haubek, D.; Poulsen, K. Intragenomic recombination in the highly leukotoxic JP2 clone of Actinobacillus actinomycetemcomitans. Microbiology 2005, 151, 3371–3379. [CrossRef] 42. Höglund Åberg, C.; Haubek, D.; Kwamin, F.; Johansson, A.; Claesson, R. Leukotoxic activity of Aggregatibacter actinomycetemcomitans and periodontal attachment loss. PLoS ONE 2014, 9, e104095. [CrossRef] 43. Kittichotirat, W.; Bumgarner, R.E.; Asikainen, S.; Chen, C. Identification of the pangenome and its components in 14 distinct Aggregatibacter actinomycetemcomitans strains by comparative genomic analysis. PLoS ONE 2011, 6, e22420. [CrossRef] 44. Jorth, P.; Whiteley, M. An evolutionary link between natural transformation and CRISPR adaptive immunity. MBio 2012, 3, e00309-12. [CrossRef] [PubMed] 45. Kittichotirat, W.; Bumgarner, R.E.; Chen, C. Evolutionary divergence of Aggregatibacter actinomycetemcomitans. J. Dent. Res. 2016, 95, 94–101. [CrossRef] 14 of 18 Pathogens 2019, 8, 243 14 of 18 46. Nedergaard, S.; Kobel, C.M.; Nielsen, M.B.; Møller, R.T.; Jensen, A.B.; Nørskov-Lauritsen, N.; The Danish HACEK Study Group. Whole genome sequencing of Aggregatibacter actinomycetemcomitans cultured from blood stream infections reveals three major phylogenetic groups including a novel lineage expressing serotype a membrane O polysaccharide. Pathogens. submitted. 47. Holm, P. The influence of carbon dioxide on the growth of Actinobacillus actinomycetemcomitans (Bacterium actinomycetem comitans Klinger 1912). Acta. Pathol. Microbiol. References Scand. 1954, 34, 235–248. [CrossRef] [PubMed] 48. Henderson, B.; Ward, J.M.; Ready, D. Aggregatibacter (Actinobacillus) actinomycetemcomitans: A triple A* periodontopathogen? Periodontology 2000 2010, 54, 78–105. [CrossRef] [PubMed] 49. Möller, Å.J.R. Microbiological examination of root canals and periapical tissues of human teeth. Scand. Dent. J. 1966, 74, 5–6. 50. Johansson, E.; Claesson, R.; van Dijken, J.W.V. Antibacterial effect of ozone on cariogenic bacterial species. J. Dent. 2009, 37, 449–453. [CrossRef] [PubMed] 51. Khurshid, Z.; Zohaib, S.; Najeeb, S.; Zafar, M.S.; Slowey, P.D.; Almas, K. Human saliva collection devices for proteomics: An update. Int. J. Mol. Sci. 2016, 17, 846. [CrossRef] [PubMed] 52. Khurshid, Z.; Mali, M.; Naseem, M.; Najeeb, S.; Zafar, M.S. Human gingival crevicular fluids (GCF) proteomics: An overview. Dent. J. 2017, 5, 12. [CrossRef] 53. Slots, J. Selective medium for isolation of Actinobacillus actinomycetemcomitans. J. Clin. Microbiol. 1982, 15, 606–609. [PubMed] 54. Höglund Åberg, C.; Kwamin, F.; Claesson, R.; Johansson, A.; Haubek, D. Presence of JP2 and non-JP2 genotypes of Aggregatibacter actinomycetemcomitans and periodontal attachment loss in adolescents in Ghana. J. Periodontol. 2012, 83, 1520–1528. [CrossRef] [PubMed] 55. Marsh, P. Microbial ecology of dental plaque and its significance in health and disease. Adv. Dent. Res. 1994, 8, 263–271. [CrossRef] [PubMed] 56. Slots, J. Salient Biochemical Characters of Actinobacillus actinomycetemcomitans. Arch. Microbiol. 1982, 131, 60–67. [CrossRef] [PubMed] 57. Couturier, M.R.; Mehinovic, E.; Croft, A.C.; Mark, A.; Fisher, M.A. Identification of HACEK clinical isolates by matrix-assisted laser desorption ionization—Time of flight mass spectrometry. J. Clin. Microbiol. 2011, 49, 1104–1106. [CrossRef] [PubMed] 58. Zambon, J.J.; DeLuca, C.; Slots, J.; Genco, R.J. Studies of leukotoxin from Actinobacillus actinomycetemcomitans using the promyelocytic HL-60 cell line. Infect. Immun. 1983, 40, 205–212. 59. Haubek, D.; DiRienzo, J.J.; Tinoco, E.M.; Westergaard, J.; Lopez, N.J.; Chung, C.P.; Poulsen, K.; Kilian, M. Racial tropism of a highly toxic clone of Actinbacillus actinomycetemcomitans associated with juvenile periodontitis. J. Clin. Microbiol. 1997, 35, 3037–3042. 60. Sampathkumar, V.; Velusamy, S.K.; Godboley, D.; Fine, D.H. Increased leukotoxin production: Characterization of 100 base pairs within the 530 base pair leukotoxin promoter region of Aggregatibacter actinomycetemcomitans. Sci. Rep. 2017, 7, 1887. [CrossRef] 61. Poulsen, K.; Ennibi, O.K.; Haubek, D. Improved PCR for detection of the highly leukotoxic JP2 clone of Actinobacillus actinomycetemcomitans in subgingival plaque samples. J. Clin. Microbiol. 2003, 41, 4829–4832. [CrossRef] 2. Kirakodu, S.S.; Govindaswami, M.; Novak, M.J.; Ebersole, J.L.; Novak, K.F. References Teng, Y.T.; Hu, W. Expression cloning of a periodontitis-associated apoptotic effector, cagE homologu Actinobacillus actinomycetemcomitans. Biochem. Biophys. Res. Commun. 2003, 303, 1086–1094. [CrossRef] 75. Ramasubbu, N.; Thomas, L.M.; Ragunath, C.; Kaplan, J.B. Structural analysis of dispersin B, a biofilm-releasing glycoside hydrolase from the periodontopathogen Actinobacillus actinomycetemcomitans. J. Mol. Biol. 2005, 349, 475–486. [CrossRef] [PubMed] 76. Karched, M.; Ihalin, R.; Eneslätt, K.; Zhong, D.; Oscarsson, J.; Wai, S.N.; Chen, C.; Asikainen, S.E. Vesicle-independent extracellular release of a proinflammatory outer membrane lipoprotein in free-soluble form. BMC Microbiol. 2008, 8, 18. [CrossRef] [PubMed] 77. Rosan, B.; Slots, J.; Lamont, R.J.; Listgarten, M.A.; Nelson, G.M. Actinobacillus actinomycetemcomitans fimbriae. Oral. Microbiol. Immunol. 1988, 3, 58–63. [CrossRef] 78. Harano, K.; Yamanaka, A.; Okuda, K. An antiserum to a synthetic fimbrial peptide of Actinobacillus actinomycetemcomitans blocked adhesion of the microorganism. FEMS Microbiol. Lett. 1995, 130, 279–285. [CrossRef] 79. Fine, D.H.; Furgang, D.; Schreiner, H.C.; Goncharoff, P.; Charlesworth, J.; Ghazwan, G.; Fitzgerald-Bocarsly, P.; Figurski, D.H. Phenotypic variation in Actinobacillus actinomycetemcomitans during laboratory growth: Implications for virulence. Microbiology 1999, 145, 1335–1347. [CrossRef] 80. Inoue, T.; Tanimoto, I.; Ohta, H.; Kato, K.; Murayama, Y.; Fukui, K. Molecular characterization of low-molecular-weight component protein, Flp, in Actinobacillus actinomycetemcomitans fimbriae. Microbiol. Immunol. 1998, 42, 253–258. [CrossRef] 81. Kachlany, S.C.; Planet, P.J.; Desalle, R.; Fine, D.H.; Figurski, D.H.; Kaplan, J.B. flp-1, the first representative of a new pilin gene subfamily, is required for non-specific adherence of Actinobacillus actinomycetemcomitans. Mol. Microbiol. 2001, 40, 542–554. [CrossRef] 82. Haase, E.M.; Zmuda, J.L.; Scannapieco, F.A. Identification and molecular analysis of rough-colony-specific outer membrane proteins of Actinobacillus actinomycetemcomitans. Infect. Immun. 1999, 67, 2901–2908. 3. Wang, Y.; Chen, C. Mutation analysis of the flp operon in Actinobacillus actinomycetemcomitans. Gene 2 351, 61–71. [CrossRef] [PubMed] 84. Perez, B.A.; Planet, P.J.; Kachlany, S.C.; Tomich, M.; Fine, D.H.; Figurski, D.H. Genetic analysis of the requirement for flp-2, tadV, and rcpB in Actinobacillus actinomycetemcomitans biofilm formation. J. Bacteriol. 2006, 188, 6361–6375. [CrossRef] [PubMed] 85. Schreiner, H.C.; Sinatra, K.; Kaplan, J.B.; Furgang, D.; Kachlany, S.C.; Planet, P.J.; Perez, B.A.; Figurski, D.H.; Fine, D.H. Tight-adherence genes of Actinobacillus actinomycetemcomitans are required for virulence in a rat model. Proc. Natl. Acad. Sci. USA 2003, 100, 7295–7300. [CrossRef] [PubMed] Acad. Sci. USA 2003, 100, 7295–7300. [CrossRef] [PubM 86. Wang, Y.; Liu, A.; Chen, C. Genetic basis for conversion of rough-to-smooth colony morphology in Actinobacillus actinomycetemcomitans. Infect. Immun. 2005, 73, 3749–3753. [CrossRef] [PubMed] 87. References Optimizing qPCR for quantification of periodontal pathogens in a complex plaque biofilm. Open Dent. J. 2008, 2, 49–55. [Cross 63. Yoshida, A.; Ennibi, O.K.; Miyazaki, H.; Hoshino, T.; Hayashida, H.; Nishihara, T.; Awano, S.; Ansai, T. Quantitative discrimination of Aggregatibacter actinomycetemcomitans highly leukotoxic JP2 clone from non-JP2 clones in diagnosis of aggressive periodontitis. BMC Inf. Dis. 2012, 12, 253. [CrossRef] [PubMed] 64. Vˇetrovský, T.; Baldrian, P. The variability of the 16S rRNA gene in bacterial genomes and its consequences for bacterial community analyses. PLoS ONE 2013, 8, e57923. [CrossRef] [PubMed] 65. Suzuki, N.; Nakano, Y.; Yoshida, Y.; Ikeda, D.; Koga, T. Identification of Actinobacillus actinomycetemcomitans serotypes by multiplex PCR. J. Clin. Microbiol. 2001, 9, 2002–2005. [CrossRef] [PubMed] 66. Kaplan, J.B.; Perry, M.B.; MacLean, L.L.; Furgang, D.; Wilson, M.E.; Fine, D.H. Structural and genetic analyses of O polysaccharide from Actinobacillus actinomycetemcomitans serotype f. Infect. Immun. 2001, 69, 5375–5384. [CrossRef] 15 of 18 Pathogens 2019, 8, 243 15 of 18 67. Tomich, M.; Planet, P.J.; Figurski, D.H. The tad locus: Postcards from the widespread colonization island. Nat. Rev. Microbiol. 2007, 5, 363–375. [CrossRef] 68. Rose, J.E.; Meyer, D.H.; Fives-Taylor, P.M. Aae, an autotransporter involved in adhesion of Actinobacillus actinomycetemcomitans to epithelial cells. Infect. Immun. 2003, 71, 2384–2393. [CrossRef] 69. Danforth, D.R.; Tang-Siegel, G.; Ruiz, T.; Mintz, K.P. A nonfimbrial adhesin of Aggregatibacter actinomycetemcomitans mediates biofilm biogenesis. Infect. Immun. 2018, 87, e00704-18. [CrossRef] 70. Yue, G.; Kaplan, J.B.; Furgang, D.; Mansfield, K.G.; Fine, D.H. A second Aggregatibacter actinomycetemcomitans autotransporter adhesin exhibits specificity for buccal epithelial cells in humans and Old World primates. Infect. Immun. 2007, 75, 4440–4448. [CrossRef] f 71. Vega, B.A.; Belinka, B.A.; Kachlany, S.C. Aggregatibacter actinomycetemcomitans Leukotoxin (LtxA; Leukothera®): Mechanisms of Action and Therapeutic Applications. Toxins 2019, 11, 489. [CrossRef] 72. Sugai, M.; Kawamoto, T.; Pérès, S.Y.; Ueno, Y.; Komatsuzawa, H.; Fujiwara, T.; Kurihara, H.; Suginaka, H.; Oswald, E. The cell cycle-specific growth-inhibitory factor produced by Actinobacillus actinomycetemcomitans is a cytolethal distending toxin. Infect. Immun. 1998, 66, 5008–5019. 73. Boesze-Battaglia, K.; Alexander, D.; Dlaki´c, M.; Shenker, B.J. A Journey of cytolethal distending to through cell membranes. Front. Cell. Infect. Microbiol. 2016, 6, 81. [CrossRef] [PubMed] 73. Boesze-Battaglia, K.; Alexander, D.; Dlaki´c, M.; Shenker, B.J. A Journey of cytolethal distending toxins through cell membranes. Front. Cell. Infect. Microbiol. 2016, 6, 81. [CrossRef] [PubMed] 74. Teng, Y.T.; Hu, W. Expression cloning of a periodontitis-associated apoptotic effector, cagE homologue, in 74. References Pei, Z.; Niu, Z.; Shi, S.; Shi, L.; Tang, C. Phenotypic changes in nonfimbriated smooth strains of Aggregatibacter actinomycetemcomitans grown in low-humidity solid medium. Ultrastruct. Pathol. 2013, 37, 121–126. [CrossRef] 16 of 18 16 of 18 Pathogens 2019, 8, 243 88. Fine, D.H.; Velliyagounder, K.; Furgang, D.; Kaplan, J.B. The Actinobacillus actinomycetemcomitans autotransporter adhesin Aae exhibits specificity for buccal epithelial cells from humans and old world primates. Infect. Immun. 2005, 73, 1947–1953. [CrossRef] 89. Li, L.; Matevski, D.; Aspiras, M.; Ellen, R.P.; Lépine, G. Two epithelial cell invasion-related loci of pathogen Actinobacillus actinomycetemcomitans. Oral. Microbiol. Immunol. 2004, 19, 16–25. [CrossRef 90. Mintz, K.P. Identification of an extracellular matrix protein adhesin, EmaA, which mediates the adhesion of Actinobacillus actinomycetemcomitans to collagen. Microbiology 2004, 150, 2677–2688. [CrossRef] 91. Tang, G.; Kitten, T.; Munro, C.L.; Wellman, G.C.; Mintz, K.P. EmaA, a potential virulence determinant of Aggregatibacter actinomycetemcomitans in infective endocarditis. Infect. Immun. 2008, 76, 2316–2324. [CrossRef] 92. Cornelissen, C.N. Subversion of nutritional immunity by the pathogenic Neisseriae. Pathog. Dis. 2018, 76, 91. Tang, G.; Kitten, T.; Munro, C.L.; Wellman, G.C.; Mintz, K.P. EmaA, a potential virulence determinant of Aggregatibacter actinomycetemcomitans in infective endocarditis. Infect. Immun. 2008, 76, 2316–2324. [CrossRef] 91. Tang, G.; Kitten, T.; Munro, C.L.; Wellman, G.C.; Mintz, K.P. EmaA, a potential virulence determinant of Aggregatibacter actinomycetemcomitans in infective endocarditis. Infect. Immun. 2008, 76, 2316–2324. [CrossRef] 92. Cornelissen, C.N. Subversion of nutritional immunity by the pathogenic Neisseriae. Pathog. Dis. 2018, 76, ftx112. [CrossRef] 92. Cornelissen, C.N. Subversion of nutritional immunity by the pathogenic Neisseriae. Pathog. Dis. 2018, 76, ftx112. [CrossRef] 93. Benz, R. Channel formation by RTX-toxins of pathogenic bacteria: Basis of their biological activity. Biochim. Biophys. Acta 2016, 1858, 526–537. [CrossRef] [PubMed] Biochim. Biophys. Acta 2016, 1858, 526–537. [CrossRef] [PubMed] 94. Frey, J.; Kuhnert, P. RTX toxins in Pasteurellaceae. Int. J. Med. Microbiol. 2002, 292, 149–158. [CrossRef] [PubMed] 94. Frey, J.; Kuhnert, P. RTX toxins in Pasteurellaceae. Int. J. Med. Microbiol. 2002, 292, 149–158. [CrossRef] [PubMed] 95. Taichman, N.S.; Tsai, C.C.; Baehni, P.C.; Stoller, N.; McArthur, W.P. Interaction of inflammatory cells and oral microorganisms. IV. In vitro release of lysosomal constituents from polymorphonuclear leukocytes exposed to supragingival and subgingival bacterial plaque. Infect. Immun. 1977, 16, 1013–1023. [PubMed] 96. Lally, E.T.; Golub, E.E.; Kieba, I.R.; Taichman, N.S.; Rosenbloom, J.; Rosenbloom, J.C.; Gibson, C.W.; Demuth, D.R. Analysis of the Actinobacillus actinomycetemcomitans leukotoxin gene. Delineation of unique features and comparison to homologous toxins. J. References Biol. Chem. 1989, 264, 15451–15456. [PubMed] 97. Kolodrubetz, D.; Dailey, T.; Ebersole, J.; Kraig, E. Cloning and expression of the leukotoxin gene from Actinobacillus actinomycetemcomitans. Infect. Immun. 1989, 57, 1465–1469. [PubMed] 98. Balashova, N.V.; Crosby, J.A.; Al Ghofaily, L.; Kachlany, S.C. Leukotoxin confers beta-hemolytic activity to Actinobacillus actinomycetemcomitans. Infect. Immun. 2006, 74, 2015–2021. [CrossRef] [PubMed] 99. Claesson, R.; Gudmundson, J.; Åberg, C.H.; Haubek, D.; Johansson, A. Detection of 640-bp deletion in Aggregatibacter actinomycetemcomitans leukotoxin promoter region in isolates from an adolescent of Ethiopian origin. J. Oral Microbiol. 2015, 7, 26974. [CrossRef] 100. He, T.; Nishihara, T.; Demuth, D.R.; Ishikawa, I. A novel insertion sequence increases the expression of leukotoxicity in Actinobacillus actinomycetemcomitans clinical isolates. J. Periodontol. 1999, 70, 1261–1268. [CrossRef] 101. Höglund Åberg, C.; Kwamin, F.; Claesson, R.; Dáhlen, G.; Johansson, A.; Haubek, D. Progression of attachment loss is strongly associated with presence of the JP2 genotype of Aggregatibacter actinomycetemcomitans: A prospective cohort study of a young adolescent population. J. Clin. Periodontol. 2014, 41, 232–241. [CrossRef] 102. Burgess, D.; Huang, H.; Harrison, P.; Aukhil, I.; Shaddox, L. Aggregatibacter actinomycetemcomitans in African Americans with localized aggressive periodontitis JDR Clin Trans Res 2017 2 249 257 [CrossRef] 101. Höglund Åberg, C.; Kwamin, F.; Claesson, R.; Dáhlen, G.; Johansson, A.; Haubek, D. Progression of attachment loss is strongly associated with presence of the JP2 genotype of Aggregatibacter actinomycetemcomitans: A prospective cohort study of a young adolescent population. J. Clin. Periodontol. 2014, 41, 232–241. [CrossRef] 102. Burgess, D.; Huang, H.; Harrison, P.; Aukhil, I.; Shaddox, L. Aggregatibacter actinomycetemcomitans in African Americans with localized aggressive periodontitis. JDR Clin. Trans. Res. 2017, 2, 249–257. [CrossRef] 103. Leung, W.K.; Ngai, V.K.; Yau, J.Y.; Cheung, B.P.; Tsang, P.W.; Corbet, E.F. Characterization of Actinobacillus actinomycetemcomitans isolated from young Chinese aggressive periodontitis patients. J. Periodontal Res. 2005, 40, 258–268. [CrossRef] [PubMed] 104. van der Reijden, W.A.; Bosch-Tijhof, C.J.; van der Velden, U.; van Winkelhoff, A.J. Java project on periodontal diseases: Serotype distribution of Aggregatibacter actinomycetemcomitans and serotype dynamics over an 8-year period. J. Clin. Periodontol. 2008, 35, 407–492. [CrossRef] [PubMed] 105. Pahumunto, N.; Ruangsri, P.; Wongsuwanlert, M.; Piwat, S.; Dahlen, G.; Teanpaisan, R. Aggregatibacter actinomycetemcomitans serotypes and DGGE subtypes in Thai adults with chronic periodontitis. Arch. Oral Biol. 2015, 60, 1789–1796. [CrossRef] [PubMed] 106. Alaluusua, S.; Asikainen, S. Detection and distribution of Actinobacillus actinomycetemcomitans in the primary dentition. J. Periodontol. 1988, 59, 504–507. [CrossRef] [PubMed] 107. References Slots, J.; Reynolds, H.S.; Genco, R.J. Actinobacillus actinomycetemcomitans in human periodontal disease: A cross-sectional microbiological investigation. Infect. Immun. 1980, 29, 1013–1020. [PubMed] 108. Frisken, W.; Higgins, T.; Palmer, J.M. The incidence of periodontopathic microorganisms in young children. Oral Microbiol. Immunol. 1990, 5, 43–45. [CrossRef] 109. Kononen, E.; Asikainen, S.; Jousimies-Somer, H. The early colonization of gram-negative anaerobic bacteria in edentulous infants. Oral Microbiol. Immunol. 1992, 7, 28–31. [CrossRef] 17 of 18 17 of 18 Pathogens 2019, 8, 243 110. Merglova, V.; Polenik, P. Early colonization of the oral cavity in 6-and 12-month-old infants by cariogenic and periodontal pathogens: A case-control study. Folia Microbiol. 2016, 61, 423–429. [CrossRef] 111. Preus, H.R.; Zambon, J.J.; Dunford, R.G.; Genco, R.J. The distribution and transmission of Actinobacillus actinomycetemcomitans in families with established adult periodontitis. J. Periodontol. 1994, 65, 2–7. [CrossRef] 111. Preus, H.R.; Zambon, J.J.; Dunford, R.G.; Genco, R.J. The distribution and transmission of Actinobacillus actinomycetemcomitans in families with established adult periodontitis. J. Periodontol. 1994, 65, 2–7. [CrossRef] 112. Asikainen, S.; Chen, C.; Slots, J. Likelihood of transmitting Actinobacillus actinomycetemcomitans and Porplyromonna gingivalis in families with periodontitis Oral Microbiol Immunol 1996 11 387–394 111. Preus, H.R.; Zambon, J.J.; Dunford, R.G.; Genco, R.J. The distribution and transmission of Actinobacillus actinomycetemcomitans in families with established adult periodontitis. J. Periodontol. 1994, 65, 2–7. [CrossRef] 112. Asikainen, S.; Chen, C.; Slots, J. Likelihood of transmitting Actinobacillus actinomycetemcomitans and Porplyromonna gingivalis in families with periodontitis. Oral Microbiol. Immunol. 1996, 11, 387–394. [CrossRef] 112. Asikainen, S.; Chen, C.; Slots, J. Likelihood of transmitting Actinobacillus actinomycetemcomitans and Porplyromonna gingivalis in families with periodontitis. Oral Microbiol. Immunol. 1996, 11, 387–394. [CrossRef] 113. Rêgo, R.O.; Spolidorio, D.M.; Salvador, S.L.; Cirelli, J.A. Transmission of Aggregatibacter actinomycetemcomitans between Brazilian women with severe chronic periodontitis and their children. Braz. Dent. J. 2007, 18, 220–224. [CrossRef] 114. Van Winkelhoff, A.J.; Boutaga, K. Transmission of periodontal bacteria and models of infection. J. Clin. Periodontol. 2005, 32, 16–27. [CrossRef] [PubMed] 115. Do˘gan, B.; Kipalev, A.S.; Okte, E.; Sultan, N.; Asikainen, S.E. Consistent intrafamilial transmission of Actinobacillus actinomycetemcomitans despite clonal diversity. J. Periodontol. 2008, 79, 307–315. [CrossRef] [PubMed] 116. Saarela, M.H.; Do˘gan, B.; Alaluusua, S.; Asikainen, S. Persistence of oral colonization by the same Actinobacillus actinomycetemcomitans strain(s). J. Periodontol. 1999, 70, 504–509. [CrossRef] [PubMed] 117. Haubek, D.; Ennibi, O.K.; Vaeth, M.; Poulsen, S.; Poulsen, K. Stability of the JP2 clone of Aggregatibacter actinomycetemcomitans. J. Dent. Res. 2009, 88, 856–860. References [CrossRef] [PubMed] 118. Sun, R.; Kittichotirat, W.; Wang, J.; Jan, M.; Chen, W.; Asikainen, S.; Bumgarner, R.; Chen, C. Genomic stability of Aggregatibacter actinomycetemcomitans during persistent oral infection in human. PLoS ONE 2013, 8, e66472. [CrossRef] 119. van Winkelhoff, A.J.; Slots, J. Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in nonoral infections. Periodontology 2000 1999, 20, 122–135. [CrossRef] 120. Paju, S.; Carlson, P.; Jousimies-Somer, H.; Asikainen, S. Heterogeneity of Actinobacillus actinomycetemcomitans strains in various human infections and relationships between serotype, genotype, and antimicrobial susceptibility. J. Clin. Microbiol. 2000, 38, 79–84. 21. Paturel, L.; Casalta, J.P.; Habib, G.; Nezri, M.; Raoult, D. Actinobacillus actinomycetemcomitans endocard Clin. Microbiol. Infect. 2004, 10, 98–118. [CrossRef] 22. Clarridge, J.E.; Zhang, Q. Genotypic diversity of clinical Actinomyces species: Phenotype, source, and dis correlation among genospecies. J. Clin. Microbiol. 2002, 40, 3442–3448. [CrossRef] 123. Kaplan, A.H.; Weber, D.J.; Oddone, E.Z.; Perfect, J.R. Infection due to Actinobacillus actinomycetemcomitans: 15 cases and review. Rev. Infect. Dis. 1989, 11, 46–63. [CrossRef] [PubMed] 124. Brouqui, P.; Raoult, D. Endocarditis due to rare and fastidious bacteria. Clin. Microbiol. Rev. 2001, 14, 177–207. [CrossRef] [PubMed] 125. Nørskov-Lauritsen, N. Classification, identification, and clinical significance of Haemophilus and Aggregatibacter species with host specificity for humans. Clin. Microbiol. Rev. 2014, 27, 214–240. [CrossRef] [PubMed] 126. Lützen, L.; Olesen, B.; Voldstedlund, M.; Christensen, J.J.; Moser, C.; Knudsen, J.D.; Fuursted, K.; Hartmeyer, G.N.; Chen, M.; Søndergaard, T.S.; et al. Incidence of HACEK bacteraemia in Denmark: A 6-year population-based study. Int. J. Infect. Dis. 2018, 68, 83–87. [CrossRef] [PubMed] 127. Yew, H.S.; Chambers, S.T.; Roberts, S.A.; Holland, D.J.; Julian, K.A.; Raymond, N.J.; Beardsley, J.; Read, K.M.; Murdoch, D.R. Association between HACEK bacteraemia and endocarditis. J. Med. Microbiol. 2014, 63, 892–895. [PubMed] 128. Armitage, G.C.; Offenbacher, S. Consensus Report. Periodontal diseases: Epidemiology and diagnosis. Ann. Periodontol. 1996, 1, 16–22. [CrossRef] 129. Fine, D.H.; Markowitz, K.; Fairlie, K.; Tischio-Bereski, D.; Ferrendiz, J.; Furgang, D.; Paster, B.J.; Dewhirst, F.E. A consortium of Aggregatibacter actinomycetemcomitans, Streptococcus parasanguinis, and Filifactor alocis is present in sites prior to bone loss in a longitudinal study of localized aggressive periodontitis. J. Clin. Microbiol. 2013, 51, 2850–2861. [CrossRef] 130. Ebbers, M.; Lübcke, P.M.; Volzke, J.; Kriebel, K.; Hieke, C.; Engelmann, R.; Lang, H.; Kreikemeyer, B.; Müller-Hilke, B. Interplay between P. gingivalis, F. nucleatum and A. actinomycetemcomitans in murine alveolar bone loss, arthritis onset and progression. Sci. Rep. 2018, 8, 15129. References [CrossRef] Pathogens 2019, 8, 243 18 of 18 18 of 18 131. Armitage, G.C. Development of a classification system for periodontal diseases and conditions. Northwest Dent. 2000, 79, 31–35. [CrossRef] 132. Tonetti, M.S.; Greenwell, H.; Kornman, K.S. Staging and grading of periodontitis: Framework and proposal of a new classification and case definition. J. Clin. Periodontal. 2018, 45, S149–S161. [CrossRef] 133. Fine, D.H.; Patil, A.G.; Loos, B.G. Classification and diagnosis of aggressive periodontitis. J. Periodontol. 2018, 89, s103–s119. [CrossRef] [PubMed] 134. Teughels, W.; Dhondt, R.; Dekeyser, D.; Quirynen, M. Treatment of aggressive periodontitis. Periodontology 2000 2014, 65, 107–133. [CrossRef] [PubMed] 135. Hamad, C.; Haller, B.; Hoffmann, T.; Lorenz, K. Five-year results of nonsurgical generalized aggressive periodontitis. Quintessence Int. 2019, 50, 104–113. [PubMed] 136. Dopico, J.; Nibali, L.; Donos, N. Disease progression in aggressive periodontitis patients. A restrspective study. J. Clin. Periodontol. 2016, 43, 531–537. [CrossRef] [PubMed] 137. Kulik, E.M.; Lenkeit, K.; Chenaux, S.; Meyer, J. Antimicrobial susceptibility of periodontopathogenic bacteria. J. Antimicrob. Chemother. 2008, 61, 1087–1091. [CrossRef] [PubMed] 138. van Winkelhoff, A.J.; Herrera, D.; Oteo, A.; Sanz, M. Antimicrobial profiles of periodontal pathogens isolated from periodontitis patients in the Netherlands and Spain. J. Clin. Periodontol. 2005, 32, 893–898. [CrossRef] [PubMed] 139. Akrivopoulou, C.; Green, I.M.; Donos, N.; Nair, S.P.; Ready, D. Aggregatibacter actinomycetemcomitans serotype prevalence and antibiotic resistance in a UK population with periodontitis. J. Glob. Antimicrob. Resist. 2017, 10, 54–58. [CrossRef] 140. Jensen, A.B.; Haubek, D.; Claesson, R.; Johansson, A.; Nørskov-Lauritsen, N. Comprehensive antimicrobial susceptibility testing of a large collection of clinical strains of Aggregatibacter actinomycetemcomitans does not identify resistance to amoxicillin. J. Clin. Periodontol. 2019, 46, 846–854. [CrossRef] 141. Iversen, K.; Ihlemann, N.; Gill, S.U.; Madsen, T.; Elming, H.; Jensen, K.T.; Bruun, N.E.; Høfsten, D.E.; Fursted, K.; Christensen, J.J.; et al. Partial oral versus intravenous antibiotic treatment of endocarditis. N. Engl. J. Med. 2019, 380, 415–424. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

https://openalex.org/W4287251953

https://zenodo.org/records/5571086/files/Manor.8T.V1.4.pdf

English

8T - The Grand Theory of Everything

Zenodo (CERN European Organization for Nuclear Research)

cc-by

𝟖𝐓−𝑽𝒂𝒓𝒊𝒂𝒕𝒊𝒐𝒏𝒂𝒍 𝑴𝒂𝒏𝒊𝒇𝒐𝒍𝒅𝒔 Manor O −29.3.2021 𝟖𝐓−𝑽𝒂𝒓𝒊𝒂𝒕𝒊𝒐𝒏𝒂𝒍 𝑴𝒂𝒏𝒊𝒇𝒐𝒍𝒅𝒔 Manor O −29.3.2021 𝟖𝐓−𝑽𝒂𝒓𝒊𝒂𝒕𝒊𝒐𝒏𝒂𝒍 𝑴𝒂𝒏𝒊𝒇𝒐𝒍𝒅𝒔 Fermions, Manifolds and Arbitrary Variations Define a Lorentz manifold, which is the connected manifold with (3,1) signature: s = (M, g) Invoke it to be stationary by Euler Lagrange operator, 𝑆= 𝑆0 × ℝ: Invoke it to be stationary by Euler Lagrange operator, 𝑆= 𝑆0 × ℝ: ℓ= (s, s′,t) ℓ= (s, s′,t) ℓ= (s, s′,t) ∂ℓ ∂s −( d dt) ∂ℓ ∂s′ = 0 Develop the last equation: ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) If the Lorenztion manifold to be stationary and no data is attainable from the first three terms, we can require the manifold to those two conditions: If the Lorenztion manifold to be stationary and no data is attainable from the first three terms, we can require the manifold to those two conditions: ∂g ∂t = 0 , − ∂2g′ ∂t2 = 0 ∂g ∂t = 0 , − ∂2g′ ∂t2 = 0 If these two are hold to be true, we have areas of extremum curvature on the manifold and negative time invariant acceleration. The demand of extrunum curvature to stay as they are overtime means the acceleration cannot affect them – if so, directed away from them. This in agreement with what we speculate as "dark energy". Notice that M is now the matric tensor, g is the Ricci flow. That is the result of parametrizing the manifold to "s" variable and inserting it to EL operator, yielding agreement with Einstein principle of equivalence. ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t δg − ∂ℓ ∂s ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 δg′ = 0 Equation reads, length to manifold, manifold to matric, matric to flow, flow to time. δg as amount of arbitrary variations, which by demands of stationarity we require to vanish. Discretizing and partitioning the term 𝛿𝑔 into a series of sub elements, we can derive the existence of fermions, i.e. showing that it must have an even amount of elements, which differ in sign and create nine threefold combination, and no more than two distinct elements. Equation reads, length to manifold, manifold to matric, matric to flow, flow to time. δg as amount of arbitrary variations, which by demands of stationarity we require to vanish. Discretizing and partitioning the term 𝛿𝑔 into a series of sub elements, we can derive the existence of fermions, i.e. 𝟖𝐓−𝑽𝒂𝒓𝒊𝒂𝒕𝒊𝒐𝒏𝒂𝒍 𝑴𝒂𝒏𝒊𝒇𝒐𝒍𝒅𝒔 Manor O −29.3.2021 Fermions, Manifolds and Arbitrary Variations Outward Acceleration from Extremum Curvatures ____________________4 Fermions as Variations of a Lorentz Manifold ______________________5 Series s Coupling Constant the rimes and Bosons, P Deriving the Coupling Constants Series ____________________________7 Majestic (3) Possible Meanings __________________________________11 Majestic (3) Spin (1/2) & Nature and Morphisms ____________________12 Majestic (3) as an Electron & Stationary Manifolds_____________________15 Weak Interaction Left Orientation & (3) is the Electron_________________18 Duality of the First Three Interactions at 26 __________________________21 Pauli Exclusion Principle ________________________________________23 Series of Masses and Generations & SEW unification __________________25 Cosmological Predictions Arrow of Time & Continuous Discrete Aspects__________________________30 The Homogenous Universe & Commutivity______________________________32 Universe Future & Gauge Fields & Higgs & Antimatter & PW Duality The Future Universe -& Limits___________________________________________34 Gauge fields Prediction & Fluid Turbulences _______________________________37 Quark Masses Mixing & Orthogonal Curvatures______________________________39 The Higgs Field & Proof PNP___________________________________________42 Anti-matter & Dirac Delta Variation & Odds _______________________________44 Jets & Spiral galaxies & Principle of Least Curvature ________________________ 48 PW Duality & Gravity – Acceleration__________________________________ 53 The Feynman Variation on Lorentz Manifolds ______________________________56 Time & Rings & Fermion Gravity Gravity as a Combination of Particles______________________________________58 Galaxies growth & Predicting the Graviton Mass______________________________59 Fermions are Imperfect Rings_____________________________________________60 1 Manor O The Primorial and The Arrow ____________________________________________62 Universe Packet Creation & Flatness_______________________________________63 Prime-Fold Quark Chains & Quark Confinement _____________________________ 65 Infinite Dimensions & Einstein Equivalence ________________________________ 66 Gluon confinement & Universe packet Symmetry ___________________________69 Proof: Quarks are fundamental &Yang Milles Conjecture_______________________71 heory Final T des of a ha S Golden 8T& QFT Axioms & QFT Weaknesses______________________________________74 The Three Critical Theorems______________________________________________76 Refuting Magnetic Monopoles_____________________________________________77 Most Symmetrical Interaction is the Weak____________________________________78 Hermitian Conjugation and Primes__________________________________________79 Final Shot at Quantum Relativity___________________________________________ 80 Total Variations Pairing___________________________________________________82 Boson & Fermion Propagations____________________________________________83 The Lagrangian Variation ________________________________________________84 Cluster Decomposition__________________________________________________85 Hadronic Symmetries & Densities_________________________________________86 Feynman Diagrams & Freezing Time_______________________________________88 The Axis of Evil______________________________________________________90 Bosonic Probabilities___________________________________________________91 Conservation of Variation________________________________________________92 Bosonic Strings & Curvature Absorptions ___________________________________93 Proof: The Riemann Hypothesis__________________________________________95 Visualizations________________________________________________________99 Strong Interaction and the Electron _______________________________________103 Virtual Curvatures & Curvature Knots ________________________________________104 Birth of Universes & EMT symmetry_____________________________________106 Primorial and Probability_______________________________________________109 Asymptotic Freedom & Manifold Jumpps__________________________________110 Curvature Eruptions __________________________________________________114 2 The Primorial and The Arrow ____________________________________________62 Universe Packet Creation & Flatness_______________________________________63 Prime-Fold Quark Chains & Quark Confinement _____________________________ 65 Infinite Dimensions & Einstein Equivalence ________________________________ 66 Gluon confinement & Universe packet Symmetry ___________________________69 Proof: Quarks are fundamental &Yang Milles Conjecture_______________________71 heory Final T des of a ha S Golden 8T& QFT Axioms & QFT Weaknesses______________________________________74 The Three Critical Theorems______________________________________________76 Refuting Magnetic Monopoles_____________________________________________77 Most Symmetrical Interaction is the Weak____________________________________78 Hermitian Conjugation and Primes__________________________________________79 Final Shot at Quantum Relativity___________________________________________ 80 Total Variations Pairing___________________________________________________82 Boson & Fermion Propagations____________________________________________83 The Lagrangian Variation ________________________________________________84 Cluster Decomposition__________________________________________________85 Hadronic Symmetries & Densities_________________________________________86 Feynman Diagrams & Freezing Time_______________________________________88 The Axis of Evil______________________________________________________90 Bosonic Probabilities___________________________________________________91 Conservation of Variation________________________________________________92 Bosonic Strings & Curvature Absorptions ___________________________________93 Proof: The Riemann Hypothesis__________________________________________95 Visualizations________________________________________________________99 Strong Interaction and the Electron _______________________________________103 Virtual Curvatures & Curvature Knots ________________________________________104 Birth of Universes & EMT symmetry_____________________________________106 Primorial and Probability_______________________________________________109 Asymptotic Freedom & Manifold Jumpps__________________________________110 Curvature Eruptions __________________________________________________114 2 Manor O 8T Versus M-T ______________________________________________________115 Universe Packet Density _______________________________________________116 The Commuter & The Curvature Code __________________________________118 Degrees of Freedom& Curvature Spectra's ________________________________120 The Ghost Neutrino ___________________________________________________122 Alternative for Dark Matter___________________________________________123 Curvature Spikes & Graviton Illusion ____________________________________125 Curvature Terminators _________________________________________________128 Fermion Distributions & Next Planck Constant ______________________________129 Primorial Nature & Imaginary Couplings ___________________________________131 The Chameleon Particle & Higgs Field ____________________________________134 The Vacuum & Star's Life ______________________________________________136 The Arch of Time Arrows & Net Versus Spin ______________________________139 Spin Symmetries – Free Electrons________________________________________141 Exotic Charges & Unbounded Quarks _____________________________________142 Spikes Stability & Curvature Vortexes____________________________________145 Nested Curvatures & Energy and Mass____________________________________147 Sphere Shape of Stars & Inner Curvature __________________________________149 Higgs as Measurement Tool & The Action_________________________________151 Ripping Space-Time & Interference & Quantum Manifolds ____________________153 Homeomorphisms& Amplitudes & Quantum Entanglement ____________________158 Electron W Boson Duality & Objects in Class________________________________ 164 High Energy Vacuums& W Duality ________________________________________165 Odd Photons Absorptions & Co-products_____________________________________168 Variational Distributions & Co-Variance____________________________________170 Spin Chronicles & Gravity Classes ________________________________________ 173 HE Paradox & N-Tuples & Observables & Devils Gift_________________________176 Interacting Fields & Five-Vector& Heredity & Dark Matter______________________181 Bosonic Mass Pattern & Strong Paradox & Multiverse Uncertainties_______________186 Primorial wave & Fermion Superposition's & Matter Transfer____________________190 Discrete Curvature Ripples & Fields Mixture & Epilogue________________________192 3 Manor O Manor O Fermions, Manifolds and Arbitrary Variations Fermions, Manifolds and Arbitrary Variations showing that it must have an even amount of elements, which differ in sign and create nine threefold combination, and no more than two distinct elements. δg1 + δg2 … = ∑𝛿g𝑖 𝑁 𝑖=1 ∑𝛿g𝑖= 0 𝑁 𝑖=1 4 Manor O Manor O Given four elements distinct: Given four elements distinct: Given four elements distinct: Given four elements distinct: δg1 + δg2 > 0 δg3 + δg4 < 0 δg1 + δg2 > 0 δg3 + δg4 < 0 If δg1 + δg2 + δg3 + δg4 ≠0 δg1 + δg2 + δg3 + δg4 ≠0 Than the overall series cannot vanish, by that logic we need even amounts of equal elements of pluses and minuses. The amount must be even and summed as zero, ensuring stationary Lorentz manifold. Suppose that we had three distinct elements, two pluses and minus: Than the overall series cannot vanish, by that logic we need even amounts of equal elements of pluses and minuses. The amount must be even and summed as zero, ensuring stationary Lorentz manifold. Suppose that we had three distinct elements, two pluses and minus: δg1 + δg2 + δg3 > 0 or δg1 + δg2 + δg3 < 0 δg1 + δg2 + δg3 < 0 Demanding the series to vanish this will exclude this result, and so there could not be three distinct elements in the series, else the overall series will not vanish to zero. As a result of those sceneries, we require the series to have an even amount of variation elements, manifesting as two distinct elements in the series, which differ in sign. If we allow those sub elements in the series to vary as well, and by the above reasoning, there are only two elements in the series, they are varying in a discrete way, or forming a group. Let it be only four elements in the series and one of the pluses just changed its nature 𝐎: δg1 →δg2 δg1 + δg1 + δg2 + δg2 = 0 δg1 + δg1 + δg2 + δg2 = 0 To: To: δg1 + δg2 + δg2 + δg2 = 0 There must be a way to bring it back to where it was, so the overall series can vanish, it takes another map, on the varying element to bring it back to where it was. Y∶ δg2 →δg1 5 Manor O Therefore, we have Lorenztion manifold with arbitrary variations, which vanish into matter based on that idea. One does not know whether these are the actual variations, as the mathematics does not entail any details about that. Therefore, the graph could be inaccurate in elements order. The colors meant to elements pairing. Reader does not have to agree with what one did, but as one will calculate the ratios of all the forces known, one kindly asks the reader to keep reading as some truth seem to obey the reasoning one is suggesting. Manor O Therefore, to bring an element to itself given only two varying elements in the series we need two distinct maps, which attach a varying element to itself, by a threefold combination. δg1(O)δg2(Y)δg1 For example. Even though the sub elements in the series are varying, the overall series can vanish. Now, count all the ways of possible combinations of those elements. We are going to analyze by the integral signs. Since it is a group, there is a natural map, which change an element to itself. One built his analysis firstly on those natural maps. So: So: (1(e)1(e)1) 2(e)2(e)2 (221) (112) (211) (122) (212) (121) The first two combinations are by the natural maps and one used them to build the other combinations. Overall, there are eight such combinations and additional one arrow combination, which yield (333). Here is how one built it, starting from those two natural maps. (Arrows to variations, colors to pairings): The first two combinations are by the natural maps and one used them to build the other combinations. Overall, there are eight such combinations and additional one arrow combination, which yield (333). Here is how one built it, starting from those two natural maps. (Arrows to variations, colors to pairings): The first two combinations are by the natural maps and one used them to build the other combinations. Overall, there are eight such combinations and additional one arrow combination, which yield (333). Here is how one built it, starting from those two natural maps. (Arrows to variations, colors to pairings): 6 2𝟏𝟏 −−− 212 𝟏𝟐𝟐 −−−− 121 221 −−−−− −−− 112 222 −−−− 111 333 2𝟏𝟏 −−− 212 𝟏𝟐𝟐 −−−− 121 221 −−−−− −−− 112 222 −−−− 111 333 2𝟏𝟏 −−− 212 𝟏𝟐𝟐 −−−− 121 221 −−−−− −−− 112 222 −−−− 111 333 6 Bosons, Primes and the Coupling Constants Series Theorem (1) – nature will not allow a prime amount of variation to appear by itself. Define prime to be (2n+1) variations. 1.1) Prime amounts appear in pairs. Theorem (2): Nature will generate force if a prime net amount of arbitrary variation will appear. Net variations will appear when combine two amounts of prime variations. Two does not appear, as it is an even amount of variations, which vanish. Define 𝑁𝑉 as the series of prime net variations and the number one. 𝑁𝑉= 2𝑉+ 1 𝑉≥0 𝑁𝑉= 2𝑉+ 1 Count all the prime pairs of variations, (3,3) (3, 5) (3,7) (3,11), (3,13) … (5, 3) (5,5) (5,7) (5,11) (5,13) … (7, 3) (7,5) (7,7) (7,11) (7,13) … That is a tedious work, but here is the great part. We only need to do it twice to find what nature does repeatedly. Manor O Since we have only two varying elements in the series, we can eliminate almost all the number yields a and after a sum that is divisible by two options, as we require obtaining divisible by three. By The following reasoning: Two as we have only two varying elements. Three as these elements create a certain amount of threefold combinations. The sums satisfying the condition is (5,13) or (7,11) and (29,31). The sums satisfying the condition is (5,13) or (7,11) and (29,31). Of course, there are more as prime pairs are infinite, but as one mentioned, it took two pairs to understand the principle: ) : 3 ( Theorem Each prime pair should have a net variation element 𝑁𝑉 proportional to Total Variations value divided by two. This will be vivid with actual examples: This will be vivid with actual examples: Analyze the (𝟕, 𝟏𝟏) total variations pair with NV = (+𝟏): Total variations sum is divisible by two: 18/2 = 9 And then by three And then by three 9/3 = 3 We know that we have 𝑁𝑉 = (+1) so it can be extracted to yield: 𝐹1 = 8 + 1 𝐹1 = 8 + 1 However, even amounts of variations vanish so we can ignore the element 8 and write: However, even amounts of variations vanish so we can ignore the element 8 and write: 𝐹1 = 1 Analyze the next pair of total variations (𝟐𝟗 , 𝟑𝟏) with 𝐍𝐕= (+𝟑) 29 + 31 = 60 60/2 = 30 Analyze the next pair of total variations (𝟐𝟗 , 𝟑𝟏) with 𝐍𝐕= (+𝟑) 29 + 31 = 60 60/2 = 30 In addition, three devisible. We know we have three net variations so extract: 27 + 3 In addition, three devisible. We know we have three net variations so extract: 27 + 3 In addition, three devisible. We know we have three net variations so extract: 27 + 3 Now that is all you need to complete the series and calculate the next element: 8 Manor O Notice: 27 = 24 + (3) (8 ∗3) = 24 Obtain the ratio: [8 + 1]: [27 + 3] = [8 + 1]: [24 + (3)] + 3 [8 + 1]: [27 + 3] = [8 + 1]: [(𝟖∗𝟑) + (𝟑)] + 3 Next element 𝑉= 2 and 𝑁𝑉= +5 so if the overall idea to be correct we take this element, multiply by the even sum of the previous element in the series, add extra invariant three, and we know we need add to the sum the extracted 𝑁𝑉. [(𝟐𝟒∗𝟓) + (𝟑)] + 𝟓 = 𝟏𝟐𝟖. Next in line: Next in line: [(120 ∗7) + (3)] + 7 = 850 [(840 ∗11) + (3)] + 11 = 9254 Nature is than the interplay between averages of total arbitrary variations pairs to net variations/curvature. To calculate the magnitude of an element R: Nature is than the interplay between averages of total arbitrary variations pairs to net variations/curvature. To calculate the magnitude of an element R: 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. Analyze the (𝟕, 𝟏𝟏) total variations pair with NV = (+𝟏): (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 9 9 Manor O 𝒫0 = 8 + (1) 𝒫𝑁 # = (2ℳ∗ ∏𝒫𝑉 𝑉=𝑁 𝑉=1 + (ℳ)) + 𝒫𝑉= 30:128: 850: 9254.. (1.2. 𝐴) 𝒫0 = 8 + (1) 𝒫𝑁 # = (2ℳ∗ ∏𝒫𝑉 𝑉=𝑁 𝑉=1 + (ℳ)) + 𝒫𝑉= 30:128: 850: 9254.. (1.2. 𝐴) Equation (1.2.A) is another way of representation. ℳ As the first letter of the word as possible " ℳ " . ial function primor for classification as a meant Sign # . ' ajestic M ' magnitudes. Notice the strong symmetry pattern of this equation. Option 2 The invariant three as a result-There are perfect clusters of variations such (8 ∗3), (24 ∗5) , which experience additional net variation causing them to destabilize. The result is manifested in the invariant three. The additional variation could affect them could be external. Less likeable option. It is less likeable as we can them create mixtures (8*3) to destabilize by five net variations, and yield invariant three and all the beauty in which we attained than will be lost. Option 1 The invariant three as a cause. Notice that all the element within the closed term (8 ∗. . ) Are two and three divisible to vanish into matter. The invariant three prevents it completely and then as a result, a net variation will appear. The net variation is proportional to the right element in the bracket(8 ∗ 3) ∝3 𝑎𝑛𝑑 (24 ∗5) ∝5 . 𝐎𝐯𝐞𝐫𝐯𝐢𝐞𝐰 𝐨𝐟 𝐫𝐞𝐚𝐬𝐨𝐧𝐢𝐧𝐠: Axiom – prime amount of arbitrary variations pair to each other Their overall sum must be dividable by two and three Two distinct elements, which create threefold combinations Define generated force as prime net variation in which we associate 𝑁𝑉 element total variations 2 ∝ to 𝑁𝑉 element by the relative size of total pairing Net variation function cannot contain an even, as it will vanish We searched for the first two prime pairs and derived 8 + 1 and 27 + 3 We saw that nature multiply the even sum by the next element of 𝑁𝑉 We found the invariant three element. We obtained a number to which we add the extracted net variation We calculated the next element to be exactly 128 and the two next interactions: 8 + (1): (24 + (3)) + 3:(120 + (3)) + 5:(840 + (3)) + 7 … (1): (30): (128): (850):(9254) … _____________________________________________________________ Axiom – prime amount of arbitrary variations pair to each other Their overall sum must be dividable by two and three We calculated the next element to be exactly 128 and the two next interactions: We calculated the next element to be exactly 128 and the two next interactions: 8 + (1): (24 + (3)) + 3:(120 + (3)) + 5:(840 + (3)) + 7 … 8 + (1): (24 + (3)) + 3:(120 + (3)) + 5:(840 + (3)) + 7 … (1): (30): (128): (850):(9254) … (1): (30): (128): (850):(9254) … 10 Predictions and Conclusions There are infinite Bosonic fields, or Lorentz manifold net curvature. Prime isomorphic. The clusters of total variations grow much more rapidly than the net variations. The larger the cluster, the weaker the interaction. There are infinite Bosonic fields, or Lorentz manifold net curvature. Prime isomorphic. The clusters of total variations grow much more rapidly than the net variations. The larger the cluster, the weaker the interaction. The magnitude of interactions is manifested an in infinite series of ratios The magnitude of interactions is manifested an in infinite series of ratios 1: 30: 128: 850: 9254…by the expressions, notice that (1.2) differ by an additional term: 1: 30: 128: 850: 9254…by the expressions, notice that (1.2) differ by an additional term: 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.1) (1.2) ) 3 ( Majestic Possible meanings of the ) 3 ( Majestic Possible meanings of the Option 3 both appear at the same time and they are related to - and net variation as duals Invariant three The each other by more fundamental relation, which is not attainable nor explainable. Even though we found a jewel, many questions still stand unanswered. Why the invariant three appear as it is and do not change is another question. Of course that the real answer to that question is that one does not know. However, one can guess and say that three is the smallest prime. If we assume that nature is Lagrangian oriented, it might be the minimal way to destabilize the cluster of potential matter. Why add thirty seven additional variations when only three is needed? It's a logical argument not a proof, and therefore rightfully argued by reader. One was trying to argue that three is a Prime minima, that is the reason for its invariant in the series. Remember that even variations vanish, so two is not an option. 11 Manor O Manor O Nature as a Set of Morphisms Nature as a Set of Morphisms ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g ∂t2 = 0 (1) (1) 𝐹𝑉=0 = 8 + (1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … Now we can define a functor to switch the setting, from a topological setting to a setting of a set. By doing so we can analyze nature in a completely different, hopefully simpler way. ∧:𝑇𝑜𝑝⟶𝑆𝑒𝑡 Now, we have a set with two elements as presented in equation below: K = (𝑀, 𝑔) The set has certain subsets. The first subset is the subset of primes or the number one. The manifold, or the set K, is generating the subset P. this subset is responsible for fermions clustering and bosonic propagations. ℙ= (2𝑛+ 1 ∪(+1) ); ℙ ∈ K (3.11) (3.11) The second subset is of even amount of curvature, which vanish into matter by threefold combination of two distinct elements that differ in sign. That is the subset described in the 8T by the arbitrary variation term presented in equation (3.21). Ε = (2𝑛 ); Ε ∈ K (3.21) (3.21) Finally, there is a morphism, isomorphism in particular given by the same equation that validate Einstein principle of equivalence. Finally, there is a morphism, isomorphism in particular given by the same equation that validate Einstein principle of equivalence. ∂g ∂t ≡𝜕2𝑔′ ∂t2 ∈ K (3.3) (3.3) The set will generate time invariant acceleration from subsets of the matric tensor that has extremum amounts of curvature that stay as they are over time. Changing the setting of nature into a set category and then partitioning the set makes things, as the author believes easier to grasp. 12 Manor O 1 2. Overall: Correlating the Majestic (3) To Spin (1/2) In the paper about primes, we have shown that they create a non-abelian group with 1/2 as generator, by using the anti-commutation relation and vanishing of even amounts of variation. It recently become evident to one that we can represent each element in the series in the following way: Since three is a prime, and aligned on the prime ring located on critical line of 1 2. The sums alongside of it are even sums such as 8, 24, 120 and so on. These expressions are interesting, as one believes they represent the notion of matter or fermions. Notice that we omitted the additional net variation, which is also prime. Meaning it is also on the Prime ring located on 1 2. Overall: [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 13 13 Manor O So the construction within the parenthesis is prime but the overall additional net is changing it, and making it: (1/2 + 1/2) = 1. So the overall 1: 30: 128 will have to do with certain elements that have element one.We already know these are Bosons, as we found the coupling constants series. If so, than the rest of the terms are Fermions, as only (1/2) is there. So it is the Majestic three, in this paper is the one half element to destabilize perfect clusters of variations and causing a net variation to appear. Notice that one chose the first option in regards to the meaning of the invariant three, as we had in part two three ideas to it possible meaning. We have proved that the Majestic three is Spin. We also proved, that bosons will propagate within variation clusters destabilized by one-half, or matter. These are non-trivial statements. We only used one equation, not experiment nor inherited knowledge. Using that framework, we can see why Bosons will propagate from Fermions. Since its invariant, all matter must have the same spin one-half So (2N) are variation clusters, the majestic three is really a destabilizing factor which is spin one half yielding matter. Because of that process, a boson will propagate from within the fermion. The nature of the boson is correlated to right element of the term: (8 ∗3) →3 (weak particle),(24 ∗5) →5 or a photon, so on. 14 14 Manor O Manor O Majestic (3) as the Electron 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.2) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 ..... [(8 ∗3) + (3)] →[2𝑁1 + 1 2] [(24 ∗5) + (3)] →[2𝑁2 + 1 2] [(120 ∗7) + (3)] →[2𝑁3 + 1 2] In previous paper, (part three) we called the (1/2) an element to destabilize Perfect clusters of variations and causing a net variation to appear. In this part, we can call it the Electron. Later in the thesis, we will prove it by putting inside the equation of the fine structure constant. In previous paper, (part three) we called the (1/2) an element to destabilize Perfect clusters of variations and causing a net variation to appear. In this part, we can call it the Electron. Later in the thesis, we will prove it by putting inside the equation of the fine structure constant. In previous paper, (part three) we called the (1/2) an element to destabilize Perfect clusters of variations and causing a net variation to appear. In this part, we can call it the Electron. Later in the thesis, we will prove it by putting inside the equation of the fine structure constant. 15 Manor O Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 0: 2𝑁() variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations When we first discovered the coupling constants equation, we only saw the analytical aspect, by and the ratio between the total variations to net variations. However, by setting the equation on the geometrical realm and examining the critical line of the primes, we can get a deeper insight to what is going on. We are able to analyze the trait of spin, we can understand why Bosons have spin one and the Invariant three or spin one-half. Therefore, it is the electron, which causes the boson propagation from clusters of potential matter. Majestic (3) as the Electron Sure, we knew that, but we did not have the mathematical equation to describe it. The coupling constants equation has than another powerful use; it describes what it going on in elementary level, not just the magnitude of the interactions. It was only available to us when we examined the geometrical realm. Please notice that the electron is inside potential cluster [2𝑁2 + 𝟏 𝟐] so we would not be able to know where it is within the cluster, it blends in [120+3] = 123. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ Spin 0: 2𝑁() variations – perfect clusters of variations Spin 1 2 : 2𝑁() + 3 variations – destabilized by the invariant three. Electron for the third coupling. Spin 1: 2𝑁() + 3 + 𝑁𝑉 - resulting in net variation of prime discrete amount. Spin 1: 2𝑁() + 3 + 𝑁𝑉 - resulting in net variation of prime discrete amount. Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations – Such as gravity. We have taken the third element in the series, as we are familiar with the nature of the electrons due to the great minds of the past century, but the following result would apply to each element in the series from the second and above. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ 16 Manor O Manor O Universe Packets - Stationary Manifolds The main equation: ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) In agreement with our model of the universe. Negative time invariant acceleration From areas of extremum curvatures on the manifold. Validating the Einstein equivalence principle between gravity and acceleration. Again, we assume no data is available from the first three terms, no indication They agree with a stationary Lorentz manifold. Now we can represent the equation (1) in a different way, if there are many stationary Lorentz manifolds we can write: 𝜕ℓ 𝜕𝑆1 − 𝜕ℓ 𝜕𝑆2 = 0 (1.52) (1.52) Alternatively: 𝜕ℒ 𝜕𝑆𝑀 − 𝜕ℒ 𝜕𝑛 = 0 (1.53) 𝜕ℒ 𝜕𝑆𝑀 − 𝜕ℒ 𝜕𝑛 = 0 (1.53) (1.53) 17 𝜕𝑆𝑀 −𝜕𝑛 = 0 (1.53) 𝜕ℒ 𝜕𝑠1 𝜕𝑠1 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−𝜕ℒ 𝜕𝑠2 𝜕𝑠2 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2) 𝜕ℒ 𝜕𝑠1 𝜕𝑠1 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−𝜕ℒ 𝜕𝑠2 𝜕𝑠2 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2) (2) 𝜕𝑠1 𝜕𝑀𝜕𝑔𝜕𝑡 𝜕𝑠2 𝜕𝑀𝜕𝑔𝜕𝑡 17 Manor O Manor O Manor O Notice that the only term which is not a prime after added the Majestic three or spin one half is the second element in the series, in which we associate with the weak interaction. [(8 ∗3) + (3)] = 27 As the series is increasing and each term inside the parenthesis is creating a higher prime than the previous element, in order of weak interaction to be of the same nature of the rest of the forces, we would need that the sum of the parenthesis to be a prime, we look for the closest higher prime: [(8 ∗3) + (3)] →29 [(8 ∗3) + (3)] →29 So in order to be like the rest of the forces. Meaning to have a prime inside a parenthesis, it lacks a certain amount of variation. If we associate each interaction to be invariant to direction – and the Cause of such a trait could be the prime term inside the parenthesis, than the weak interaction would differ by its nature. The fact that the term inside the parenthesis is not on the critical line of the primes, but left to it, can explain why the weak interaction is left oriented and differ by its nature by the rest in terms of its spin. We have proved that the majestic three is really a different representation of spin, which destabilizes clusters of perfect variations causing the 𝑁𝑉 to appear, which overall yield a propagation of a Boson from the fermion, and therefore gives us the beautiful series of coupling constants. If all the Terms on the critical line of primes are yielding interactions that are invariant to direction, than one could predict the weak interaction to be spin oriented to the left by the ratio below, since the strong interaction is also not on the critical line, such orientation could exist in its regards as well. 123 843 9243 27 Th f t th t th t i id th th i i t th iti l li f th i b t l ft t 123 843 9243 27 The fact that the term inside the parenthesis is not on the critical line of the primes, but left to it, can explain why the weak interaction is left oriented and differ by its nature by the rest in terms of its spin. Weak Interaction Negative Left orientation Weak Interaction Negative Left orientation 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.1) (1.2) 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 …. [(8 ∗3) + (3)] →[2𝑁1 + 1 2] [(24 ∗5) + (3)] →[2𝑁2 + 1 2] [(120 ∗7) + (3)] →[2𝑁3 + 1 2] [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 Notice that each term in the series within the parenthesis is prime → (123, 843,9243 … )…as one did not calculate the entire series he is going to assume that is would be true concerning each higher element in the series. We are leaving out the net variation in this part. Notice that each term in the series within the parenthesis is prime → (123, 843,9243 … )…as one did not calculate the entire series he is going to assume that is would be true concerning each higher element in the series. We are leaving out the net variation in this part. Notice that each term in the series within the parenthesis is prime → (123, 843,9243 … )…as one did not calculate the entire series he is going to assume that is would be true concerning each higher element in the series. We are leaving out the net variation in this part. 18 Manor O We have proved that the majestic three is really a different representation of spin, which destabilizes clusters of perfect variations causing the 𝑁𝑉 to appear, which overall yield a propagation of a Boson from the fermion, and therefore gives us the beautiful series of coupling constants. If all the Terms on the critical line of primes are yielding interactions that are invariant to direction, than one could predict the weak interaction to be spin oriented to the left by the ratio below, since the strong interaction is also not on the critical line, such orientation could exist in its regards as well. 27 −29 = −2 (1 2 −2) = −3 2 19 19 Manor O Manor O Manor O The Coupling Constants Series – Majestic Three is the Electron Recall that arbitrary variations vanish in pairs of even numbers. That axiom in our framework related to fermions and allowed us to make a transformation regarding the strong interaction: 8 + (1) → (1) 8 + (1) → (1) So we can use it to prove that the majestic three is indeed an electron and solidify our theory and its validity: 32 128 = 8 + (1) 128 32 128 = 8 + (1) 128 32 128 = 8 + (1) 128 32 128 = 8 + (1) 128 Even amount of variations taken to vanish so the final form of equation above is exactly like the equation in the beginning of the paper, with the electron. Even amount of variations taken to vanish so the final form of equation above is exactly like the equation in the beginning of the paper, with the electron. 8 + (1) 128 → (1) 128 = ℯ2 4𝜋𝓀𝑐 20 Manor O Mathematical Duality of Forces-Virtual Variations We will take the equation built and first three developments: 8 + (1): [(8 ∗3) + (3)] + 3: [(24 ∗5) + (3)] + 5 We will take the equation built and first three developments: 8 + (1): [(8 ∗3) + (3)] + 3: [(24 ∗5) + (3)] + 5 The idea: we will allow the net variations to vary, and when they have the same value, than the expressions inside the parentheses will become scalar multiple. This will be done by using the idea of virtual variations: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (3)] + 3 Notice that now the third is a scalar multiple of the second by a factor of five: [(24 ∗5) + (3)] + 3 [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 3 [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 3 [(8 ∗3) + (3)] + 3 Therefore, the weak and the electric are differing now by a scalar. That is simply beautiful. However, the strong force just accepted that extra two variations so it is just become: Therefore, the weak and the electric are differing now by a scalar. That is simply beautiful. However, the strong force just accepted that extra two variations so it is just become: 8 + (1) + 2 →8 + (1). As Even amounts of variations vanish. It does not affect it. We can try something more interesting, and that is the real purpose of the part: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (3)] + 𝟐 8 + (1) + 3 21 Proof: The Pauli Exclusion Principle 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850:9254.. (1.2) (1.2) 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 We have seen that we can change the term outside the parenthesis, and so we can reach duality between the forces. When we did it in the first three terms, we saw that their duality is exactly on 24+2 variations, which is in agreement with what we know in other theories of GUT. We briefly mention in that paper, that we cannot touch the invariant three. This will be the subject of this part. If we for example combine: [(24 ∗5) + (3)] + 5 ± INTEGER… We can switch and change the terms outside the parenthesis, as those are net variations and they do not seem to obey to any strict rules. However, we could not touch the invariant three and now we will examine deeply the reason. [(24 ∗5) + (3) + (3)] + 5 = [(24 ∗5) + 𝐸𝑣𝑒𝑛)] + 5 𝐸𝑣𝑒𝑛= 0 [(24 ∗5) + 0)] + 5 →Impossible [(24 ∗5) + (3) + (3)] + 5 = [(24 ∗5) + 𝐸𝑣𝑒𝑛)] + 5 𝐸𝑣𝑒𝑛= 0 [(24 ∗5) + 0)] + 5 →Impossible As even amount of variations vanish. Recall that the invariant three is the cause; It is the destabilizing factor yielding a net variation. In the case of the third element, it is the Electron. So using that framework, we can see why we cannot combine two electrons or invariant three elements together. The term than becomes meaningless, a photon cannot propagate from nowhere and the coupling constant series does not makes sense anymore. So the invariant three cannot be combined, it will repel each other. The net variation however can be changed and switched, which makes the flexibility and duality of the forces. The equation is with complete agreement with our understanding; we are just examining additional meaning of it. Manor O Manor O Manor O Now this will ruin the duality and the series, the weak and the electric are not isomorphic, and the strong just got a prime amount of variations that cannot vanish. To solve that we can define a virtual exchange of variation →(1𝑣). [8 + (1)] + 3 −(𝟏𝒗): [(24 ∗5) + (3)] + 𝟑 The real variations are (+3) but to ensure the nature of the strong force, there is a virtual exchange of one variation, marked in bold. For a very short time period, the strong is now a scalar multiple of the other two. Overall, they have the same prime amount of net variations – will mean they are at equivalence relation. For the first three forces: 𝑁𝑉= +(3). [8 + (1)] + 3 −(𝟏𝒗): [(8 ∗3) + (3)] + 3 ∶ [(24 ∗5) + (3)] + 𝟑 [8 + (1)] + 3 −(𝟏𝒗): [(8 ∗3) + (3)] + 3 ∶ [(24 ∗5) + (3)] + 𝟑 We can say that there are three real exchanges and one virtual, so overall four exchanges, which causes all the forces to align on the 𝑁𝑉= +(3). Taking the average of the Sum: 4/2 = 2 𝑛𝑒𝑡. We can say that there are three real exchanges and one virtual, so overall four exchanges, which causes all the forces to align on the 𝑁𝑉= +(3). Taking the average of the Sum: 4/2 = 2 𝑛𝑒𝑡. We can say that there are three real exchanges and one virtual, so overall four exchanges, The converging value of the those exchanges will modify the middle element: [(8 ∗3) + (3)] + 3. Since we want to keep the prime net variation as it is, to ensure duality, and we can't touch the invariant three, we add this (+2), the first term: ((8 ∗3) + 2) = 26. The point where they three aligned will be 24 + 2 variations. certain agreement with this number exist. 22 22 Manor O Proof: The Pauli Exclusion Principle It allows us to examine it from a deeper, more profound view. Now we can understand why fermions do not commute – because even variations vanish and so bosons will not be propagated. 23 Manor O If we eliminate the electron, than no boson will be propagate at all. However, consider the following: [(24 ∗5) + (3)] + 5 + [(24 ∗5) + (3)] + 5+. . = [(24 ∗5) + (3)] + 7 + [(24 ∗5) + (3)] + 3 +. . = While we cannot touch the terms inside the parenthesis, we can change and combine the net variation, there seems to be no limitation in regards to that operation, we have done it before, and showed that the forces can be scalar multiples. We can cluster the net variations, which means that many electrons can emit net variations together, That is bosons, which agrees to what we know as laser, or what we know as bosons commutation relation in QFT. However, using the 8-theory framework we can get a new and fresh insight on why those things are the way they are using the coupling constant equation. As we mentioned in part four of the paper series on coupling constants, the invariant three blends in the total cluster of the fermions, so we cannot know where he is. That is in agreement with the Heisenberg principle of uncertainty. 24 Curvature is Not Allowed ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂g ∂t = 0 ∩ 𝜕2g′ ∂t2 = 0 𝜕ℒ 𝜕𝑠1 − 𝜕ℒ 𝜕𝑠2 = 0 Curvature is Not Allowed ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 24 24 Manor O We partitioned and discretized the arbitrary variation term and derived the existence of Fermion. In particular, we have shown that it must have an even amount of elements, which differ in sign and create nine threefold combination, and no more than two distinct elements. δg1 + δg2 … = ∑𝛿g𝑖 𝑁 𝑖=1 ∑𝛿g𝑖= 0 𝑁 𝑖=1 (2.12) (2.12) The point that was not analyzed before is that, the term in equation is indicating that fermion clusters must have zero curvature. Curvature is not allowed in fermion clusters. That is because in the 8T, the term is the arbitrary variation of the Ricci flow. That is in contrast to Albert Einstein theory of general relativity that associate matter formation to curvature, curvature in the 8T is only allowed as part of the Bosonic interactions, given by the Primorial. Those Bosonic interactions are propagating from fermion clusters, but it is not the fermion clusters which bends the four-dimensional space-time configuration. Keeping that in mind, even when we allow net curvature to appear on the manifold, its magnitude is relativity small and insignificant given by the principle of least variation. The most significant and strong interaction are those with the smallest net amount of curvature, given by (2). The strongest interactions are perfectly ordered by the sequence of the Primorial. 𝑁𝑉 𝑇𝑉 ⟶𝑅 (2) 𝑁𝑉 𝑇𝑉 ⟶𝑅 (2) 0.111 > 0.1 > 0.039 > 0.008… ⟶0 (2) 0.111 > 0.1 > 0.039 > 0.008… ⟶0 0.111 > 0.1 > 0.039 > 0.008… ⟶0 Summing up, those two features of the 8T, a theory that deals with varying manifolds and varying curvature, ironically indicate that curvature is "not allowed". In fermion clusters it must vanish, and it vanish into matter. When it does appear as net curvature on the matric tensor, which Is discrete amount isomorphic to primes, it is very small amount compared to total variations. Those two points indicate that the universe should be, flat. We reached the same conclusion without using the second representation of the universe packet. 25 Manor O Strikingly Beautiful Relation of Three Generations Masses The idea, which is followed by the last paper, is that if 8 + (1) to generate force, and force is extended outward, (short or long ranged) than 8 −(1) would be to generate mass, or arbitrary variations converging inward. Equipped with this idea we can search for a mathematical pattern. First, take all the masses, accurate as they can and combine them according to generation: [1.9] [1320] [172,770] [4.4] [87] [4240] 1.9 + 4.4 ≈6 1 3 1320 + 87 = 1407 172,770 + 4240 = 177010 [4.4] Seemingly nothing in common, luckily we can change it. Soon one will reason why the following exactly, multiple equation one by factor of nine and divide the third family by a factor of nine. 6 1 3 ∗9 = 57 = 50 + 7 1320 + 87 = 1407 = 1400 + 7 177010 9 = 19,667 = 19,660 + 7 26 Manor O 50 ∗28 = 1400 1400 ∗14 = 19,600 (60 𝑀𝑒𝑉 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒−0.03% 50 ∗28 = 1400 1400 ∗14 = 19,600 (60 𝑀𝑒𝑉 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒−0.03%) but but 28 = 7 ∗4 14 = 7 ∗2 28 = 7 ∗4 14 = 7 ∗2 so to go from first to second: so to go from first to second: (7 ∗4) ∗50 + (7) And from second to third (7 ∗2) ∗1400 + (7) Notice that it is a decreasing by an even factor of two. In addition, if we go from low to high it does not make sense physically, it should be Lagrangian oriented, nature is devising by increasing amount to minimize the arbitrary variations, so if correct we should go from three to one by devising: 19,660 + (7) 7 ∗2 = 1400 + (7) 1400 + (7) 7 ∗4 = 50 + (7) ∗1 9 Next, we can predict that total mass for fourth to sixth families: 50 + (7) 7 ∗8 ∗1 9 = 0.113 𝑀𝑒𝑉 0.113 7 ∗16 ∗9 = 0.000113 𝑀𝑒𝑉 𝒐𝒓 0.113 7 ∗16 = 0.00100 𝑀𝑒𝑉 0.000113 7 ∗32 ∗9 = 5.95 ∗10−8 𝑀𝑒𝑉 𝒐𝒓 0.00100 7 ∗32 = 0.0000045 𝑀𝑒𝑉 value of the forth which he t to converging a see can We . 𝑀𝑒𝑉 0.1140 or 0.113113 6 families: - Summing 4 is 55.25-55.69 lighter than first family: value of the forth which he t to converging a see can We . Manor O So according to the above reasoning and mathematical notion, one will predict infinite family is as family's Mev ≈ 0.113113 f o D masses, converging to total value - forming below the masses of the U below the six are neglected due to little contribution the total sum. So overall, we can write: M𝑁+1 = M𝑁+ (7) 7 ∗∏ 𝑁𝐸+1 r E=1 ∗ 1 9 (1.3) M𝑁+1 = M𝑁+ (7) 7 ∗∏ 𝑁𝐸+1 r E=1 (1.31) 𝑁𝐸+1 = 2 ∗𝑁𝐸 ; 𝑁𝐸= 2𝐸; 𝐸≥1 M𝑁+1 = M𝑁+ (7) 7 ∗∏ 𝑁𝐸+1 r E=1 ∗ 1 9 (1.3) M𝑁+1 = M𝑁+ (7) 7 ∗∏ 𝑁𝐸+1 r E=1 (1.31) 𝑁𝐸+1 = 2 ∗𝑁𝐸 ; 𝑁𝐸= 2𝐸; 𝐸≥1 (1.31) Strikingly Beautiful Relation of Three Generations Masses 𝑀𝑒𝑉 0.1140 or 0.113113 6 families: - Summing 4 is 55.25-55.69 lighter than first family: 6.3 0.1131130595 = 55.696 𝑜𝑟 6.3 0.1140 = 55.26 Note that we needed to readjust the scale by the factor of 8 + (1) as we manipulated the data, in a search for a pattern. Adjust it in the third family, by Multiplication and in the first and by division. The following reason, T-B family has much more mass, thus much more arbitrary variation converging inward, that might by the reason it has 8 + (1) factor in the nominator, and in the first, the arbitrary variations are so small, we need to adjust it in the opposite direction, to increase by 8 + (1). Whether in the fifth family and below, additional rescales are needed is unknown, we do include two options, with the 8 + (1) or without it. 27 Strong Electroweak Unification In the 8T thesis, page twenty-one and twenty-two, the author presented the strong electroweak unification based on the primorial coupling series, which resulted in the accurate prediction of alignment on 26 variations. The unification was done via four exchanges, three real and virtual exchange. That was in rigor: [8 + (1)] + 3 −(𝟏𝒗): [(8 ∗3) + (3)] + 3 ∶ [(24 ∗5) + (3)] + 𝟑 However, there is a simple way to do exact same thing without the virtual exchange of variation and taking the average of sum of exchanges. That is just by two real exchanges of variation from the third coupling term to the first coupling term. This will lead to the same result presented in the thesis, the unification of the strong electroweak interactions. [8 + (1)] ⟶8 + (1) + 2 [(24 ∗5) + (3)] + 5 ⟶[(24 ∗5) + (3)] + 3 [(24 ∗5) + (3)] + 5 ⟶[(24 ∗5) + (3)] + 3 The new, simpler way to unification does not include virtual exchange of variation; The new, simpler way to unification does not include virtual exchange of v 8 + (1) + 2: [(8 ∗3) + (3)] + 3 ∶[(24 ∗5) + (3)] + 𝟑 8 + (1) + 2 ⟶8 + (3) The two real exchanges between the third and the first will modify the same middle element the exact same manner as presented in the 8T thesis. The only term we can vary is the left, as we want to ensure duality among the forces; we cannot touch the net variation, marked in black; [(8 ∗3) + (3)] + 𝟑. [(8 ∗3) + (3)] + 𝟑. We cannot vary the invariant three; the modification will be to the left term in the coupling series; (8 ∗3) + 2 = 26 The restrictions imposed on such variation on the strong are the same as presented in the thesis. I.e. it must be to an infinitesimal interval. The physical meaning of such equivalence relation in high energy is a morphism between the Bosons. A gluon morphism the weak interaction 𝑊+, 𝑊−,𝑍 Bosons, and photon morphism to the 𝑊+, 𝑊−,𝑍 Bosons. γ ⟶𝑊− [(24 ∗5) + (𝑒)−] + 𝑊− [8 + (𝔤+ 2)] →8 + 𝑊− (1) At high energies there exist a morphism among the photon and the Gluon to the Boson of the weak interaction. Overview of ideas Mass is a variation of the manifold converging inward. Just like force but opposite in direction. Nature is eliminating the arbitrary amount of variations by devising in increasing amounts. That prediction is the rule of dark matter in our theory. It suits the fact that very quickly the families total is converging to zero. The rate in which the conserving to zero is made is unknown. The theory provides two options. First, with the rescaling factor to each family and second option without it. Rescaling only Once. Both options agree on the value of the total mass of the fourth, which is about 56 Times lighter than first. M𝑁+1 = M𝑁+ (7) 7 ∗∏ 𝑁𝐸+1 r E=1 ∗ 1 9 (1.3) M𝑁+1 = M𝑁+ (7) 7 ∗∏ 𝑁𝐸+1 r E=1 (1.31) As we combined the net masses of the two elements, the value should be again, decomposed to the two separate elements. There are an infinite variety of families whose mass is decreasing, thus below first generation of quarks, this could agree with so-called, dark matter. Cosmologists to decide whether the mass values predicted agree with the data. 28 Manor O The Rise of the Arrow of Time ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) (1.2) In our framework we have a Lorentz manifold inside an Euler- Lagrange equation. The manifold experience arbitrary variations, which vanish into, matter, we proved it in previous papers. Each time net variation appear on the manifold, a boson is manifested into our matric. That was the idea, which derived the coupling constants equation. Net variations are prime, and for each prime, there is a boson, unique boson: However, how does that relate to the arrow of time? Recall that the coupling constant equation is really a built upon a ratio between total variations divided by two and net variations which are prime. We saw that the total variations grew much more rapidly than the net, and we required a Sequence, that it will go from low to high. Therefore, the arrow of time should go from low to high as well. There could not be a photon propagation without electron, which propagate from the nuclei, or cluster of so-called quarks. The sequence of The coupling constant equation is the sequence of time it allows us to build from the elementary to the massive, first arbitrary variations eliminate and vary themselves, create protons and neutrons which vary as well, propagate electrons, which vary as well, yielding photons and electromagnetism.Nature as the interplay of total variations to net variations, which grow in number and gets weaker from one element to another, explain why the forces at a large scale are much weaker than those at smaller scale, here are much more total variations and the net is divided across the whole cluster. So starts and galaxies must appear only after the strong, weak and electromagnetic. Nature is going from high to low, from small amount or strong variations to weak amounts of net variations over bigger clusters of total variations. Keep in mind that when one say variation he means curvature as we built the 8- theory upon a Lorentz manifold. However, if we look at each element in itself, like electromagnetism for example we will not see any clues for the arrow of time, as it's not telling anything about the arrow. Strong Electroweak Unification The Gluon at high energy can become a longer-range meditator (assuming we consider weak as longer ranged). (1) At high energies there exist a morphism among the photon and the Gluon to the Boson of the weak interaction. The Gluon at high energy can become a longer-range meditator (assuming we consider weak as longer ranged). 29 Manor O The Rise of the Arrow of Time It is only when we found the series of Coupling constants and the intimate relation of the boson to primes and putted them in a row, than and only than we can see the rise of the arrow of time. In other words, we can reason why galaxies and cluster of galaxies can form only after the strong, weak and the electric. We are also able to reason the weakness of gravity and the interactions in higher terms in the series. 1 > 3 30 > 5 128 > 7 850 … 1 > 3 30 > 5 128 > 7 850 … 1 > 3 30 > 5 128 > 7 850 … 1 > 3 30 > 5 128 > 7 850 … 1 > 3 30 > 5 128 > 7 850 … 1 > 3 30 > 5 128 > 7 850 … 30 Manor O Continuous and Discrete Aspects of Nature ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) By analyzing equation (1 it is vividly clear that the setting is continuous, we have a smooth manifold which is the connected manifold. As both 8T and Einstein GR are composed upon a continuous (3,1) matric tensor. However, 8T is also discrete in a sense that the Bosons are associated with discrete amounts of curvature, prime or one, in their nature. That was the idea that lead to the discovery of the primorial coupling series. 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128: 850:9254.. (1.2) (1.1) So taken from that point of view the universe has an element that is discrete. This element comes to an agreement with the fundamental Planck constant, which state can the Quantum oscillator can change only by discrete amounts. Therefore, the 8T setting is continuous, but this continuous setting has certain quantities that are discrete and are of grand importance. Another element that could be regarded as discrete is the number of universes in the packet. It is possible to regard, and maybe it is even the case, to each newborn manifold in the packet as a descended of a more ancient manifold in the packet, which was born due to matric tensor fluctuations. Classification can be made based upon the location in the packet. It is possible (theoretically that is) to numerate the manifolds in the packet, assuming it is finite but still aspiring infinity. That is an additional element which is discrete, despite each manifold is continuous. So based on this short analysis of the main two equations of the framework 8T, we have a mixture of both continuous setting, given by infinite smooth manifolds interacting with each other, and at the same time, discrete features such as prime numbers, isomorphic to Bosonic fields and discrete number of universes in the packet – which could be described as a graph if we can correlate the newborn manifolds to those they emerged from. 31 Manor O Manor O Manor O The Almost homogenous Universe g ∂s ∂g ∂t 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) The reason the universe is not completely homogenous based on the framework is that the manifold experience arbitrary variations – which than vanish into fermions. marked in black. The reason the universe is not completely homogenous based on the framework is that the manifold experience arbitrary variations – which than vanish into fermions. marked in black. ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t δg −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 δg′ = 0 Those variations are arbitrary amount of curvature of a manifold, and they are subject to net variations, which yielded the coupling constant equation. We saw that nature is really the interplay between total arbitrary variations to net variations. Net variations are prime in their nature, and so in the 8- theory Framework for each prime number there exist a boson. 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 The series gives rise to the arrow of time; we should see more interactions as time goes on and so, bigger and bigger structures which makes the manifold less and less homogenous. The bigger the cluster of total variations the weaker the force, as it is divided across the whole cluster. By looking at those two equations we can see exactly why the universe or the Lorentz manifold in The 8-theory framework is not homogenous, because of those arbitrary variations and the additional net variations. The first accounts for fermions, known as quarks, the other known as bosons. Using that framework, we can see why the manifold cannot be homogenous, it is almost obvious. Of course, the question of the homogenous structure is a question in which we cannot really answer, as it has no numerical data, it’s a question revolving around a theory in which the lack of Homogeny is a feature of the main axioms and equations. We can see it in the framework of the 8-theory, or any Lagrangian oriented theory, which includes arbitrary variations, which must vanish at border. The beauty and innovative part in the 8-theory is that, all life forms, galaxies, clusters of galaxies are those arbitrary variations. 32 The Commutativity of the Coupling Constants Series There is a symmetry we can impose on those terms, that is by changing the order of the elements. Changing the order of the elements makes no difference to the overall value of the coupling. The series in equation (1.2) will still hold either way. [2𝑁1 + 1 2] + 1 2 ⟶[1 2] + 2𝑁1 + 1 2 [(8 ∗3) + (3)] + 3 ⟶[3] + (3) + (8 ∗3) Now its matter clusters unbound due to the net curvature, which is the first in order. The point is not the physical meaning of such an event, but rather the commutativity of the primorial equation. If we take the final values of each coupling as the main objective, that the equation is order invariant, or commutative. The same applies for each higher element and lower as well in the coupling term. Another point regarding the strong interaction is that, it implies that the gluons are unbound. They must come from somewhere and as they are net curvature on the manifold isomorphic to one, each gluon pulls or increase the probability of arrival to other gluons. The same applies to each boson in each coupling term. For example the photon: Now its matter clusters unbound due to the net curvature, which is the first in order. The point is not the physical meaning of such an event, but rather the commutativity of the primorial equation. If we take the final values of each coupling as the main objective, that the equation is order invariant, or commutative. The same applies for each higher element and lower as well in the coupling term. Another point regarding the strong interaction is that, it implies that the gluons are unbound. They must come from somewhere and as they are net curvature on the manifold isomorphic to one, each gluon pulls or increase the probability of arrival to other gluons. The same applies to each boson in each coupling term. For example the photon: ∑𝛾𝑖 𝑀 𝑖=1 = ∑𝛿g𝑖 𝑀 𝑖=1 > 0 (3.13. 𝐵) Suppose you had a set of 𝐾 gluons bounded to a Quark Triplet, in that sense it does not matter in which order they clustered to the sea of gluons. We can vary the set as much we would like order wise. Therefore, from that angle there is a symmetry there as well. The Commutativity of the Coupling Constants Series 𝐾= ∑𝔤𝑖⟶ 𝑖=𝐾 𝑖=1 ∑(+1)𝑖 𝑖=𝐾 𝑖=1 𝐾= ∑𝔤𝑖⟶ 𝑖=𝐾 𝑖=1 ∑(+1)𝑖 𝑖=𝐾 𝑖=1 that the main representation in which bosons are propagating from fermion clusters with spin one- half is the most reasonable and seemingly best way to understand nature. This short assay does not indicate that the opposite is correct, but rather present the coupling constants series from viewpoint of symmetry, order invariance or commutativity. 33 Manor O Manor O Therefore, we have an inward acceleration and areas of negative curving on the Manifold, which agrees with the description of a compressed Lorentz manifold. However, is it reasonable physically to make such a transformation from (1) to (2)? Suppose it is reasonable to change the direction of the acceleration. By looking at the second term: + 𝜕𝑔 𝜕𝑡→− 𝜕𝑔 𝜕𝑡 Meaning, all the galaxies, clusters of galaxies, which represent extremum curvature on the manifold, must be eliminated and revert their direction inward, toward the manifold. Such shift will be along an inward acceleration and a process of manifold compression. The process than is synonymous to going from a lower energy state, colder state, to a much higher state of energy. It is a higher state of energy as it is a process of immense masses compressing inward, toward a converging Lorenz manifold, such process will be encompassed by friction, heat and high entropy. It is not Lagrangian oriented and not likeable scenario in our framework. There is no need for calculation of hydrogen atoms per unit space when we have the mathematical equation. We can also analyze the subject of expansion or collapse by using the coupling constant equation in its third representation, the arrow of time. 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 1 > 1 30 > 1 128 > 1 850 … 1 > 1 30 > 1 128 > 1 850 … A universal collapse would be to revert the side of the arrow. From weaker and weaker interactions at mega scales, to go for smaller interactions much stronger: 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 1 1 1 (1.1) (1.2) A universal collapse would be to revert the side of the arrow. From weaker and weaker interactions at mega scales, to go for smaller interactions much stronger: A universal collapse would be to revert the side of the arrow. From weaker and weaker interactions at mega scales, to go for smaller interactions much stronger: 1 > 1 30 > 1 128 > 1 850 … The physical meaning would be than, stars, galaxies and clusters of galaxies to deform and in an endless succession until we reach quarks and gluons. The Future Universe ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 and −𝜕2𝑔′ ∂t2 = 0 𝜕𝑔 𝜕𝑡= 0 and −𝜕2𝑔′ ∂t2 = 0 𝜕𝑔 𝜕𝑡= 0 and −𝜕2𝑔′ ∂t2 = 0 This equation describes dark energy or time invariant acceleration from areas of extremum curvature on the Lorenz manifold. We assume no data is available from the first three terms, which describe a varying matric in spatial dimensions. To ensure universe collapse, we need to revert the signs so we will get: In other words, the acceleration is now directed inwards, and the new equation is: In other words, the acceleration is now directed inwards, and the new equation is: ∂ℓ ∂s ∂s ∂M ∂M ∂g 𝜕2g′ ∂t2 −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ ∂g ∂t = 0 (1.4) (1.4) 34 34 Manor O Such process would require immense amount of energy and it has to happen across all the spectra of the foreseeable universe. In our framework, it means less manifold net variations (positive curving) over time. Physically it does not make sense, it's not Lagrangian oriented. To go from low state of energy and aspire the highest level. There is no indication that such process could accrue in nature, without artificial intervene. As far as one knows, it comes to an agreement with the laws of thermodynamics. Nevertheless, more importantly, in our framework, there is no reason for such unnatural thing to happen. 35 Manor O Does The Universe has Limits? 8T ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂g ∂t = 0 ∩ 𝜕2g′ ∂t2 = 0 Manor O ∂g ∂t = 0 ∩ 𝜕2g′ ∂t2 = 0 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) The question at the heart of this assay is whether the manifold, i.e. the universe has borders. It is finite or infinite in its nature ? according to the above framework, Since it is a defined object within a set of distinct objects of the same class, a set of universes which flatten each other and interact at areas of extremum curvatures, we will prove it later in the thesis, it is finite. On the other hand, since the interaction is ever accruing causing the matric tensor between those areas of extremum curvatures to expand from them, in that sense the finite object is varying in size and ever increasing, aspiring to infinity. So according to the 8T, similar to ideas suggest by scientists of the 20-th century, the manifold is closed, but it has no limit. If one is correct it was Einstein who suggested that definition. It is finite, but aspiring to infinity due to the pressure exhorted from other manifolds. We can make a prediction according to this new framework; Prediction (1): The degree of universe flatness is proportional to time. universe flatness is inversely proportional to temperature The degree of (2): Prediction 36 Manor O The Coupling Constants Equation and Gauge Fields The Coupling Constants Equation and Gauge Fields The coupling constant equation: 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.2) Each term individually: Each term individually: 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 Let us look at the first term: 8 + (1) 8 + (1) 8 + (1) Remember back in the day, when we concluded that we could ignore the eight, since even amount of variations vanish, and just write that the first element is one. We also know that there are eight gluon fields. These are meditating the strong interaction and color charge. However, this could be just a coincidence. Let us examine the next term in the series: [(8 ∗3) + (3)] + 3 [(8 ∗3) + (3)] + 3 [(8 ∗3) + (3)] + 3 This term describe the nature of the weak interaction. Notice the right inside the parenthesis: This term describe the nature of the weak interaction. Notice the right inside the parenthesis: (8 ∗𝟑) We also know that there are three gauge fields meditating the weak interaction. The massive W the Z bosons. which we correlate to SU(2) and isospin. If the right term inside the parenthesis is a reflection on the number of fields meditating an interaction than we can examine the next term on the series, electromagnetism: [(24 ∗𝟓) + (3)] + 5 That is a daring statement to make, but if the assumption to hold true, There Should be five gauge fields meditating the electric interaction. Five distinct kinds of photons. It is really an absurd statement to make, given the fact that there are no indication that there is an agreement with experiment regarding that idea. However, sometimes in theoretical physics, bold risks must be taken, and so the author of this paper will allow his belief regarding the great power of the equation to guide him and State: 8T predicts five gauge fields meditating electromagnetism. Whether such thing could be correct, only time and experiment will tell. The Coupling Constants Equation and Gauge Fields 37 Manor O Proof: Fluid Turbulence ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) Define flow of fluid as an amount of arbitrary curvature on a Lorentz manifold, marked in green. Define flow of fluid as an amount of arbitrary curvature on a Lorentz manifold, marked in green. ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t δg −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 δ𝑔′ = 0 Assume it is a continuous process in time, meaning we can break it to an infinite sequence, in contrast to the original idea, each discrete is representing flow of matter at large scale, a fluid: 𝛿g = δg1 + δg2 … To each we can associate an appropriate time To each we can associate an appropriate time δg1 ⟶𝑡1 δg2 →𝑡1 + ∆𝑡= 𝑡2 That is in agreement with fluid flow, a sequence of vanishing curvature spikes, turning into a cluster of matter which will form a fluid flow. The fluid flow has a continuous sequence in time. Analyze the first element alone of the fluid flow by breaking it to infinitesimal time intervals: ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t δg1 −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 δg′1 = 0 The first element is causing the metric to accelerate outward, so the space in which the second element of the flow is not the same as the metric in which the first element was moving in. the first element is causing the metric to vary, and its length to vary as well, the motion cannot be put in terms of vectors Therefore, the second element itself is doing the same, it is an endless process of metric variation causing the motion to be chaotic as the metric itself varying in accordance to curvature flow. Therefore, if we break down the motion of fluid to an infinite sequence, we can derive the reason the motion of fluid cannot be put in vector form, each subset of curvature is causing an outward acceleration of the metric, and the next subset is moving in a different metric than the first, it could revert sideways, sideways inwards, sideway outwards, and same Applies for each additional element. 38 Manor O Manor O Manor O Take the masses of all the generations and combine them: Take the masses of all the generations and combine them: [4.4] [87] [4240] [4.4] 1.9 + 4.4 = 6.3 1320 + 87 = 1407 1.9 + 4.4 = 6.3 1320 + 87 = 1407 172,760 + 4240 = 177000 172,760 + 4240 = 177000 172,760 + 4240 = 177000 The idea by Quark mixture we mean multiplication of masses of the first and second to yield the total mass of third, times a scalar. Therefore, a total mass of the first family multiplied by the total mass of the second family, both multiplied by a scalar, will yield the total mass of the third. We can proof that is the almost case exactly for the values of the masses above: 6.3 ∗1407 = 8864.1 177,000 8864.1 = 19.96 6.3 ∗1407 = 8864.1 177,000 8864.1 = 19.96 6.3 ∗1407 = 8864.1 177,000 8864.1 = 19.96 If we can allow a slight variation of the first masses to be 6.29 Mev and not 6.3, it will be If we can allow a slight variation of the first masses to be 6.29 Mev and not 6.3, it will be 6.29 ∗1407 = 8850 177,000 8850 = 20 39 Manor O Therefore, just a slight variation of 0.01 Mev and we have a beautiful number and a way to combine the total mass of the first and the second, mix them and multiply by the scalar, to reach the total mass of the third. Reader should argue that it could be just a coincidence, a choice of certain values to yield the scalar and he might be right as the masses are not measured or known as exact, they could divert. Assuming the mixing will accrue at scalar numbers only, we can build correction angles to ensure the scalar number will hold. So if the masses of the first divert or measured at a higher value that 6.29, there will be a correction angle to retain the same scalar we obtained. The correction angles could have more than one value and they can be positive or negative. Take the mass of the up quark to be average between 1.9 to 2.2 Mev, which is 2.05 Mev. 1.9 + 2.2 2 = 2.05 𝑀𝑒𝑣 2.05 + 4.4 = 6.45 𝑀𝑒𝑣 6.45 ∗1407 = 9075.15 177,000 9075.15 = 19.503 1.9 + 2.2 2 = 2.05 𝑀𝑒𝑣 2.05 + 4.4 = 6.45 𝑀𝑒𝑣 6.45 ∗1407 = 9075.15 177,000 9075.15 = 19.503 The correction angle to reach desired number would be: The correction angle to reach desired number would be: The correction angle to reach desired number would be: 19.503 + cos (11.5) ≈ 20 There could be many more, the correction angles are not limited in number and depend upon the masses values taken of the first, second, and the third as well. The idea behind stay the same. The correction angle will be added to yield a scalar multiple. 20 ∗(𝑇𝑀1 ∗𝑇𝑀2) ≈𝑇𝑀3 Among all the topics can be explained by the 8-theory, and there has been quite a few, the question of Quark mixing seems to be among the hardest ones, and among the topics not within reach. This part is not a proof of any sort but a mathematical idea, the reader should rightfully argue and doubt it. One was trying to reason in the simplest and most elegant way, the weird phenomenon of Quark mixing. Whether it makes sense or not, readers should decide after analyzing the Paper. 40 Manor O The Coupling Constants series Orthogonal Curvatures We have a proven that the coupling constants series is the same under sign reversal, which gives rise to the existence of anti-matter. [2𝑁3 + 1 2] + 1 2 ⟶[−2𝑁3 − 1 2] − 1 2 (1.45) (1.45) Since the one-half is a representation of net curvature on the manifold, and the electron is represented by the one-half inside the bracket, we can represent the positron and the electron as curvature oriented in orthogonal way, leading to an inner product that is zero. ⟨𝛿𝑔𝑖|−𝛿𝑔𝑖⟩= 0 (1.46) (1.46) The fact their inner product is zero, is indicating an energy release. The pairing can be thought as two orthogonal pulls leading to peer pressure on the matric tensor. such pressure could lead to the matric be ripped apart, and by doing so we will observe a gate to the base space of raw energy, the Ricci flow, given by 𝜕𝑔/𝜕𝑡 on the main equation (1). We can use equation (1.46) with leptons as elements of the inner product such as the electron and positron: ⟨+3𝑖|−3𝑖⟩= 0 ⟨+3𝑖|−3𝑖⟩= 0 In addition, at the same time with bosons: ⟨γ𝑖|γ− 𝑖⟩= 0 Which are net curvature unbounded on the matric tensor in contrast to the electron, bounded by the nuclei, given by the fact it is within the bracket. From that point of view, it is clear that Anti-Matter is the perfect source of energy as it is leading to a pure release of energy, given by the orthogonality of the curvatures participating as given by (1.46). Notice that the summation is holding in (1.46), we can eliminate clusters of inverse curvature elements as long is the index is the same. 41 Manor O Manor O Manor O The Coupling Constant Equation and Higgs Mechanism 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.2) Let us look at the first term describing the strong. We saw that the eight vanish since it's an even in our framework. Let us look at the first term describing the strong. We saw that the eight vanish since it's an even in our framework. 8 + (1) →(1) We also know that from physics the gluons are massless. Let us examine the second term. (24 + (3)) + 3 We know that the bosons that meditate the weak interaction do carry mass. Moreover, we know that the symmetry of SU(2) forbids mass terms in the Lagrangian, and the solution which allows us to include mass terms without ruining the symmetry is the Higgs idea. This idea works by including extra terms. In our framework, the extra term is the majestic three. Therefore, the Higgs field is responsible for the lack of order in our series, which could have been a beautiful Series of eight multiples. In a sense of the standard model, we can say it is "breaking the symmetry" by inserting the invariant three. So overall, we move from spin 0 – perfect clusters of variations. With the Majestic three Inserted by the Higgs Field we move to a matter with spin one-half, we did so by setting the equation on the critical line of the primes. This 𝑡ℎ𝑟𝑒𝑒 is really a destabilizing factor than yields a net variation, which is prime as well. For example – Electromagnetism: For example – Electromagnetism: Perfect clusters of variations→2N Perfect clusters of variations→2N Destabilize the perfect 2N is the Majestic (3) →( 1 2) →Electron. Destabilize the perfect 2N is the Majestic (3) →( 1 2) →Electron. Blends in the potential cluster to yield in that case→123. Blends in the potential cluster to yield in that case→123. The result is the net variation, which is also prime: N(V) →( 1 2) → +(5) The overall frame yields: [2𝑁+ ( 1 2)] + ( 1 2) → 𝟏𝟐𝟑+ 𝟓= 𝟏𝟐𝟖. Magnitude of an interaction. [2𝑁+ ( 1 2)] + ( 1 2) → 𝟏𝟐𝟑+ 𝟓= 𝟏𝟐𝟖. Magnitude of an interaction. [2𝑁+ ( 1 2)] + ( 1 2) → 𝟏𝟐𝟑+ 𝟓= 𝟏𝟐𝟖. Magnitude of an interaction. The main point of the part is that the Majestic three is a result of the Higgs field. It is the reason the majestic three appears. So overall, our framework does not contradict the Higgs Idea but support it and allow us an additional view on how the mechanism work. As the Higgs is responsible for additional terms in the Lagrangian, and in the 8-theory we see that the first elements in the series of coupling constant differ by an additional term, the Majestic three or spin (1/2) . 42 Manor O Manor O K ∶ A ⟶ B K ∶ A ⟶ B Let B = {𝑎1 … . 𝑎𝑚} a subset of A which satisfy the condition K. m < n. Allocate: Proof: P ≤NP Let it be a set - Let it be a set - A = {𝑎1 … 𝑎𝑛} A = {𝑎1 … 𝑎𝑛} Define a condition on the set: Manor O Manor O Anti-Matter & Dirac Delta Variation ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 , − 𝜕2𝑔′ ∂t2 = 0 [∂ℒ 𝜕𝑠 𝜕𝑠 𝜕𝑀 𝜕𝑀 𝜕𝑔] 𝜕g 𝜕𝑡𝛿g −[∂ℒ 𝜕𝑠′ 𝜕𝑠′ 𝜕𝑀 𝜕𝑀 𝜕𝑔′] 𝜕2g′ ∂t2 𝛿𝑔′ = 0 (1.1) ∑𝛿g𝑖 𝑁 1=1 = 0 (2.12) ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 , − 𝜕2𝑔′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 , − 𝜕2𝑔′ ∂t2 = 0 [∂ℒ 𝜕𝑠 𝜕𝑠 𝜕𝑀 𝜕𝑀 𝜕𝑔] 𝜕g 𝜕𝑡𝛿g −[∂ℒ 𝜕𝑠′ 𝜕𝑠′ 𝜕𝑀 𝜕𝑀 𝜕𝑔′] 𝜕2g′ ∂t2 𝛿𝑔′ = 0 (1.1) (1.1) ∑𝛿g𝑖 𝑁 1=1 = 0 (2.12) (2.12) Reader should be familiar with the procedure. Now we have seen that we can derive the nature of fermions and the quark model by allowing the series, which contain two distinct elements to vary. So overall we obtain eight threefold combinations of those elements. Therefore, even though the elements are varying the series could vanish. That is in agreement with a stationary Lorentz manifold. There could be however, another way to ensure a stationary Lorenz manifold. Which will match each element in the series its mirrored element. That is 𝛿𝑔1 + 𝛿∃𝑔1 = 0 𝛿𝑔2 + 𝛿∃𝑔2 = 0 𝛿𝑔1 + 𝛿∃𝑔1 = 0 𝛿𝑔2 + 𝛿∃𝑔2 = 0 𝛿𝑔1 + 𝛿∃𝑔1 = 0 𝛿𝑔2 + 𝛿∃𝑔2 = 0 By mirror, it means the same but opposite in sign. So the overall sum of the Series will hold as zero. In the 8- theory framework, Quarks are regarded as arbitrary amount of curvature on a manifold. Based on this view, anti-quarks and anti-matter is arbitrary curvature with opposite direction. Same magnitude just different direction. So overall, that framework would allow the existence of anti-matter. That is in agreement with quantum field theory and with the Dirac equation for spinors. In fact, the moment of Singularity could be a result of the series not equal to zero. Allocate: K ⟶𝑡1 Time in which the subset B was obtained after running the condition. Allow the elements of A to vary over time. Δt ∶A ⟶ A′ Δt: B ⟶ B′ Δt ∶A ⟶ A′ Δt: B ⟶ B′ Δt: B ⟶ B′ Let an isomorphism exit between the sets after the operation Δt. Define a functor on the subset B: ℘∶ set ⟶ Top ℘∶ set ⟶ Top In order to obtain an EL equation of the subset 𝓛 (B, B', t) on a topological space. Set the space to be complex analytical to ensure differentiation is possible at all time. 𝛛𝓛 𝛛𝐁−𝛛𝓛 𝛛𝐁′ ∗𝐝 𝐝𝐭= 𝟎 Or Or B − B′ ∗ Δt = 0. B − B′ ∗ Δt = 0. Since we allocated to obtaining the subset B the time 𝑡1– we can write: (𝑡1)B − B′ ∗(𝑡1 + Δt) = 0 For a given condition we impose on a set, which yield a subset to satisfy it, in order to ensure the subset to be a valid solution we are required to examine it will stay invariant under time translations after we operate a functor on it and switch to a topological space. In other words, the variations of the subset to vanish at border. One can say that the subset has to be close with respect to time. Thus, time obtaining a suggested solution will always to shorter than the time required deciding the existence of a solution. the time of making a decision regarding the existence of a solution and obtaining the solution will be equal if the set is not varying over time. Δt = 0. 𝑬𝒏𝒅 𝒐𝒇 𝑷𝒓𝒐𝒐𝒇. 43 Manor O Anti-Matter & Dirac Delta Variation 𝜹𝒈≠𝟎 The moment the series is not equal to zero than means that we have net curvature, or maximal curvature on the manifold, which will yield a negative extremum time invariant acceleration from The moment the series is not equal to zero than means that we have net curvature, or maximal curvature on the manifold, which will yield a negative extremum time invariant acceleration from it. ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 (1) (1) In other words, the moment of asymmetry in the series yielding net curvature on the manifold could be the reason for singularity and so called among the masses "big bang". It is just an idea of course, but up until now the 8- theory was on point in regards to Issues on other theory could explain. In other words, the moment of asymmetry in the series yielding net curvature on the manifold could be the reason for singularity and so called among the masses "big bang". It is just an idea of course, but up until now the 8- theory was on point in regards to Issues on other theory could explain. 44 Manor O The Primorial Coupling Constants Series – Odds versus Primes Shifting to spin representations: [(8 ∗3) + (3)] + 3 →[2N1 + 1 2] + 1 2 = 2𝑁+ 1 [(24 ∗5) + (3)] + 5 →[2N2 + 1 2] + 1 2 = 2𝑁+ 1 [(120 ∗7) + (3)] + 7 →[2N3 + 1 2] + 1 2 = 2𝑁+ 1 So to proof the uniqueness of primes compared to odds or any other kind of a ring different from primes we can try associate 𝑁𝑉∉ ℙ and construct just for means of making the point, the following term: [(24 ∗5) + (3)] + 9 → [2N2 + 1 2] + ℛ ℛ≠1 2 1 2 < ℛ+ 1 2 < 1 [(24 ∗5) + (3)] + 9 → [2N2 + 1 2] + ℛ Therefore, as a result we will have a total spin that is neither one-half nor one. that is against experiment and against other leading theories such as quantum field theory. The point of this short assay is that the prime is a subgroup of the real, which in a sense is much smaller and so it is imposing a restriction on the values that can be regarded as net curvature on the matric tensor. Such a framework is resembling a symmetry limitations by physical theories. In addition, when compared to string theory that allow an infinite variety of particles, some with exotic traits, the number of bosonic curvature is indeed infinite but at the same time, cannot be associated with any number. The number of options is smaller than the entire field of the reals,ℝ, as we need to take into account the spin trait given by the second representation. Therefore, as a result we will have a total spin that is neither one-half nor one. that is against experiment and against other leading theories such as quantum field theory. The point of this short assay is that the prime is a subgroup of the real, which in a sense is much smaller and so it is imposing a restriction on the values that can be regarded as net curvature on the matric tensor. Such a framework is resembling a symmetry limitations by physical theories. In addition, when compared to string theory that allow an infinite variety of particles, some with exotic traits, the number of bosonic curvature is indeed infinite but at the same time, cannot be associated with any number. Manor O Manor O Manor O The Primorial Coupling Constants Series – Odds versus Primes The number of options is smaller than the entire field of the reals,ℝ, as we need to take into account the spin trait given by the second representation. Therefore, given by the term in equation with the ℛ to those readers who wondered whether there should be a coupling term after the seven, with 𝑁𝑉= +(9), the suggest answer is the following: given by spin representation considerations and by taking as an axiom that no spin between one half and one is allowed in nature, there should not be a coupling term associated with this number. Primes are imposing a strict limitation leading to a smaller infinity of bosonic ripple fields on the Einstein matric tensor. Therefore, given by the term in equation with the ℛ to those readers who wondered whether there should be a coupling term after the seven, with 𝑁𝑉= +(9), the suggest answer is the following: given by spin representation considerations and by taking as an axiom that no spin between one half and one is allowed in nature, there should not be a coupling term associated with this number. Primes are imposing a strict limitation leading to a smaller infinity of bosonic ripple fields on the Einstein matric tensor. 45 45 Dirac Delta Variation Our main equations in the framework: ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s ∂M ∂M ∂g′ 𝝏𝟐𝐠 𝛛𝐭𝟐= 0 (1) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) The Dirac delta in our framework is an interference on the Lorenztion manifold. An arbitrary Amount of curvature 𝛿𝑔 on the manifold. Since it is not allowed and must vanish, we require 𝛿𝑔= 0, as we did previously in this framework. The Dirac delta in our framework is an interference on the Lorenztion manifold. An arbitrary Amount of curvature 𝛿𝑔 on the manifold. Since it is not allowed and must vanish, we require 𝛿𝑔= 0, as we did previously in this framework. 𝛿g ≠0 𝑎𝑡 𝑡= 0 𝛿g = 0 𝑎𝑡 𝑡> 0 So the Dirac delta in our framework describe the process in which arbitrary amount of curvature appear, and vanish into matter. However, there is no restriction with regard to Time. Arbitrary amount of curvature can appear at any time, so we must modify the idea of the Dirac in our framework. 𝛿g ≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) 𝛿g ≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) We also require that ∆𝑡→0 as just after the arbitrary amount or interference will appear, it will immediately vanish into matter. Therefore, in this framework is rich in delta functions. The difference is that the delta can appear at time that is not null. In a sense, we have more flexibility with the delta. After the delta appeared and as a result fermions were manifested into the metric. Those fermions could still vary, and experience a net curvature or net variation. As was analyzed in this paper those net curvatures were taken to be prime numbers and that was the reasoning behind the construction of the coupling constant equation. Those net variations of the manifold are another interference, but and interference which propagate from fermions, and is prime number. Therefore, in that sense it cannot turn into fermions. Fermions vanish in even amount of variations. The result is a propagation across the manifold Ripples on the metric all across. 46 Manor O Manor O 𝛿g = 0 𝑎𝑡 𝑡1 = 𝑄(𝑡+ ∆𝑡) At later continuation of time: At later continuation of time: 𝑡2 > 𝑡1 This condition is satisfied: 𝛿g ≠0 𝑎𝑡 𝑡2 = 𝑄(𝑡+ ∆𝑡+ ∆𝑡) 𝛿g ≠0 𝑎𝑡 𝑡2 = 𝑄(𝑡+ ∆𝑡+ ∆𝑡) Moreover, the amount of variations is either prime or one: Moreover, the amount of variations is either prime or one: 𝛿𝑔= 2 (𝑛+ 1 2) ; 𝑛≥0 𝛿𝑔= 2 (𝑛+ 1 2) ; 𝑛≥0 Than we have a ripple on the manifold which propagate all across, toward all directions. The Laplacian operator than is vital to description for a mathematical description of the Manifold ripples, or bosonic fields. Important point to take is that the underlining reason for the Boson propagation all across the metric is their prime number feature. Define a bosonic ripple across the Lorentzian metric: ∇2= 𝜕2𝑀𝑥 𝜕2g + 𝜕2𝑀𝑦 𝜕2g + 𝜕2𝑀𝑧 𝜕2g (1.41) That is curvature propagation across all metric spatial dimensions as: 𝑀𝜇 ∈𝑆 𝑆= (𝑀, 𝑔) 𝑆= (𝑀, 𝑔) 47 47 Manor O Manor O Reasoning for Spiral Structures of Galaxies ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 (1) ∂g ∂t = 0 , −𝜕2𝑔′ ∂t2 = 0 [∂ℒ 𝜕𝑠 𝜕𝑠 𝜕𝑀 𝜕𝑀 𝜕𝑔] 𝜕𝑔 𝜕𝑡𝛿𝑔−[∂ℒ 𝜕𝑠′ 𝜕𝑠′ 𝜕𝑀 𝜕𝑀 𝜕𝑔′] 𝜕2𝑔′ ∂t2 𝛿𝑔′ = 0 (1) ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 (1) ∂g ∂t = 0 , −𝜕2𝑔′ ∂t2 = 0 ∂g ∂t = 0 , −𝜕2𝑔′ ∂t2 = 0 [∂ℒ 𝜕𝑠 𝜕𝑠 𝜕𝑀 𝜕𝑀 𝜕𝑔] 𝜕𝑔 𝜕𝑡𝛿𝑔−[∂ℒ 𝜕𝑠′ 𝜕𝑠′ 𝜕𝑀 𝜕𝑀 𝜕𝑔′] 𝜕2𝑔′ ∂t2 𝛿𝑔′ = 0 (1) (1) Notice the first requirement: Notice the first requirement: 𝜕𝑔 𝜕𝑡= 0 𝜕𝑔 𝜕𝑡= 0 In addition, the second requirement: In addition, the second requirement: 𝛿g = 0 𝛿g = 0 Those two simple requirements combined together can allow us to a deep Insight into the structure of galaxies. In the 8-theory framework, we have a Lorenz manifold, the manifold has areas of extremum curvature that stay as they are over time. That is given by the first requirement. The manifold also experience arbitrary variations, the second requirement. Those arbitrary variations vanish into matter in agreement with a stationary Lorentz manifold. The combination of both condition than implies that in order for the areas of extremum curvature to stay as they are, the arbitrary variations cannot appear inside them. That is by the combination of the two requirements. However, those arbitrary variations still appear in the framework. In addition, the areas of extremum curvature are a vital part of this theory. The combination of both requirement is than resulting in areas of extremum curvatures surrounded by arbitrary variations that could not affect them. The following model of the 8-theory is than intersecting with the large-scale geometrical shape of galaxies. However, it is known that so called, black holes in the center of galaxies are absorbing matter and nothing can escape them. So in a Second glance the first requirement will not hold in such case. However, that is not a real problem if we assume that those black holes, which we regard as areas of extrunum curvature inside galaxies also omit matter. We know it is the case, as we call it the hawking radiation −𝛿𝑔(𝐻). Spin zero: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠) Spin zero: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠) Spin zero: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠) Spin zero: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠) Bosons with spin one: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) + 𝑁𝑉 Bosons with spin one: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) + 𝑁𝑉 Bosons with higher spin integers: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) + 𝑁𝑉1 + 𝑁𝑉2 Bosons with higher spin integers: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) + 𝑁𝑉1 + 𝑁𝑉2 Suppose we have two photons pairing, photon and anti-photon, both were emitted from fermion clusters with opposite sign: [2N2 + 1 2] + 1 2 = 2𝑁2 + 1 [2N2 −1 2] −1 2 = 2𝑁2 −1 [2N2 + 1 2] + 1 2 = 2𝑁2 + 1 [2N2 −1 2] −1 2 = 2𝑁2 −1 The result of combining the photons would be again, a cluster with zero spin as we analyzed in the theory. Since the higgs boson has spin zero, the conclusion is that two opposite in charge sign photons, can give rise to a spin zero particle such as the higgs. It is the case with photon jests, but here analysis is via the primorial coupling constants series, which makes it easy to understand. 2𝑁2 −1 + 2𝑁2 + 1 = 4𝑁2 γγ−→𝐻0 48 Manor O Measuring the Electron – 8T [(24 ∗5) + (3)] + 5 →[2N2 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ [(24 ∗5) + (3)] + 5 →[2N2 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ Electrons in this framework are represented by the majestic three. Since it is not an even number it cannot vanish into matter. Since it is trapped on the bracket it can't propagate like the net variation 𝑁𝑉= γ, which are net curvature on the matric tensor or ripples. The conclusion is that the electron is propagating across the nuclei, the hadron structure which is two and three divisible to vanish into matter. Now in physics there is the problem of measuring the energy of the electron, and the problem is due the varying the radius, the smaller the radius the higher energy of the electron. So at radius aspiring zero, infinite energy is manifested, against observations.One would like to add certain notes on the issue on measurement. First, Electrons in this framework are represented by the majestic three. Since it is not an even number it cannot vanish into matter. Since it is trapped on the bracket it can't propagate like the net variation 𝑁𝑉= γ, which are net curvature on the matric tensor or ripples. The conclusion is that the electron is propagating across the nuclei, the hadron structure which is two and three divisible to vanish into matter. Now in physics there is the problem of measuring the energy of the electron, and the problem is due the varying the radius, the smaller the radius the higher energy of the electron. So at radius aspiring zero, infinite energy is manifested, against observations.One would like to add certain notes on the issue on measurement. First, regarding the electron as a separate entity is wrong. The electron is part of the manifold, and is effected by what is going on the matric tensor. Trying to measure it solely based on radii seems to relay on too simplistic ideas, which ignore complexity. Second, measurement of the electron in a varying radii will take a certain period of time. For all this period we will need to know where the electron is which is impossible to do. Reasoning for Spiral Structures of Galaxies 𝜕𝑔 𝜕𝑡+ 𝛿𝑔+ (−𝛿𝑔(𝐻)) = 0 (1.442) (1.442) So overall those two simple requirements in our framework provide an Interesting indication to structure of large-scale matter formations in the universe. The hawking radiation is a vital part of making the two conditions hold true. For each unit of fermions absorbed or manifested inside the area of extremum curvature we require a hawking radiation Particle emitted from the area, so the first requirement will hold true. 49 Manor O Measuring the Electron – 8T So measurement of the electron propagating across the nuclei seems to be impossible to do, as modern physics regard the electron as a cloud of probability. Third, suppose it was possible to measure the electron for a certain period of time. the measurement is done via scattering photons onto the electron and by doing so varying its energy, increasing it. Of course, the electron can omit those photons to a new direction or in a different rate, but measuring the electron will affect the electron energy and so the experiment itself is part of the problem. [(24 ∗5) + (3)] + 5 (3) + 5 = 8 8 = 0 [(24 ∗5) + (3)] + 5 (3) + 5 = 8 8 = 0 The thing to take from this short assay is that the problem of measurement is not just due to radii leading to an infinite energy scales, as 𝑟→0 but also due to the time needed to perform the measurement and the influence of photons as a tool of measurement that clearly effect the measured object by varying its energy, as it get absorbed into it. Another possible problem is the existence of the measurer that is matter on the manifold. The configuration of matter on the manifold is varying the matric tensor and causing it to accelerate outward and the manifold is different due to the matter configuration, given by the main equation of the 8T. 50 Manor O Manor O The Principle of Least Variation 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … (1.1) (1.1) (1.2) We derived the coupling constant as a ration between total arbitrary variations to the net variations, 𝑁𝑉, which are outside the parenthesis. Those net variations are a different representation of curvature on the Lorenztion manifold. Notice the numerical relations between the total to net: 𝑁𝑉 𝑇𝑉 ⟶𝑅 (1.5) 1 9 = 0.111 3 30 = 0.1 5 128 = 0.039 7 850 = 0.008 0.111 > 0.1 > 0.039 > 0.008 … (1.51) (1.5) 0.111 > 0.1 > 0.039 > 0.008 … (1.51) 0.111 > 0.1 > 0.039 > 0.008 … (1.51) The reasoning was clear, as the coupling constant equation is multiplies each Even sum of the previous element in the next prime, and the net variations are the prime numbers sequence itself. In means that each element the net curvature is a smaller and smaller portion of the whole variation cluster, which reason why the sequence is getting weaker and weaker. Based on this equation we can vividly derive and predict the weakness of gravity. We can say that nature is aspiring to minimize the ratio of net to total. All the possible amount of curvature can and will appear and nature, but the most common and noticeable ones are those with the bigger ratio, or least amount of net variation: 0.111 > 0.1 > 0.039 > 0.008 … →0 The bosons in which we are already know of. The interactions associated with the number one, three and five. The two lowest primes and one. The 8 – theory principle, which is derived by this analysis, is the Principle of least variation or curvature as we are dealing with a Lorenz manifold. Just as Feynman did in quantum path integrations, all is taken into account. However, the most significant routes are the simplest ones. In this framework the most significant Interactions are those with the largest ratios between the net Variations to the total variations The largest ratios are those with the least curvature or Smallest prime numbers and the number one, and primes are representing manifold variations. 51 Manor O Electron Positron Decay and the Higgs – 8T If there is an elimination of the destabilizer, i.e. the majestic three there will be no propagation of the boson from the fermion and the result would be again spin zero. In the 8T thesis, we constructed four categories for particle classifications using the primorial coupling constants equation: Spin zero: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠) Matter with spin one-half: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) Matter with spin one-half: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) Bosons with spin one: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) + 𝑁𝑉 Bosons with higher spin integers: (2𝑁 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠+ 3) + 𝑁𝑉1 + 𝑁𝑉2 + ⋯ → According to the following framework, the pairing of electron positron pair than can also construct an emerging of the Higgs. Since the Higgs has only one term in the coupling series, prediction would be propagation similar to gravity, that is local and short ranged. ee+ →𝐻0 ee+ →4𝑁 We can expend that result and say that any amount of even inverse in sign, fermions of the kind of majestic (3), i.e. the electron and its anti-matter dual, pairing to each other will yield a spin zero particle of certain sort. This particle again can by morphed into a new distinct particle given: 2𝑁+ 2𝑁= 2 ∑𝑁𝑖 𝐾 𝑖=1 If we eliminate the destabilizer there is no need to analyze the photons pairing. As one is a mathematical physicist and not a particle physicist, the reaction suggested may have been known already for a long time. however his prediction is made according to a new theory which predict the magnitude of coupling constants, and so may shade new light on the interactions among particles and unveil at least some of the complexity in that wonderful area of research. 52 Manor O Manor O The Coupling Constant Equation And The Wave-Particle Duality 53 Manor O The Coupling Constant Equation And The Wave-Particle Duality 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … (1.1) (1.2) We can vary the 𝑁𝑉 outside of the parenthesis so by doing so, reaching duality among the three first forces at 26 variations was attained. [8 + (1)] + 3 −(𝟏𝒗): [(8 ∗3) + (3)] + 3 ∶ [(24 ∗5) + (3)] + 𝟑 By analyzing the third element in the series, the propagation of a photon from a fermion so called the electron. Certain insight from the new framework is becoming vividly clear. In the context of wave particle duality. [(24 ∗5) + (3)] + 5 [(24 ∗5) + (3)] + 5 Since it is a prime net variation outside the parenthesis, it can not vanish into matter. As fermions vanish in even amounts. The ripple field of boson across the matric is mathematically described: ∇2= 𝜕2𝑀𝑥 𝜕2𝑔+ 𝜕2𝑀𝑦 𝜕2𝑔+ 𝜕2𝑀𝑧 𝜕2𝑔 𝑀𝑥,𝑦,𝑧 ∈𝑆 𝑆= (𝑀, 𝑔) ∇2= 𝜕2𝑀𝑥 𝜕2𝑔+ 𝜕2𝑀𝑦 𝜕2𝑔+ 𝜕2𝑀𝑧 𝜕2𝑔 𝑆= (𝑀, 𝑔) Suppose, in an experiment we decide to measure the photon momenta of position. Its done by scattering an additional photon onto the photon, which already propagated Form the electron. For simplicity sake, we suppose it is one additional element that is only one photon: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (3)] + 5 + 5 [(24 ∗5) + (3)] + 5 →[2N2 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 + 5 → [2N2 + 1 2] + 1 2 + 1 2 [(24 ∗5) + (3)] + 5 + 5 → [2N2 + 1 2] + 1 2 + 1 2 Those equations are the second variation of the coupling constant equation, which is the prime critical line. By adding the additional net variation, we reach a spin that is no longer associated with boson propagation, 3/2 .Before our measurement the bosons had spin one, and by measurement with additional photon, a variance of spin has occurred, so now our Boson behave like a fermion, it has an additional half unit of spin. Overall in the 8-theory by analyzing the coupling constant equation in the second representation, it is possible to extrapolate the reason for the phenomenon of wave particle duality. Fiber Bundles In the 8T, we are analyzing a varying manifold, which is the connected manifold with (3,1) signature. This manifold as you know has two components, the matric M, and the flow. The manifold has been analyzed in a Variational framework, i.e. Euler Lagrange equation to yield the main equation: ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 (1) (1) The purpose of this short assay is to describe the relationship between the base spaces, which is the Ricci flow to the total space that is the manifold matric tensor which we are living on. The relationship between those two spaces will be described by the concept of fiber bundle. The order in which events are accruing in this framework is firstly effected by the Ricci flow space, i.e. the base space. ∂L ∂s ⟵∂s ∂M ⟵∂M ∂g ⟵∂g ∂t We can define a fiber bundle between the Ricci flow and the Matric tensor. Define the base space and the total space: ℛ→𝐵𝑎𝑠𝑒 𝑠𝑎𝑝𝑐𝑒 𝕄𝑇→𝑇𝑜𝑡𝑎𝑙 𝑆𝑝𝑎𝑐𝑒 ℛ→𝐵𝑎𝑠𝑒 𝑠𝑎𝑝𝑐𝑒 𝕄𝑇→𝑇𝑜𝑡𝑎𝑙 𝑆𝑝𝑎𝑐𝑒 ℛ→𝐵𝑎𝑠𝑒 𝑠𝑎𝑝𝑐𝑒 ℛ→𝐵𝑎𝑠𝑒 𝑠𝑎𝑝𝑐𝑒 𝕄𝑇→𝑇𝑜𝑡𝑎𝑙 𝑆𝑝𝑎𝑐𝑒 𝕄𝑇→𝑇𝑜𝑡𝑎𝑙 𝑆𝑝𝑎𝑐𝑒 ψ: 𝕄𝑇→ℛ (1.32) ψ−1: ℛ→ 𝕄𝑇 (1.32.A) ψ: 𝕄𝑇→ℛ ψ−1: ℛ→ 𝕄𝑇 54 Manor O On Gravity and Acceleration 55 Manor O Manor O Manor O On Gravity and Acceleration We are familiar with the famous idea of Einstein, which state that there exist a morphism among gravity or curvature and acceleration. That is preciously the idea behind the main equation of the 8T, equation (1). 𝜕2𝑔′ ∂t2 = ∂g ∂t The question in which one will try to answer is the following: can we go further in reasoning this relation? Can we explain why it has to be that way? and do it in a simple manner which do not involve further complications equation wise. The author believes that it is possible to do using the framework of calculus of variations. To do just that we can imagine an arbitrary variation cluster which has mass, falling onto the curvature spike non vanishing. We can make an theorem and according to this theorem we can reason the relation of the main equation (1): : nature would aspire that a fermion cluster falling into a curvature spike will reach the Theorem (1.2) minima in minimal time. That is similar to Fermat principle of least time but in a different context. Now, the key point is the following: for the fermion cluster to reach the minima of the curvature spike in minimal time, it has to gain maximal speed, which is the integration of the acceleration. 𝑣= ∫𝑑2𝑥 𝑑𝑡2 Therefore, to reach the lowest point of the curve in the minimal time, nature would accelerate the falling body to a maximal speed. It is somewhat different from the equivalence principle as it puts a cause and a result relation among those two, but that is preciously the point of the paper. Can we explain why there is a morphism between those two terms? Using extremum value demand on time allows us to reason it in terms of cause and effect. Such is needed, as up to this point in time, we are able to reason it exist, Einstein proved it first in the 20-th century, but as far as one knows, the question of why was not answered. Summing up, nature is governed by creation of extremum values, based on theorem (1.2) a body falling into a curvature spike would aspire to reach the minima at 𝑡→0 and to do just that nature would aspire it's speed to reach another extremum, 𝑣→1. Theorem (1.2) could be regarded as the reason for the equivalence principle according to the author. The Feynman Path Integral Variation on Varying Lorentzian Manifolds In the 8-theory a varying Lorentz manifold is the entity of description. The Lorentz manifold Is inserted to an Euler Lagrange equation and by doing so, the main equation of the framework is obtained. ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 (1) (1) Are the conditions, which the framework is demanding to retain a stationary manifold. Those two conditions describe a time invariant acceleration directed from extrunum areas of curvature on the Lorenztion manifold. Intersection with the so-called dark energy.in addition, if no data is attainable from the first three term, it is vividly clear that there is an agreement with Einstein equivalence principle: The coupling constant equation was obtained by demanding a stationary manifold to experience net variation,𝑁𝑉= 2(𝑉+ 1/2); 𝑉≥0; 𝑁𝑉 ∈𝑃 as 𝑃 to be is the set of primes and the number one. Those ideas yielded an infinite series, in which each distinct Boson has a distinct prime, which is the 𝑁𝑉 value. Even amount of variations vanish, 2𝑉= 0. 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.2) The Feynman variation on Lorentz manifold - the objective of this part is to find out what is the probability transition of a boson from initial to final state on the manifold .Bosons are associated with prime amount of net variation 𝑁𝑉→2 (𝑉+ 1 2) which propagates as ripples on the manifold, given by a variation of the Laplacian: ∇2= 𝜕2𝑀𝑥 𝜕2𝑔 + 𝜕2𝑀𝑦 𝜕2𝑔 + 𝜕2𝑀𝑧 𝜕2𝑔 ∇2= 𝜕2𝑀𝑥 𝜕2𝑔 + 𝜕2𝑀𝑦 𝜕2𝑔 + 𝜕2𝑀𝑧 𝜕2𝑔 56 Manor O Manor O First, we define a manifold 𝑠= (𝑀, 𝑔) and insert it to an Euler LaGrange equation and an initial state of the manifold 𝑆(0) = (𝑀, 𝑔). Than we require the manifold to experience arbitrary variations, which vanish into matter. Let the arbitrary variations appear all across the matric on the manifold. Let the arbitrary variations appear all across the matric on the manifold. 𝛿𝑔 ∈𝑀(𝑥. 𝑦, 𝑧) Define a ripple propagation of a boson from an initial point on the manifold: Define a ripple propagation of a boson from an initial point on the manifold: 𝑞1 = 𝑀(𝑥1.𝑦1,𝑧1) And a final position of the matric ripple to arrive at And a final position of the matric ripple to arrive at 𝑞2 = 𝑀(𝑥2.𝑦2, 𝑧2) 57 57 Manor O The green arrow is directing the from the initial position of the ripple wave to final Position. The blue dots are the electrons omitting bosons, in that case a photon, marked in yellow. The manifold has arbitrary amount of fermions on it, which get scatterd by the initial boson wave and omit a new boson. There are infentialy more ways than the above drawing, its vivid. The ripple wave will scatter all the arbitrary variaions, but the highest probablity of arrival will be at the path of least curvature. Each fermion which get scattered omitting a boson with random direction of propagation. the framework has no data regrading the position of the propagation. The more arbitrary Variations getting scattered, the less probable it is to reach the final position. The following can be analyzed by the equation. The more arbitrary variaions in the Path, the more curved the matric, as there is an accelaration of it outward. The accelaration outward is causing the path to be longer and less linear, and so the time to reach the final position is longer, if only.There is no gurentte a photon will reach the final position in this framework as arbitary variaions created in a random fashion, and in configurations which are not predictable. However, if a photon will reach it will be in the least curved path, or the path with the least fermions getting scattered. 𝑃= ∫ 𝑑𝑞 exp [(𝑆0) ∫ 𝐿(𝑠, 𝑠′)𝑑𝑡 𝑡(𝑓) 𝑡(𝑖) 𝑞2(𝑡(𝑓)) 𝑞1(𝑡(𝑖)) ] (10) (10) Its unclear whether (10) is solvable as the arbitrary variations themselves vary their position over time and in addition, arbitrary variaons appear in random fashion in this framework. Its given by the first equation. So in a sense we can not sum all the paths if the paths vary at all times. it’s a complication of the feynamn result, But if we ignore the complication, the probablity transition should be calcuated using (10). (1.1) We can also represent the equation in the form: 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 1 30 : 1 128 : 1 850 : 1 9254 .. Manor O Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Given that framework, we can vividly see that gravity is belonging to the bosons with higher spin integers, as modern theories predict the gravitational interaction to have spin two. In the 8-theory Given that framework, we can vividly see that gravity is belonging to the bosons with higher spin integers, as modern theories predict the gravitational interaction to have spin two. In the 8-theory framework, what does it mean? In the context of the coupling constants equation what does it mean? Since it has spin two, we can associate gravity to the category of bosons with higher spin integers, which could relate to a certain combination of elements in the coupling constant series, as the elements are getting weaker and weaker, if the gravitational coupling will not be found by keeping developing to infinity it could mean gravitational will be found as a combination of elements in the series. Since it is spin two there should be three net variations outside. Gravitation as a combination of elements, using the fact it has a boson with spin 2. which could relate to a certain combination of elements in the coupling constant series, as the elements are getting weaker and weaker, if the gravitational coupling will not be found by keeping developing to infinity it could mean gravitational will be found as a combination of elements in the series. Since it is spin two there should be three net variations outside. Gravitation as a combination of elements, using the fact it has a boson with spin 2. [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 →[2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] Using the second representation of the coupling constant equation, meaning spin. It also means that the gravitational is a lot more rare as it is requiring a combination of elements in the series to be emitted and not just a singular element. Manor O Let us analyze the third element – Electromagnetism: [(24 ∗5) + (3)] + 5 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations 58 The Growth of Galaxies The new 8T framework regard bosons as net curvature on the matric tensor, as presented in equation (3.13B). The net curvature are of discrete amounts and is isomorphic to the prime numbers or one. Using three theorems made on a Lorentz manifold with (3,1)signature, it was possible to calculate the magnitude of the fine structure constant. Fermions are associated with arbitrary amounts of curvature in which curvature must vanish, their anti-commutation relation they appear in even amounts as in equation (2.12). According to this new framework of varying curvature, we can analyze the subject of growth of galaxies. The first point is that the growth of galaxies cannot segmented in time, since there infinite amount of coupling constants, i.e. net curvatures on the matric tensor, causing fermions to cluster, the amount of fermions in the galaxy should be increasing overtime. Now in consideration of the strength of each coupling term, the majority of matter should have been clustered in a relative short period as each coupling term is getting weaker and weaker. That is by the principle of least curvature, the ratio of net to total is aspiring zero in each term. A second point is that all interactions are taking part of the formation of galaxies, not just a single interaction as gravity. In fact, gravity might be the least significant in the formation of galaxies according to its order in the series, and according to its weakness. Therefore, the first point was that the formation is a continuous process, the second point of this short assay, is that the amount of fermions being clustered is inversely proportional to the development of the coupling series. The more we develop the less matter being clustered. We can make the following predictions: (1) Galaxy matter density is inversely proportional to the distance from the core of the galaxy. (2) The amount of matter being clustered is inversely proportional to time. (2) The amount of matter being clustered is inversely proportional to time. We can try and put it in mathematical rigor, suppose we took the amount of elements which vanished into matter by equation (1.35) and parametrized it: ∑𝛿𝑔𝑖 𝑁 1=1 ⟶𝐾 In addition, we can analyze the coupling term as a continuous analytical function over time ignoring the discrete amounts of curvature. Such is a valid representation due to equivalence of time arrows: 𝐹𝑟# ⟶(𝑀, 𝑔, 𝑡) 𝐹𝑟# ⟶(𝑀, 𝑔, 𝑡) 59 Manor O Manor O 𝜕𝐹𝑟 𝜕𝑡∝−1 𝜕𝐾 𝜕𝑡 (1.36) (1.36) The term (1.36) is meant to express prediction (2), the more we develop 𝐹𝑟 the less matter being clustered to the galaxy formation. Galaxies mass distribution should get denser and denser as we getting closer to the core, and vice versa. This is vivid by the principle of least variation: The term (1.36) is meant to express prediction (2), the more we develop 𝐹𝑟 the less matter being clustered to the galaxy formation. Galaxies mass distribution should get denser and denser as we getting closer to the core, and vice versa. This is vivid by the principle of least variation: Using the Coupling Constant Equation to Predict the Exact Mass of the Graviton 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) Indication That Fermions Are Closed Circles by the Coupling Constant Primorial Function Variation Indication That Fermions Are Closed Circles by the Coupling Constant Primorial Function Variation 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 𝐹𝑉=0 = 8 + (1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … The following representation of equation (1) by replacing the invariant three with pi. 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉→ (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (𝜋)) + 𝑁𝑉 8 + (𝝅 𝟑) ∶(24 + (𝝅)) + 3: (120 + (𝝅)) + 5: (840 + (𝝅)) + 7 … That is giving up certain accuracy on the coupling constant equation in order to get an insight regarding the shape of fermions. One is going to argue that such a representation is valid as we have a varying Lorenztion manifold, there could be a slight variations in the invariant three over time, toward pi and vice versa. In other words, the electron is not a perfect circle, but close to it. It is a varying circle, not a perfect shape. Varying in physical theories could mean vibration. The fact that we have a varying framework allow us to dynamically allow such slight variations without being rigid, the fact that it is not pi, could be a positive indication. Perfect shape of a circle would be problematic in a final theory, but a varying, imperfect circle seems to be much more elegant and suitable to a framework of constant variation. So according to this representation, a boson will be emitted from something close to a perfect circle, which is the electron. We gave up certain amount of accuracy and reached an astonishing insight regarding the shape of an electron. But we can go even further by representing the net variations in pi number multiples. Using the Coupling Constant Equation to Predict the Exact Mass of the Graviton 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5:(840 + (3)) + 7 … In the 8-theory framework, even amounts of manifold variations vanish. That feature allowed the following shift: In the 8-theory framework, even amounts of manifold variations vanish. That feature allowed the following shift: 8 + (1) →(1) 8 + (1) →(1) We know that the strong interaction has eight gauge fields meditating it. Those meditating particles do not carry mass. We also know that gravity has spin two. By switching to the second representation of the equation, we can represent gravity as the following: [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦 [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦 since even amount of variations vanish we will be left with one term in the final form of the term. That is similar to the strong interactions but immensely weaker. Since the bosons mediating the strong interaction are massless, and we can represent it in one term given the coupling constant equation, and by the analysis gravitation has only one term as well, we can reach a mathematical prediction, which will state, that gravitons has no mass. In agreement with reality and agreement with quantum field theory. The only thing taken from what was known before was the fact that the bosons meditating the strong interaction are massless. 60 Manor O Indication That Fermions Are Closed Circles by the Coupling Constant Primorial Function Variation They are not a separate entity of the Lorentz manifold, but appear as part of the Lorentz manifold and its ever varying nature. The closeness of the manifold indicate the closeness of the elements that appear in it. There could be more ways to prove that the following is correct. Which describe a varying Lorentz manifold. Fermions were proved to be arbitrary variations of the manifold. If the manifold is of finite size, i.e. closed, the elements in it should be closed as well. They are not a separate entity of the Lorentz manifold, but appear as part of the Lorentz manifold and its ever varying nature. The closeness of the manifold indicate the closeness of the elements that appear in it. There could be more ways to prove that the following is correct. Indication That Fermions Are Closed Circles by the Coupling Constant Primorial Function Variation 8 + (𝝅 𝟑) ∶(24 + (𝝅)) + 3: (120 + (𝝅)) + 5: (840 + (𝝅)) + 7 → 𝝅 𝟑) ∶(24 + (𝝅)) + 𝝅: (120 + (𝝅)) + (𝝅+ 1.82): (840 + (𝝅)) + (2𝝅 + 0.716).. Such representation is beautiful but what does it mean? of course that the real answer is that one does not know. Two options come to mind. The first is regrading the probability to find a boson in varying area. the bigger variations clusters, the larger the area of possible emission and the less likable it is do detect the boson. The higher the net variations, the smaller the probability to find the boson. Another possible option is of magnitude. The boson propagate across larger areas and thus its energy is getting divided across the area, so overall it gets much weaker as we develop the coupling constant series into infinity. In agreement with the weakness of gravity. Since the 8-theory was born in 2021, there could be more variations to the coupling constant equation. Other indication fermions are of closed shape is the main equation: Such representation is beautiful but what does it mean? of course that the real answer is that one does not know. Two options come to mind. The first is regrading the probability to find a boson in varying area. the bigger variations clusters, the larger the area of possible emission and the less likable it is do detect the boson. The higher the net variations, the smaller the probability to find the boson. Another possible option is of magnitude. The boson propagate across larger areas and thus its energy is getting divided across the area, so overall it gets much weaker as we develop the coupling constant series into infinity. In agreement with the weakness of gravity. Since the 8-theory was born in 2021, there could be more variations to the coupling constant equation. Other indication fermions are of closed shape is the main equation: 61 Manor O ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 (1) (1) Which describe a varying Lorentz manifold. Fermions were proved to be arbitrary variations of the manifold. If the manifold is of finite size, i.e. closed, the elements in it should be closed as well. Primorial Coupling Constants Equation and the Rise of the Arrow of Time 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ (1.1) 𝐹𝑉=0 = 8 + (1) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ Suppose a boson was emitted from a fermion due to net variation of a certain magnitude. If the arrow of time is two sided and reversible, there must be a way to bring the photon back to the electron. However, the physics of the 20-th Century forbids us from doing that, as we don’t even know where the photon is. Momentum and position are conjugate variables in quantum field theory. So once a boson is propagated into the metric, there is no possible way to bring it where it was. An additional argument is that all bosons are indistinguishable, so even if it was possible to trace and revert the photon, in a system with more than one Photon, its again beyond reach. The reason we emphasize those arguments as to the context of the arrow of time. Manor O what is time? Time is the result of net variations being clustered to different magnitudes. The succession of bosons with decreasing magnitude converging to zero is the direction of the arrow. The fact that each element is different than is preceding is the physical manifestation of the arrow of time. This equation encompass all the interactions according to magnitude, and so as those are different, the difference is the factor that gives rise to the arrow. If all elements in the series were identical there could not be a rise to the arrow. Using that coupling constant equation, we can reason for the chronology of events from the moment of singularity to the present moment. We can reason for electrons propagation only after protons were created. We can reason gravitational interactions only after electric interactions and we possibly can reason also, how galaxies were formed. Notice that the fourth element in the series is only 6.65 weaker than the electric. That is immensely stronger than the gravitational interaction and using that element as a building block for clustering after electric interactions and it is possible to explain how relatively fast galaxies formed in a short window of time, from the manifold being too hot to being too Primorial Coupling Constants Equation and the Rise of the Arrow of Time At first, at a certain point after the singularity, there were only elements of the first Element in the coupling series on the expended manifold: 8 + (1) 8 + (1) If the expended manifold experience multiple net variations of the first element than it is possible to cluster those: ∑𝐶𝑛= 8 + (1) 𝑛=∞ 𝑛=1 We can cluster into groups of three and get: (8 ∗3) + (1) ∗3 = 24 + (3) The invariant three, in 8-theory framework is, as you already know, is the destabilizing factor yielding a net variation so overall: The invariant three, in 8-theory framework is, as you already know, is the destabilizing factor yielding a net variation so overall: 24 + (3) →[24 + (3)] + 3 24 + (3) →[24 + (3)] + 3 Therefore, we can derive the intimate relation between the coupling constant series and the direction of time. The following procedure can be done on any additional element in the series. In the 8-theory 62 Universe Packets – Creation . let us examine the equation (2) which meant to express the idea of infinite amount of universes interacting with each other. ∑∂ℓ 𝜕𝑠𝑛 ∞ 𝑛=2 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 The second term indicate that there are more universes with negative orientation of curvature than the inverse orientation given by the first term. To create a flat universe packet we need infinite amount of pairs with opposite orientation that is as presented in the 8T thesis: ∂ℓ 𝜕𝑠1 − ∂ℓ 𝜕𝑠2 = 0 ∂ℓ 𝜕𝑠1 − ∂ℓ 𝜕𝑠2 = 0 However, instead of equalizing into zero, we can parametrize the equation and consider it as a universe pair, the packet than is considered as the summation of all the universe pairs. ∂ℓ 𝜕𝑠1 − ∂ℓ 𝜕𝑠2 = ℨ1 ℨ1 + ∑ℨ1 ∞ 𝑛=2 = 0 (2. A) So the idea is to represent the packet as the summation of universe pairs with opposite curvature orientation flatting each other, the universe packet according to this idea is infinite but contain an even number of universes, i.e. manifolds flatting each other. That is because we need an even number of manifolds with inverse curvature orientation. Another way of representation is to vary the equation (2): ∑∂ℓ 𝜕𝑠𝑚 𝐾/2 𝑚=1 𝜕𝑠𝑚 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−∑∂ℓ 𝜕𝑠𝑛 𝐾 2 𝑛=1 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2.1. A) When equation (2) was originally derived, the idea was to present the infinite packet as presented in the second term. However, (2..A1) and (2.A) are representing more accurate form of description as to how the packet is constructed, how much manifolds it must contain, i.e. an even number aspiring infinity with opposite curvature orientation. In addition to how it was possibly created. The packet was possibly created as a result of universe pairs which interact with each via areas of extremum curvatures, at first only with each other. Those two universes as they interact flatten each other causing outward acceleration from those 63 Manor O extremum curvatures. Later they join to another universe inverse dual to form a packet of four which flatten each other and so on, endlessly. Those pairs could cluster immediately or gradually toward the growing packet which will contain even amount of universes, as a set of pairs flatting each other. Universe Packets – Creation That is in agreement with the demand of the stationary on the Lorentz manifold presented in the beginning of 8T thesis. So according to this idea, the parameter K is an even. One final point, equation (2.A) represent the pairs within one universe packet considered infinite. It could be finite and then the structure of the multiverse is the summation of all the packets. That is by: 𝖅𝟏+ ∑𝖅𝒏 ∞ 𝒏=𝟐 = 𝓓𝟏 (𝟐. 𝑩) 𝒟1 + ∑𝒟𝑖 ∞ 𝑖=2 = 𝜗 (𝟐. 𝑪) 𝖅𝟏+ ∑𝖅𝒏 ∞ 𝒏=𝟐 = 𝓓𝟏 (𝟐. 𝑩) 𝒟1 + ∑𝒟𝑖 ∞ 𝑖=2 = 𝜗 (𝟐. 𝑪) (𝟐. 𝑪) Cosmological Flatness Flatness than, in the 8-theory is a continuous process given by the arbitrary variation term to vanish into matter and by ever decreasing ratio of total curvature to net curvature. by the direction of the arrow. Flatness than, in the 8-theory is a continuous process given by the arbitrary variation term to vanish into matter and by ever decreasing ratio of total curvature to net curvature. Prime-Fold Quark Chains & Anti Matter Terminators We have described the arbitrary variations of the manifold by the term on the main equation: ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t δg −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 δ𝑔′ = 0 We partitioned and discretized the arbitrary variation term and derived the existence of Fermion. In particular, we have shown that it must have an even amount of elements, which differ in sign and create nine threefold combination, and no more than two distinct elements. δg1 + δg2 … = ∑𝛿𝑔𝑖 𝑁 𝑖=1 ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (2.12) δg1(𝑂)δg2(𝑌)δg1 (2.12) To take an element back to itself, we needed two maps, which created a threefold combination, and we had eight such combinations, plus one arrow combination. Please notice the subtle structure: To take an element back to itself, we needed two maps, which created a threefold combination, and we had eight such combinations, plus one arrow combination. Please notice the subtle structure: δg1(𝑂)δg2(𝑌)δg1 → 𝜉= 1 [δg1(𝑂)δg2(𝑌)δg1(𝑂)δg2](𝑌)δg1 𝜉= 2 [δg1(𝑂)δg2(𝑌)δg1(𝑂)δg2(𝑌)δg1(𝑂)δg2](𝑌)δg1 𝜉= 3 The 𝜉= 𝑘 is a winding number, counting the repeats from an element to itself. Recall that we need the exact chain in opposite order to be the paired element, so the overall curvature could vanish into zero. However, we only dealt with the simplest case 𝜉= 1. the longer the chain, the less probable it is to have any chance to be eliminated. There is however, no law that prevents it, such things could accrue in nature. We can replace the last element in the chain with a curvature terminator δg1 → δg1 𝒯, which has to be the same as the first in the chain but opposite to it to ensure the mutual elimination, similar but opposite in sign means anti-matter, so δg𝑛 𝒯 are an anti-matter terminators . [δg1(𝑂)δg2(𝑌)δg1(𝑂)δg2(𝑌)δg1(𝑂)δg2](𝑌)δg1 𝒯 We can argue that the chain itself is separating the two, so the overall structure is stable. If it is stable, it means that the two can never reach each other; they are placed or connected by opposite side of the middling chain. Cosmological Flatness In the 8-theory a varying Lorentz manifold is the entity of description. The Lorentz manifold is inserted to an Euler Lagrange equation and by doing so, the main equation of the framework is obtained. ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 (1) ∂ℒ ∂s ∂s ∂M ∂M ∂g 𝛛𝐠 𝛛𝐭−∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐠′ 𝛛𝐭𝟐= 0 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 The framework regard (1.2) as the second main construction of the theory and currently has five representation and different uses of the equation. (1.2) represent the concept of Least variation, the most significant interactions in nature, are those with the largest Value of total to net 𝑁𝑉 𝑇𝑉. By analyzing the ratios of the first elements in the series we Conclude the ratio 𝑁𝑉/𝑇𝑉 →0 as 𝑁𝑉→∞ , the result is following ratios: 1 9 = 0.111 , 3 30 = 0.1 , 5 128 = 0.039… 0.111 > 0.1 > 0.039 > 0.008… 1 9 = 0.111 , 3 30 = 0.1 , 5 128 = 0.039… The biggest rations are those with the least 𝑁𝑉 amount and thus they are the most noticeable on the manifold. The third representation of (6) revolves around to the arrow of time. The direction of the series is assumed to match the direction of Time. So as we increment the time 𝑡1 = 𝑡1 + ∆𝑡 and allowing time aspire Infinity 𝑡→∞, the manifold will experience higher number of net variations 𝑁𝑉→ ∞. and at the same time the ratio of 𝑁𝑉 𝑇𝑉 →0, which means that the matric on the manifold Is getting more and more flat. The most curved, or intense interaction than is the first, the Strong interaction due to its largest value of 𝑁𝑉 𝑇𝑉. Using the coupling constant equation ratio between net to total variations (curvature) it becomes vividly clear that gravitation will be aspiring to flatness, due to the immense value of 𝑁𝑉. Gravity will be almost not noticeable. It is also possible to derive that the manifold will become more flat followed 64 Manor O by the direction of the arrow. Prime-Fold Quark Chains & Anti Matter Terminators [δg1 −δg2−δg1 −δg2 −δg1 −δg2]) −δg1 𝒯 [ ℎ𝑖 𝑓 𝑏𝑖 𝑖 𝑖 ] [δg1 −δg2−δg1 −δg2 −δg1 −δg2]) −δg1 𝒯 δg1 −[𝑐ℎ𝑎𝑖𝑛 𝑜𝑓 𝑎𝑟𝑏𝑖𝑡𝑟𝑎𝑟𝑦 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠] − δ [δg1 −δg2−δg1 −δg2 −δg1 −δg2]) −δg1 𝒯 δg1 −[𝑐ℎ𝑎𝑖𝑛 𝑜𝑓 𝑎𝑟𝑏𝑖𝑡𝑟𝑎𝑟𝑦 𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛𝑠] − δg1 𝒯 The overall chain structures are prime, notice that they have according to the first three-winding numbers three, five and seven elements accordingly, and can go to infinity. It is really a remarkable sight to reveal how important the prime numbers are to most fundamental and intimate ways of nature. 65 Manor O Quark Confinement and the D'alembert Variation ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝛿𝑔→∑𝛿𝑔𝑖 𝑁 𝑖=1 = 0 𝑁∈ℝ, 𝑁→∞ −𝛿𝑔′ → ∑−𝛿𝑔′ 𝑖= 0 𝑁 𝑖=1 ∑𝛿𝑔𝑖 𝑁 𝑖=1 − ∑𝛿𝑔′ 𝑖= 0 𝑁 𝑖=1 ∑𝜕𝐸𝑖/𝜕𝑡 𝑁 𝑖=1 − ∑𝜕𝐸2 𝑖 𝜕2𝑡 𝑁 𝑖=1 = 0 (2.1) (1) The sum of all arbitrary variations and accelerations is taken to zero in this framework. Similar to the procedure D'alembert taken with forces and accelerations. That is an additional take on the phenomena of quark confinement, published earlier by the author. The sum of all arbitrary variations and accelerations is taken to zero in this framework. Similar to the procedure D'alembert taken with forces and accelerations. That is an additional take on the phenomena of quark confinement, published earlier by the author. 66 Manor O The Equivalence Principle in Quantum Scale ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) Suppose the matric has a fermion that is an arbitrary variation of the manifold given by the term 𝛿𝑔= 0 . What would be the consequence? Given by equation (1) the arbitrary variation will cause the matric to accelerate outward. That is in complete agreement of Einstein theory of gravitation, equation (1) implies: 𝜕𝑔 𝜕𝑡= 𝜕2𝑔′ ∂t2 Since the matric varied due to the arbitrary variation, which appeared in it, and in particular, it expended outward, the distance increased. Suppose the quark was conscious and could perform measurement, its very existence affected the matric, and the time in which a boson field will need to reach the object measured has increased because of the quark manifesting. In special relativity, the great Einstein used velocity, but here there is no velocity. There is no such thing velocity in the 8 theory. The quark may conclude that the object is moving, but what is happening is that the matric itself is varying, because of that quark. We also have in this framework the invariance of the speed of light, given by the coupling constant equation, and the fact that the propagation process is similar in all interactions. [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 General relativity implies an equivalence relation between curvature and acceleration. 8 theory implies that as well, but also in addition implies that curvature will cause outward acceleration of the matric by (1). Einstein had to add the cosmological constant in an artificial way, but here it’s the main equation. Such a condition than allow us to understand relativity in a new and elegant way. C is invariant, and every arbitrary variation of the manifold causing an outward acceleration of the matric, the matric itself varying in such way that those arbitrary variations will eventually measure different distances and times, the measured object can be standing still but it will observed as moving, but what is happening is that the matric is expending. The entire theory of Einstein is not only contained in just one equation but expended to a new horizon. General relativity implies an equivalence relation between curvature and acceleration. Dimensional Multiverse Infinite Dimensional Multiverse nfinite ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝜕ℒ 𝜕𝑆𝑀 − 𝜕ℒ 𝜕𝑛 = 0 (1.53) ∑∂ℓ 𝜕𝑠𝑚 𝐾/2 𝑚=1 𝜕𝑠𝑚 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−∑∂ℓ 𝜕𝑠𝑛 𝐾/2 𝑛=1 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2.1.A) Equation (2) is the second representation of the main equation. Notice that even though the main equation describe both Einstein theory of relativity (1) and time invariant acceleration away from extremum curvature on the manifold, it lacks providing the reason for such a process. Equation (2) is than used; our universe is wrapped in many similar, stationary manifolds, which are distinct. They are assumed topologically invariant. Such a construction than allow us to understand why each manifold can not have any number of dimensions, it is confined within many other manifolds. Its also more reasonable to assume that there are many stationary manifolds than to assume that there is only one stationary manifold. The (2) is more elaborated equation than equation (1). Suppose that each manifold has n-dimensions. ∑∂ℓ ∂sn K m=1 ∂s1 ∂M ∂M ∂g ∂g ∂t → ND Sm Take into account the number of manifolds wrapping our manifold making its matric accelerate outward and add those dimensions ∑∂ℓ ∂sn K n=1 ∂sn ∂M ∂M ∂g ∂g ∂t → ∑ND s(n) ∞ n=2 So the number of dimensions in our framework is: So the number of dimensions in our framework is: 𝑁𝐷 𝑆1 + ∑𝑁𝐷 𝑠(𝑛) ∞ 𝑛=2 = 𝑇𝐷 (2.11) (2.11) 67 67 Manor O Manor O The Equivalence Principle in Quantum Scale ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 8 theory implies that as well, but also in addition implies that curvature will cause outward acceleration of the matric by (1). Einstein had to add the cosmological constant in an artificial way, but here it’s the main equation. General relativity implies an equivalence relation between curvature and acceleration. 8 theory implies that as well, but also in addition implies that curvature will cause outward acceleration of the matric by (1). Einstein had to add the cosmological constant in an artificial way, but here it’s the main equation. (1). Einstein had to add the cosmological constant in an artificial way, but here it s the main equation. Such a condition than allow us to understand relativity in a new and elegant way. C is invariant, and every arbitrary variation of the manifold causing an outward acceleration of the matric, the matric itself varying in such way that those arbitrary variations will eventually measure different distances and times, the measured object can be standing still but it will observed as moving, but what is happening is that the matric is expending. The entire theory of Einstein is not only contained in just one equation but expended to a new horizon. Such a condition than allow us to understand relativity in a new and elegant way. C is invariant, and every arbitrary variation of the manifold causing an outward acceleration of the matric, the matric itself varying in such way that those arbitrary variations will eventually measure different distances and times, the measured object can be standing still but it will observed as moving, but what is happening is that the matric is expending. The entire theory of Einstein is not only contained in just one equation but expended to a new horizon. 68 Manor O Manor O The Coupling Constants Series and Gluons Confinement ∑𝛿g𝑖 𝑁 1=1 = 0 (2.12) (2.12) Bosons are mentioned in the first paragraph are described as net curvature, given by the term (2.12) of the photon for example. Now, we have used the visualization of the sea of gluons on the Quark triplet in the following way. ∑𝛾𝑖 𝑀 𝑖=1 = ∑𝛿g𝑖 𝑀 𝑖=1 > 0 (3.13. 𝐵) (3.13. 𝐵) When we indulge in high energy collusions that is synonymous with trying to roll the quark triplet uphill. It is possible to try as the bosons are just net curvature unbound as given by (1), however since each boson is a curvature of certain magnitude it increase the probability of arrival to its position, therefore we have a "sea" of gluons. That was the analysis in the context of Quark confinement. Now assume we have a positive summation of Gluons trapping a Quark triplet in the above hyperbole. Assume there is no restriction regarding Gluons, one of them leaves the hyperbole. ∑𝔤𝑖 𝐾 𝑖=1 = ∑𝛿g𝑖 𝐾 𝑖=1 → ∑𝛿g𝑖 𝐾−1 𝑖=1 (3.13) (3.13) Since there are a sea of gluons, and one free gluon, which just left, the gluon that just left could be replaced by another Gluon or alternatively are re-attracted to the hyperbole just as larger masses attract smaller masses, as an analog. Strong curvature clusters pull weaker curvature or free curvatures. The pull is not restricted only to fermions such as Quarks. In that way we could explain the phenomena of Gluon confinement. One final point, since there are eight gluon fields, we should be able to describe the interaction on the matric tensor between each Gluon type. In other words, given two net curvatures unbound which somehow differ in their nature, the matric tensor itself may produce a meditator in between, so this meditator may be regarded as a physical entity, which could or could not manifest as a new particle. Such description are currently not within the domains of description of the 8T. That raises another question, how can two net curvatures on the matric tensor exactly differ from one another ? +(1) . assuming they are all isomorphic to the same discrete number, i.e 69 Manor O Symmetry of a Universe Packet Symmetry of a Universe Packet ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝜕g 𝜕𝑡= 0 , − 𝜕2𝑔′ ∂t2 = 0 ∑∂ℓ 𝜕𝑠𝑛 𝐾 𝑚=1 𝜕𝑠1 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−∑∂ℓ 𝜕𝑠𝑛 𝐾 𝑛=1 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2. A1) (1) We can define a functor: We can define a functor: ∧ ∶𝑇𝑜𝑝→𝑆𝑒𝑡 To switch from a topological space, as manifold wrapping into a discrete setting. Now the entire universe packet is an open set. For simplicity sake we assume it has finite amount of elements, and so it's really a closed set for the sake of the example. The manifold itself is an open manifold due to equation (1) it has no boundary and it is uncompact, the switching into set is than meant to emphasize the object itself, confinement among many others. ℘→(𝑆1, 𝑆2 … 𝑆𝐾) (1.83) 𝑆𝐾= (𝑀𝐾,g); 𝐾≥1; 𝐾∈ℝ ; ℘→(𝑆1, 𝑆2 … 𝑆𝐾) (1.83) 𝑆𝐾= (𝑀𝐾,g); 𝐾≥1; 𝐾∈ℝ ; (1.83) Equation (1.83) meant to specify the closed set of open manifolds, causing the matric tensor of each manifold to accelerate outward. Notice that there is a symmetry in the set, we can vary each element order it won’t make a difference, equation (1) will hold. In particular, the conditions below equation (1) will hold either way, and for simplicity we assumed it closed. If there are additional manifold packets joining the set than the conditions below (1) could by adiabatically invariant, assuming that is in fact the case we can reach a new prediction: The rate of acceleration from areas of extremum curvature should increase overtime, if (1.83) is an open set. We can describe the time invariant acceleration as a product of many connected manifolds, assumed stationary which interact at areas of extremum curvature. The order is not important and thus it is possible to vary the index of each manifold. The set itself is which increase overtime is the key. 𝜕2𝑔′ ∂t2 = 𝑆1 ⊕𝑆3 ⊕… . 𝑆𝐾 𝐾⟶∞ 𝐾⟶∞ 70 70 Manor O Using the Primorial Coupling Constants Function to Derive C Invariance Yang Mills Conjecture Using the Primorial Coupling Constants Function to Derive C Invariance Yang Mills Conjecture 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ The second representation of the primorial function using the prime critical line: The second representation of the primorial function using the prime critical line: [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 Notice that beside the variation cluster which get bigger, all the interactions are taking the same form. We have the destabilizing factor, which is the electron for example, or the 1 2 inside the parenthesis, yielding a net variation which is also 1 2 or prime, according to first representation of the primorial function. So the propagation speed of all boson of this type must be similar, precisely because there is no detail regarding the speed of propagation and because there is no difference among the bosons, they are all of the same hand. We already proved their dynamical nature by varying the net variations and as a result making them scalar multiples. Based on the framework of the eight theory, we can make an additional prediction. All bosons of the above type, 2𝑛() + 1, with no consideration of their mass, will propagate across at the same speed. The same speed applies for all. Even to bosons with mass. That is the Yang Mills problem, how can a boson that carry mass move at the speed of a boson, which do not carry mass. In the 8-theory the answer is given. Since mass is associated with 8 −1 variations, and boons are of the type 8 + 1 the combination of a boson with mass will not effect on the propagation. Notice that beside the variation cluster which get bigger, all the interactions are taking the same form. Proof: Quarks are the Fundamental Building Block ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) Suppose that the two distinct elements derived in the beginning of the thesis are not fundamental and are constructed by two elements that are more fundamental: δg1 = δg𝑎+ δg𝑏 > 0 δg2 = δg𝑐+ δg𝑑< 0 δg1 = δg𝑎+ δg𝑏 > 0 δg2 = δg𝑐+ δg𝑑< 0 Since we require the series to vanish, take all the sub elements and combine them. If: δg𝑎+ δg𝑏+ δg𝑐+ δg𝑑 ≠0 The series could not vanish; there could not be four distinct elements as subsets. There could not be also three distinct elements that differ in sign, as proven in the original 8T thesis. The result of such construction, is that even if the Quark themselves are composite of certain sort according to the new scenario, the sub elements of those Quarks will appear as Quarks. Meaning they will appear as two varying elements, in even number, which differ in sign and anti-commute or summed as zero when combined. Such a simple prove that there is not anything new beyond Quarks. In addition, even if there is, the new elements will appear as Quarks. That is in agreement with the lack of experimental evidence for anything beyond Quarks, and the notion that Quarks are indeed the most fundamental. Another important point is that the reason of Quarks being the most fundamental is a result of stationary . Lorentzian manifold . Lorentzian manifold 71 71 Manor O Using the Primorial Coupling Constants Function to Derive C Invariance Yang Mills Conjecture We have the destabilizing factor, which is the electron for example, or the 1 2 inside the parenthesis, yielding a net variation which is also 1 2 or prime, according to first representation of the primorial Notice that beside the variation cluster which get bigger, all the interactions are taking the same form. We have the destabilizing factor, which is the electron for example, or the 1 2 inside the parenthesis, 1 8 −1 + 8 + 1 = 0 (3.1) (3.1) The boson that is a mass carrier, causing the matric to converge inward, will be balanced to the other direction by its very nature. As a result, he will move on a linear, not curved trajectory and his speed will not be effected by its mass. In equation six, we took even amount of variations to vanish and so the result is zero. No curving to either direction. Of course, the ideal would be to extract the actual speed of light from the 8-theory. It is currently beyond reach. The equation does not change under any condition that means that the speed of propagation does not means under any conditions. In the case of the third element, than, speed of light is invariant to all. Therefore, the 8-theory framework suggest an elegant . g mills conjecture and simple solution to the Yan 72 8T and QFT – Axiomatic Analysis Quantum field theory has certain features that play a significant rule, and repeat themselves in one way or another along each epos of the theory. Among those, we can name the commutation and anti- commutation of bosons and fermions. The Dirac delta or interference known as a field, the operators of matter creating and destructing, cluster decomposition and Lorentz invariance. In addition to Feynman path integrations and diagrams. That being said, what are the mathematical axioms in which QFT is built upon? One would like to suggest those following axioms: Axiom (1) – Nature is probabilistic Axiom (1) – Nature is probabilistic Axiom (2) – Fermions repeal, Bosons do not Axiom (3) – There is only one set of rules Axiom (3) – There is only one set of rules By the first axiom, we can include the Feynman diagrams and the Feynman path integrations. In addition to arbitrary amount of matters appear and disappear by operators we insert. By the second axiom the commutation and anti-commutation relation and the nature of spin and statistics. The third axiom, the Lorentz invariance and the entire set of symmetries and conservation laws, at quantum scale (Nother) and at large scale (Lorentz). Those three axioms also stand at the heart of 8-theory, so in essence the nature of those theories, their innate ideas about nature is the same. The difference is which ideas are describing the axioms and which objectives the theory is set to achieve. Quantum field theory searches for probability of certain occurrences, it does it amazingly well but lacks to provide the reason for those arbitrary numbers, such as coupling magnitudes. QFT uses integrations across the entire space-time that are impossible to solve. 8- Theory is also probabilistic in its nature, maybe even more than QFT. It has no data regarding any direction of motion, momenta, and location at any point and so on. Very little to no physical data is manifested in this theory. However, it does describe beautifully the magnitudes of the couplings, the reason each magnitude is what it is, the process of propagation and the dynamic nature of the forces. The methods uses are partial differential equations, and the methods also uses in quantum field theory given by axiom (2), the commuting relation of fermions and bosons. It does not currently have complicated integrations over space-time or it can specify the decays as QFT. Manor O Manor O 73 73 8T and QFT – Axiomatic Analysis However, it does describe the dark energy in an accurate fashion given by its main equation, a varying Lorentz manifold. Gravity is within its domain of description as it was built upon the work of two of the greatest minds in science Einstein and Lorentz. It is also supported by the coupling constant equation and predict that graviton will be massless and that gravity is actually a combination of three net variations. The 8T has two arbitrary numbers less than QFT; it predicts infinite bosonic fields, which relate to Lorentz net curvature on the manifold. It also predicts infinite families below first generation, and thus does not face questions as to those arbitrary numbers. 8T and QFT both are described in terms of the Dirac delta. QFT uses the delta as a description for the wave equation, as a way to describe a complete set of states, alongside with a set of amplitudes. 8T uses the Dirac delta in more flexible manner, it applies to times that are different from zero as well, and describe how an arbitrary amount of curvature vanish into matter. Any net variation at a later continuation of time than describing a bosonic ripple field across the manifold, given by a variation of the Laplacian. While QFT is mainly physical, 8T is mainly and almost completely mathematical, the axioms at the heart of those theories are the same, the methods are similar, the 8T describe phenomena not within the realms of QFT, and QFT can calculate probabilities not within the realm of 8T. 8T is just as probabilistic as QFT, if not more. It validates Pauli Exclusion Principle and the fermionic and bosonic difference between spin and statistic, and have just one set of rules. This set of rules has three axioms: Axiom (1): All universes are Lorentzian manifolds Axiom (1): All universes are Lorentzian manifolds Axiom (2): All Lorentzian manifolds are stationary Axiom (2): All Lorentzian manifolds are stationary Axiom (3): Net Curvature on the manifold is a bosonic field. Net are Primes or one. Axiom (3): Net Curvature on the manifold is a bosonic field. Net are Primes or one. 74 Manor O Weaknesses in the QFT Framework First, we can represent the QFT functional integral, equation that we cannot solve. Z = ∫𝐷𝜑𝑒𝑖 ∫𝑑4𝑥[1 2𝜕𝜑2−𝑉(𝜑)+𝐽(𝑥)𝜑(𝑥)] The QFT framework is assuming such a thing is solvable and we can not solve it. Author will argue it is incorrect. First, by integrating all over space-time, physicists make an implicit assumption that space- time is continuous and smooth. Such an assumption is invalid, in the new 8T framework in which space-time is the matric tensor varying presented in equation (2), there could be knots, deformations of the matric tensor to the flow, i.e. the base space, given by fiber bundle. ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) So already, it is the first complication on the QFT framework, which count as a weakness. The second, QFT classify notions according to fields, which is a function of space-time, at the same time it lacks providing reasoning to what those fields are, or how they were created. QFT domain of description does not include dark energy, dark matter, the moment of singularity, gravity and curvature. Its domain of description is mainly partial and limited, despite its accuracy. Another point which is quite important is that in examining a theory we should classify according to two different categories. The first, is the ideas, equations and predictions. the second is the methods in which those ideas are described. For example, invariance under shifting frames in quantum scale is described by group theory suggested by Wigner in 1930. If we classify QFT according to ideas and methods, it is vividly clear that there are few simple ideas described by complicated methods and long and unclear notation. 75 75 The Three Critical Theorems "Theorem (1) – nature will not allow a prime amount of variation to appear by itself. Define prime to be (2n+1) variations. 1.1) Prime amounts appear in pairs." Theorem (1) - The physical meaning of that theorem is that bosonic fields cannot be propagated from nowhere. The 8T correlate bosonic propagation to prime net variations of the manifold, and bosons, as we know them, propagate from fermions, which vanish in even number of variations. Theorem (1.1) – even amount of variations is the result of two prime numbers combined. So to create variation cluster vanishing into matter we need two primes to appear in a pair. "Theorem (2): Nature will generate force if a prime net amount of arbitrary variation will appear. Net variations will appear when combine two amounts of prime variations. "Theorem (2): Nature will generate force if a prime net amount of arbitrary variation will appear. Net variations will appear when combine two amounts of prime variations. "Theorem (2): Nature will generate force if a prime net amount of arbitrary variation will appear. Net variations will appear when combine two amounts of prime variations. Two does not appear, as it is an even amount of variations, which vanish." Theorem (2): In continuation of theorem (1), after variation cluster vanished into matter, two distinct elements in threefold combination, a net variation, which is prime can propagate from within it. The feature of the bosonic propagation is their prime number amount of variations, and therefore their expansion across the entire matric. A boson must propagate from an even amount of variations, which is matter. Theorem (2): In continuation of theorem (1), after variation cluster vanished into matter, two distinct elements in threefold combination, a net variation, which is prime can propagate from within it. The feature of the bosonic propagation is their prime number amount of variations, and therefore their expansion across the entire matric. A boson must propagate from an even amount of variations, which is matter. Theorem (3): "Each prime pair should have a net variation element 𝑁𝑉 proportional to Total Variations value divided by two" Theorem (3): "Each prime pair should have a net variation element 𝑁𝑉 proportional to Total Variations value divided by two" Theorem (3): Each net variation is proportional to the average of the elements in the pair. There could not be net variation 𝑁𝑉= +(101) propagating from (7,11) total variation pair. It does not make sense. The three theorems in be put in concise and simple manner: The three theorems in be put in concise and simple manner: (1) Bosonic fields cannot propagate from nowhere (2) Bosonic Fields propagate from matter clusters (3) Bosonic fields are infinite in kind and isomorphic to prime numbers or one. Theorem (3) was the critical theorem that eventually allowed calculating the value of the fine structure constant and validating the entire framework. Theorem (3) was the critical theorem that eventually allowed calculating the value of the fine structure constant and validating the entire framework. 76 76 Manor O Manor O Refuting Magnetic Monopoles Examine the term describing the electric coupling. We proved majestic three is the electron. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ Define a magnet as a set of electrons, which spin around as part of a larger cluster of matter. ∑𝑒𝑖 𝑁 𝑖=1 → ∑(3)𝑖 𝑁 𝑖=1 ∑ei N i=1 ∈∑δgk M k=1 ; M > N ∑𝛿𝑔𝑘 𝑀 𝑘=1 = 0 (2.12) (2.12) As we did in the 8T thesis, the elements in the term describing matter anti commute, appear in an even number that differ in sign and vanish to zero when summed. However, the spinning electrons are added to a positive summation: ∑(3)𝑖 𝑁 𝑖=1 > 0 We have two conditions that are not aligned and contradict each other. Both were proven in the 8- Theory to be correct. ∑(3)𝑖 𝑁 𝑖=1 > 0 ∩ ∑δgk 𝑀 𝑘=1 = 0 ∑(3)𝑖 𝑁 𝑖=1 > 0 ∩ ∑δgk 𝑀 𝑘=1 = 0 The only way to satisfy the second term is to add an opposite spinning cluster so the term would vanish into zero, meaning spinning cluster of electrons in the opposite direction, so (2.12) would be satisfied. ∑(−3)𝑖 𝑇 𝑖=1 < 0 ∑(−3)𝑖 𝑇 𝑖=1 + ∑(3)𝑖 𝑁 𝑖=1 = 0; 𝑇= 𝑁 ∑δgk M k=1 = 0 77 77 Manor O The Most Symmetrical Interaction is The Weak Interaction We have proven that the majestic (3), in the case of the electric coupling is the electron. The destabilizing factor yielding a net variation. Overall, the thesis main example was the third element in the series. Therefore the weak interaction did not get enough interaction regrading a very interesting feature it possess. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= [(8 ∗3) + (3)] + 3 We can replace the net variation by the majestic three and the correctness of the term will retain. It could explain why the weak interaction is different in terms of its spin, and also allow us to make prediction regarding a fermion, which is analogous to the electron, which can get propagated by the boson of the weak interaction, , 𝑁𝑉= +(3). The overall value is the same; there is a "symmetry" in such a variation, which is not attainable in any term of the coupling constant series. It could mean that the majestic three regarding the weak and the boson, which is propagated, are isomorphic to each other. [(8 ∗3) + (3)] + 3 → [(8 ∗(3)) + 3] + (3) (8 ∗ ∏(3) 𝑉=𝑅 𝑉=1 + 𝑁𝑉) + (3) 78 78 Manor O Manor O Manor O Manor O Hermitian Conjunction and Prime Numbers ∑δgi N i=1 = 0 N → ∞ N = 2𝑛; 𝑛∈ℝ There is no limitation concerning such measurement, we have an even amount of arbitrary variations, which differ in sign and summed as zero. Suppose we had an odd amount of arbitrary variations. N = 2𝑛+ 1; 𝑛∈ℝ; 2n + 1 ∈ℂ ∑δgi N+1 i=1 ≠0 N = 2𝑛+ 1; 𝑛∈ℝ; 2n + 1 ∈ℂ So now, the measurement of the fermion cluster become impossible as the manifold is no longer stationary. An elimination of that extra variation must be made. Nature can eliminate it by mirror projections, i.e. Hermitian conjugation. By doing so, the measurement of the fermion cluster will become possible again, or transitioned back to the real field from the complex field. ∑δgi N+1 i=1 + ∑δgi N−1 i=1 = 0 2𝑛+ 1 + 2𝑛−1 = 0 ∑δgi N+1 i=1 + ∑δgi N−1 i=1 = 0 2𝑛+ 1 + 2𝑛−1 = 0 2𝑛+ 1 + 2𝑛−1 = 0 So even amount of variation is measurement, additional variation causing the measurement to become impossible, and transition it to the complex field which makes the measurement impossible. To retain the previous state, a mirror projection will be taken. 2𝑛 ∈ℝ 2n + 1 ∈ℂ 2n + 1 + 2n −1 ∈ ℝ; 2n + 1 ∈ℂ 2n + 1 + 2n −1 ∈ ℝ; Define Hermitian as: Define Hermitian as: ℋ ∶ℂ→ ℝ 79 79 Final Shot at Quantum Relativity Define an observer, distinct observer, as an arbitrary amount of curvature on the manifold. An infinite series of fermions. ∑𝛿𝑔𝑖 𝑁 𝑖=1 = 0 (2.12) 𝑁→ ∞ ∑𝛿𝑔𝑖 𝑁 𝑖=1 = 0 (2.12) 𝑁→ ∞ (2.12) 𝑁→ ∞ Define an additional observer, distinct, which differ in the amount of curvature it creates on the matric. The observer is an infinite series of fermions which overall vanish into matter. ∑𝛿𝑔𝑟 𝑀 𝑟=1 = 0 𝑀→ ∞ ∩ 𝑀! = 𝑁 ∑𝛿𝑔𝑟 𝑀 𝑟=1 = 0 𝑀→ ∞ ∩ 𝑀! = 𝑁 𝑀→ ∞ ∩ 𝑀! = 𝑁 Now, analysis of the two observers on equation (1.2). Assume they are measuring the same object, and the entire matric is null, the entire matric contain each observer and the measured object. The setting chosen for simplicity sake, as those things will be too complex to analyze in a real physical scenario. Defined the measured object for both observers as: ∑𝛿𝑔𝑘 𝑇 𝑘=1 = 0 ∑𝛿𝑔𝑘 𝑇 𝑘=1 = 0 Now for the first observer and the measured object, the total arbitrary variation summed as: ∑δgi N i=1 + ∑δgk T k=1 = 0 Now for the second observer and the measured object, the total arbitrary variation summed as: ∑δgr M r=1 + ∑δgk T k=1 = 0 ∑δgi N i=1 + ∑δgk T k=1 ≠∑δgr M r=1 + ∑δgk T k=1 [ ∂ℒ ∂s ∂s ∂M ∂M ∂g ] ∂g ∂t δgik −[ ∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′] ∂2g′ ∂t2 δg′ ik = 0 [ ∂ℒ ∂s ∂s ∂M ∂M ∂g] ∂g ∂t δgrk −[ ∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′] ∂2g′ ∂t2 δg′ rk = 0 (2.13) [ ∂ℒ ∂s ∂s ∂M ∂M ∂g] ∂g ∂t δgrk −[ ∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′] ∂2g′ ∂t2 δg′ rk = 0 (2.13) (2.13) 80 80 Manor O Those observers will cause the matric to accelerate outward so the object will be observed moving. His velocity is dependent upon the amount of curvature the observer is creating, and so two different observers, different by the above definition, will measure two different distances crossed and two different times for the same object. The reason however, is not for the object itself, it’s the different nature of the observers, and in particular the amount of curvature they possess. Now since we proved the yang mills conjecture we have the same propagation speed for all bosons: [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 The time needed to cross the same matric which accelerated outward in different amounts is different. So, measured time which is different for each observers is quite vivid and a must by using (1.2) and the 8T framework. In fact, using such framework makes relativity notoriously complicated, as everything needs to be taken into account. Everything is causing the matric to vary; it is at a verge of impossible to do at the real world. Our best theories are radically simplified. By "everything", one means every arbitrary variations of fermion in the matric needs to be taken into account, which was not done in that analysis for simplicity sake. The majority of the paper was known to the reader. What is different is the reason of relativity and the analysis of this beautiful idea in the 8T framework, which imposes additional complications, in quantum scale. 81 Manor O The Coupling Constants Series and Total Variations Pairing 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.1) (1.2) We have obtain the net variation, 𝑁𝑉, as part of a total variation pair,(𝑝1, 𝑝2), which we required the sum to be two and three divisible. We gave two examples for the strong interaction: (𝑝1, 𝑝2) = (5,13) (𝑝1, 𝑝2) = (7,11) (𝑝1, 𝑝2) = (7,11) Two points with regard to those pairs. First, it is commutative, we can replace the elements in the pair and nature will be invariant, the coupling series will hold: (𝑝1, 𝑝2) → (𝑝2, 𝑝1) (𝑝1, 𝑝2) → (𝑝2, 𝑝1) Nature is invariant to the actual value of the elements; we can choose any two primes, as long as their sum creating an even number, two and three divisible of certain magnitude, the coupling constant will hold as well. In the 8T thesis we chosen the first pair, it could have worked exactly as well with the second pair. (𝑝1 + 𝑝2) = 𝑆1 (𝑝3 + 𝑝4) = 𝑆1 (𝑝1 + 𝑝2) = 𝑆1 (𝑝3 + 𝑝4) = 𝑆1 (𝑝3 + 𝑝4) = 𝑆1 An additional point that was not mentioned in the thesis, the coupling series will hold with any additional amount of primes clustering. We chose the simplest one, two primes in a pair. It could have been four, six or any even number of primes pairing. Any even amount of primes added will yield an even number. Of course the adjustment needed to be made regarding to the division, so we can reach the average value. ∑ 𝑃𝑖 𝑁 𝑖=1 N = 𝑆𝐴𝑣𝑒𝑟𝑎𝑔𝑒 (2.14) (2.14) If we had four primes pairing, divide by four, six primes divide by six, to reach the average. Of course the average must be two and three divisible, so it could get harder and less likely to find higher numbers of primes pairing which satisfying the condition. It will be impossible to reach the smallest sum in the series with a hundred primes pairing. So for the beginning of the series there could be a limitation. 82 Manor O The Lagrangian Variation ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 − 𝜕2𝑔′ ∂t2 = 0 ℒ= 𝑇−𝑉 (1) How can a representation of the Lagrangian be made on the varying Lorentz manifold, is the main question one will analyze in this short assay. Of course, the real answer to that question is one does not know. However, educated guess will be made. The kinetic term could be associated to the outward matric expansion, due to (𝜕𝑔/𝜕𝑡) term, which is synonymous with energy in physical theories. (𝜕𝑔 𝜕𝑡) = (𝜕𝐸 𝜕𝑡) = 𝜕2𝐸′ ∂t2 That is the kinetic term. A Ricci Tensor overtime, yielding an energy expansion outward causing a matric acceleration on the object generating the energy. That is the main equation that was derived by putting an Einstein manifold in Euler Lagrange Equation. Now, what is the potential energy in the 8T framework? In physical theories, the potential is associated with the mass, which is certain feature of the object itself. In the 8T we do not have any objects, the objects are manifestations of discretizing and partitioning the term 𝛿𝑔→ ∑ 𝛿𝑔𝑖 𝐾 𝑖=1 = 0 in equation (1) to vanish into fermions. How can we translate that into a potential term? [∂ℒ ∂s ∂s ∂M ∂M ∂g] ∂g ∂t δg −[∂ℒ ∂s ∂s′ ∂M ∂M ∂g′]∂2g′ ∂t2 δg′ = 0 (1) (1) Since we would like to measure how much arbitrary variations the object we measure contain, and those arbitrary variations are two and three divisible to vanish into matter, two distinct elements which created threefold combinations, to get a measure of the amount of arbitrary variations, the action needed is the following Transformation: Since we would like to measure how much arbitrary variations the object we measure contain, and those arbitrary variations are two and three divisible to vanish into matter, two distinct elements which created threefold combinations, to get a measure of the amount of arbitrary variations, the action needed is the following Transformation: ∑𝛿𝑔𝑖 𝐾 𝑖=1 = 0 →∑|𝛿𝑔𝑖 𝐾 𝑖=1 | = 𝒱8𝑇 (2.12) The idea, whether correct or not, was to take the absolute number of varying elements participating in the construction of the object. The Lagrangian Variation ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 − 𝜕2𝑔′ ∂t2 = 0 ℒ= 𝑇−𝑉 The Lagrangian Variation ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) 𝜕𝑔 𝜕𝑡= 0 − 𝜕2𝑔′ ∂t2 = 0 ℒ= 𝑇−𝑉 Fermionic & Bosonic Propagations [∂L ∂s ∂s ∂M ∂M ∂g] ∂g ∂t δg −[∂L ∂s′ ∂s′ ∂M ∂M ∂g′] ∂2g′ ∂t2 δg′ = 0 (1) ∑δgk M k=1 = 0 (2.12) (1) (1) (2.12) We partitioned and discretized into a series of arbitrary variations that vanish into matter. We do not have any data regarding the position on the manifold in which those arbitrary variations appear, nor can we assume they possess momenta, as we invoked stationarity on the Lorenztion. 𝑀0 is the connected manifold. 𝑀= 𝑀0 × 𝑅 𝑀= 𝑀0 × 𝑅 In other words, arbitrary variations, which vanish into matter, can be regarded and described by scalar fields that are real, since they have an even amount of variations. ∑δgk M k=1 ∈ℝ ∑δgk M k=1 ∈ℝ Those arbitrary variations, still a subject to additional variance. Such a variance is either prime or one in our framework. These are the variations associated with bosonic propagation. One associated with the strong and each prime with additional coupling term, weak, electric and so on. Because of their prime number feature, they are not vanishing like a fermion scalar but rather as a vector propagation all across. The propagation is associated with a variation of the 𝛻2 operator to the setting of the stationary manifold. The bosonic ripple field is than described by: ∇2= ∂2Mx ∂2g + ∂2My ∂2g + ∂2Mz ∂2g In other words, it is a vector field propagating all across the matric, due to its prime number feature, for the second element in the coupling constant series and above. Since the bosonic propagation is associated with prime amount of variations, we can associate it to a complex field, which than require a Hermitian conjunction in order to perform measurement upon. In other words, we can associate bosonic fields to complex vector fields. 𝑁𝑉= 2𝑉+ 1; 𝑁𝑉∈ℙ 𝑁𝑉∈ℂ 83 Manor O The Lagrangian Variation The operation was done via the insight we gained in previous paper, two distinct elements which differ in sign, so by eliminating the minus sign we can estimate how many arbitrary variations appear in the cluster. To sum up, The energy, causing outward acceleration minus the total amount of arbitrary variations constructed in the cluster. ℒ= (𝜕𝑔 𝜕𝑡) −∑|𝛿𝑔𝑖 𝐾 𝑖=1 | (3) (3) 84 Manor O The Perfect Symmetry of Hadrons 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5:(840 + (3)) + 7 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5:(840 + (3)) + 7 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5:(840 + (3)) + 7 Now, in the 8T framework, up to this point, we nicknamed the left term of each element in the coupling as a variation cluster, which is divisible by two and three to perfectly vanish into matter. This variation cluster is destabilized by the majestic three, causing a net variation to appear, or in other words, to boson propagation from the fermion. Recently one noticed a very interesting fact, that the left terms of the coupling constant series for the weak interaction are identical to densest packing 𝐷4 highest kissing number that is 24. So all the left coupling terms are actually 4D spheres, leading to a propagation of the electron. That may sound outrageous but not in the 8T, as we only have 4D manifold, three spatial and one temporal. By looking at the coupling constant in that light, we can also regard the hadron as possessing an extreme density, as it has the highest kissing number in 4D, and the electron is not bound to it but revolves around it, as the majestic is a separate term. The following apply to each other term: 24 ∗5 →120 24 ∗5 ∗7 →840 Notice that those numbers are associated with highest kissing numbers in higher dimensions. Notice that those numbers are associated with highest kissing numbers in higher dimensions. 𝐸8 →240 = 120 ∗2 𝑝12 →840 𝐸8 →240 = 120 ∗2 𝑝12 →840 Of course, ignoring the higher dimensions complexity and focusing on the part of the highest kissing numbers, we can reach an insight, those fermionic clusters in each term are most dense, in agreement with what we know about the structure of the fermions, and in particular the hadron. Also, notice that those higher dimensions are scalar four multiples, which as one believes, means that should appear on the manifold eventually. Cluster Decomposition In quantum field theory, one learns that the connected part of the S matrix must vanish. Distinct events do not effect each other. 𝑆𝛽𝛼 𝐶→0 𝑆𝛽𝛼 𝐶→0 What is the equivalent of the cluster decomposition principle on the Lorentz manifold (𝑀, 𝑔𝐸) with signature (3,1), invoked stationary, 𝑀= 𝑀0 × 𝑅, is the subject of this paper. ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) Since the manifold experience arbitrary variations that vanish into matter, all across the matric, the smoothness of the matric must be taken into account. Bosonic propagation described by the delta must cross the metric before reaching a distinct event on the manifold. The result of such a construction would be that only arbitrary variations that vanished relativity closed to each other, will have an effect on each other. Suppose we had two distinct arbitrary variations, that is by discretizing and partitioning the term 𝛿𝑔 in equation (1.11), as was done in previous papers of the 8T, to proof that these are fermions: 𝛿𝑔= ∑𝛿𝑔𝑖 ∞ 𝑖=1 = 0 (2.12) (2.12) We impose two conditions equivalent to the cluster decomposition in QFT. Those conditions are synonymous with saying that distinct events will not affect each other. Consider two arbitrary variations 𝛿𝑔𝑖+ 𝛿𝑔𝑖+1 𝛿𝑔𝑖+ 𝛿𝑔𝑖+1 Suppose those appeared at distinct parts of the matric, 𝑀𝜇 is a four vector isomorphic to the arbitrary variation with the matching index 𝛿𝑔𝑖: 𝑀𝜇→𝑀(𝑥𝑖 , 𝑦𝑖,𝑧𝑖, 𝑡𝑖) 𝑀𝜇→𝑀(𝑥𝑖 , 𝑦𝑖,𝑧𝑖, 𝑡𝑖) 𝛿𝑔𝑖→𝑀𝜇 𝑀𝜇→𝑀(𝑥𝑖 , 𝑦𝑖,𝑧𝑖, 𝑡𝑖) 𝛿𝑔𝑖→𝑀𝜇 Same for the additional variation, 𝛿𝑔𝑖+1, a four vector 𝑀𝜈, the condition than requires that: Same for the additional variation, 𝛿𝑔𝑖+1, a four vector 𝑀𝜈, the condition than requires that: 𝑀𝜇−𝑀𝜈≤ 𝜖 𝜖→0 𝑀𝜇−𝑀𝜈≤ 𝜖 𝜖→0 In other words, two arbitrary variations must appear close to each other on the matric, at very short time interval. That is synonymous with the quantum field theory statement of the connected part of the amplitudes to vanish. The two conditions are synoptic in the four vector. The arbitrary variations should appear close on the matric spatial dimensions and at a short time interval. 85 Manor O The Perfect Symmetry of Hadrons The highest kissing number in 𝐷4 is the base to all other kissing numbers at those higher dimensions. By looking at the coupling constant series, than we can correlate the manifold and validate it has only four dimensions, since all higher terms are the dimension four multiples of the kissing number, 24. And thus there could not be more than four dimensions on our manifold. There are of course other manifolds, which according to the series are four dimensional as well, interacting with our own as given by the main equation of the 8T. But by coupling constant series, it is possible to derive why the manifold has exactly four dimensions, because of the kissing number of the second term and above. In addition, the number 24 is associated with the leech lattice, which has most density within a certain dimensional range, is intimately related to this number. In the 8-Theory however there is no use of any lattices. Rather we use variations. Notice the 24 is perfectly to and three divisible to vanish into matter. There is no additional variation left alone. The hadron is perfectly compact and most dense because of that trait. Than it is destabilized by additional term, the element in which we called the majestic three. The point one was trying to make is that the perfect symmetry of the hadronic structure is preserved along each coupling term, i.e. each interaction. In addition, it is than lessen by the electron, i.e. the additional element in the third coupling term. And either the electron is also the cause of that symmetry break in all other terms or electron analogues field. 24 ∗𝑁𝑉 𝑀𝑂𝐷 (6) = 0; (3.12) 𝑁𝑉= 2𝑉+ 1; 𝑉≥1 𝑁𝑉 ∈ℙ 𝑁𝑉= 2𝑉+ 1; 𝑉≥1 𝑁𝑉 ∈ℙ 86 The Feynman Diagrams 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 (1.1) Examine the term describing the electric coupling. We proved majestic three is the electron in the 8- Theory thesis. Examine the term describing the electric coupling. We proved majestic three is the electron in the 8- Theory thesis. Theory thesis. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ e ↘ ⟶ (𝛾) ⟶ e ↗ (+3) ⟶(𝛾) ⟶ (+3) (+3) ⟶(+5) ⟶ (+3) (+3) + 𝑁𝑉= (+3) + 5 = 8 8 = 𝑒𝑣𝑒𝑛 𝑒𝑣𝑒𝑛= 0 The electron, represented as the majestic three combined with the net variation yielding an even amount of variation that vanish. That is synonymous with saying that the electron has absorbed the photon. The conservation of variation ensure that no electron can disappear from the manifold. However, as the combination of 𝑁𝑉 and the electron, i.e. the three yielding an even, there has to be a vanishing of certain sort into the electron. It is moved into an exited state, vanishing of curvature,(𝛾) = (+5) into the receiving electron, which causes the deflection in trajectory. Using the numerical trait and insight gained by the coupling constant series, by the 8T framework, it is possible to add an additional liar to the Feynman diagrams and interactions among bosons and fermions in what seems as a very simple and elegant manner. What can be derived about the nature of the electron using the coupling constant representation? First of all, it is bounded by the bracket, it cannot escape and behave as the net variation, i.e. the photon. Despite the fact that both elements represented by a prime. Second, the electron is represented as a prime number, three, which cannot vanish into matter, but also cannot propagate as a bosonic fields across the matric its behavior than would propagation across the nuclei, in agreement with current understanding about the probabilistic behavior of that particle. Manor O If it was any other number than 24, than the symmetry of the hadrons was not perfect, as equation (1.2) will not hold. The symmetry is breaking due to an external element added by the higgs field from the second element and above, the majestic three. It is currently unclear whether this element is the same for each of the coupling terms. For the electric, it was proven the electron. However, for the weak interaction term and higher terms it could be an electron analogues particle manifested in the element three as mentioned in the previous paragraph and again, it's so important one wanted to emphasize it here as well. There are two main points two take from this short assay. The first is the perfect symmetry of the hadronic structure due to its numerical features. The second point is that the symmetry is breaking from an external element not from within the hadronic structure, due to the higgs field, inserting the majestic three. 87 87 Manor O The Feynman Diagrams There is no data regarding the current position, momenta, orbitals, no physical data of any sort is manifested in the 8- 88 Manor O theory. An additional way to analyze it is to say that the electron blends in the hadronic cluster, [(24 ∗5) + (3)]. The hadronic cluster is closed and represented in a closed term within the bracket. The summation of the term is perfectly suitable to vanish into matter. theory. An additional way to analyze it is to say that the electron blends in the hadronic cluster, [(24 ∗5) + (3)]. The hadronic cluster is closed and represented in a closed term within the bracket. The summation of the term is perfectly suitable to vanish into matter. Freezing Time ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) The subject of this paper would be the forth term of the main equation. The question in hand would be it's physical meaning, what does it mean ∂g/ ∂t = 0 as we required for a stationary manifold. In previous papers author considered it as a gate to space that is flat, the Ricci flow space which is in between two manifolds and is accessible in extremum energies, high or low, so in that idea the term ∂g/ ∂t = 0 referred to curvature in the context of energy . This term is also considering areas of maximal curvature such as black holes and galaxies, as presented above. If we consider black holes as an area of extremum curvature and correlate it to the term, ∂g/ ∂t = 0 it means that time vanishes in a black hole; it is the same at all for an observer inside a critical range, and only pass for observer outside a critical range. When derived the primorial coupling series in March 21, very soon later the author correlated the direction of the series to the direction of time. That is in page thirty of the 8T thesis. The Axis of Evil The main equations in our new framework: ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂g ∂t = 0 −∂2g′ ∂t2 = 0 ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂g ∂t = 0 −∂2g′ ∂t2 = 0 Describing the Lorentz manifold (𝑀, 𝑔𝐸) with signature (3,1), invoked stationary, 𝑀= 𝑀0 × 𝑅. Equation (1 1) satisfy the Einstein principle of equivalence and expends it to a cause and effect Describing the Lorentz manifold (𝑀, 𝑔𝐸) with signature (3,1), invoked stationary, 𝑀= 𝑀0 × 𝑅. Equation (1.1) satisfy the Einstein principle of equivalence and expends it to a cause and effect l ti hi I ki t ti if ld t f t it ill i ld t d Describing the Lorentz manifold (𝑀, 𝑔𝐸) with signature (3,1), invoked stationary, 𝑀= 𝑀0 × 𝑅. Equation (1.1) satisfy the Einstein principle of equivalence and expends it to a cause and effect relationship. Invoking a stationary manifold, any amount of curvature on it, will yield an outward acceleration of the matric. In that sense, it is different from general relativity, as there is no need to insert the cosmological constant as a separate entity. Using that equation, we built a new way to explain relativity by saying that two distinct observers will cause different accelerations of the matric, and so, by measuring the same object, will reach different times and distances. In our theory, the manifold has a varying matric according to a varying topology. The subtle idea is that the manifold has a compact topological space that is accessible from every point given high enough energy. Such space covers every point in matric space. Such a space is what makes the theory works, it is the space keeping the manifold stationary and with the second condition causing it to accelerate outward. Since there are no coordinate to such space, it is the same everywhere, and since every point in the matric is connected to it, there could be the illusion that each point in space was the point in which something cosmologically significant has accrued at singularity. The Feynman Diagrams 1 > 3 30 > 5 128 > 7 850 … Therefore, in other sense, we can take ratios of net variation and state in the primorial the following term apply: Therefore, in other sense, we can take ratios of net variation and state in the primorial the following term apply: Therefore, in other sense, we can take ratios of net variation and state in the primorial the following term apply: ∂g ∂t ≠0 From gluons, clustering triplets of arbitrary variations we move to heavy weak interaction Bosons with mass due to the additional element inserted, than to photons and so on. The direction of the series is the direction of time, but it does not answer what time really is. Of course that the real answer is that the author does not know. In the 8T thesis, the author considered time as a parameter that is intimately connected to the variation of the net element which create a difference in what there is, different amount of clustering leading to different objects, bigger clusters. In that sense we have ∂g/ ∂t ≠0. Since black holes swallow matter, but also omit radiation, the amount of curvature they contain is not varying and can be described by , ∂g/ ∂t = 0 which also implies that the arrow of time freeze or that time does not pass. ∂g ∂t + δg + (−δg(H)) = 0 Therefore, time, despite being so elusive can be correlated to two different features, the set of prime net variations that differ from one another, in particular the jumps from coupling to coupling. The second is that time is correlated to energies that are not extremum ∂g/ ∂t ≠0 and in extremum energies , ∂g/ ∂t = 0 time does not pass, or at least time does not seem as passing. An observer generating energies at the level of , ∂g/ ∂t = 0 in a mathematical sense is to create a maximal curve, the maximal curve does not vary with time, as it is a maximal point and so to an outside observer time is standing still. We can make a prediction: (1) An observer able to generate energies at the magnitude ∂g/ ∂t = 0 will freeze time, at the area of generation for an unknown radius. 89 Manor O Reasoning Bosonic Probabilities 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1):(24 + (3)) + 3:(120 + (3)) + 5: (840 + (3)) + 7 … (1.1) (1.2) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1):(24 + (3)) + 3:(120 + (3)) + 5: (840 + (3)) + 7 … (1.2) 8 + (1):(24 + (3)) + 3:(120 + (3)) + 5: (840 + (3)) + 7 … 8 + (1):(24 + (3)) + 3:(120 + (3)) + 5: (840 + (3)) + 7 … Examine the term describing the electric coupling. We proved majestic three is the electron in the 8- Theory thesis. The photon is represented as net variation, which is unbound. It is free to propagate all across the manifold. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ Suppose such a photon just propagated from the electron. i.e. the majestic three. The meaning of such an occurrence is that there is a net curvature that is unbound on the manifold. Such curvature will effect all other potential propagation toward itself. It will create a pull effect on other potential boson propagating from fermions. That is in agreement with what we know about the commutation relation of bosons, and the fact that the probability to find a boson increase if there is already a boson in a certain position of the matric. The innovative part of this paper and the main point to take is the new setting, a in which a photon itself is a net curvature causing other curvature propagating at later time to converge to its position. When analyzed via the new framework it than becomes quite easy to understand what is going on at that fundamental level. ∑𝛾𝑖 𝑁 𝑖=1 > 0 (3.13) ∑γi N i=1 = ∑δgi N i=1 > 0 (3.13. B) The point of view presented is not presented in quantum field theory framework, the methods they use to describe the commutation and anti-commutation is VOA, vertex of algebra, and there is simply no way to imagine or to grasp the intuitive reason for the such a behavior. By using an approach combining manifolds and variation, i.e. Euler Lagrange, it is possible to explain the behavior of bosons in an intuitive and simpler fashion. The Axis of Evil Not the whole topological space is satisfying the condition, 𝜕𝑔/𝜕𝑡= 0 there are arbitrary variations in that space which vanish into matter on the matric, we have proved it in previous papers. Each net variation than is isomorphic to the prime numbers or to the number one, and thus we were able to prove the coupling constant series, presented in equation(1.1) and (1.2). The point of this short assay is the fact that there is an underlining space, which is invariant to matric coordinate and covers the entire matric. We know it covers the entire matric as the manifold is connected to the topological space but no spatial coordinates are given in equation (1.1). The topological space is than invariant, and the equation is really a right to left chain of the order. Notice that the chain in equation (3.12) is exactly describing the order in which things are happening in cosmological scales. coordinates are given in equation (1.1). The topological space is than invariant, and the equation is really a right to left chain of the order. Notice that the chain in equation (3.12) is exactly describing the order in which things are happening in cosmological scales. ∂L ∂s ⟵∂s ∂M ⟵∂M ∂g ⟵∂g ∂t (3.12) (3.12) 90 Manor O The Conservation of Variation 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … (1.1) (1.1) (1.2) 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … 8 + (1) + 3 −(1𝑣) ∶ (24 + (3)) + 3: (120 + (3)) + 3 8 + (1) + 3 −(1𝑣) ∶ (24 + (3)) + 3: (120 + (3)) + 3 In the paper about the interactions dynamic nature, we varied the first and the third interactions, i.e. the strong and the electric, in their 𝑁𝑉 element, so all the net variations will align on the same integer. The important point, which was not mentioned, is that the net variations varying their position among the terms are confined within the manifold. In other words, it is conserved. That is also the case with the gravitational coupling, which as far as the 8T can predict, is a result of two net variations added to the original net variation. The data regarding the nature of gravity came from the second representation, i.e. the spin representation of the coupling constant equation. [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 = [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 can put the conservation law in rigor and construct an appropriate theorem: We can put the conservation law in rigor and construct an appropriate theorem: Theorem (1.0) – The sum of net variations on all the coupling elements cannot escape the manifold. Theorem (1.0) – The sum of net variations on all the coupling elements cannot escape the manifold. Theorem (1.1): The sum of all net variations increase with time. Theorem (1.0) – The sum of net variations on all the coupling elements cannot escape the manifold. Theorem (1.1): The sum of all net variations increase with time. Theorem (1.1): The sum of all net variations increase with time. Reasoning Bosonic Probabilities It is possible to state that each boson is creating a "gravitational effect", i.e. curvature on the manifold, and thus increase the probability of arrival for other bosons to itself. 91 Manor O The Conservation of Variation ∮ (𝑑𝑀)(𝑀0 × 𝑅)( 𝑡=𝑍 𝑡=0 ∑𝑁𝑉) ∞ 𝑉=0 ∈𝑀 (3.16) Z → ∞ If one constructed properly, one summation of the net variation to each V across the entire manifold matric, over time, must belong to the manifold itself and cannot decrease. It could be related to the second rule of thermodynamic, the entropy rise alongside the net variations overtime. Of course, the total variations grow much faster, but that was not the subject of this paper. The point was to emphasize that the sum of net variation is bounded to the manifold, despite the fact it grows with time. 92 Manor O Bosonic Strings - Cyclic Groups Bosonic Strings - Cyclic Groups (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … 8 + (1) [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … 8 + (1) [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 8 + (1) [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 Cyclic groups in mathematics are represented by the following, if a set of elements is generated by one single element, than we have a cyclic group. Since all the bosonic fields or net variations in the 8T are generated by the same element, i.e. the majestic three, than there is in this framework an infinite cyclic group. Define the majestic three as the generator: (3) → ℳ ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(5)ℳ… . } (3.17) (3) → ℳ ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(5)ℳ… . } (3.17) (3.17) By representing the propagation in such fashion, we can state that since the bosons are propagations are part of an infinite cyclic group, the sub elements of that cyclic group are cycles themselves. We have proven the representation of the coupling constant series in the thesis: By representing the propagation in such fashion, we can state that since the bosons are propagations are part of an infinite cyclic group, the sub elements of that cyclic group are cycles themselves. We have proven the representation of the coupling constant series in the thesis: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(γ)ℳ… . Bosonic Strings - Cyclic Groups } Therefore, that is a proof that bosonic net variations are cycles, or in physical theories, bosonic particles are in fact closed strings. That is because they are generated by the same element. Therefore, that is a proof that bosonic net variations are cycles, or in physical theories, bosonic particles are in fact closed strings. That is because they are generated by the same element. 93 Manor O Manor O Curvature Absorptions [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ (3) → ℳ ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(5)ℳ… . } ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(γ)ℳ… . } All of this was covered in previous papers. The majestic three is a generator of a cyclic group of bosonic net curvature propagations, isomorphic to the primes or one. We made a Feynman diagram using the new framework: e ↘ ⟶ (𝛾) ⟶ e ↗ (+3) ⟶(+5) ⟶ (+3) (+3) + 𝑁𝑉= (+3) + 5 = 8 →0 (+3) + 𝑁𝑉= (+3) + 5 = 8 →0 The point of this paper is that in order to understand how the manifold vary, there are to be a summation of all curvature absorptions and emissions. As an electron absorb a photon, the manifold gets more flat, as 𝑁2 = +(5) just vanished into the electron and vice versa. By looking at clusters of photons in unit matric, it is also possible to estimate how much curvature exits on the manifold. As bosons are net variations unbound, it was derived that preciously for that reason the probability of boson arrival after a boson is propagated. ∑𝛾𝑖 𝑁 𝑖=1 > 0 (3.13) ∑γi N i=1 = ∑δgi N i=1 > 0 (3.13. B) The point is, we can use space- time summation and in particular, the distribution of fermions to bosons to estimate how curved the matric, or how it varies over time. It is vividly clear that a real world estimation is at the verge of impossible, but a rough evaluation is always within reach. The point is, we can use space- time summation and in particular, the distribution of fermions to bosons to estimate how curved the matric, or how it varies over time. It is vividly clear that a real world estimation is at the verge of impossible, but a rough evaluation is always within reach. ∑𝛾𝑖 𝑁 𝑖=1 → 𝓅 𝜕𝓅 𝜕𝑡−𝜕𝑀 𝜕𝑔= 0 (3.18) 94 Manor O Light is Bending Space-Time ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (2.12) ∑𝛾𝑖 𝑀 𝑖=1 > 0 (3.13) ∑γi M i=1 = ∑δgi M i=1 > 0 (3.13. Curvature Absorptions B) (2.12) (3.13) By putting the Lorentz manifold in Euler- LaGrange framework and allowing arbitrary variations to appear, in which we require to vanish, in the 8T we discretized and partitioned the term (2.12) and were able to prove that arbitrary variations of the manifold vanish into matter. Each net variation or net curvature is isomorphic to a bosonic field propagation. In particular the boson associated with photon propagation is 𝑁𝑉= +(5). Such propagation is than yielding a positive summation, i.e. a positive curvature by (1.15), so fermion clusters are flat according to the 8T framework, but bosonic propagations are curvature on the manifold. The weird and unexpected result is that bosonic fields are deflecting fermion clusters and not the opposite as believed by GR. It is an unexpected result, but up until recently we thought there are only four forces, and such thought lead to thinking that physics can be unified. 95 Manor O ⱷ𝐬𝟏+ ⱷ𝐬𝟏+ ⱷ𝐬𝟐+ ⱷ𝐬𝟐= 0 Proof – The Riemann Hypothesis Define a Lorentz manifold 𝐬= (𝐌, 𝐠) Use it to assemble a Lagrangian and require it to be stationary: L = (s, s′, t) 𝛛𝐋 𝛛𝐬− 𝛛𝐋 𝛛𝐬′ ∗ 𝐝 𝐝𝐭= 𝟎 Allow arbitrary variations of the manifold. Ensure it will vanish: Allow arbitrary variations of the manifold. Ensure it will vanish: ⱷ𝐬= 0 ⱷ𝐬= 0 ⱷ𝐬= 0 Turn it to a series of arbitrary variations: Turn it to a series of arbitrary variations: ⱷ𝐬= ⱷ𝐬𝟏+ ⱷ𝐬𝟐+ ⱷ𝐬3… If there are only four elements in the series, and we require them all to vanish, than we can allocate two pluses and two minuses: ⱷ𝐬𝟏+ ⱷ𝐬𝟑> 0 ⱷ𝐬𝟐+ ⱷ𝐬𝟒< 0 If If ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐+ ⱷ𝐬𝟒 ≠0 ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐+ ⱷ𝐬𝟒 ≠0 ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐+ ⱷ𝐬𝟒 ≠0 Than the overall series cannot vanish, by that logic we need equal amounts of plus and minuses. The overall amount must be even and summed as zero. Than the overall series cannot vanish, by that logic we need equal amounts of plus and minuses. The overall amount must be even and summed as zero. Suppose that we had three distinct elements, two pluses and minus: ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐> 0 ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐> 0 or or ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐< 0 ⱷ𝐬𝟏+ ⱷ𝐬𝟑+ ⱷ𝐬𝟐< 0 Demanding the series to vanish this forbid this result, and so there could not be three distinct elements in the series, else the overall series will not vanish. As a result of those sceneries, we require the series to have an even amount of variation elements, manifesting as two distinct elements in the series, which differ in sign. If we allow those sub elements in the series to vary as well, and by the above reasoning, there are only two elements in the series, they are varying in a discrete way, or forming a group. Let it be only four elements in the series and one of the pluses just changed its nature 𝐎: ⱷ𝐬𝟏→ⱷ𝐬𝟐 ⱷ𝐬𝟏+ ⱷ𝐬𝟏+ ⱷ𝐬𝟐+ ⱷ𝐬𝟐= 0 To: 96 V: set  ring Analyze any multiplication or addition combination of Pẖ on the ring. Let the ring exist on a Lorentz manifold, a topological space. Y∶ ⱷ𝐬𝟐→ⱷ𝐬𝟏 Y∶ ⱷ𝐬𝟐→ⱷ𝐬𝟏 Therefore, to bring an element to itself given only two varying elements in the series we need two distinct maps, which attach a varying element to itself, by a threefold combination. ⱷ𝐬𝟏(O) ⱷ𝐬𝟐(Y) ⱷ𝐬𝟏 For example. Even though the sub elements in the series are varying, the overall series can vanish. Now, count all the ways of possible combinations of those elements. We are going to analyze by the integral signs. Since it is a group, there is a natural map, which change an element to itself. One built his analysis firstly on those natural maps. The first two combinations are by the natural maps and one used them to build the other combinations. Overall, there are eight such combinations and additional one arrow combination, which yield (333) here is how one built it, starting from those two natural maps. (Arrows to variations, colors to pairings): 2𝟏𝟏 −−− 212 𝟏𝟐𝟐 −−−− 121 221 −−−−− −−− 112 222 −−−− 111 333 2𝟏𝟏 −−− 212 𝟏𝟐𝟐 −−−− 121 221 −−−−− −−− 112 222 −−−− 111 333 333 Now that we have a series of 2N elements, varying to one another and forming threefold combinations, which we require to vanish at end, we can set the stage for a proof of primes. Define: Pᵐ as the set of {2 , 3} as "minimal primes". In addition, all the other primes to be in a set of Pẖ as meant "prime higher". Define Pẖ = {2n + 1} not divisible by Pᵐ as "prime higher" set – 2n taken as amount of Lorentz manifold arbitrary variations. Define Pẖ = {2n + 1} not divisible by Pᵐ as "prime higher" set – 2n taken as amount of Lorentz manifold arbitrary variations. {2n + 1} as an odd amount of variations not divisible by minimal primes {2n + 1} as an odd amount of variations not divisible by minimal primes Pṱ = Pẖ + Pᵐ ; to be the set of all primes Pṱ = Pẖ + Pᵐ ; to be the set of all primes Pṱ = Pẖ + Pᵐ ; to be the set of all primes Manor O ⱷ𝐬𝟏+ ⱷ𝐬𝟐+ ⱷ𝐬𝟐+ ⱷ𝐬𝟐 ≠0 ⱷ𝐬𝟏+ ⱷ𝐬𝟐+ ⱷ𝐬𝟐+ ⱷ𝐬𝟐 ≠0 There must be a way to bring it back to where it was, so the overall series can vanish, it takes another map, on the varying element to bring it back to where it was. Manor O 2 [T ((n1 n2 … )) + (n1 + n2 + n3 … ) + 1 2] = 2([T ((n1 n2 … )) + N(s) + 1/2] N(s) = (n1 + n2 + n3 … ) = 0 As sums of even amounts of arbitrary variations vanish. Since all the elements are two multiples, they all vanish. Final form: 2 ([T (n1 n2 … )] + 1 2) Add any infinite even series of distinct higher primes to obtain Add any infinite even series of distinct higher primes to obtain nfinite even series of distinct higher primes to obtain (2n1 + 1) + (2n2 + 1) + (2n3 + 1) … = [2(n1 + n2 … ) + even] = [2(n1 + n2… )] as even = 0. Prime cannot form, as even amount of variations vanish exactly to zero. That is the reason the paper begins with deriving fermions, their anti-commutation relation. Even amount of distinct higher primes added will never form a prime. Add any infinite odd series of distinct higher primes to obtain (2n1 + 1) + (2n2 + 1) + (2n3 + 1) … = [2(n1 + n2 … ) + odd] = [2(n1 + n2 … ) + (even + 1)] (10) However, even amounts of arbitrary variations vanish: However, even amounts of arbitrary variations vanish: However, even amounts of arbitrary variations vanish: However, even amounts of arbitrary variations vanish: even = 0 [2(n1 + n2 … ) + 1] or: 2[n1 + n2 … + 1/2] (11) (11) 𝐌𝐮𝐥𝐭𝐢𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧: Define T to be a number aspiring infinity: T ∞ Multiply an even or odd series aspiring infinity of distinct higher primes to obtain: [(2n1 + 1)(2n2 + 1)(2n3 + 1). . . (2n + 1]) = [(2n1 + 1)(2n2 + 1)(2n3 + 1). . . (2n + 1]) = 97 𝐂𝐚𝐭𝐞𝐠𝐨𝐫𝐲 𝐭𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐨𝐧𝐬 Define a functor on "Primes higher" ring G: ring  group All "primes higher" are forming a closed non-abelian group with 1/2 as generator. The condition to group forming is to have an odd amount of primes under addition and eliminating even amounts of arbitrary variations taken as an axiom. Define additional functor G′: group  set Add the sets: Add the sets: Add the sets: Pẖ + Pᵐ = Pṱ ; 98 Manor O Define a functor on Pṱ: G′′: set  group All primes are forming a non-abelian group of generator 1/2. Minimal primes are part of the group by nature of the proof, defined technically to be prime. Primes are forming a non-abelian group under addition and multiplication. The condition to satisfy is to have an odd amount of primes under operation of addition. No matter how far into infinity we will go, the framework of vanishing of even amount of variations will ensure that all primes take the same form – aligned on 1 2. Setting the stage and examining primes not as numbers, but rather as arbitrary variations of a manifold, which vanish in pairs of even variations, we are able to show primes to form a non-abelian closed group under 2(n+1/2). Final functor on the total group of primes: Riemann: Group  ring All primes are forming an infinite ring on the critical line of 1/2 and only there. 𝑬𝒏𝒅 𝒐𝒇 𝒑𝒓𝒐𝒐𝒇. All primes are forming an infinite ring on the critical line of 1/2 and only there. All primes are forming an infinite ring on the critical line of 1/2 and only there. 𝑬𝒏𝒅𝒐𝒇𝒑𝒓𝒐𝒐𝒇 𝑬𝒏𝒅 𝒐𝒇 𝒑𝒓𝒐𝒐𝒇. 99 Manor O Visualization - Photon Emission [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ (𝑒) = (3) → ℳ ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(5)ℳ… . } (4.1) ℬ= {𝑁1 = (+3)ℳ, 𝑁2 = +(γ)ℳ… . } (4.1) Equation (4) to describe the process of emission due to the invariant three, proven the electron, assumed the electron for each higher term in the coupling series. equation (4.1) describe the invariant three as the generator of a cyclic group, meaning all bosonic propagations are sub elements of that group and so we prove they are closed cycles themselves. Therefore, we can draw the interaction between two electrons and a photon emission in the following way: (𝑒) (𝑒) As was proven, they cannot move at the same orientation of the distortion due to their prime number feature, combined together there will be a vanishing and so the coupling series than would not make sense. The end conclusion would than imply that the boson propagated from nowhere which is impossible. (𝑒) (𝑒) (𝑒) As was proven, they cannot move at the same orientation of the distortion due to their prime number feature, combined together there will be a vanishing and so the coupling series than would not make sense. The end conclusion would than imply that the boson propagated from nowhere which is impossible. 100 Manor O Interference [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ The image above represent a net curvature on the Lorentz manifold, in that specific case, it’s the photon associated with 𝑁𝑉= (+5) net variations, and total 128 variations. Suppose that we perform the two slits experiment and open an additional route for net curvature. this is the visualization of what could happen according to our new theory: Manor O Interference The image above represent a net curvature on the Lorentz manifold, in that specific case, it’s the photon associated with 𝑁𝑉= (+5) net variations, and total 128 variations. Suppose that we perform the two slits experiment and open an additional route for net curvature. this is the visualization of what could happen according to our new theory: The image above represent a net curvature on the Lorentz manifold, in that specific case, it’s the photon associated with 𝑁𝑉= (+5) net variations, and total 128 variations. Suppose that we perform the two slits experiment and open an additional route for net curvature. this is the visualization of what could happen according to our new theory: The image above represent a net curvature on the Lorentz manifold, in that specific case, it’s the photon associated with 𝑁𝑉= (+5) net variations, and total 128 variations. Suppose that we perform the two slits experiment and open an additional route for net curvature. this is the visualization of what could happen according to our new theory: There are two ways to explain. The first is to say that two opposite but similar in magnitude curvature occupying the same space will have a segment of mutual cancelation. If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. There are two ways to explain. The first is to say that two opposite but similar in magnitude curvature occupying the same space will have a segment of mutual cancelation. If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. There are two ways to explain. The first is to say that two opposite but similar in magnitude curvature occupying the same space will have a segment of mutual cancelation. If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. ≬: 𝐴→ 𝐵 ≬:𝐴′ →𝐵 ≬:𝐴′ →𝐵 Interference will accrue at the manifold segment that is mutual to both starting point. Interference Define the interference operator: ≉:𝐴∩𝐴′ (3.18) (3.18) 101 Manor O Quark Visualization Imagine constant variation so the overall construction is curvature varying, according to the combination where will be a pairing according the graph presented in the 8 theory thesis or the group suggested by the particle physicist Gell Mann. Each arrow in the visual is a representation of the gluon, or the first element in the coupling constant primorial function. δg1 δg2 δg1 Quark Visualization δg1 Imagine constant variation so the overall construction is curvature varying, according to the combination where will be a pairing according the graph presented in the 8 theory thesis or the group suggested by the particle physicist Gell Mann. Each arrow in the visual is a representation of the gluon, or the first element in the coupling constant primorial function. Imagine constant variation so the overall construction is curvature varying, according to the combination where will be a pairing according the graph presented in the 8 theory thesis or the group suggested by the particle physicist Gell Mann. Each arrow in the visual is a representation of the gluon, or the first element in the coupling constant primorial function. 102 Manor O Manor O Manor O Visualization of the Multiverse ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) Visualization of the Multiverse Visualization of the Multiverse (2.1. A) 103 Manor O Strong Interaction – The Electron g 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ 𝐹𝑉=0 = 8 + (1) (1.1) The main argument of this short assay is that it is possible to regard each higher coupling terms as the strong interaction being destabilized in ever-growing fermion formations. It’s the electron that has so much significance in the coupling constants series. Back in the day, when author derived the coupling series, in the thesis he believed that each term would have unique destabilizer, but now it seems very clear that such an assumption is quite likely wrong and eventually will lead to complexity that is not needed. Another way to state it is that three is isomorphic to itself. What is varying is the size of the fermionic cluster and the magnitude of the net curvature. The shift in understanding manifested itself in toward the end of the thesis but still it is important to clarify to avoid confusion among readers. It is also possible to represent the coupling, as you already know, in the form of spin representations by setting it on the prime critical strip. [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 To solidify the statements made in previous papers, the variance in that representation is the fermionic clusters, represented in the right of each term, and the net variation, or net curvature that is prime or one. The conclusion if one is correct is the electron is destabilizing larger and larger fermionic cluster yielding an infinite succession of net curvature on the manifold, which causes the endless process of clustering. One prefer that version, as it is simpler than to assume that each term would have a unique destabilizer. As the fermionic cluster gets much more massive in rate, the net curvature than becomes less significant, preciously the idea behind the principle of least variation. 104 Virtual Curvatures In calculus of variations, we have the procedure of the following for the vanishing of virtual displacements within a massive cluster. Such a procedure makes description of motion rather simple, as we do not need to describe the innate motion of a static body. Similar in a sense to the Laplace operator. ∑𝐹𝑖𝑑𝑟𝑖 𝑁 𝑖=1 = 0 (3.19) (3.19) What would be the equivalent statement in the 8-theory? As we do not use force in the innate description of the theory, all we have is net curvature, 𝑁𝑉, on the Lorentz manifold, which was invoked stationary by the Lagrangian operator. We also did not use radius per se, it is different from the Riemann line element in which we associate curvature. One will suggest the following analogue for the equation (3.19): ∑𝛿𝑔𝑖𝜕𝐿𝑖= 0 𝑁 𝑖=1 (3.19. 𝐴) The sum of all arbitrary variations per varying manifold unit length is summed as zero. As we say variations, we mean curvature, so the sum of arbitrary curvatures is taken to zero. We can similarly use that construction in the same manner and for the same purposes used in calculus of variations, to avoid describing the inner motions of a static body. 105 Manor O Curvature Knots Curvature Knots 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ (1.1) (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ Suppose that instead of a prime number as in equations (1.2) describing the weak and the electric, we would have a number that is odd, which could be a composite of odd number primes. Define the odd number function: Φ𝑛= 2𝑛+ 1 (3.2) 𝑛∈ℝ; Φ𝑛∉ℙ (3.2) 𝑛∈ℝ; Φ𝑛∉ℙ ℙ⟶𝑠𝑒𝑡 𝑜𝑓 𝑝𝑟𝑖𝑚𝑒𝑠 So Φ𝑛 is a series of odd numbers that replace only the external 𝑁𝑉 in the coupling constant series. The new series is now described by: So Φ𝑛 is a series of odd numbers that replace only the external 𝑁𝑉 in the coupling constant series. The new eries is now described by: [2N1 + 1 2] + Φ𝑛1 (3.21) [2N2 + 1 2] + Φ𝑛2 (3.22) (3.22) Since Φ𝑛 is not a prime it cannot act as a bosonic ripple field on the matric tensor. Since it is on an even number, devisor of modulo six it cannot vanish into matter. It is a composite of prime, or a composite of net curvature, and because it is a composite, which is stable on the matric tensor, we will have a curvature which is time- invariant, not matter like nor boson like. In other words, a knot. The main point is if one is correct, a knot is composite of net curvature, associated with odd numbers. That is an expansion of the 8T, which did not analyze the odd numbers, but rather referred only to prime numbers and even numbers, isomorphic to primes and evens respectively. Since odds are not on the prime critical line the expressions on terms (2.1) and (2.11) would not have spin one, but neither spin one-half, that is to say they cannot be associated with a particle of any sort. According to the size of the odd numbers we should be able to observe those knots on the matric tensor. Below an example to such knot. 106 Manor O Matric Tensor Fluctuations and the Birth of Universes ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) ∂g ∂t = 0 , − ∂2g′ ∂t2 = 0 𝜕ℒ 𝜕𝑆𝑀 − 𝜕ℒ 𝜕𝑛= 0 (1.53) (1) (1.53) ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (2.1. A) The matric tensor experience arbitrary variations that vanish into matter. We describe the process of arbitrary variations vanishing into matter in the thesis, by the variation of the Dirac Delta function. The matric tensor experience arbitrary variations that vanish into matter. We describe the process of arbitrary variations vanishing into matter in the thesis, by the variation of the Dirac Delta function. 𝛿𝑔≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) δg1 + δg2 … = ∑𝛿𝑔𝑖 𝑁 𝑖=1 ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 𝛿𝑔≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) δg1 + δg2 … = ∑𝛿𝑔𝑖 𝑁 𝑖=1 ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 There is always a chance net curvature will appear at later continuation of time. That is bosonic fields given by the primorial coupling series: There is always a chance net curvature will appear at later continuation of time. That is bosonic fields given by the primorial coupling series: 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) 𝛿𝑔≠0 𝑎𝑡 𝑡2 = 𝑄(𝑡+ ∆𝑡+ ∆𝑡) ∆𝑡→0 ∆𝑡→0 Moreover, the amount of net curvature is either prime or one: Moreover, the amount of net curvature is either prime or one: 𝛿𝑔= 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 107 Manor O Manor O Manor O Now that we presented the 8T foundation, we can visualize the birth of new universe. By assuming a segment of the matric tensor to experience a certain amount of curvature it could lead to a departing from the original manifold. One can try to put it in visual means. This idea is synonymous with the vacuum fluctuations in QFT. The main point of this short assay is that the net curvature led to a departing from the original matric tensor to a new entity. The outer shell of this new manifold will accelerate due to other manifolds wrapping around it given by equation (2). That is in agreement with QFT prediction of infinite universes. The entire evolution of the universes from singularity to complete flatness in given by the main equation (1). The stage and actual flattening moment in different in each manifold. That is an elegant way to eliminate the question – why 13.7B years? The main point of this short assay is that the net curvature led to a departing from the original matric tensor to a new entity. The outer shell of this new manifold will accelerate due to other manifolds wrapping around it given by equation (2). That is in agreement with QFT prediction of infinite universes. The entire evolution of the universes from singularity to complete flatness in given by the main equation (1). The stage and actual flattening moment in different in each manifold. That is an elegant way to eliminate the question – why 13.7B years? 108 Manor O Manor O EMT Symmetry suppose that the electron has absorbed a discrete amount of net curvature, its energy increased. Since we are familiar with the equivalence relation between mass and energy, as presented by Albert Einstein, energy increase is synonymous with mass increase. Suppose its mass increased in such way that now instead of the electron, it is a Muon or a Tau. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝜇−)] + γ In addition, the Tau: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝜏−)] + γ Mass is curvature converging inward, so if the electron has absorbed net curvature its mass increased. That is supported by the Quark masses series of the 8T. Those higher generation particle according to coupling series are representing a symmetry. The magnitude will stay as it is, invariantly of the actual particle, we can call it the EMT symmetry, first letter of each generation particle name. What will vary as a result of the particle varied is energy of the photon emitted. The heavier the particle, the more energy the emitted net curvature should contain. That is again implied by equivalence between mass and energy. Such a construction allow us to make two predictions regarding the energy of the net curvature, i.e. the photon in the case of the third coupling term: (1) The Energy of the photon emitted is proportional to Lepton generation. (2) The coupling constants series is invariant to generation – what is varying is the energy of the net curvature. (2) The coupling constants series is invariant to generation – what is varying is the energy of the net curvature. 109 The Coupling Constants Series and Probability First, we can represent the original equation, which regard Bosonic fields to be net curvature on the varying Lorentz manifold. Those Bosons are isomorphic to prime numbers or one - ℙ∪+(1), and propagating from matter clusters destabilized by the majestic three, which is the electron, from the second element and above. Associate a probability of certain sort to the first element, 𝑁𝑉= (+3). the majestic three and the invariant multiplier eight will be presented as a constants, ℳ, Κ. 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ (1.1) (1.2) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ Correlating the net curvature element to a certain probability. Correlating the net curvature element to a certain probability. 𝑁𝑉= (+3) ⟶𝑃(𝐴) 𝑃(𝐴) < 1 Now, for simplicity sake assume that the probability is the same for all each higher element in the series. As we do not really know what is the probability of such an event, it is possible to assume that is the case. We can represent the equation in means of probability. 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 + ℳ) + 𝑃(𝐴) (3.3) 𝐴∈ℙ; For each higher term than there is a dependence, the next element in the series can only arise after a previous probability was satisfied, as it is a series. So the longer we develop, the smaller the probability to detect the boson as it is depended upon longer chain of events, with probability smaller than one. We can represent it in a simpler fashion by ignoring the constants: 110 Manor O 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 ) (3.33𝐴) Let 𝐴→ ∞ 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 ) ⟶0 (3.1) Manor O 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 ) (3.33𝐴) (3.33𝐴) Let 𝐴→ ∞ Let 𝐴→ ∞ 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 ) ⟶0 (3.1) (3.1) Such a representation of the primorial series than makes it easier to understand how hard it will be to detect those higher term coupling bosons, and why they have not found up to this day. The Coupling Constants Series and Probability However, it scientists have detected gravitational waves they should be able to detect the next elements in the coupling series, as they are about seven, and seventy two weaker than the electric. Therefore, despite each term is an individual element which have a unique boson isomorphic to ℙ for the second and above, there is an implicit dependence given by the fact that is a mathematical series and each even sum is a scalar multiple of the next prime. If we represent the series from an angle of the arrow of time, the higher the coupling term, the more time it will need to develop it. Weakest interactions appear than after longer periods of time, and the strongest most common ones appear at the beginning. We can make a prediction: (1) The probability of locating the boson of the third term is significantly higher than the sixth term. Manifold Jumps and Pharrell Transport ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 (1) ∂ℓ 𝜕𝑠1 − ∂ℓ 𝜕𝑠2 = 0 ∑∂ℓ 𝜕𝑠𝑚 𝐾/2 𝑚=1 𝜕𝑠𝑚 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−∑∂ℓ 𝜕𝑠𝑛 𝐾/2 𝑛=1 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2.1. A) ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 (1) (1) (2.1. A) The implicit assumption of equation of (2) is that in order of the universe packet to flatten each other the curvatures on the manifolds must be interfacing. That is synonymous with stating that the universe packet must have topologically invariant manifolds, or manifolds in which the extremum curvature distribution is identical on the matric tensor. That is because in the manifolds flatten each other due to the interaction of those areas, if they are not interacting the flattening, i.e. dark energy would not be correct. It is possible to prove that if there were only two manifolds, which are not interacting with each other via those areas, equation (2.1A) will not be correct; the universe would not be flat as we measure it today. The requirement of the universe packet than imposes a symmetry in a sense that only topologically invariant manifolds are "allowed" on the packet. We do not know whether it is actually the case but so it seems by equation (2.1A) and the "thought experiment" of only two manifolds interacting in the packet, assumed different topology. Another point to mention is same topology does mean same matter distribution on each manifold. Distinct manifold can have a dust of gas of certain curvature, which is equivalent to the mass of a certain galaxy on another manifold. Those universes differ from each other in a distance measure which is not known, can could vary as other topologically invariant manifolds enter the packet. Between each manifold pair there is the same base space, Ricci flow, given by the fourth term of (2). everything written up to this point was already covered in 8T previous papers. From here we a have a completely new paper. which is equivalent to the mass of a certain galaxy on another manifold. Those universes differ from each other in a distance measure which is not known, can could vary as other topologically invariant manifolds enter the packet. Asymptotic Freedom Bosons were proven discrete amount of net curvature on the matric tensor, we can represent them by the term in equation (1.3): ∑𝛿𝑔𝑖> 0 𝑀 𝑖=1 ; 𝑀⟶∞ (3.13) ∑𝛿𝑔𝑖 ∈(+1) ∪ℙ 𝑀 𝑖=1 ℙ ⟶𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 ∑𝛿𝑔𝑖> 0 𝑀 𝑖=1 ; 𝑀⟶∞ (3.13) ∑𝛿𝑔𝑖 ∈(+1) ∪ℙ 𝑀 𝑖=1 ℙ ⟶𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 (3.13) Now, we have used the visualization of the sea of gluons on the Quark triplet in the following way. 111 Manor O In the context of asymptotic freedom, when we indulge in high energy collusions, that is synonymous with trying to roll the quark triplet uphill. It is possible to try as the bosons are just net curvature unbound as given by (1), however since each boson is a curvature of certain magnitude it increase the probability of arrival to its position, therefore we have a "sea" of gluons. For example, in the third coupling term presented in equations (3) to (3.1): ∑𝛾𝑖 𝑀 𝑖=1 = ∑𝛿𝑔𝑖 𝑀 𝑖=1 > 0 (3.13.𝐵) [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ Taken from that point of analysis, asymptotic freedom is a result of curvature converging to a point, or the existence of gluons on the quark triplet. If the number of bosons is ever increasing on the quark triplet, so does the overall curvature of the magnitude. To roll a quark uphill an infinite curve is at the verge of impossible. The attempt to roll the quark triplet elements uphill will eventually lead to a the quark reaching the minima, lowest point on the curve. Similar to other physical phenomena aspiring minima. Overall the 8T from birds eye overview, allow us to explain phenomena which is considered "advanced" such as Pauli Principle, asymptotic freedom, Spin, the commuter, the reason for the coupling magnitudes, dark energy and probability of arrival in rather simple and elegant way. All we need is just two equations, (1) and the coupling constants series. 112 Manor O Manor O Manor O Manifold Jumps and Pharrell Transport Between each manifold pair there is the same base space, Ricci flow, given by the fourth term of (2). everything written up to this point was already covered in 8T previous papers. From here we a have a completely new paper. Since the manifolds have the same curvature distribution, they have the same energy given by the term of the Ricci flow, if you can switch from the matric tensor of one manifold to its flow, and the flow is the same for all the manifolds in the packet, than you can jump or get into the matric tensor of another manifold. In other words, the Ricci flow is the kernel of the entire manifold packet. That is by equation (2.1A) and the fact that each manifold, which flatten each other interact by the areas of extremum curvatures 𝜕𝑔/𝜕𝑡= 0. So to switch from manifold to manifold, it will require an immense amount of energy, and such an energy level would lead to a deformation of the matric tensor to the kernel, the Ricci flow, and from the Ricci flow we can reach again the matric tensor of a distinct manifold. 8T than regard the matric tensor of each manifold to be a map to another matric tensor. below are two illustrations, the first is the universe packet, the matric is the dark part and the flow is the white part between each two manifolds. the second illustration is an interaction of areas of extremum curvatures between two distinct manifolds, between each two there is the Ricci flow, which is the same for both. The jump from manifold to manifold can be done via this base space. 113 Manor O ⇋∶𝑀1 ⟶𝑀𝑁 Those universes differ from each other in a distance measure, which is not known. As the illustration above suggest, they are very close. The packet could vary as other topologically invariant manifolds enter the packet, also known as cosmological singularity. The main point of this short assay between each manifold pair, and actually all the manifolds in the packet is the same base space, Ricci flow, given by the fourth term of (2.1A), which allows the jumps, as illustrated above). It is currently unclear whether there are infinite manifold packets or just one manifold packet which is infinite. Manifold Jumps and Pharrell Transport It is also unclear whether the question of distance is applicable in the base space, The Ricci flow, as it pure energy oriented. Those universes differ from each other in a distance measure, which is not known. As the illustration above suggest, they are very close. The packet could vary as other topologically invariant manifolds enter the packet, also known as cosmological singularity. The main point of this short assay between each manifold pair, and actually all the manifolds in the packet is the same base space, Ricci flow, given by the fourth term of (2.1A), which allows the jumps, as illustrated above). It is currently unclear whether there are infinite manifold packets or just one manifold packet which is infinite. It is also unclear whether the question of distance is applicable in the base space, The Ricci flow, as it pure energy oriented. Those universes differ from each other in a distance measure, which is not known. As the illustration above suggest, they are very close. The packet could vary as other topologically invariant manifolds enter the packet, also known as cosmological singularity. The main point of this short assay between each manifold pair, and actually all the manifolds in the packet is the same base space, Ricci flow, given by the fourth term of (2.1A), which allows the jumps, as illustrated above). It is currently unclear whether there are infinite manifold packets or just one manifold packet which is infinite. It is also unclear whether the question of distance is applicable in the base space, The Ricci flow, as it pure energy oriented. 114 8T versus MT 8-Theory, revolves around varying curvature to the M-theory that is considered to be an elevated version of string theory, and includes additional dimension and unification of so called five distinct string theories. The two theories differ in noticeable and subtle difference. The first difference is that the M-theory also describe alongside the first three interactions, the interaction of gravity. In the 8T, all interactions are distinct amounts of gravity. That is discrete amount of curvature on the matric tensor. That is given by the Primorial coupling constants series and in the main equation, which is in agreement with the equivalence principle. ∂g ∂t = 𝜕2𝑔′ ∂t2 Second difference is the subject of description, 8T only aspire to describe a varying manifold. It does not include any particles motion or any particles of any sort, and all the particles were derived with no a-priori data regarding their nature. That was done in the three critical theorems that yielded the primorial in March 2021. M-Theory aspire to describe the behavior, vibration and motion of different "strings" or infinitesimal quantities, in three and higher dimensions. Such bold entities of description have not yielded testable predictions to date. Such an analysis is also has an implicit axiom – understating the way those infinitesimal things vary can tell us something about physics. The beginning of the M-theory is describe by the five distinct kinds of strings, and that is the subject of description in birds eye view. A third difference is the number of arbitrary numbers appearing in the theory. 8-Theory has three arbitrary numbers less than any other theory. The number of Bosons is infinite and isomorphic to prime numbers. The number of families is also infinite given by the Quark masses series, which provides us with additional prediction of fourth family below first generation, causing the matric tensor to have additional amounts of light mass particles. The third number is the number of dimensions, as the universe is part of a packet, each with its own set of finite dimensions; the overall number of dimensions is infinite as well. Those are distinct and do not get mixed into one manifold. It is the reason the manifold is flat and the reason each manifold can not be infinite in dimension, as it is confined by others. M- theory does the opposite and describe nature by additional arbitrary number which is 11D. Manifold Volcanos and Curvature Eruptions Imagine that instead of having just one electron in the primorial, we will have an entire surface full of electrons. Each of them is emitting a net curvature of prime magnitude, and they all emit that magnitude at the same temporal segment on the surface of the matric tensor M. [(24 ∗5) + ∑(𝑒𝑖) 𝑁 𝑖=1 ] + γ𝑖 (3.41) ⊚: ∑(𝑒𝑖) 𝑁 𝑖=1 →γ𝑖 (3.41) The result of this construction is an immense eruption of net curvature off the manifold, similar to a volcano eruption in geo-physics, its concentrated amount of net curvature eruptions due to a positive summation of electrons that emit together, ⊚ as a time operator of all elements in the matric tensor. The eruption could be linearly polarized. In physics it is also known as "lasers". The volcano is the summation of electrons, and the magma is the timed eruptions of photons. It is the same main equation just a different variation – applicable to many particles propagation. The energy of the eruption ray is proportion to the electron summation on the surface, which emit together and to inversely proportional to the area scattered by the eruption ray. The volcano is the electrons on the surface and the magma is the photons, in their concentrated from can melt and cut steal. An analogy makes it easier to describe. So overall the "geo-surface" of the matric tensor is flat, due to the net curvature being relativity small portions and due to the fact arbitrary amount of curvature vanish into matter. The "geo-surfaces" or matric tensors in the 8T have dormant volcanos, which could suddenly become active, causing curvature eruptions of immense magnitude at a timed moment, analogous to magma eruptions. 115 8T versus MT If it is 11D, there has to be a reason it has to be that way. Why not 13D? What makes 11D special? The answer is – nothing. As a number of dimensions, it is good as any other. The fact that seemingly certain traits of Quantum physics are in agreement with this number does not make it special, it could work for a higher dimensional number of certain sort. Another way to put it, this number could be part of a subgroup of numbers. Another arbitrary number of the M-theory is the five "distinct" kind of strings, and the overall emphasis on those strings, makes the theory very weak. As one wrote above, it is building upon the implicit assumption that those strings, and in particular their shape, are important. So 8T has three arbitrary numbers less, MT has two arbitrary numbers more. A forth difference is that 8T is described in terms of spaces, extra spaces. The Matric space and the Riccy flow space, which is the base space. The relation among the two is described by a fiber bundle, since all the manifolds are topologically invariant; it is possible to jump from one manifold into under by switching to the Ricci flow. This space does not obey the rules of distance, and is compact. M-theory describe physics in terms of additional dimensions. So overall, it is much longer description, as you have to describe a-lot more according to each extra dimension. One theory describe spaces, which are two. The other dimensions which are infinite. The fifth difference and the last one, is the number of testable predictions as part of the length of description. 8T has description of dark energy, the equivalence principle, The Primorial coupling constants series and all the known Bosons to be prime amounts of net curvature, fermions as arbitrary variations that vanish. It includes the Quark masses series, curvature knots, matric tensor deformations to the base space, the duality of the thirst forces at 26, and it does so using only one equation (2.1A). It is that simple in can be encompassed in one equation. Second difference is the subject of description, 8T only aspire to describe a varying manifold. It does not include any particles motion or any particles of any sort, and all the particles were derived with no a-priori data regarding their nature. 8T versus MT That was done in the three critical theorems that yielded the primorial in March 2021. M-Theory aspire to describe the behavior, vibration and motion of different "strings" or infinitesimal quantities, in three and higher dimensions. Such bold entities of description h t i ld d t t bl di ti t d t S h l i i l h i li it i 116 Manor O Manor O ∑∂ℓ 𝜕𝑠𝑚 𝐾/2 𝑚=1 𝜕𝑠𝑚 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−∑∂ℓ 𝜕𝑠𝑛 𝐾/2 𝑛=1 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2.1. A) (2.1. A) M-theory does need a larger amount of numerical description. Alongside, the number of testable predictions given its mains equations and power of predictions – is as far as one knows, stand at zero or very close to it at the levels of energy we can reach for today. Another way to state it, it needs a-lot more time and space (on paper) to describe the M-theory, and it gives little to no testable predictions. It was the best we had up-until recently, but according to the analysis and comparing to the new 8T, it seems to have been surpassed. . M-theory does need a larger amount of numerical description. Alongside, the number of testable predictions given its mains equations and power of predictions – is as far as one knows, stand at zero or very close to it at the levels of energy we can reach for today. Another way to state it, it needs a-lot more time and space (on paper) to describe the M-theory, and it gives little to no testable predictions. It was the best we had up-until recently, but according to the analysis and comparing to the new 8T, it seems to have been surpassed. . Universe Packet Density ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 ∂ℒ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℒ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2𝑔′ ∂t2 = 0 (1) ℨ1 + ∑ℨ1 ∞ 𝑛=2 = 𝒟1 (1.2𝐵) ℨ1 + ∑ℨ1 ∞ 𝑛=2 = 𝒟1 (1.2𝐵) (1.2𝐵) 117 Manor O Up to this point in the thesis, we assumed that there is only one packet of stationary Lorentz manifolds, which grow in number. Each manifold has a distinct arrow of time, which is a unique moment of singularity or a unique age. The older the universe the flatter is should be, as it was a subject of pressure from other manifolds for longer temporal periods. That is currently the only option presented in the 8T thesis. However, it now becomes evident that it could be wrong. There could be a limitation of the number of stationary manifolds that composes the packet. Such that if that limit is reached, any matric tensor fluctuations volatile enough will ignite a manifold, which will join a distinct packet. Similar to wave packets, which comes in an infinite number. As far as one can see, the current equations of the 8T indicate that the universe has a "sphere packing" structure, an unknown number of thin liars stacked or compressed together in a packet. If the number is infinite than we have one packet of stationary manifolds. If there exist a limit, there are multiple. Another interesting point, if the number of manifolds in the packet is finite, than the degree of acceleration outward from areas of extremum curvatures is also finite, which is what we required for a stationary manifold. If the number of manifolds increases without a density limit, than the outward acceleration should increase overtime, as more stationary manifolds is in the packet. That seems more correct as we know that the so-called "dark energy" is time invariant. Therefore, that could imply that there is a limitation of density in the packet. We can define this density limit by varying the main equation: ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) K ∈ℝ ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) We can parameterize the manifolds presented in (2.1A) to put the idea of stationary manifold packets, which are distinct in rigor. We can parameterize the manifolds presented in (2.1A) to put the idea of stationary manifold packets, which are distinct in rigor. ℨ1 + ∑ℨ𝑛 ∞ 𝑛=2 = 𝒟1 (2. 𝐵) (2. Manor O 𝐵) Moreover, the new structure of the multiverse is the summation of all the universe packets: 𝒟1 + ∑𝒟𝑖 ∞ 𝑖=2 = 𝜗 (2.C) (2.C) In the 8T we assumed there is just one infinite packet, and the dark energy could be an adiabatic variant, which vary very slowly. This paper analyzed the structure of the multiverse by imposing a limitation on the density of the packet, leading to infinite number of distinct packets as described by equations (2.B) and (2.C). 118 Manor O Manor O The Curvature Code - 8T ∑𝛿𝑔𝑘 𝑀 𝑘=1 = 0 (2.12) ∑𝛾𝑖 𝑀 𝑖=1 = ∑𝛿𝑔𝑖 𝑀 𝑖=1 > 0 (3.13. 𝐵) (2.12) (3.13. 𝐵) Suppose we allocate to the terms, additional terms according to each variable in the main equation. Now as a result we have those fourfold terms for fermions and Bosons accordingly: Suppose we allocate to the terms, additional terms according to each variable in the main equation. Now as a result we have those fourfold terms for fermions and Bosons accordingly: [𝛿ℓ𝛿𝑠𝛿𝑀]𝛿𝑔= 0 [𝛿ℓ𝛿𝑠𝛿𝑀]𝛿𝑔> 0 [𝛿ℓ𝛿𝑠𝛿𝑀]𝛿𝑔= 0 [𝛿ℓ𝛿𝑠𝛿𝑀]𝛿𝑔> 0 We do not know what are the values of the three terms inside the bracket, however since we know to associate the conditions in equations (2.12) and (3.13B) to be equal to zero and larger than zero accordinaly, these are in essence constraint to the rest of the unknown chained terms. For fermions, we can deduce: [𝛿ℓ𝛿𝑠𝛿𝑀] = 0 Because of the ∑ 𝛿𝑔𝑖= 0 𝑁 𝑖=1 auxiliary condition, which impose a constraint on the chained terms. The fermions will receive the form of points, which are flat and are infinitesimal in length, on the matric tensor of the manifold. Now analyze the Bosons, with the auxiliary condition ∑ 𝛿𝑔𝑖 𝑀 𝑖=1 > 0 , the chained term is: [𝛿ℓ𝛿𝑠𝛿𝑀] > 0 The Commuter In QFT one of the most important ideas which emphasize the difference between fermions to bosons is the mathematical expression commuting/anti commuting relations for bosons and fermions respectively. The term is presented in the following form: [𝐴𝑖 ,𝐵𝑖] ± = 0 Fermions anti commute, summed as zero when combined and bosons commute, the only they to be summed as zero as if they are subtracted from one another. The actual way of QFT representation is not important in this paper. The idea of the commuting anti-commuting relations of bosons and fermions is in perfect agreement with the 8T. As was presented in the thesis, the arbitrary variations term is associated with fermions. We require the term to vanish, so when partitioned we needed an even amount of two distinct elements which differ in sign. ∑𝛿𝑔𝑘 𝑀 𝑘=1 = 0 (2.12) (2.12) On the other hand, the bosons were regarded as net curvature of discrete prime amounts as described by the primorial, which add up to a positive summation, so they only way to eliminate them is to subtract from one another. That is in agreement with QFT idea of commutation relation. The term describing Bosons is (3.13.B): ∑𝛾𝑖 𝑀 𝑖=1 = ∑𝛿𝑔𝑖 𝑀 𝑖=1 > 0 (3.13. 𝐵) Overall, one of the most important ideas in Quantum Field Theory is in perfect intersection with the 8T. We can visualize it and reason why that is from an angle of curvature on the matric tensor using the main equation and primorial. We can even use the commuter on the two terms. Overall, one of the most important ideas in Quantum Field Theory is in perfect intersection with the 8T. We can visualize it and reason why that is from an angle of curvature on the matric tensor using the main equation and primorial. We can even use the commuter on the two terms. [𝛿𝑔𝑖 ,𝛿𝑔′𝑖 ] ± = 0 (1.6) (1.6) The first term in the commuting relation (1.6) is describing the partitioned terms, the second is the acceleration. Fermions will accelerate toward each other, in agreement with vanishing curvature. Bosons will accelerate to a joint point on the matric tensor. That is because each bosons is a net curvature that increase the probability of arrival to itself. As was analyzed before, 8T and QFT does not contradict one another. 119 Manor O [𝛿ℓ𝛿𝑠𝛿𝑀] > 0 Bosons will receive the form of non-local propagation on the matric tensor of the manifold. The opposite of infinitesimal scales, that is because they cannot vanish into matter, and isomorphic to prime numbers. Similar to how we presented the process of emission. Summing up, we do not know what are the chained three terms are, but we have proven the ideas two-pillar ideas of the 8T: (2.12) and (3.13.B) in which we can use, as auxiliary conditions. Those auxiliary conditions are used on the chained three terms, which we do not know, and thus they are the key to solve the entire chain. Those two conditions are the vital key to the curvature code – the language of nature. 120 Manor O Degrees of Freedom - 8T We have derived the main equation (1) by EL operator. The following way: ℓ= (s, s′,t) We can state that the 8T analysis in that form has one degree of freedom. ℓ= (s1, s′1, t) since we have proven the second representation in equation (1.2), and thus we can represent the EL operator as a system of differential equations with an infinite degrees of freedom. Those differential equations describe a system of stationary manifolds. That is a different way to state that we are dealing with an infinite dimensional universe, using the original operator. ℓ= (s1 … s𝑛,s′1 … s′𝑛, t1→𝑛) (1.61) (1.61) (1.61) the time operator is of course present in each manifold, but since each manifold has a unique moment of singularity, each manifold is getting flattened in different temporal moment, we have to index the time parameter, so to indicate that the arrow is in different stages for each manifold. Such a representation allows us to eliminate the question regarding the arbitrary number of 13.7B billion years. Equation (1.61) is another way to represent the structure of the multiverse, infinite manifolds that are stationary, and interact with each other. Since each manifold is part of the packet, it is confined by it and cannot escape the variation of the manifold than can be presented only within the domain of the packet. Such an analysis also eliminate the question of three dimensional universe, by representing infinite degrees of freedom, we can elevate the universe to infinite dimensions. We can represent the packet in a discrete way, for example: s1 ⟶𝑑𝑖𝑚𝑒𝑛. . (1 →3) + 𝑡1 s2 ⟶𝑑𝑖𝑚𝑒𝑛. . (4 →6) + 𝑡2 s3 ⟶𝑑𝑖𝑚𝑒𝑛. . (𝐾→𝐾+ 2) + 𝑡𝐾 s3 ⟶𝑑𝑖𝑚𝑒𝑛. . (𝐾→𝐾+ 2) + 𝑡𝐾 Manor O Manor O 𝐾∈ ℝ Since we already presented a symmetry regarding the universe packet, we can change the index of the summation with no effect. Residents of the "second manifold" regard themselves as first, and thus count their dimensions as first to third, if we are residents of the "first" manifold, we count our three as the first to the third, and "theirs" as fourth to six. Each resident of distinct manifold regard "his" dimensions as the lowest, i.e. first to third plus a unique arrow. 121 8T- Curvature Spectra's we have seen the multiplier of each term from the second and above, is reflecting the number of so- called "fields" of each interaction. The first coupling term has eight gluon fields: 8 + (1) The second term has three fields, the massive W and Z bosons, in accordance to the right multiplier, marked in black: The second term has three fields, the massive W and Z bosons, in accordance to the right multiplier, k d i bl k [(8 ∗𝟑) + (3)] + 3 Since all we have in the 8T its curvature, author would like to coin the term – "Curvature spectra", that is each interaction has Bosonic, net curvatures which differ from one another in certain orientation. It is currently unclear which kind of a physical difference it is, it could be a difference in a orientation of the curvature, or a more obvious difference related to mass or both. The features of the W and Z bosons differ from one another supporting the idea of the spectra. Therefore, it is possible to represent the right multiplier as means of a spectrum that is to parametrize it. ∑Ψ𝑖 𝑖=𝑁𝑉 𝑖=1 = 𝑁𝑉 ∑Ψ𝑖 𝑖=𝑁𝑉 𝑖=1 = 𝑁𝑉 Therefore, in the 8T, instead of having a certain finite number of fields, we have an infinite amount of curvature orientations, all appear on the matric tensor and are isomorphic to prime numbers and one. The curvature spectra is parametrized and counting the number of orientations which in physical theory account the different kinds of particles associated with each coupling term. The new elevated form of the primorial is: 𝐹𝑅# = (2ℳ∗ ∏Ψ𝑖 𝑖=𝑁𝑉 𝑖=1 + (ℳ)) + 𝑁𝑉 (1.2𝐵) (1.2𝐵) 122 Alternative Explanation for Dark Matter we presented the Quark masses series, and predicted an infinite series of families with total mass aspiring zero. Mass is considered arbitrary amount of curvature converging inward, with a symmetry break of the 8 −(1) variations. That is the inverse to the primorial, associated with curvature diverging, or 8 + (1) variations. In the case of mass generation, nature is devising in increasing amounts to eliminate the arbitrary amounts of curvature. We predicted the total mass of the fourth to be 0.113 Mev, 55-56 times lighter than first. The advantage of this idea is that we no longer need to explain why there are three families. 19,600 ⟶1400 ⟶56 ⟶0.113 … 19,600 ∗9 ⟶1400 ⟶56 9 = 176,400 ⟶1400 ⟶6.3 176,400 ⟶1400 ⟶6.3 The two versions are presented in the thesis as it is unknown whether the factor of nine is repeating itself for the fifth family and below. Keeping that in mind, assuming this idea is wrong, what alternative explanation can we offer for the issue of dark matter? Notice that according to the main equation (1) or (1.2) we have an infinite packet of universe which interact at areas of extremum curvatures, that means that there two distinct manifolds (if we regard each manifold to be somewhat of a thin liar), whose extremum curvature interact with our own. Since we are familiar with the equivalence principle between mass and energy, the dark energy as given by equation, can be regarded as dark mass.Those masses of distinct manifolds may have an additional gravitational interaction. If each manifold has distinct subspaces, which are newer manifolds that rose from the original manifold, those subspaces may interact with the original manifold that means a distinct set of mass, interacting with our own. The advantage of this idea is that, there could not be any additional trait of matter it the matter is own distinct (yet very close to our own) manifold. It seems to be suitable to the fact that dark matter do not do anything other than to exhort gravity. The weakness of the original idea is that if there is a fourth family below first, it could behave like original matter, omit and absorb light, which is not in agreement with what we speculate. However, if it is matter on a distinct space, or a infinite spaces of the packet, than the features of dark matter could be explained easier. The Ghost Neutrino From experiment, we know that the electron does not propagate by itself but rather with another ghost particle, the electron neutrino. What kind of numerical traits in the 8T this particle possess? In other words, we need to add it the coupling term of the electric without changing the magnitude of the coupling. If reader is familiar with the 8T two symmetry breaking – inward to generate mass: 8 −(1) Moreover, outward to generate a ripple on the matric tensor given by the term: 8 + (1) The answer is clear the ghost particle, the electron neutrino cannot be associated with neither symmetry breaking classes. It has the be a particle which has no effect on the coupling term, we can represent it but it will vanish. The answer than is that the electron neutrino is represented by the following numerical trait that associate with vanishing in the 8T: 𝑣𝑒⟶8𝑛 ; 𝑛∈ℝ [(24 ∗5) + 8 + (3)] + 5 → [(24 ∗5) + 𝑣𝑒+ (𝑒−)] + γ [(24 ∗5) + 𝑣𝑒+ (𝑒)] + γ = 128 𝑣𝑒= 0 [(24 ∗5) + 𝑣𝑒+ (𝑒−)] + γ ⟶[(24 ∗5) + (3)] + 5 Based on the 8T we can predict that the electron neutrino will be massless, in order for the coupling term to stay as it is. The same apply to each higher generation neutrino according to the EMT symmetry. The fact it has no mass does not mean it cannot exhort pressure. The photon is massless, it can extort pressure. If the photon will propagate on a tiny mass measuring scale it will cause the measuring scale to differ from zero due to the pressure it exhorts, and effective mass as it is raw energy. Summing up, it is possible to represent the electron neutrino by using a perfect symmetry multiplier that does not affect the coupling term. The fact it is a prefect symmetry multiplier means the electron neutrino has no mass. That is in agreement with experiment. 123 Alternative Explanation for Dark Matter Summing up, the alternative explanation of dark matter is gravitational effect from a distinct manifold, which interact at areas of extremum curvature. There is advantage to taking the point of view, as it could agree with the features of dark matter behavior. However, using that viewpoint, we still need to explain why there are only three fermions generations. The explanation is not part of this new idea, which is the disadvantage comparing to the original idea. The gravitational effect of dark matter should not be strong, as we have immense fermion clusters, according to the primorial, the ratio of net to total should be very small, aspiring zero, so if dark matter would be explained that route, the gravitational magnitude effect it should have should be weak, that is compared to the first elements in the primorial. 𝑁𝑉 𝑇𝑉 ⟶𝑅 0.111 > 0.1 > 0.039 > 0.008… ⟶0 0.111 > 0.1 > 0.039 > 0.008… ⟶0 124 Manor O We can take the original illustration and modify it We can take the original illustration and modify it 125 The Canonical Equations of Curvature Spikes Suppose we would like to present a simple way to create an analog for the canonical equations of motion, presented by Hamilton. How can we do it in a simple way on a varying Lorentz manifold, with four chained terms in the differential equation? This is an interesting question, and the real answer is one does not know. However, here is an educated guess. The idea is to use the terms in equations (1.48) and (1.49) to derive something fundamental about the momenta of Fermions and Bosons. Suppose we replace the known variable of Hamilton by: ∂qi ⟶∂gi And we know from the equivalence principle that ∂gi = ∂g′i ∂gi = ∂g′i We can present the canonical equation of curvature spikes, if one intuition is correct: ṗ i = ∂ℓ ∂gi ṗ i = ∂ℓ ∂gi In addition, since we know it vanish into zero, we cannot present it as being in the denominator. We can than represent the canonical equation of curvature spikes for Fermions: ∂gi = ∂ℓ ṗ i = 0 (1.7) (1.7) And since we can derive the beautiful result, which comes to an agreement with previous results of the 8T, fermions momenta will vanish to zero. That is another way to state that they will accelerate toward one another. Therefore, configurations of Fermions must appear stationary, similar to Quark Triplets in H d f l N i h h h i i b h h f And since we can derive the beautiful result, which comes to an agreement with previous results of the 8T, fermions momenta will vanish to zero. That is another way to state that they will accelerate toward one another. Therefore, configurations of Fermions must appear stationary, similar to Quark Triplets in Hadrons for example. Notice that the emphasis is not on constant rate but rather on the momenta of arbitrary variation set. We can do the exact same thing for Bosons, since we know that they are isomorphic to prime numbers that cannot vanish into matter, the canonical equation of curvature spikes for Bosons is the following: Hadrons for example. Notice that the emphasis is not on constant rate but rather on the momenta of arbitrary variation set. The Canonical Equations of Curvature Spikes We can do the exact same thing for Bosons, since we know that they are isomorphic to prime numbers that cannot vanish into matter, the canonical equation of curvature spikes for Bosons is the following: 𝜕𝑔𝑖= 𝜕ℓ 𝑝̇𝑖 > 0 (1.71) (1.71) Meaning that the Bosonic configuration must have some net momenta, we cannot find a Boson at rest. That is in agreement with the 8T ideas. Fermions have opposite signs, demands by stationary Lorentz manifold. Bosons are all positive summations, net curvature on the matric tensor isomorphic to prime numbers, they cannot cancel one another as Fermions do. That is the similar to the idea construction in that led to the 8T commuter for fermions, plus, and Bosons minus respectively: [δgi , δg′i ] ± = 0 [δgi , δg′i ] ± = 0 126 The Graviton Illusion - 8T We have analyzed the term of Graviton as a combination of three net curvature that could be either distinct or identical. That is by the spin two trait. That spin two trait, in net curvature representation means that gravity is short ranged, as the three net curvature and the invariant three sums to be a positive number. [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 (2.2) [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →[2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦] (2.3) (2.3) Up until now, reader is probability familiar with every equation presented, as those are 8T thesis fundamentals. From here on out, we have a completely new paper. Since nature does not impose a restriction on the kind of particles to which are describing the term (2.2) and (2.3), it is possible to predict that there are infinite classes of Gravitons of distinct magnitudes. Alternatively, that if we take an even sum or certain sort, add a generator and three net curvature of certain magnitude, which belong to the prime ring, that combination will result a "Graviton like" particle. [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1…𝐾𝑁 ∈ ℝ [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1…𝐾𝑁 ∈ ℝ Since spin two vanishes due to being an even number, the difference between each Graviton is the term that does not vanish in spin representation - (2𝑁()). The smaller this term the stronger the Graviton should be. Therefore, if one analysis is correct, Gravitons classes are infinite in kind, and they are, in contrast to the first three interactions that are in a sense independent, is not independent and depends upon the composite elements. As previously mentioned, Gravitons are a superposition of net curvatures (equivalent or distinct is currently not known), which means that in order to sustain Graviton on quantum scale, it requires aligning three net curvatures in time and position. If one of the net curvature terms is not there, we no longer have the Graviton. The main point of this paper is to make a prediction about the nature of Graviton, and here it is the prediction: (1) Gravitons are infinite in kind. (1) Gravitons are infinite in kind. (1) Gravitons are infinite in kind. It is a daring statement to make given the fact that we did not detect even a single graviton to date, but the 8T is a daring theory. It also provides us a practical way to test whether Graviton like particles can be created in an artificial way. For example, for the electromagnetic coupling, we need the term in (2.4) to create a "Graviton like" particle, the Graviton is a matter of illusion, and it is everywhere and nowhere at the same time, as it is quite rare to create the term in (2.4) as far as one can see. [(2𝑁(2)) + (𝑒−)] + γ + γ + γ (2.4) [(2𝑁(2)) + (𝑒−)] + γ + γ + γ (2.4) Suppose that we are a given the gravitation coupling as the following term: [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 However, the net variation 𝑁𝑉𝐾3 suddenly vanish from the combination and being replaced by a distinct net curvature element: However, the net variation 𝑁𝑉𝐾3 suddenly vanish from the combination and being replaced by a distinct net curvature element: [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 127 Manor O Manor O Curvature Terminators and Conservation of Energy - 8T The point of the first 8T proof presented in pages 3-4, which was only briefly mentioned, is that nature is aspiring to eliminate the curvature. The result of the elimination is yielding the group that allowed physicists to predict the existence of the omega minus (333) in the 8T. However even if (2.12) is vanishing to zero, there is constant creation of matter. Arbitrary variation of the manifold are not obeying a time limit, they can and are created in a random fashion. So one way to put it is that energy is not conserved. That is because matter is constantly being created, and matter is synonymous with energy. The only way to ensure that the energy will be conserved is to present a new way of curvature terminators that is anti-matter. We allow the existence of anti-matter as the coupling magnitudes are preserved under sign reversal. [2𝑁3 + 1 2] + 1 2 ⟶[−2𝑁3 − 1 2] − 1 2 (1.45) (1.45) We analyzed the subject of anti-matter in previous paper, and in particular the subject of orthogonal curvature, which has inner product zero. So to ensure the conservation of energy, we will have to present the set of arbitrary curvature terminators, for fermions, it has two inverse elements. We analyzed the subject of anti-matter in previous paper, and in particular the subject of orthogonal curvature, which has inner product zero. So to ensure the conservation of energy, we will have to present the set of arbitrary curvature terminators, for fermions, it has two inverse elements. Χ = [−𝛿𝑔1, +𝛿𝑔2] ⟨𝛿𝑔𝑖|−𝛿𝑔𝑖⟩= 0 (1.46) (1.46) So overall, there are two main stages of curvature elimination. First arbitrary variation vanish into matter, as presented in the 8T thesis and the prove above. Secondly, to ensure the conservation of energy, anti-matter terminators are presented. Whether energy is actually conserved is unknown, author tend to belief is not. That is due to the asymmetry of matter to anti-matter in the universe. If for each matter created there is also an anti-matter particle, anti-matter should be more common. We have presented the same procedure of orthogonal curvatures to leptons and Bosons. Manor O 𝑁𝑉𝐾3 ≠𝑁𝑉𝐾4 That seems as a trivial change, but it really is not. 8T predicts that the Gravitons are infinite in kind. That is the example of that idea. We previously mentioned gravity is different because it is a composite interaction due to the spin two trait. That is in contrast to interactions of the primorial which are not a composite but contain one net element. That means that gravity coupling magnitude could vary over time. In particular, it means that net curvature elements can replace other elements that were part of the threefold composite. Since the total spin is invariant, i.e. two, there is not a change in the nature of gravity; the structure is the same while the composite element could be different. [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 →2𝑁() + 2 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 →2𝑁() + 2 Prediction: The Gravitational coupling constant is not a constant at all. Prediction: The Gravitational coupling constant is not a constant at all. Prediction: The Gravitational coupling constant is not a constant at all. 128 Manor O Manor O Curvature Terminators and Conservation of Energy - 8T We used equation (1.46) with leptons as elements of the inner product such as the electron and positron: ⟨+3𝑖|−3𝑖⟩= 0 In addition, with bosons, described by the term (2.12) as they were proven discrete amount of prime curvature on the matric tensor: In addition, with bosons, described by the term (2.12) as they were proven discrete amount of prime curvature on the matric tensor: ∑𝛿𝑔𝑖 𝑀 1=1 > 0 (3.13. 𝐵) ame rules apply for the photon as an example: Same rules apply for the photon as an example: Same rules apply for the photon as an example: ⟨γ𝑖|γ− 𝑖⟩= 0 Summing up, if require the conservation of energy we must present the arbitrary curvature terminators, i.e. Anti-matter. If the number of anti-matter terminators is smaller than the number of arbitrary variations which vanish into matter, which seems to be the case on our manifold, than energy is not conserved, as matter is constantly being created. 129 Manor O Manor O Direction Invariant Fermion Distributions - 8T The sole mathematical discipline of the 8T is calculus of variations. As reader assumed familiar with it, one of the major features of this theory is the vanishing of variation. ∂ℒ δ𝑞𝑖 = ∂ℒ ∂𝑞𝑖 −∂ℒ ∂𝑞𝑖̇ ( 𝑑 𝑑𝑡) = 0 Since in our theory we have the arbitrary variation term in equation (1.48) to vanish into matter, we can represent the idea as: ∂ℒ δgi = ∂ℒ ∂gi −∂ℒ ∂gi̇ ( d dt) = 0 (1.8) (1.8) If so, Bosons as net curvature isomorphic to prime numbers are interfering with the stationarity of the manifold, hence their name "Agrarian", as they cannot vanish into matter, they cause the matter clustering. For Bosons: ∂ℒ δgi = ∂ℒ ∂gi −∂ℒ ∂gi̇ ( d dt) > 0 (1.81) (1.81) One final point, since the primorial coupling series is invariant to direction: One final point, since the primorial coupling series is invariant to direction: 𝒫0 = 8 + (1) 𝒫𝑁 # = (2ℳ∗ ∏𝒫𝑉 𝑉=𝑁 𝑉=1 + (ℳ)) + 𝒫𝑉= 30: 128:850: 9254.. ( 1.2.𝐴) ( 1.2.𝐴) As presented in the idea of probability variation of the (1.2A): As presented in the idea of probability variation of the (1.2A): 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 + ℳ) + 𝑃(𝐴) The matter configuration of the manifold should invariant to direction, that is preciously because it is not possible to determine where the lepton is going to emit or absorb, or even which kind of bosons are in play. Put another way, because it is impossible to know where the net curvature that violate stationarity will appear, the fermion distribution across all directions of the manifold is the same, there is no special direction of any sort. That is preciously the current modern picture of cosmology, the universe look everywhere the same. Same idea we presented in earlier paper of the sphere shape of starts, but now to much larger Fermion clusters. 130 Manor O Manor O Minimizing the Laws The last part of this paper will revolve around a feature of nature which was mentioned briefly in previous papers, and in the 8T thesis. Lagrangian oriented theories are based upon the principle of least action, which deals with minima of certain classes, and this is the most significant feature of those theories. There is one additional minima in the 8T and in a final theory that should get our attention, as it is just as important. That is minimizing the number of laws that govern everything. In every universe, at every stage of development of the manifold, from the flattening by the packet to complete coldness, the number of laws should stand at minima. In other words, the number of equations or ideas in which we use to describe everything should be minimal, and that is a significant feature of a final theory. The minima is not only path-oriented such as in classical mechinqes or QED, it is also manifested in the number of laws. 8T is that kind of theory as all we have achieved, from dark energy to the coupling series and the Quark masses series, is encompassed in just one equation and two conditions. ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ ∂2g′ ∂t2 = 0 (1) (1) In addition, all the standard model of particle physics is encompassed in just two conditions. For fermions: ∑δgi = 0 N i=1 (2.12) (2.12) Bosons as net curvature of discrete prime amounts: Bosons as net curvature of discrete prime amounts: Bosons as net curvature of discrete prime amounts: ∑δgi M i=1 > 0 ∑δgi M i=1 > 0 (3.13. B) (3.13. B) 131 Predicting the Value of the Next Planck Constant The paper main point is to provide a theoretical prediction regarding the fourth interaction. Since from measurement we know the value of the Planck constant, and in our theory, it is associated with the net variation element of the third coupling term; we can predict the next value of the Planck constant for the fourth coupling term based on the ratio of the net variation of the two coupling terms, as they are the discrete amounts which get emitted or absorbed into the lepton. ℏ→+5 Define the next Planck constant as: In agreement with what we expect, as each net variation is larger than the preceding, now we can take the actual value of the Planck constant and multiply by the ratio to reach the exact prediction – the next Planck Constant should be 1.4 larger than the original Planck is and stand as: 9.27649806 × 10−34 𝑚2 𝑘𝑔 /𝑠 9.27649806 × 10−34 𝑚2 𝑘𝑔 /𝑠 132 132 The Nature of the Primorial – 8T In previous papers, author presented the claim that the primorial coupling series is invariant, both across the manifold packet and both in time. The reason for that invariance was the invariance of the prime ring. It is possible to solidify the nature of this claim from a different angle of analysis, that is by classifying the primorial as a scalar function. A scalar function as reader probably knows is a real function, defined within a region and which values are invariant to any coordinate transformation. Because of the invariant prime ring, we can classify the primorial as a scalar function. The gradient of Because of the invariant prime ring, we can classify the primorial as a scalar function. The gradient of a scalar function is a covariant vector. ∂ℒ ∂𝜙𝛽 = ∂ℒ ∂𝜙𝑎 ∂𝜙𝑎 ∂𝜙𝛽 As an example of a covariant vector. The primorial is answering the criteria of a scalar function as her values are bounded to ℝ, it is defined within the boundary of the expanding manifold as given by the main equation (1). Since the 8T is built upon the EL operator and the action principle, the invariance under shifting frames is already embedded in the nature of the theory. However, it is also important to emphasize in the context of the primorial coupling series. Two final points, the first, is the primorial does not conation time parameter and thus is not varying time for independent interactions – i.e. only one distinct prime as net variation. The last argument does not include gravity as it is a composite interaction as given by equations (2.2) and (2.3) : [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 (2.2) [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →[2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦] (2.3) (2.3) We have proven it is possible to replace one of the composite elements and keep the nature of the gravity invariant, and thus gravity coupling could vary overtime, by replacing 𝑁𝑉𝐾3 →𝑁𝑉𝐾4 . Therefore, despite the primorial being a scalar function whose nature is invariant concerning elements which are independent, i.e. contain only one distinct prime, it does not apply to elements which are composite such as Gravity. Which vary over time. The Nature of the Primorial – 8T [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1…𝐾𝑁 ∈ ℝ [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≠𝑁𝑉𝐾4 The gravity could be described by infinite distinct composites, which are time variant and still retain the inner nature of the Graviton: The gravity could be described by infinite distinct composites, which are time variant and still retain the inner nature of the Graviton: [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 →2𝑁() + 2 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 →2𝑁() + 2 Therefore, despite the primorial being a scalar function whose nature is invariant concerning elements which are independent, i.e. contain only one distinct prime, it does not apply to elements which are composite such as Gravity. Which vary over time. 133 Manor O Manor O The Chameleon Particle - 8T We have taken two routes in the meaning of the invariant three, back in march author believed that the invariant three is different for each term. Later, a shift in perspective accrued and author stated that it is the electron for each higher term, which destabilize ever-growing fermion clusters causing net variations to appear in different magnitudes. That is because the invariant three is isomorphic to itself. In this paper, we will analyze the meaning of those options. If it is the electron for each higher term, which seems to be the more reasonable option, than there should be a set of Planck constants. The original Planck constants that describe the numerical term of photon absorption and emission is not special but part of an infinite set. Η = [ℏ1 … ℏ𝐾] Each Planck constant is isomorphic to a prime number according to the primorial coupling series. Another prediction that should be made. The prediction is the following: Each higher term in the coupling series should be bigger than the preceding. That is because those higher terms are representing bigger quanta in the series. The statement is not in contradiction with the fact that each element in the series is weaker than the preceding as we calculated the ratio of net to total. Here we only interested in the net. So according to this viewpoint, which is the electron for each higher term, we reached a prediction regarding the discovery of Max Planck. Now we can expend the earlier option, which regard the destabilizer, i.e. the invariant three to be different for each term. Since it is the invariant three for each term, but it appears again as different for each coupling, it is again resembles a chameleon. If it is in fact the case, author does not lean to this direction, but would like to cover the spectra of options. Either option we take, we have an element which is either same for all, causing an emission of different bosons, according to the current thought tides. Alternatively, we have distinct particles manifested by the same number, causing net curvature of distinct amounts to appear on the matric tensor. We presented those two options. Author is strongly leaning toward the first in this paper, i.e. Imaginary Coupling Constants Suppose the matric tensor has two interactions on it, which are studied by an observer. This observer does not know that the Bosons are isomorphic to the prime ring, and there are only two interactions, the electric interaction and the fourth interaction. [(24 ∗5) + (𝑒)] + γ [(120 ∗7) + (3)] + 7 Assuming the net curvature appear not in a superposition but rather as distinct propagations on the matric tensor. if the observer is not familiar with the series, he could for example take the average of the two net curvature as a new coupling term. That is, associate Bosons to the ring of the integers and not to the ring of the primes, in that case to the integer six, the average. He could decide that there is a coupling constant, whose magnitude lies in between the range 𝓇: 128 < 𝓇 < 850 γ + 7 2 = 6 While in fact, he is measuring the average net curvature of two distinct prime amounts of net curvature. That is somewhat resembles the pseudo-forces measured from certain frames of reference in Einstein theory of relativity. We previously stated that in the 8T, the coupling magnitudes are invariant as the prime ring itself is observer invariant. It is also invariant across the manifold packet, different universe will possess the same coupling magnitudes, and as a result the same particles. That is due of the invariance of the prime ring. We can not associate a Boson to an even number, which vanish. In that sense it is imaginary. 134 Higgs Stealth Field The analysis of the Higgs field will be done via the spin representation. In the 8T thesis, page sixteen, we presented the following classification: Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations In other words, the Higgs field is represented by the first term and is affecting the series from the weak interaction and above, as it is responsible to the additional term appearing in the coupling term of the weak and above, i.e. the invariant three. The two key points, which are at the heart of this paper, are the following. According to spin representation, there is more than one Higgs particle. That is because, if one idea is correct, there is no restriction imposed on the term of the spin zero. That is in rigor, spin zero can be parametrized; 2𝑁( ) →2𝑁𝜇 𝜇∈ℝ 2𝑁𝜇∈ℝ 2𝑁( ) →2𝑁𝜇 𝜇∈ℝ 2𝑁𝜇∈ℝ Because of the parametrization of the first term in spin representation, we can create infinite terms that are distinct, that is: 2𝑁𝜇≢2𝑁𝜇+1 ≢2𝑁𝜇+2 … Each corresponds to a unique Higgs operator if one intuition is correct. It is again a bold risk as spin representation and net variation representation are different. The idea was to take a certain feature of the net variation representation, which is the ever-increasing variation terms, and use it in spin representation to predict that there are infinite Higgs particles. The second main point is the following. Since the 2𝑁𝜇 coupling terms are always present in the coupling series, the effect of Higgs, or the interaction of the Higgs with the fermions and Bosons is constant. Hence, its name in the paper title, it resembles a stealth field, which is unfelt and yet is always there. That is only evident in spin representation. In addition, since the Higgs field is part of the primorial coupling series, i.e. a scalar function, that do not include a time parameter, we can predict that the Higgs is time invariant. The Chameleon Particle - 8T it is the electron for all of those higher terms, as it was proven the invariant three to be an electron by putting it on the formula of the fine structure constant. However, there is always a reasonable chance that one's intuition is wrong and it could be a new particle for each term. The "proper" term for this element is the chameleon particle, both option describe its chameleon trait. To sum things up, three predictions were made: There is an infinite set of Planck constants. Each is isomorphic to a prime ( 1 ) (2) Those Planck constants are larger and larger from one interaction to another. (3) The invariant three is the electron for each higher coupling term. The electron is the chameleon particle. It emits different bosons for each coupling term. 135 Manor O Manor O Higgs Stealth Field If the higgs field is associated with the 2𝑁𝜇 term of the weak interaction as an example, it should be massless. If it is not the Higgs field itself is going via a process of a symmetry break. Either that or the idea of the mass symmetry break of the 8 −1 variations is incorrect. To summarize four predictions were made: (1) Higgs are infinite in kind (1) Higgs are infinite in kind (2) Higgs are in constant interaction with Fermions and Bosons, it is a stealth field. (3) Higgs particles are time invariant 4) Higgs should manifested as Massless particle. If it is not, it is going via a symmetry break. (4) Higgs should manifested as Massless particle. If it is not, it is going via a symmetry break 136 The Vacuum – 8T The idea was presented in the canonical equations of curvature spikes, (1.8) for fermions and (1.81) for Bosons, vanishing and non-vanishing curvature spikes accordingly: ∂ℒ δgi = ∂ℒ ∂gi −∂ℒ ∂ġ i ( d dt) = 0 (1.8) ∂ℒ δgi = ∂ℒ ∂gi −∂ℒ ∂ġ i ( d dt) > 0 (1.81) (1.8) (1.81) The summation of the prime pairing to zero is resulting in an infinite set of zeros. That is synonymous with the vacuum idea of quantum field theory: ∑0𝑖 𝑇 𝑖=1 = 𝒱 (2.16) (2.16) Summing up in a concise manner. The vacuum is the result of prime pairing which do not have a net variation element, as they are not sums identical to (2.15) in their devisors. Thus, they vanish into zero. The sum of all vanishing zeros is the vacuum of the 8T, as presented in equation (2.16). All prime pairs appear, as previously mentioned, we can pair any even number of primes, we chose 𝑁= 2 for simplicity sake. The idea of the vacuum in this theory is somewhat hard to grasp, as it requires knowing beforehand where those violations of stationarity will appear, which is impossible to know. What is possible to know is those violations of stationarity has appeared, that is simple, where there is stars and galaxies. The vacuum idea than is more appropriate to describe in terms of short to infinitesimal time intervals, it is not a continuous entity in time. The Vacuum – 8T We derived the primorial by using total variations pairing, we searched for pairs that have certain features we know about fermions. In particular, the total sums of the pairs had to be two and three divisible. Below marked in black are the pairings we used to derive the series. (3,3) (3, 5) (3,7) (3,11), (3,13) … (5, 3) (5,5) (5,7) (5,11) (𝟓,𝟏𝟑) … (7, 3) (7,5) (7,7) (𝟕, 𝟏𝟏) (7,13) … … . (29, 19)(29,23), (29, 29), (𝟐𝟗,𝟑𝟏)… We calculated the sums of those prime pairing using the simple formula: ∑𝒫𝑖 𝑖=𝑁 𝑖=1 = 𝑆; (2.14) 𝑁= 2 (2.14) And each of those pairs to theorized based on theorem three, have a net curvature element proportional to the average, we searched for the first two pairs, derived the third coupling term using the formula of the primorial, without the prime pairing, and concluded the idea was correct. (𝑝1, 𝑝2) = (5,13) → 𝑁𝑉= +1 (𝑝3, 𝑝4) = (29,31) → 𝑁𝑉= +3 (𝑝5, 𝑝6) = (? , ? ) → 𝑁𝑉= +5 The fact that those prime pairs are in agreement with the coupling magnitudes does not mean that those pairs are exclusive or special. There is no law suggesting that these are the only pairs appearing in our theory and that is a good thing. Therefore, all prime pairing are appearing but because we have the condition of (1.48) those prime pairs of variations are taken to vanish. Therefore, we can define the prime pairs that are not suitable for the coupling criteria: (𝑝𝑁, 𝑝𝑁+𝐾) = 𝑆𝑁 𝑆𝑁≢S [2 ,3] | S (2.15) [2 ,3] | S (2.15) (2.15) Assuming we required the original condition, for the sum to be divisible by two and three. Therefore, the majority of those pairs do not answer the condition. However since the still vanish due to being an even number and using equation, each pair could be regarded as a single distinct zero. So those prime pairs vanishing are the playing the rule of the vacuum in the 8T. 137 Manor O 𝑇⇢∞ We started the summation as the primes indexed from one to six does answer the criteria of coupling constants, and cannot regarded as part of the vacuum. That is because they have a non-vanishing element 𝑁𝑉 of certain kind. Those 𝑁𝑉 elements are violations of stationarity causing fermions to cluster. The Coupling Constants Series – a Star's Stability and Collapse It is possible to reason the stability of the star in two ways, which are identical almost. The first is more general, that is by the opposing symmetry breaking of mass generation and force generation. Those two eliminate each other perfectly to achieve stability. By the primorial, we have proven the curvature diverging to be associated with the term 8 + 1 and the Quark masses series with the symmetry breaking of the inverse kind, 8 −1 given by the series of total masses of each fermion generation: 19,600 ⟶1400 ⟶56 ⟶0.113 … 19,600 ∗9 ⟶1400 ⟶56 9 = 176,400 ⟶1400 ⟶6.3 176,400 ⟶1400 ⟶6.3 That is to say that the curvature diverging inward is equal to the curvature diverging outward, and so the matter formation described by the term (1.48) is stable. If so, so does the star, as it is a cluster of fermions. We can choose a more direct root to describe the stability of a star. That Is by comparing gravity to the forces extending or radiating from the star outward. The key point is that with time, gravity of the star can get stronger. As we currently regard gravity as a composite element, as time goes 138 Manor O by, the primorial is generating larger and larger net variations, which could change the ratio of the gravitational "constant" of a star given by equations (2.2) and (2.3): [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 (2.2) [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →[2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦] (2.3) [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1…𝐾𝑁 ∈ ℝ (2.3) [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 However, the net variation 𝑁𝑉𝐾3 suddenly vanish from the combination and being replaced by a distinct net curvature element that is larger and in agreement with the arrow of time. now the gravitational constant of the star is stronger while the forces extending outward are the same. [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≠𝑁𝑉𝐾4 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≠𝑁𝑉𝐾4 The structure of the gravity is invariant to the change of the element. Since the total spin is invariant, i.e. two, there is not a change in the nature of gravity; the structure is the same while the composite element could be different. The Arch of Time Arrows Is it possible to explain the three "distinct" time arrows using one idea? Author will argue that by using the primorial it is easily within reach. Starting with the radiation arrow, the primorial is perfectly suitable, as we regard the Bosons to be discrete amount of energy or radiation emitted from the lepton. As was presented in the 8T thesis, pages thirty and thirty-one, the time arrow is evident. That is because each coupling term is weaker than the preceding given by the ratio of net to total pair averages. The direction of the arrow is the direction of time. 1 9 > 3 30 > 5 128 > 7 850 … 1 9 > 3 30 > 5 128 > 7 850 … The Coupling Constants Series – a Star's Stability and Collapse [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 →2𝑁() + 2 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 →2𝑁() + 2 While the forces expending which are not a composite are the same, at a certain stage those gravitational interactions will supersede the forces extending outward, which will mean that the curvature converging will exceed the curvature diverging, and so this will ignite the collapse. That analysis is more detailed than the first root, given by the inverse symmetries and a lot more complicated as gravity is a composite interaction that seems to be only partially understood even with the recent advancement of the coupling series and the main equation (1). The key point to take from that analysis is that gravity due to being a composite and time variant can get larger over time, while the independent interactions, given by the primorial, which is a scalar function that do not include the time parameter, are the same in magnitude. That creates a long-term advantage toward the gravitational effect over the independent interactions. The result of such a framework is such that with large time increments, the probability for a collapse of a star is ever increasing. That agrees to what we have previously stated about gravity. That Gravitons are infinite in kind, and are short ranged due to spin two trait, which vanishes. 139 0.111 > 0.1 > 0.039 > 0.008 … We already have the radiation arrow and the cosmological arrow unified by the primorial. Now the last arrow, the thermodynamics arrow. How can we present the idea of thermodynamics within the context of the primorial? As one believes, there are several ways to do just that. Among the set of potential ideas, we can state that as the primorial has more options, meaning more distinct primes will be propagated over time. That is because the direction of the arrow is the direction of time. so, over time we have more and more distinct elements, alongside constant matter creation given by equation (2.12), the result of such framework seems to be with an agreement with the second law of thermodynamics and therefore with the thermodynamics arrow. We can use functors to present that idea in rigor; For simplicity sake, we can use a setting of a partitioned set: For simplicity sake, we can use a setting of a partitioned set: ⊎: 𝑇𝑜𝑝→set 𝚺:[∑δgi = 0 N i=1 , ∑Zi ∑NVi K i=1 = K i=1 Z1NV1.. ZKNVK,t1] (2.17) 𝚺:[∑δgi = 0 N i=1 , ∑Zi ∑NVi K i=1 = K i=1 Z1NV1.. ZKNVK,t1] (2.17) (2.17) The set in equation (2.17) includes all the arbitrary variations that appeared on the manifold, all the net curvature classes according to their kind, given by the index summation, and according to the amount of times each appeared, given by the scalar multiples 𝑍1 → 𝑍𝐾 . At later continuation of time, according to the primorial, we will find that the new set is presented by (2.17.1): The set in equation (2.17) includes all the arbitrary variations that appeared on the manifold, all the net curvature classes according to their kind, given by the index summation, and according to the amount of times each appeared, given by the scalar multiples 𝑍1 → 𝑍𝐾 . At later continuation of time, according to the primorial, we will find that the new set is presented by (2.17.1): ψ ∶[∑δgi = 0 N+∆N i=1 , ∑Zi ∑NVi K+∆K i=1 = K+∆K i=1 Z1NV1. . ZKNVK, t1 + ∆t] (2.17.1) ∆𝑁 ,∆𝐾 ∈ ℝ In other words, more matter was created, the number of non-vanishing distinct curvature increased, and their kind increased as well. We have more elements of distinct kind. That does not contradict the flatness, as those are getting weaker and weaker; the point of the above equations is to present the thermodynamic picture in a simple way, which intersect with the Primorial. The primorial is the arch, which according to the 8T propagate all the three time arrows. Radiation are the Bosons, the cosmological is given by the ratios, and the TM arrow is given by the rise of entropy at infinitesimal time increments, these are different fingers of the same hand. 140 Manor O ∑∂ℓ 𝜕𝑠𝑚 𝐾/2 𝑚=1 𝜕𝑠𝑚 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡−∑∂ℓ 𝜕𝑠𝑛 𝐾/2 𝑛=1 𝜕𝑠𝑛 𝜕𝑀 𝜕𝑀 𝜕𝑔 𝜕𝑔 𝜕𝑡= 0 (2.1. A) 0 = 8 + (1) (1.1) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 8 + (1) [(8 ∗3) + (3)] + 3 [(24 ∗5) + (3)] + 5 [(120 ∗7) + (3)] + 7 This is the first representation of the primorial, discrete amount of net curvature on the manifold. It is a detailed representation as we have leptons, Bosons as separate entities. This does not exist in spin representation of the primorial, and that is preciously how we derived the particle wave duality, due to spin variation. In spin representation, we used the prime critical line. That is the transformation for matter: [(8 ∗3) + (3)] →[2𝑁1 + 1 2] [(24 ∗5) + (3)] →[2𝑁2 + 1 2] [(120 ∗7) + (3)] →[2𝑁3 + 1 2] The only thing we care about in this representation is the prime critical line. Matter is associated with one-half, while Boson configuration is associated with one. The spin representation ignore the lepton and regard all the coupling as a spin compass. We do not make a clear cutting classification to particles in this representation. Manor O In spin representation, we have one entity, the total spin of the element, either Fermionic or Bosonic. The photon before measurement had spin one, now we measured it and it varied to one-half, no longer bosonic spin. That was mentioned in the 8T thesis. However, it is important to emphasize the difference among the representations. In net curvature representation, we analyze each element separately, while in spin representation we care only about the total of elements in the prime critical line. The [2𝑁 + 1 2] is matter, [2𝑁 + 1] is for Bosons. ∆𝑁 ,∆𝐾 ∈ ℝ For Bosons: [(8 ∗3) + (3)] + 3 →[2𝑁1 + 1 2] + 1 2 = 2𝑁1 + 1 [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 1 2 = 2𝑁2 + 1 [(120 ∗7) + (3)] + 7 →[2𝑁3 + 1 2] + 1 2 = 2𝑁3 + 1 The key point, despite the spin representation is including matter in its coupling term, we don’t care about this, we regard this whole term as spin one, and therefore only to bosons. That is in contrast to the net curvature representation that makes a difference among each element in the coupling term. From spin representation it was quite simple to derive the particle wave duality for Bosons. In particular the particle wave duality is a result of total spin variation by half unit, caused by additional Boson, hitting the original Boson. [(24 ∗5) + (𝑒)] + γ + γ →2𝑁2 + 3 2 141 Spin Symmetries and Free Electrons – 8T [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ Shifting to spin representations for the third element in the series, which is electromagnetism: [(24 ∗5) + (3)] + 5 →[2N2 + 1 2] + 𝟏 𝟐 Replacing the bold element with the inner element of one-half would count as an invariant transformation that preserve the original structure in spin representation: [2N2 + 1 2] + 𝟏 𝟐→[2N2 + 𝟏 𝟐] + 1 2 [(24 ∗5) + (𝑒)] + γ → [(24 ∗5) + (γ)] + (𝑒) Such a variation of the coupling series does not affect the overall magnitude of the element, but using it we can reason for the existence of free charges in nature. Since this are not bound to matter, they do not have to vanish so nature will allow it. In previous paper we showed that if the original structure would be analyzed the electrons will add up to a positive summation of curvature, which must vanish. Nature than will generate an opposite set of spinning charges to ensure it will and that was the reason monopoles can not exist. ∑𝑒𝑖 𝑁 𝑖=1 → ∑(+3)𝑖 𝑁 𝑖=1 > 0 ∑𝑒𝑖 𝑁 𝑖=1 → ∑(+3)𝑖 𝑁 𝑖=1 > 0 Those two conditions are in contradiction. The left is a positive curvature summation within a cluster of arbitrary variations which curvature must vanish. The solution is to represent an additional cluster of spinning the inverse direction within the cluster of matter to solve the contradiction of (1.37). ∑(3)𝑖 𝑁 𝑖=1 > 0 ∩ ∑𝛿𝑔𝑘 𝑀 𝑘=1 = 0 (1.37) ∑(−3)𝑖 𝑇 𝑖=1 < 0 ∑(−3)𝑖 𝑇 𝑖=1 + ∑(3)𝑖 𝑁 𝑖=1 = 0; 𝑇= 𝑁 (1.39) (1.37) (1.39) 142 Manor O ∑δgk M k=1 = 0 (2.12) (2.12) Summing up, when the electron is bounded by the bracket as, nature will not allow to exist by itself, however by symmetry of spin leading to replacement of the elements, now the electron is free and such a vanishing of the summation is no longer valid. The equation than suggest an elegant and simple explanation to one of the most interesting enigmas of modern physics – the enigma of free electrons and lack of monopoles within matter. The Equation for Exotic Charges - 8T to bring an element to itself given only two varying elements in the series we need two distinct maps, which attach a varying element to itself, by a threefold combination. δg1(O)δg2(Y)δg1 For example. Even though the sub elements in the series are varying, the overall series can vanish. Now, count all the ways of possible combinations of those elements. We are going to analyze by the integral signs. Since it is a group, there is a natural map, which change an element to itself. One built his analysis firstly on those natural maps. So: (1(e)1(e)1) 2(e)2(e)2 (221) (112) (211) (122) (212) (121) Up to this point, reader is most likely familiar with the everything as these 8T fundamentals exactly as presented in pages 4-5 From here on out, we have a completely new paper. We have that in order the series to vanish and given the threefold combination, the charge of each particle must be a devisor of three. In order the series to vanish, given an even of elements, the charges we derived must summed as positive or negative, integer, plus or minus one. Combined with the condition of the threefold, we reached: δg1δg2δg1 = +1 δg2δg1δg2 = −1 δg1δg2δg1 ⇔δg2δg1δg2 (1.32) (1.32) The pair in equation (1.32) will be permitted as it. Will pair exactly to zero, that is in agreement with the charges of elementary quarks and in the 8T arbitrary variations of curvature on the matric tensor. suppose that instead of three threefold combination, it took five to bring an element to itself, than the charge of each particle must be a five devisor. The new five-fold combination is given by (1.31) δg1δg2δg1δg2δg1 (1.33) (1.33) (1.33) δg1δg2δg1δg2δg1 The charge of each arbitrary variation, if one is correct should be 143 Manor O δg1 = + Θ 5 δg2 = −Ζ 5 Manor O δg1 = + Θ 5 δg2 = −Ζ 5 Manor O δg1 = + Θ 5 δg2 = −Ζ 5 δg1 = + Θ 5 δg2 = −Ζ 5 In such way that the amount of each object in the set multiplied must summed as one. In the above example, the first element is appearing three times, and the second element appearing twice, so the overall combination, we can write: In such way that the amount of each object in the set multiplied must summed as one. The Equation for Exotic Charges - 8T In the above example, the first element is appearing three times, and the second element appearing twice, so the overall combination, we can write: (+Θ 5) ∗3 + (−Ζ 5) ∗2 = 1 (1.34) (1.34) If one is correct, the first pair of exotic charges is If one is correct, the first pair of exotic charges is If one is correct, the first pair of exotic charges is If one is correct, the first pair of exotic charges is δg1 = + 3 5 δg2 = −2 5 δg1 = + 3 5 δg2 = −2 5 Such that (1 32) would be satisfied Such that (1.32) would be satisfied. Such that (1.32) would be satisfied. + 9 5 −4 5 = 1 If seven elements δg1δg2δg1δg2δg1δg2δg1 δg1 = + 4 7 δg2 = −3 7 + 16 7 −9 7 = 1 We can see that there is a pattern, first of all it takes a prime-fold quark chain to bring an element to itself. Starting from threefold combination with certain charges, the numerator is increasing by one each prime-fold chain, starting from the first threefold combination. So in order to find out the charges we need to know just how many elements are in the chain. For 𝑛1 = 1 we have threefold combination of elements, so the charges are presented in the pair 𝑛1 2𝑛1 + 1 ⟶( 2𝑛1 2𝑛1 + 1), ( −𝑛1 2𝑛1 + 1) 𝑛1 2𝑛1 + 1 ⟶( 2𝑛1 2𝑛1 + 1), ( −𝑛1 2𝑛1 + 1) For 𝑛2 = 2 𝑛2 2𝑛2 + 1 ⟶(𝟐𝒏𝟏+ 𝟏 2𝑛2 + 1), (−𝒏𝟏−𝟏 2𝑛2 + 1 ) For 𝑛3 = 3 𝑛3 2𝑛3 + 1 ⟶(𝟐𝒏𝟏+ 𝟐 2𝑛3 + 1), (−𝒏𝟏−𝟐 2𝑛3 + 1 ) For 𝑛2 = 2 𝑛2 2𝑛2 + 1 ⟶(𝟐𝒏𝟏+ 𝟏 2𝑛2 + 1), (−𝒏𝟏−𝟏 2𝑛2 + 1 ) 𝑛3 2𝑛3 + 1 ⟶(𝟐𝒏𝟏+ 𝟐 2𝑛3 + 1), (−𝒏𝟏−𝟐 2𝑛3 + 1 ) The formula to represent the charge of each prime fold chain pair is the following The formula to represent the charge of each prime fold chain pair is the following 144 𝑛𝑘 2𝑛𝑘+ 1 ⟶𝟐𝒏𝟏+ (𝒌−𝟏) 𝟐𝒏𝒌+ 𝟏 , −𝒏𝟏−𝒌−𝟏 𝟐𝒏𝒌+ 𝟏 (𝟓) 𝑛𝑘= 𝑘; 𝑘∈ℝ 𝑛𝑘 2𝑛𝑘+ 1 ⟶𝟐𝒏𝟏+ (𝒌−𝟏) 𝟐𝒏𝒌+ 𝟏 , −𝒏𝟏−𝒌−𝟏 𝟐𝒏𝒌+ 𝟏 (𝟓) (𝟓) 𝑛𝑘= 𝑘; 𝑘∈ℝ Manor O Unbounded Quarks Since it was proven that fermions are described by the term: ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (2.12) (2.12) It was possible to construct the visualization of Quark confinement as the following: The question is whether it is possible to create a scenario in which the Quark elements in the triplet is unbound. Suppose that the amount of net curvature of the first coupling term is constant, that the sea of Gluons is of finite size over time. If that assumption to hold true we can parametrize the sea of Gluons: ∑𝛿(+1)𝑖= 𝐾 𝑁 𝑖=1 (2.12. A) 𝜕𝐾 𝜕𝑡= 0 (2.12. 𝐵) (2.12. 𝐵) If we accept as an axiom that the Quarks triplet is bounded by the sea of Gluons, which is finite in size. Than in order to examine Quarks as free particles, there has to be a vanishing of the net curvature or the sea of Gluons. The vanishing can be presented by an inverse set of elements, which in physics is regarded as Anti-matter. Curvature in the orthogonal direction, in such way that the inner product of the two curvatures is zero. ⟨𝛿𝑔𝑖|−𝛿𝑔𝑖⟩= 0 Given the asymmetry in between anti-matter to matter and the over simplistic assumption of the sea of Gluons to stay as it is over time, ignoring the fact that each net curvature increasing the probability of arrival to its position on the matric tensor, it is very unlikely that such a phenomena of unbounded Quarks can be observed. That is given by two reasons, the first, if the sea is in fact finite, there must be a way to count how many Gluons are presented in between the Triplet. The second, than, we will need to find a way to take the exact inverse amount of anti-particles and inject it into the sea of Gluons, to eliminate it. As far as we understand, generating anti-particles in infinitesimal amount is almost beyond our technological reach, let alone multi-particle set. That being said, ignoring those complexities of the real world, theoretically if (1.49) is correct, it should be possible, given advanced enough technology. The Growth and Decay of Curvature Spikes 145 Manor O δg = 0 at t = Q(t + ∆t) Manor O δg = 0 at t = Q(t + ∆t) Bosons are mentioned in the first paragraph are described as net curvature, given by the term (3.13): ∑δgi M i=1 > 0 (3.13) (3.13) Now, we since they are associated with prime numbers given by the primorial coupling series –that cannot vanish into matter, their lifetime is stable and in fact infinite. They propagate all across the matric tensor, causing fermions to cluster. Overtime, more and more ripples across the matric tensor should appear, they should be weaker in the elements in the beginning of the series. The bosonic spikes are described by the equation marked in black;. δg = 0 at t = Q(t + ∆t) δg ≠0 at t = Q(t + ∆t + ∆t) ∆𝑡→0 𝛿𝑔= 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 δg = 0 at t = Q(t + ∆t) δg ≠0 at t = Q(t + ∆t + ∆t) ∆𝑡→0 𝛿𝑔= 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 𝛿𝑔= 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 The first main point of this short assay is that according to the 8T, the are two main kinds of curvature spikes, the stable ones, associated with long lifetime and independence over the matric tensor. These are Bosons, which are infinite in kind proved by the primorial. The second are the exact opposite, the spikes vanish immediately and has short lifetime. These are curvature spikes unstable, associated with fermions. Another interesting question is whether the total amount of spikes both stable and unstable grow overtime. Regarding the second kind, the Bosonic spikes, it should grow overtime as the primorial is related to the arrow of time. that does not mean that the manifold gets more curved but rather more flat, given by ratio of net to total, aspiring zero. The same assumption could be made regarding unstable curvature spikes or fermions. There should be matter creation at all stages of development of the matric tensor. The term in equation (3.13) is not limited to a certain era of time. that is similar to operators of creation and destruction in QFT but much simpler as it only contains one term. Overall, this paper objective was to describe the features of each curvature spikes in terms of their stability and longevity. The Growth and Decay of Curvature Spikes Three main ideas were presented (1) Stable curvature spikes with long lifetime are bosonic fields – independent on the matric tensor. (2) Vanishing curvature spikes of short life time – fermions. Two distinct elements, threefold combinations. 3) The matric tensor should experience curvature spikes of both kind with each stage of development. f the matric tensor increase in size, so does the amount of the spikes. Moreover, in spin representation: [(24 ∗5) + (3)] + 5 →[2N2 + 1 2] + 𝟏 𝟐 The visualization of the Boson in the theory: 146 Manor O Manor O Alternatively, Alternatively, [(840 ∗11) + (3)] + 3 + 3 + 5 Since those are equivalent to the net curvature of the fifth term, the can represent the fifth term to be a composite of nested curvature of lower magnitude. We have proven the photon to by associated with 𝑁𝑉= (+5) and the bosons of the weak interaction to be 𝑁𝑉= (+3) [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + (𝑒)] + 𝒲+ [(24 ∗5) + (3)] + 5 ⟶[(24 ∗5) + (𝑒)] + γ So the fifth term can be represented the bosonic interactions of the lower coupling terms, nested to one term 𝑁𝑉= (+11): So the fifth term can be represented the bosonic interactions of the lower coupling terms, nested to one term 𝑁𝑉= (+11): [(840 ∗11) + (𝑒)] + g + γ + γ [(840 ∗11) + (𝑒)] + g + γ + γ Two photons and one gluon nested together. Alternatively two 𝒲+ bosons (can be the minus as well or the Z boson), and one photon, nested exactly to 𝑁𝑉= (+11). Two photons and one gluon nested together. Alternatively two 𝒲+ bosons (can be the minus as well o the Z boson), and one photon, nested exactly to 𝑁𝑉= (+11). [(840 ∗11) + (𝑒)] + 𝒲+ + 𝒲+ + γ In other words, take all the composite variations by lower magnitude primes associated with bosons and represent them inside the higher term. It is possible to do with every higher term and solidify the validity of the framework as curvature is all there is. We can think about the higher terms as nested net curvature of different amount. Similar to how we can represent any point in space using a set of independent vectors, we might represent each higher coupling term by a set of independent primes nested together in different combinations. This new coupling term than is an exotic new particle with is a composition of primes of lower magnitude, so despite it is a composition it will appear as a single entity with spin one as far as one believes. Manor O Manor O Manor O Nested Curvatures 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.2) Now, we can analyze the fifth term in the coupling constants series as an example of nested curvature: Now, we can analyze the fifth term in the coupling constants series as an example of nested curvature: [(840 ∗11) + (3)] + 11 = 9254 [(840 ∗11) + (3)] + 11 ⟶[2𝑁5 + 1 2] + 1 2 [(840 ∗11) + (3)] + 11 = 9254 [(840 ∗11) + (3)] + 11 ⟶[2𝑁5 + 1 2] + 1 2 Notice that we can represent the net curvature unbound, i.e. outside of the parenthesis as the following: [(840 ∗11) + (3)] + 5 + 5 + 1 Since it has spin, the net curvature is than a vortex of certain amount: Since it has spin, the net curvature is than a vortex of certain amount: In addition, interference than could be constructed as two opposite curvature vortexes interfacing with one another. The area of cancelation is the area in which the opposite ripples on the matric tensor interest. The spinning curvature vortex is a more compete version of the phenomena of interference as it takes into account the two representations of the coupling constants series. The net curvature on the matric tensor given by equations (1.1-1.2) and the prime critical line, i.e. spin. 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) (1.1) (1.2) So now, we can visualize the phenomena of interference in the following way by the two representations: If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. ≬: 𝐴→ 𝐵 ≬: 𝐴→ 𝐵 ≬: 𝐴′ →𝐵 ≬: 𝐴′ →𝐵 Interference will accrue at the manifold segment that is mutual to both starting point as previously mentioned. ≉: 𝐴∩𝐴′ (1.42) 147 𝐸= 𝑀𝐶2 148 Manor O Einstein idea is than expressing a certain morphism between converging curvature to diverging curvature, and also from the new framework we can simplify the idea of Energy. Energy is a measure of curvature on the matric tensor. Energy converging is mass, energy diverging is synonymous to the coupling constants. Energy is absorbed and emitted in discrete amounts, isomorphic to primes or one for the coupling constant series. In contrast to Einstein theory, our definition of energy is inclusive of particle masses and of Bosons. We have proven Bosons to be net curvature on the manifold. So bosons according to our definition is diverging energy on the matric tensor, in agreement with the phenomena of photon pressure for example. The reversed process is of course possible, it is possible to combine diverging energies toward a morphism of mass. We can represent Einstein idea in a new way, maybe not calculative but calculation is not the point in the 8T as it almost merely mathematical. We can parametrized the patterns of converging and diverging curvatures 8 −(1) ⟶ℊ𝑐 8 + (1) ⟶ℊ𝑑 Curvature diverging ℊ𝑑 is equal to curvature converging, ℊ𝑐, times the square of speed of light. A new version of the Einstein equation, equation (1.9). ℊ𝑑= ℊ𝑐𝑐2 (5.1) 8T – The Sphere Shape of Stars 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 + ℳ) + 𝑃(𝐴) 𝑃𝐴# = (Κ ∗ ∏𝑃(𝐴) 𝐴=2𝑛+1 𝐴=3 ) ⟶0 Since the probability is not known, and the direction of propagation of net curvature, i.e. boson is not known, we can assume that each segment of the matric tensor in one dimension will have the same probability of net curvature reaching to it from a certain fermion entity. In other words, bosons can propagate to all directions without any laws in equal probability. Boson propagation means fermion clustering in larger and larger amounts as presented by delta function. arbitrary variations vanish in even number represented in the equations 𝛿𝑔≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) 𝛿𝑔≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡= 𝑄(𝑡+ ∆𝑡) There is always a chance net curvature will appear at later continuation of time. That is bosonic fields given by the primorial coupling series ; δg = 0 at t = Q(t + ∆t) δg ≠0 at t = Q(t + ∆t) δg = 0 at t = Q(t + ∆t) δg ≠0 at t = Q(t + ∆t) Since we have 𝑁𝑉= 𝑃(𝐴) the probability of net curvature to appear from matter cluster in a certain direction is the same for all directions, and thus the result in one dimension is a circle. 149 Manor O Take three dimensional matric tensor and the result is a sphere. The conclusion if one is correct is the following. Because the probability of emission is unknown to all directions, it means it is equal to all direction or invariant to directions. In one dimension, it is a circle that the center represents the fermion which the net curvature is propagating, and in three dimensions it is a sphere. We can state the idea in simple and elegant fashion: The sphere shape of a star is due to invariance to the direction of the net curvature propagation – i.e. bosonic fields causing fermion to cluster. Manor O Take three dimensional matric tensor and the result is a sphere. The conclusion if one is correct is the following. Because the probability of emission is unknown to all directions, it means it is equal to all direction or invariant to directions. Inner Curvatures– Where 8T and GR Differ Einstein beautiful theory of general relativity is correlating matric tensor to the Stress Energy tensor by the famous equation; 𝐺𝜇𝜈= 8𝜋𝑇𝜇𝜈 The theory implies a morphism between matter, which causes the bending of space-time, and the bending of space-time dictates the trajectory of matter. This idea is correct but only up to a certain extent. In the new 8T, the fermion cluster itself is not allowed having curvature given by (2.12) but rather it is the inner curvature within the fermion clusters that causes the bending of space-time. Einstein theory is correct in the major sense of curvature and space-time bending, but the key point and were the 8T and GR differ is the source and the nature of that bending. GR correlates to (2.12) while the 8T correlates to (3.13.B), prime amounts of distict net curvature, supported by the primorial coupling series. The inner curvatures inside the fermion cluster are deflecting linearly polarized curvature rays, not the fermion cluster itself, matter itself is not the cause for bending, what is propagating within matter is the cause of bending. Those Bosons are violations of stationarity causing matter to cluster, which is manifested in their isomorphism to prime numbers. Another major and significant difference is that in Quantum scale, we currently regard Gravitation to be a composite interaction that have infinite variations. This prediction was constructed on the primorial. While Einstein and GR regard the Gravitational constant as a constant, in the 8T it is a subject to a constant variation. That is because the structure of Gravity is preserved, i.e. invariant to different composition of net variation elements, given by the equations (2.2) and (2.3) below. [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 (2.2) [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →[2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦] (2.3) [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1…𝐾𝑁 ∈ ℝ (2.3) Another possible composition, among infinity of others: Another possible composition, among infinity of others: [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≢𝑁𝑉𝐾4 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≢𝑁𝑉𝐾4 The structure of the gravity is invariant to the change of the element. Since the total spin is invariant, i.e. two, there is not a change in the nature of gravity; the structure is the same while the composite element could be different. 8T – The Sphere Shape of Stars In one dimension, it is a circle that the center represents the fermion which the net curvature is propagating, and in three dimensions it is a sphere. We can state the idea in simple and elegant fashion: The sphere shape of a star is due to invariance to the direction of the net curvature propagation – i.e. bosonic fields causing fermion to cluster. 150 Manor O That is the reasoning the 8T suggest to the proven correct and beautiful result and prediction made by the one and only - Einstein. As was mentioned above page alongside in previous papers, 8T does not associate gravity as presented in equations (2.2) and (2.3) to long range due to vanishing spin two in net variation representation. That means that the gravitational interactions among stars is mediated by different coupling. The 8T suggested that the gravitation in long ranged is meditated by light, as photons are net curvature diverging long ranged due to spin one trait that do not vanish. Manor O those matter formations were created in the first place. The only disadvantage is 8T is not computational in a sense, other than the primorial and the mass series it seems at the verge of impossible to do calculation with the main equation of the 8T, similar to the integrations presented in QFT all over space-time. On the other hand, similar to Einstein approach, ideas are more important than calculations and a search for beauty is more important than a search for numbers. So the predictions made about light rays bending, or linearly polarized curvature rays is absolute correct, it’s the cause to that bending which need to be revised, the inner curvatures, short ranged, and isomorphic to the higher coupling terms in the primorial as many elements are varying, (also count for the weakness of gravity) which cause the bending of light, not the matter per-se. That is the reasoning the 8T suggest to the proven correct and beautiful result and prediction made by the one and only - Einstein. As was mentioned above page alongside in previous papers, 8T does not associate gravity as presented in equations (2.2) and (2.3) to long range due to vanishing spin two in net variation representation. That means that the gravitational interactions among stars is mediated by different coupling. The 8T suggested that the gravitation in long ranged is meditated by light, as photons are net curvature diverging long ranged due to spin one trait that do not vanish. those matter formations were created in the first place. The only disadvantage is 8T is not computational in a sense, other than the primorial and the mass series it seems at the verge of impossible to do calculation with the main equation of the 8T, similar to the integrations presented in QFT all over space-time. On the other hand, similar to Einstein approach, ideas are more important than calculations and a search for beauty is more important than a search for numbers. So the predictions made about light rays bending, or linearly polarized curvature rays is absolute correct, it’s the cause to that bending which need to be revised, the inner curvatures, short ranged, and isomorphic to the higher coupling terms in the primorial as many elements are varying, (also count for the weakness of gravity) which cause the bending of light, not the matter per-se. Inner Curvatures– Where 8T and GR Differ The spin two indicate short range, which agrees with the idea of inner curvature, and with the lack of detecting the graviton. The spin two vanish to an even number in net curvature representation. As equation (2.3) indicate, that is how we derived the Graviton is massless. [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 →2𝑁() + 2 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 →2𝑁() + 2 Other major differences between the GR (1918) and the 8T (2021) is that GR does not include flatness while 8T flatness is and immediate result, given by (2.12) and the main equation (1). Einstein had to insert the cosmological constant that dictates that negative acceleration. Suffice to say Einstein theory does not include any of the other interactions, while 8T predicts all under the primorial series. Other major differences between the GR (1918) and the 8T (2021) is that GR does not include flatness while 8T flatness is and immediate result, given by (2.12) and the main equation (1). Einstein had to insert the cosmological constant that dictates that negative acceleration. Suffice to say Einstein theory does not include any of the other interactions, while 8T predicts all under the primorial series. Therefore, despite 8T and GR both are assembled by manifolds and curvature as the main pillars, they also differ in incredible manners in explaining the reason for that curvature. A major difference in the spectra of phenomena both theories can provide an explanation to, 8T includes Quantum interactions alongside Cosmological formations while GR as impressive as it is does not provide an answer to how 151 Manor O Up until now, reader is probability familiar with every equation presented, as those are 8T fundamentals. From here on out, we have a completely new paper. let us revisit the last sentence: "So it is impossible to measure a photon without interfering with his nature, shifting it from wave to a particle". Is that really impossible? What if instead of a photon, which located on the prime critical line, we would measure with spin zero particle, which is not on the prime critical line. Such theoretical measurement would not vary the spin of the photon, and therefore could be a better measurement tool. Suppose it someday would become possible to measure with the Higgs, instead of the photon. We know the Higgs has spin zero, and therefore we scatter the Higgs onto the photon to perform the measurement, the result according to the primorial will look: [(24 ∗5) + (𝑒)] + γ + H0 →2𝑁2 + 1 The spin of the photon has not changed; it is invariant to the Higgs particle, as it is not on the prime critical line. We can therefore make a prediction: (1) By replacing a photon by the Higgs as a measuring tool, we could measure a photon without changing its nature, from wave like to a particle like. Higgs Particle as Tool for Overcoming Measurement Problem We partitioned and discretized the arbitrary variation term and derived the existence of Fermion. In particular, we have shown that it must have an even amount of elements, which differ in sign and create nine threefold combination, and no more than two distinct elements. δg1 + δg2 … = ∑𝛿𝑔𝑖 𝑁 𝑖=1 δg1 + δg2 … = ∑𝛿𝑔𝑖 𝑁 𝑖=1 ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (2.12) ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (2.12) (2.12) In addition, with bosons, described by the term (1.49) as they were proven discrete amount of prime curvature on the matric tensor: ∑δgi M 1=1 > 0 (3.13) ∑δgi M 1=1 > 0 (3.13) (3.13) The 8T thesis, page sixteen, the author presented the spin classification of the primorial: Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations In page fifty-five in the 8T thesis, the primorial explained the phenomena of particle wave duality, by additional photon causing a shift in the spin by additional half unit. Such a shift is leading to a spin no longer Bosonic. So it is impossible to measure a photon without interfering with his nature, shifting it from wave to a particle. [(24 ∗5) + (𝑒)] + γ + γ →2𝑁2 + 3 2 152 8T – The Action These are far from trivial statement and in complete contrast to Quantum Field Theory. Which in trying to keep the S matrix unvaried, as it is present an anti-matter particle to each particle of matter created. The problem with the QFT idea of anti-matter paring to each matter created, is that if that were the case, anti-matter would be found in much higher amounts, equal to matter in fact, and it would not be that rare to detect. Therefore, QFT idea in that sense is problematic, as we know that there exist an asymmetry in matter to anti-matter distributions toward the first.8T suggest matter creation and stationarity of the action at the same time, it is the Bosonic propagation, which violate the stationarity of the manifold. Those violations are the result, as you probability know by now, of the prime number feature, i.e. a number which is neither two nor three devisable, each prime is isomorphic to a distinct Boson. We have presented the idea of violations of stationarity in equations (1.8) and (1.81) for Fermions and Bosons respectively: the manifold is still stationary. These are far from trivial statement and in complete contrast to Quantum Field Theory. Which in trying to keep the S matrix unvaried, as it is present an anti-matter particle to each particle of matter created. The problem with the QFT idea of anti-matter paring to each matter created, is that if that were the case, anti-matter would be found in much higher amounts, equal to matter in fact, and it would not be that rare to detect. Therefore, QFT idea in that sense is problematic, as we know that there exist an asymmetry in matter to anti-matter distributions toward the first.8T suggest matter creation and stationarity of the action at the same time, it is the Bosonic problematic, as we know that there exist an asymmetry in matter to anti-matter distributions toward the first.8T suggest matter creation and stationarity of the action at the same time, it is the Bosonic propagation, which violate the stationarity of the manifold. Those violations are the result, as you probability know by now, of the prime number feature, i.e. a number which is neither two nor three devisable, each prime is isomorphic to a distinct Boson. 8T – The Action Taking the main equation (2), and not (1) (to avoid second derivatives) as the Lagrangian of the theory, and using integration to get to the action, the "Hamiltonian", we can reach an interesting option. The most significant difference between the 8T and QFT, if one is correct, is that matter can be created while keeping the manifold stationary. That is because matter pairs in such way that the result is no curvature, given by (2.12). Another way to put it, it is presented in sums two and three devisable to vanish into matter, the overall result is zero. Therefore, as long as matter is created in random fashion the manifold is still stationary. These are far from trivial statement and in complete contrast to Quantum Field Theory. Which in trying to keep the S matrix unvaried, as it is present an anti-matter particle to each particle of matter created. The problem with the QFT idea of anti-matter paring to each matter created, is that if that were the case, anti-matter would be found in much higher amounts, equal to matter in fact, and it would not be that rare to detect. Therefore, QFT idea in that sense is problematic, as we know that there exist an asymmetry in matter to anti-matter distributions toward the first.8T suggest matter creation and stationarity of the action at the same time, it is the Bosonic propagation, which violate the stationarity of the manifold. Those violations are the result, as you probability know by now, of the prime number feature, i.e. a number which is neither two nor three devisable, each prime is isomorphic to a distinct Boson. We have presented the idea of violations of Taking the main equation (2), and not (1) (to avoid second derivatives) as the Lagrangian of the theory, and using integration to get to the action, the "Hamiltonian", we can reach an interesting option. The most significant difference between the 8T and QFT, if one is correct, is that matter can be created while keeping the manifold stationary. That is because matter pairs in such way that the result is no curvature, given by (2.12). Another way to put it, it is presented in sums two and three devisable to vanish into matter, the overall result is zero. Therefore, as long as matter is created in random fashion g the manifold is still stationary. 8T – The Action We have presented the idea of violations of problematic, as we know that there exist an asymmetry in matter to anti-matter distributions toward the first.8T suggest matter creation and stationarity of the action at the same time, it is the Bosonic propagation, which violate the stationarity of the manifold. Those violations are the result, as you probability know by now, of the prime number feature, i.e. a number which is neither two nor three devisable, each prime is isomorphic to a distinct Boson. We have presented the idea of violations of stationarity in equations (1.8) and (1.81) for Fermions and Bosons respectively: ∂ℒ δgi = ∂ℒ ∂gi −∂ℒ ∂ġ i ( d dt) = 0 (1.8) ∂ℒ δgi = ∂ℒ ∂gi −∂ℒ ∂ġ i ( d dt) > 0 (1.81) (1.8) (1.81) The subject of the action taken from that point of view turns to be quite complicated and requires additional analysis. That is it because the features of the Bosonic propagations must be taken into account. If we associate the Bosonic "fields" to independent, stable curvature spikes, as the author suggest in the 8T thesis that means that the stationarity cannot be preserved, if we keep developing the main equation using Ricci curvature: ∂g ∂t = −2𝑅𝑖𝑐 ∂g ∂t = −2𝑅𝑖𝑐 Than the sign of (3.13.B) for Bosons reverse: Than the sign of (3.13.B) for Bosons reverse: 153 153 Manor O ∑δgi M 1=1 > 0 ⟶ ∑δgi M 1=1 < 0 (3.13. C) (3.13. C) If we require the condition of stationarity to be (2.12) than we can examine (3.13.C) as the term which does not interfere with the action as it is smaller than zero. So taken from this point of view, Bosons are not in violating the action as well as they now reversed in sign. It is just an idea of course, the author is not included the action in the 8T thesis as it is quite a different framework than QFT or General of the subject matter, can we created a theory in which random particles of key question relativity. The all kind appear while keeping the manifold stationary? We know we can do it for Fermions, it was proven. However, can we do it for Bosons as well? Ripping Apart Space-Time The 8T setting is a Lorentz manifold, s = (M, g), with (3,1) signature. The manifold is the connected manifold, invoked stationary, 𝑠= 𝑠0 × ℝ.. The manifold has areas of extremum curvatures that remain as they are overtime, this are yielding time invariant acceleration from them on the matric tensor M, given by two conditions below (1). The reason for the acceleration in the 8T is that the manifold is a part of an infinite packet of universes, which interact at areas of extremum curvatures, as g is the Ricci flow, and as a result flatten each other matric tensor causing it to accelerate in a time invariant rate, given by equations (1). By (2) those manifolds are topologically invariant. Flatness is an immediate result of this framework as given by the illustration below. ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ ∂2g′ ∂t2 = 0 (1) ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ ∂2g′ ∂t2 = 0 (1) ∂g ∂t = 0 ∩ 𝜕2𝑔′ ∂t2 = 0 𝜕ℒ 𝜕𝑆𝑀 − 𝜕ℒ 𝜕𝑛 = 0 (1.53) ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (1) ∂g ∂t = 0 ∩ 𝜕2𝑔′ ∂t2 = 0 𝜕ℒ 𝜕𝑆𝑀 − 𝜕ℒ 𝜕𝑛 = 0 (1.53) (1.53) ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (2.1. A) Suppose that instead of the original representation of the original coupling series, we would vary it. We could take the Boson of the first interaction and split it, to any number of sub- elements. Suppose that instead of the original representation of the original coupling series, we would vary it. We could take the Boson of the first interaction and split it, to any number of sub- elements. 8T – The Action (3.13.C) also could suggest that there is a symmetry and for each violations of positive prime, there exist a negative violation represented using the Ricci curvature. 𝑁𝑖≫1 𝑁𝑖≫1 Since those magnitudes implies Bosons stronger than the strong interaction, which do not exist or else would have been easily detected, by their effects, those fractions can not be associated with a Bosonic particle. Those fractions however are not forbidden by nature as long as they can rejoin to the formation of the original net variation, which is one. If one intuition is correct in that case, that means that space can be ripped apart at high energies, and can re-merge to original formation. That is because nature does not forbid splitting the net variation element of the strong interaction to any amount of sub elements, which correspond to much higher strength in physical meaning, which can't be a Boson. If space-time can bend, and the strongest bend is isomorphic to the strong interaction, which is one, by splitting this element and using the relation of (1.4.𝐵) we have created higher energies which can not be isomorphic to a Boson. We thus created such an immense of curvature which is diverging outward, that space time itself could be ripped apart for some summation of 𝑁𝑖. Suppose that there is a limit on this parameter, space time has been ripped apart, that ripping apart means highest amount of energy, ∂g/ ∂t = 0, since all the manifolds in the packet share that condition, which we required by the main equation, ∂g/ ∂t = 0 means we have reached the kernel, and we can jump from manifold to manifold. If one intuition is correct ∂g/ ∂t = 0 is the space in between two distinct manifolds flatting each other. It also means that at extremum low energies, space-time would be ripped apart to allow a gate to this space. Such a construction allow us to reason the physical phenomena of "light balls". Since the manifold is actually a flat surface getting flatter and flatter, so does this space must appear flat, and not varying over time, as ∂g/ ∂t = 0 means does not vary overtime. So suppose some traveler would like to travel to another point on another manifold, assuming that long enough travel would get him there, he decides to travel to a radius 𝑅, 𝑅→∞ and but that is only because he does not understand that those manifolds are very close. 𝑁𝑖≫1 A more knowledgeable traveler decides to use high energy or a natural light ball to reach the kernel, at a distance of (1/ 𝑅) from him, he gets in and within no time, he is at the point of another distinct manifold. It does not have to the inverse of 𝑅 but the idea was to the demonstrate that idea of distance does not apply within that space. It is currently unclear whether it is possible to jump from one point to another on the same manifold. If there exist two areas are extremum curvature are existing on the same manifold, it means that it is possible to jump from one to another again by changing to the kernel, which is the same for all. These ideas are so against intuition and hard to grasp as we are used to think in terms of linearity as means of reaching from one point to another. Since those magnitudes implies Bosons stronger than the strong interaction, which do not exist or else would have been easily detected, by their effects, those fractions can not be associated with a Bosonic particle. Those fractions however are not forbidden by nature as long as they can rejoin to the Manor O Manor O In addition, from here: 𝑁𝑖≫1 Ripping Apart Space-Time 8 + ∑( 1 𝑁𝑖 ) 𝐾 𝑖=1 (1.4.A) ∑( 1 𝑁𝑖 ) 𝐾 𝑖=1 = 1 8 + ∑( 1 𝑁𝑖 ) 𝐾 𝑖=1 (1.4.A) ∑( 1 𝑁𝑖 ) 𝐾 𝑖=1 = 1 However, in physics, the coupling constants as presented in the 8T, exist in the form: However, in physics, the coupling constants as presented in the 8T, exist in the form: 𝛼𝑠:𝛼𝑤:𝛼−1 →1: 30;128 𝛼𝑠:𝛼𝑤:𝛼−1 →1: 30;128 So if we split the strong into sub elements we have created in a sense magnitude which are of the order: ( 1 𝑁𝑖 ) −1 ⟶𝑁𝑖 (1.4.𝐵) 154 Manor O γ + γ = 2 That is not the case according to the primorial, so that is the same procedure taken, but on another phenomena we know exists in waves. The fact that the total spin of the two photons is less than the summed spin of each individual photon implies that there is a cancelation. The particle wave duality emphasize the total spin of the single element, but here we analyze the number of elements total spin. So the primorial clearly shows: So the primorial clearly shows: γ + γ = 3 2 thus taken from this point of view, shades light on the relation between spin and wave interference. But we can go even further, and by the difference make a prediction. According to the difference, author would make a prediction: (1) The wave summation of the two photons would lead to a cancellation of 25% the wave of each photon individually. Spin and Interference We have presented the spin classification in the 8T thesis, page seventeen, while using the prime critical line: Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations In page fifty-five in the 8T thesis, the primorial explained the phenomena of particle wave duality, by additional photon causing a shift in the spin by additional half unit. Such a shift is leading to a spin no longer Bosonic. So it is impossible to measure a photon without interfering with his nature, shifting it from wave to a particle. [(24 ∗5) + (𝑒)] + γ + γ →2𝑁2 + 3 2 [(24 ∗5) + (𝑒)] + γ + γ →2𝑁2 + 3 2 The same equation which we used for the particle-wave duality on a single photon getting scattered by additional photon can shade light on the structure of interference. That is because the above term includes two photons, normally a physicist summing their spin would expect their spin to be summed as an integer: 155 Quantum Manifolds s = (M, g) ⟶(M, g, ℱ ) 𝛗: 𝑵𝑽⟶𝓹𝒊 (𝟐. 𝟒) 𝑁𝑉= 2 (𝑉+ 1 2) ; 𝑉≥0 (1.42) 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ ℱ= ∑𝓅𝑖 𝑖=𝑁𝑣 𝑖=1 𝑁𝑉= +3 ⟶𝓅𝑖=3 𝓅𝑖∈[0,1] 𝓅𝑖∈[0,1] To each net variation element, 𝑁𝑉 there exist a parameterized unique probability of emission or absorption onto the lepton from the second term (for simplicity sake the first term is ignored) and above given by (2.4), the summation of all the probability than taken onto ℱ, which was chosen as tribute to one of the all-time greats, Richard Feynman. The new manifold can be than considered as the Feynman manifold. For simplicity sake, we will analyze the third coupling term, electromagnetism. [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ (1.44) (1.44) Suppose that the electron just emitted a Boson, a net variation of discrete prime amount, and a distinct electron has just absorbed it. The electron that just absorbed it is described by the process: Suppose that the electron just emitted a Boson, a net variation of discrete prime amount, and a distinct electron has just absorbed it. The electron that just absorbed it is described by the process: (𝑒⟵γ) →𝑒 The important point is that the electron, which absorbed and the electron that emitted, now differ in terms of their probability. The probability of the electron that absorbed is higher as it has additional term of distinct prime curvature within it. The result of all this is that we can introduce a superscript on the electron to sum the number of elements, i.e. prime bosons it has within it, and according to this number the probability of emission/absorption is varying. 𝑒⟶𝑒𝒦 156 Manor O 𝒦∈ ℝ 𝒦∈ ℝ For the electron that absorbed a photon, the new parameterization will be: For the electron that absorbed a photon, the new parameterization will be: 𝑒𝒦= 𝑒+1 𝑒𝒦= 𝑒+1 For the electron that emitted the photon the probability in the new parameterization will be: For the electron that emitted the photon the probability in the new parameterization will be: 𝑒𝒦= 𝑒0 𝑒𝒦= 𝑒0 We need to introduce a sub-script to differentiate the two electrons, so overall: 𝑒𝒦= 𝑒+1 →𝑒 1 +1 𝑒𝒦= 𝑒0 →𝑒0 0 𝑒𝒦= 𝑒+1 →𝑒 1 +1 𝑒𝒦= 𝑒0 →𝑒0 0 𝑒𝒦= 𝑒0 →𝑒0 0 The point is now that each of those leptons has a distinct probability, we need another superscript on the probability parameter to differentiate between two elements of the same coupling kind: 𝑒 1 +1 ⟶𝓅𝑖=5 𝑒=1 𝑒0 0 ⟶𝓅𝑖=5 𝑒=0 𝑒 1 +1 ⟶𝓅𝑖=5 𝑒=1 𝑒0 0 ⟶𝓅𝑖=5 𝑒=0 Since that superscript is the summation of the absorbed net curvature of distinct amount, we can easily conclude that the probability of this lepton to emit is higher, because of the summation of the superscript. That is: 𝓅𝑖=5 𝑒=1 > 𝓅𝑖=5 𝑒=0 𝓅𝑖=5 𝑒=1 > 𝓅𝑖=5 𝑒=0 That the probability of emission is higher due to the higher subscript. It is also proportional to the superscript, the higher it is, and the higher should be the probability: 𝓅𝑖∝𝒦 𝓅𝑖∝𝒦 𝓅𝑖∝𝒦 The summation of all probabilities across all the coupling terms on the manifold is manifested in the summation: ℱ= ∑𝓅𝑖 Χ 𝑖=𝑁𝑣 𝑖=1 ℱ= ∑𝓅𝑖 Χ 𝑖=𝑁𝑣 𝑖=1 The subscript is the kind of net variation: The subscript is the kind of net variation: The subscript is the kind of net variation: 𝛗: 𝑵𝑽⟶𝓹𝒊 (𝟐. 𝟒) 𝛗: 𝑵𝑽⟶𝓹𝒊 (𝟐. 𝟒) (𝟐. 𝟒) The superscript is the element which absorbed The superscript is the element which absorbed Χ = 𝑒𝑖; 𝑖∈ℝ Χ = 𝑒𝑖; 𝑖∈ℝ ℱ= ∑𝓅𝑖 Χ 𝑖=𝑁𝑣 𝑖=1 = 1 (2.41) (2.41) In (2.41) we need to sum all the elements in Χ. Χ = ∑𝑒𝑖 𝐾 𝑖=1 ⟶ Χ𝑆 This results in the final form of (2.41): This results in the final form of (2.41): 157 Manor O ℱ= ∑𝓅𝑖 Χ𝑆 𝑖=𝑁𝑣 𝑖=1 = 1 (2.41. 𝐴) (2.41. 𝐴) The end result is a varying manifold which take into account the probably of emission due to absorption, that is due to a superscript summation on the lepton. Manor O The final form of (2.41) sums over all the leptons of a certain kind which injected onto ℱ. These are Quantum manifolds, in other words. We can also make a prediction that the electron would aspire to the lowest summation on the superscript, which means to the lowest energy level possible, or to the least amount of prime distinct curvature within it. 𝑒 𝒩 𝒦⟶𝑒 𝒩 0 For some time parameter: 𝑡→∞ 𝑒 𝒩 𝒦⟶𝑒 𝒩 0 𝑡→∞ For some time parameter: Another way to state is the exact same thing: 𝜕𝓅𝑖 𝜕𝑡≠0 That is to state that leptons has a varying probability of emission over time, and if we aspire to be more brave, according to the superscript prediction, the probability of emission should be lower, and aspire zero over time. One can only consider emission has the superscript is describing how much distict prime amounts the lepton contains. The prediction about absorption seems to be a somewhat more complicated, and depends upon the expansion of the manifolds as an example, thus it will be left out of this paper. 8T – Homomorphism's [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 (2.2) [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 2] →[2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦] (2.3) [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1 𝐾𝑁∈ℝ (2.3) [𝑁𝑉𝐾1,𝑁𝑉𝐾2 ,𝑁𝑉𝐾3] ∈ ℙ 𝐾1…𝐾𝑁 ∈ ℝ Summing up, we have defined two kinds of homomorphism in the paper; the first is from prime pairs to zero: Summing up, we have defined two kinds of homomorphism in the paper; the first is from prime pairs to zero: ℧: 𝑆𝑁⟶0𝑖 (2.18) (2.18) The second is from lower composition of primes to reach a distinct higher prime: Ζ: ∑𝑁𝑉𝐾 𝑁 𝐾=1 ⟶𝑁𝑉(𝐾1+𝐾2… ) (2.19) 𝑁= 2𝑛+ 1; 𝑛∈ℝ Ζ: ∑𝑁𝑉𝐾 𝑁 𝐾=1 ⟶𝑁𝑉(𝐾1+𝐾2… ) (2.19) (2.19) Those two process are positive indication that the arrow of time is not reversible and that there is constant 'loss" of information as the manifold develops. Lost in a sense that it is impossible to retrace how we reached a certain situation, not lost in a sense that some net variation has vanished from the manifold, we have presented the conservation of variation to eliminate such scenarios. 8T – Homomorphism's The summation of the prime pairing to zero is resulting in an infinite set of zeros. That is synonymous with the vacuum idea of quantum field theory, all the prime pairs, which do not have net variation element, that is non-vanishing element, 𝑁𝑉, are the composite of the vacuum of the 8T: ∑0𝑖 𝑇 𝑖=1 = 𝒱 (2.16) (2.16) Up to this point was introduction, reader is probability familiar with every equation presented, as those are 8T fundamentals. From here on out, we have a completely new paper. Suppose we have a new zero appears, that zero could be a result of two distinct prime pairs, in that sense we don’t have an isomorphism but an homomorphism which indicate a loss for information. 0𝑖=1 ⟶(𝑝𝑁=1, 𝑝𝑁=2) 0𝑖=1 ⟶(𝑝𝑁=3, 𝑝𝑁=4) 0𝑖=1 ⟶(𝑝𝑁=1, 𝑝𝑁=2) 0𝑖=1 ⟶(𝑝𝑁=3, 𝑝𝑁=4) 0𝑖=1 ⟶(𝑝𝑁=3, 𝑝𝑁=4) The loss of information in that sense is indicating that the arrow of time is not reversible, that is because it is impossible to indicate to which pair the zero is correlated. Additional feature of loss of information is part of the primorial higher primes which composite of lower magnitude primes. In the proof of the Riemann hypothesis, author showed that primes are forming non-abelian group under addition and multiplication. The condition under addition is to have odd amount of higher primes, to reach new higher prime. Since prime are isomorphic to a Boson, we create a unique prime in more than one combination. Take as an example the prime, i.e. Boson 𝑁𝑉4 = +101, the first prime composition is: 𝑁𝑉4 = 91 + 7 + 3 𝑁𝑉4 = 31 + 67 + 3 𝑁𝑉4 = 91 + 7 + 3 The second composition is an example: 𝑁𝑉4 = 31 + 67 + 3 158 Manor O There are several unique compositions for distinct higher primes, which indicate that it is impossible to correlate an higher Bosons to a constant structure, that is in fact a major feature of gravity in the 8T, and the reason we consider it to be a time variant interaction. Manor O 𝑁𝑉4 = 91 + 7 + 3 The second composition is an example: The second composition is an example: 𝑁𝑉4 = 31 + 67 + 3 𝑁𝑉4 = 31 + 67 + 3 There are several unique compositions for distinct higher primes, which indicate that it is impossible to correlate higher Bosons to a constant structure that is in fact a major feature of gravity in the 8T, and the reason we consider it to be a time variant interaction. The non-Abelian feature of Bosonic particles indicate that is homomorphic, and there is a loss of information. Just as nature creates discrete amount of curvature on a continuous smooth setting, it also has features both Abelian and non-Abelian according to each spin classification. Bosons are non-Abelian, violations of stationarity and Fermions are vanishing curvature spikes, forming an Abelian group of two distinct elements and their product. The beauty is that we can reason for the Bosonic non-Abelian trait and do it with ease as we understand now given by the primorial how to represent them. That is only because 8T started with the ideas and theorems, derived the series and reasoned the aspiring infinity terms. If we just kept measuring magnitudes or kept searching for new particles, while adding new Bosons to the standard model, we would not be able to reason for their non-abelian nature. The key point is that there exist a time in science in which ideas exceed measurements, as measurements can not explain to us why things are the way they are. If a race has a very strong technical abilities, which manifested in highly sensitive measuring equipment, it is able to detect all the first fifty interactions, but does not know where does numbers are coming from or how many of those numbers exist, how much does it now about nature ? How much effort they invested in measurements versus a race who only needs a mathematical series and a calculator? Abelian versus Non-Abelian Since we have proven that the arbitrary variation term contain only two distinct element which vary to one another to form a group, and nine combination of two distinct elements, we can consider matter to abelian theories. Such is in fact the case as the number of combinations from the omega minus to the proton and neutron is finite. The two elements and their joint product which is the omega minus. 𝜘: 𝑇𝑜𝑝→𝑆𝑒𝑡 Τ = [δg1, δg2,δg1 × δg2] 𝜘: 𝑇𝑜𝑝→𝑆𝑒𝑡 Τ = [δg1, δg2,δg1 × δg2] Matter formations than is described by abelian group theory and a finite set of transformation on finite number of elements. In contrast to theories which tries to predict how those arbitrary variations vary such as string theory, the arbitrary variations of the terms in the set is to the other element, slight variations of each term, from itself to itself do not generate a new particle, or else the number of combinations would be immensely bigger and that is not the case. Theories of that kind are destined to fail, as if one correlates each slight variation to a new particle you are heading to infinite amount of particles and no laws of nature of any sort. δg1′ ⟷δg1 The condition of stationarity imposes a restriction, any variation of the term must be accompanied with the inverse variation on the second element in the set, so that the total series would vanish into zero. In other words, the stationarity demand (1.48) is responsible for the finite number of Fermions, and the fact that they from an Abelian group. The same does not apply to Bosons. We have used the proof of the Riemann hypothesis to demonstrate they form a non-abelian group, that is evident as the primes are infinite in kind, and each Boson is prime isomorphic. As an example of the non-Abelian features of the Bosons we created an higher Bosons as a combination of odd number of lower magnitude primes: 159 Curvature Spikes Amplitudes In page sixty six of the 8T thesis we have presented a possible variation of the primorial, replacing the electron by pi, to derive it is imperfect circle close to pi. We presented additional variation, with the net variation as demonstrated in the page below. 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉→ (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (𝜋)) + 𝑁𝑉 8 + (𝝅 𝟑) ∶(24 + (𝝅)) + 3:(120 + (𝝅)) + 5: (840 + (𝝅)) + 7 … 8 + (𝝅 𝟑) ∶(24 + (𝝅)) + 𝝅: (120 + (𝝅)) + (𝝅+ 1.82): (840 + (𝝅)) + (2𝝅 + 0.716).. 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉→ (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (𝜋)) + 𝑁𝑉 Up to this point was introduction, reader is probability familiar with every equation presented, as those are 8T fundamentals. From here on out, we have a completely new paper. we have analyzed the fact that each element in the series is weaker than the preceding to the aspiring zero ration of net to total. This is relevant because the idea one would like to present in this paper is the following: the pi terms of the net variation are representing the area of propagation and the numerical terms of the net variation such as 1.82, 0.716. ., are representing the amplitude, the height of the spike. As we keep developing the series the amplitude gets weaker and weaker, i.e. lower and the area of propagation gets wider. Highest amplitude (from the second and above to avoid the complexity of the first term) is correlative to the second term. We can make it rigorous: Up to this point was introduction, reader is probability familiar with every equation presented, as those are 8T fundamentals. From here on out, we have a completely new paper. we have analyzed the fact that each element in the series is weaker than the preceding to the aspiring zero ration of net to total. This is relevant because the idea one would like to present in this paper is the following: the pi terms of the net variation are representing the area of propagation and the numerical terms of the net variation such as 1.82, 0.716. ., are representing the amplitude, the height of the spike. As we keep developing the series the amplitude gets weaker and weaker, i.e. Manor O 𝜂𝑛𝜋⟶∞ E𝑛⟶0 𝑛→∞ 𝑛→∞ Such a classification is beneficial, as we would like to insert and include vital features as amplitudes and spikes areas, which are fundamental importance in physical theories. The terms were not possible to include in net variation, as it contain only one term and certainly not possible to include in spin representation. For those reasons, we can use the pi representation which trade off the accuracy but allows us to expand the scope of the 8T to new horizons. A beautiful visualization of the idea of the spike amplitude and area, is the water ripple illustration: will aspire lower and lower height As time goes by, the amplitude and the circular areas will get larger and aspire infinity, similar to what are primorial is indicating. This pi representation allows us to vividly observe the wave features of the primorial; the difference is that instead of water wave we have diverging curvature on the matric tensor, which is isomorphic to prime numbers or one. The prime number feature is indicating the independency of those waves and lack of dependency on matter. The aspiring zero spike is an indicator to the weakness of interaction from term to term, it can also be used to explain the particle wave-duality, the top of the spike can be viewed by an observer as a particle, while at the same time the pi multiples are the part which represents the waves. The complication is that the spike should travel with the wave itself, and that is not the case in the water illustration. However, the key point is that the pi representation allows us to make a classification according to ever-increasing spike area and ever-decreasing spikes height, which could be an analogue to wave amplitude and wave propagation in space that fills space overtime. will aspire lower and lower height Curvature Spikes Amplitudes lower and the area of propagation gets wider. Highest amplitude (from the second and above to avoid the complexity of the first term) is correlative to the second term. We can make it rigorous: 𝜂𝑛𝜋⟶∞; 𝜂∈[0, ℝ] 𝑁𝑉−𝜂𝜋 ⟶E𝑛; 𝑁𝑉−𝜂𝜋 ⟶E𝑛; The result of this idea will vary the primorial in the following way: The result of this idea will vary the primorial in the following way: 8 + (𝝅 𝟑) ∶(24 + (𝝅)) + 𝝅:(120 + (𝝅)) + (𝜂𝑛𝝅+ E𝑛):(840 + (𝝅)) + (𝜂𝑛+1𝝅 + E𝑛+1).. For the weak interaction the amplitude and the area are embedded in the term 𝝅, the classification is more vivid from the coupling term of the Electric and above. As the series develop we can see the inverse relation among the two components: 160 Manor O 𝜂𝑛𝜋⟶∞ E𝑛⟶0 𝑛→∞ 8T – Quantum Entanglement Suppose we had two photons which are propagated in the same moment in time and each photon is moving in the opposite direction of the other: [(24 ∗5) + (3)] + 5 + 5 → [(24 ∗5) + (𝑒)] + γ + γ (1.44) (1.44) For simplicity for the first time we can use subscript on the photons, we can use another notation to specify the direction of propagation in the following way. [(24 ∗5) + (3)] + 5 + 5 → [(24 ∗5) + (𝑒)] + γ1 ⃖ + γ2 (1.44. 𝐴 (1.44. 𝐴) 161 [(24 ∗5) + (3)] + 5 + 5 → [(24 ∗5) + (𝑒)] + γ1 ⃖ + γ2 (1.44. 𝐴) 161 161 Manor O Because the photons are net curvature of distinct amount, which propagate in all directions, and once liberated from the lepton are independent due to their prime number feature given by the primorial function, these curvature spikes are non-vanishing and have long lifetime. Due to the particle wave duality these can be considered particles as well, and from here we can reason the phenomena of quantum entanglement. If we consider those two entities as particles which is valid perspective according to the primorial, notice that the photon without the invariant three is described by spin one-half: [(24 ∗5) + (3)] + 5 →[2𝑁2 + 1 2] + 𝟏 𝟐 Than we have two seemingly disconnected photons in space which move in opposite direction which instantly effect each other and thus be considered as entanglement, or ghostly action at a distance. However, if we take into account the fact that the photons are net curvature diverging to all directions, even directions that the net curvature wave backward in time than these photons, no matter how far away in space are always connected. That is because there is always an intersection of the waves, so once we measure one of those two, the other is immediately modified. We have introduced the curvature code for Fermions and Bosons accordingly: Bosons accordingly: [δℓδsδM]δg = 0 [δℓδsδM]δg > 0 Which is to indicate that fermions are finite in size while Bosons vary in size overtime, that is due to the last term which in this case is used as an auxiliary condition given by equations (1.48) and (1.49) for Fermions and Bosons accordingly: put another way it is impossible to separate two photons in net curvature representation. The idea of two photons separated is an illusion of the particle picture. We can present it in a different angle, those two waves which propagate to opposite directions will always have a connection, if started at a joint point those waves will propagate outward to that point and by doing so cancel each other, as we did with interference. If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. Manor O Manor O In the context of Quantum entanglement we can modify the first two equations to present the idea of propagating to opposite directions on the matric tensor M: In the context of Quantum entanglement we can modify the first two equations to present the idea of propagating to opposite directions on the matric tensor M: ≬: γ1 ⃖ ⟶ 𝑀1 ≬:γ2 ′ ⟶𝑀2 𝑀1 ≢𝑀2 ≉: γ1 ⃖ ∩ ≬: γ2 Quantum entanglement is the result of waves intersecting and moving to all directions, including the directions which are the opposite to the trajectory of the particle in particle spin representation, those trajectories however are canceled due to another waves, this cancelation implies that the photons are always connected by some area of intersection. When we measure the first photon, we immediately measure the second as well, as they are connected by: 0 < 𝑡< ∞ Some of those ideas were mentioned before, as an example the motion of a photon in all directions including those opposite in time is mentioned by Feynman in path integrations formulation of QED. The wave features of photon is well known among all, but the key reasoning of the primorial is the following: photons are net curvature that are independent, their size increase and propagate to all directions; two photons which start at a joint point cannot be separated due to those features, the intersection means it is impossible to measure a single photon in the first place. Manor O If we define ripple operators ≬ from a starting area to another area, the mutual area of both will be the amount of interference. ≬: 𝐴→ 𝐵 ≬: 𝐴′ →𝐵 ≬: 𝐴→ 𝐵 ≬: 𝐴′ →𝐵 Interference will accrue at the manifold segment that is mutual to both starting point as previously mentioned. ≉:𝐴∩𝐴′ (1.61) (1.61) 162 Manor O Manor O way we have Bosons emitting Bosons. The emission is described by the third coupling term, given by the equivalence relation between the electron and 𝑊− Boson, W duality in short: way we have Bosons emitting Bosons. The emission is described by the third coupling term, given by the equivalence relation between the electron and 𝑊− Boson, W duality in short: 𝑊−≡(𝑒−) [(24 ∗5) + (𝑒)] + γ ≡[(24 ∗5) + (𝑊−)] + γ 𝑊−≡(𝑒−) [(24 ∗5) + (𝑒)] + γ ≡[(24 ∗5) + (𝑊−)] + γ Since the coupling constants series is commutative, we can go further and make an additional prediction: Since the coupling constants series is commutative, we can go further and make an additional prediction: [(24 ∗5) + (𝑊−)] + γ ⟶[(24 ∗5) + γ] + (𝑊−) This prediction is breaking the rules of the primorial in which we regard the invariant three to be the destabilizer that lead to a net variation, but since the coupling magnitude is preserved, it is possible to change the order of the terms themselves. The order is a reflection on the element that is being propagated. So according to the last prediction, a (𝑊−) will be propagated from a photon, a massless particle. It is a wild prediction, but at the same time very interesting, maybe even correct as the photon as energy which could morph onto mass. 8T – Objects in Class Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations 8T – Objects in Class Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Up to this point was introduction, reader is probability familiar with every equation presented, as those are 8T fundamentals. From here on out, we have a completely new paper. up to this day we have taken Bosons to be discrete amount of curvature which is prime and belong to the ring of real integers. Such views comes to an agreement with the ideas behind Quantum mechanics. The discrete amounts and the terms with one net variation are time invariant, that is contrast to composite interactions such as gravity. Composite interaction can contain many distinct combinations with keep the spin invariant. 8T – The Coupling Constants Series and the W Duality Let us analyze the first coupling term in the primorial: [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑒−] + 𝑊− [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑒−] + 𝑊− The electron and the Boson of the weak interaction are represented by the same element, the majestic three is for the electron, and the three net variations are for the 𝑊+ boson. The kernel is the three and the image is both the electron and the 𝑊− Boson. Such a duality among those two is in agreement with modern particle physics that states that the electron and the 𝑊− Bosons have the same charge. The same applies to the opposite charges. Because of the duality of those two. Because there is no numerical difference, we can replace the order as mentioned in the 8T thesis, page seventy-eight. [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑊−] + 𝑒− We can also replace the actual elements: We can also replace the actual elements: [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑊−] + 𝑊− [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑒−] + 𝑒− The coupling term with two electrons as an example, could describe the motion of free electron that could join another atom, the electron tradeoff that ignite the entire chemistry. In addition, we can make a prediction regarding photon emission. In particular, the author predict that photon can be emitted from the Boson of the weak interaction. Therefore, in a The coupling term with two electrons as an example, could describe the motion of free electron that could join another atom, the electron tradeoff that ignite the entire chemistry. In addition, we can make a prediction regarding photon emission. In particular, the author predict that photon can be emitted from the Boson of the weak interaction. Therefore, in a 163 Manor O Now, those primes were chosen in order to make a point, the total sum of the three primes is equivalent to a photon squared. [(24 ∗5) + (3)] + 25 ⟶[(24 ∗5) + (3)] + γ2 The photon squared can be decomposed to matter and anti-matter: γ = ±5 Instead of looking at the actual number, we can again state that the combination of three object in class are equal to one object multiplied and that each object in the class has an inverse object yielding a unitary object, that is synonymous with anti-matter. The universe itself is an object in a class, the class of stationary manifolds flattening each other via areas of extremum curvatures. The objects in the class of universes are also infinite and ever increasing, but the class is the same for all, it obeys the same rules and the same sub-objects, i.e. Bosons appear in the same order of each sub-element, i.e. universe. That is because the invariance of the prime ring under time shifts. We can define the class of universes; each universe is a manifold of the same class, Lorentz with (3,1) signature: 𝒮⟹{𝑠1 … 𝑠𝑛} 𝑠1 ≡𝑠2 … ≡𝑠𝑛 It is implicitly assumed that there is only one class of universes, but whether that is actually the case is unknown. However it is reasonable to assume that nature would generate the minimal number of distinct elements, which is one, rather than maximal amount of distinct elements. Another point that is important is that if each manifold in the packet would be different in kind, the flattening or the interaction between each two manifold pairs with opposite curvature orientation could lead to complications as the manifolds are different. It was implicitly assumed that nature is oriented to the minima in the kind of manifolds it generate, and if one manifold is Lorentz manifold so does the rest. Manor O Now, the author would like to analyze the 8T construction n from a different angle, which is more mathematical and maybe do not have any physical meaning, and that is representing the Bosons as objects in class. That way we ignore the numerical values as numeric and regard them as different objects of the same class. The objects differ from one other, but not in quanta but the numbers serve as classification to the object not as numerical value. That is another way to analyze the 8T construction. In a certain sense it is valid to analyze the 8T according to such view, as the Bosons are different fingers of the same hand, they are all part of the ever changing geometry of space time. At high energy they can turn into one another, as proven in the thesis. To make it more mathematically rigorous we can put the idea of Bosons as objects in class. ℬ⟹{𝑁𝑉1 … 𝑁𝑉𝐾} 𝑁𝑉1 ≢𝑁𝑉2. … The two operations we used, or can use on this class are multiplication and addition. In the operation of addition, we combined net curvature of distinct amount to reach higher distinct primes, in such way we created the gravitational term. That was the idea that allows us the associate the nature of Bosons to non-abelian group. In the context of the idea of object in class, that is mean that the class has infinite objects. In the operation of multiplication, we have combined net variations of certain amount and reached equivalence between addition and multiplication, so the author presented the idea of Quadratic curvature whose physical meaning is unclear. [(24 ∗5) + (3)] + 5 + 13 + 7 → [(24 ∗5) + (3)] + 25 [(24 ∗5) + (3)] + 5 + 13 + 7 → [(24 ∗5) + (3)] + 25 164 (𝑝𝑁, −𝑝𝑁) ∈[ℝ,−ℝ] (𝑝𝑁, −𝑝𝑁) ∈[ℝ,−ℝ] However, two important points, first such a construction would lead to immense energy release, if nature is oriented to the lowest energy state configuration, such pairs would violate it, and thus should be quite rare to detect. That was the idea which used to describe the outward acceleration by QFT, and which led to the immense difference among the observed rate of acceleration to the expected rate using that idea. The second point is that even if an element of mirrored curvature would exist, the chances of it pairing it the exact opposite are rather small, that is in contrast to the idea of the original prime pairing which impose no limitation. The bottom line is that according to each idea we can make a classification of vacuums, each vacuum differ in the amount of energy in contains, even though each vacuum is described by an infinite set of zeros, the way those zeros were created is an indication to the level of energy it contains. If each prime amount of variations is isomorphic to a net energy, than the mirrored pair with a minus sign would be isomorphic to negative energy, which could explain the rarity of those elements. The final point is that the summation of all the prime pairs and their mirrored elements is much larger than the rate of so-called "dark energy" that is by measurements made and the idea presented in QFT that led to the immense observed difference. ∑ ∞ 𝑘=1 ∑(𝑝𝑁𝐾, −𝑝𝑁𝐾) ∞ 𝑁=1 ≫𝜕2𝑔′ ∂t2 8T – High Energy Vacuums The summation of the prime pairing to zero is resulting in an infinite set of zeros. That is synonymous with the vacuum idea of quantum field theory: ∑0𝑖 𝑇 𝑖=1 = 𝒱 (2.16) (2.16) The vacuum is the result of prime pairing which do not have a net variation element, as they are not sums identical to (2.15) in their devisors. Thus, they vanish into zero. The sum of all vanishing pairs yielding zeros is the vacuum of the 8T, as presented in equation (2.16). All prime pairs appear, as previously mentioned, we can pair any even number of primes, we chose 𝑁= 2 for simplicity sake. The idea of the vacuum in this theory is somewhat hard to grasp, as it requires knowing beforehand where those violations of stationarity will appear, which is impossible to know. What is possible to know is those violations of stationarity has appeared, that is simple, where there is stars and galaxies. The vacuum idea than is more appropriate to describe in terms of short to infinitesimal time intervals, it is not a continuous entity in time.. We considered those prime pairs vanishing to a set of zero to be part of a certain domain: (𝑝𝑁, 𝑝𝑁+𝐾) = 𝑆𝑁 𝑆𝑁∈[0, ℝ] (𝑝𝑁, 𝑝𝑁+𝐾) = 𝑆𝑁 𝑆𝑁∈[0, ℝ] The idea of a vacuum can be presented in a different manner, that is by taking an a prime element and its mirrored pair, which is synonymous with matter-anti matter pairs vanishing into zero. While the original pairs vanishing into matter, which contains energy, the new pair containing one prime and it’s mirrored element would vanish to total energy. That is because it's sum is not two and three devisable but rather exactly zero, indicating a release of energy. The idea can be presented as: (𝑝𝑁, −𝑝𝑁) = 0 165 Manor O [(24 ∗5) + (𝑊−)] + γ ⟶[(24 ∗5) + γ] + 𝑊− In the 8T thesis, page twenty-nine we have presented the SEW unification by aligning the net variation element of the three couplings. That was by an exchange of two net variations from the third to the first coupling term, so all three would be at 𝑁𝑉= +3. The key point in the context of SUSY is the following, since there exist the W duality, the morphisms presented in SEW unification applies to the electron. 8 + (1) + 2: [(8 ∗3) + (3)] + 3 ∶[(24 ∗5) + (3)] + 𝟑 8 + (1) + 2 ⟶8 + (3) 8 + (1) + 2: [(8 ∗3) + (3)] + 3 ∶[(24 ∗5) + (3)] + 𝟑 8 + (1) + 2 ⟶8 + (3) The two real exchanges between the third and the first will modify the same middle element the exact same manner as presented in the 8T thesis. The only term we can vary is the left, as we want to ensure duality among the forces; we cannot touch the net variation, marked in black; [(8 ∗3) + (3)] + 𝟑. We cannot vary the invariant three; the modification will be to the left term in the coupling series; (8 ∗3) + 2 = 26 The restrictions imposed on such variation on the strong are the same as presented in the thesis. I.e. it must be to an infinitesimal interval. The physical meaning of such equivalence relation in high energy is a morphism between the Bosons. A gluon morphism the weak interaction 𝑊+, 𝑊−,𝑍 Bosons, and photon morphism to the 𝑊+, 𝑊−,𝑍 Bosons. γ ⟶𝑊+/𝑊−/𝑍 [(24 ∗5) + (𝑒)] + 𝑊+ [8 + (𝔤+ 2)] →8 + 𝑊+/𝑊−/𝑍 [(24 ∗5) + (𝑒)] + 𝑊+ [8 + (𝔤+ 2)] →8 + 𝑊+/𝑊−/𝑍 (1) At high energies there exist a morphism among the photon and the Gluon to the Bosons of the weak interaction. The Gluon at high energy can become a longer-range meditator (assuming we consider weak as longer ranged). Now take the duality relation manifested in the term: 𝑾−≡(𝒆) (𝟏. 𝟔𝟏) (𝟏. 8T – SUSY and W Duality Let us analyze the first coupling term in the primorial: [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + (𝑒)] + 𝑊− The electron and the Boson of the weak interaction are represented by the same element, the majestic three is for the electron, and the three net variations are for the 𝑊+ boson. The kernel is the three and the image is both the electron and the 𝑊+ Boson. Such a duality among those two is in agreement with modern particle physics that states that the electron and the 𝑊+ Bosons have the same charge. The same applies to the opposite charges. Because of the duality of those two. Because there is no numerical difference, we can replace the order as mentioned in the 8T thesis, page seventy-eight. [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑊−] + (𝑒) We can also replace the actual elements: [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + 𝑊−] + 𝑊− [(8 ∗3) + (3)] + 3 ⟶[(8 ∗3) + (𝑒)] + (𝑒) The coupling term with two electrons as an example, could describe the motion of free electron that could join another atom, the electron tradeoff that ignite the entire chemistry. In addition, we can make a prediction regarding photon emission. In particular, the author predict that photon can be emitted from the Boson of the weak interaction. Therefore, in a way we have Bosons emitting Bosons. The emission is described by the third coupling term, given by the equivalence relation among the electron and 𝑊− Boson, i.e., W duality: 𝑾−≡(𝒆) (𝟏. 𝟔𝟏) [(24 ∗5) + (𝑒)] + γ ≡[(24 ∗5) + (𝑊−)] + γ (𝟏. 𝟔𝟏) [(24 ∗5) + (𝑒)] + γ ≡[(24 ∗5) + (𝑊−)] + γ Since the coupling constants series is commutative, we can go further and make an additional prediction: 166 Manor O 𝟔𝟏) And the new morphisms would be: And the new morphisms would be: γ ⟶𝑒− [8 + (𝔤+ 2)] →8 + 𝑒− γ ⟶𝑒− [8 + (𝔤+ 2)] →8 + 𝑒− γ ⟶𝑒− [8 + (𝔤+ 2)] →8 + 𝑒− An extension on the main prediction: An extension on the main prediction: (1.1) Because of the Duality (1.61), at high energies there exist a morphism between The Bosons of the first and third interaction into matter. A photon into an Electron, and a Gluon into an Electron. (1.1) Because of the Duality (1.61), at high energies there exist a morphism between The Bosons of the first and third interaction into matter. A photon into an Electron, and a Gluon into an Electron. This version of SUSY does not include the Super partners of all particles but rather a morphism between particles we already know exist. The entire idea of SUSY is contained in the Primorial coupling series in the first and second representations. 167 Manor O Odd Photon Absorptions and The Graviton ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ Suppose that the photon now getting onto the electron, since the electron has no definite location but rather itself is a cloud of probability, the chances of a photon to get scattered onto the electron as a particle are rather small. However taking into account the prime number feature and the propagation all across the matric in all directions, the photon will get scattered onto the electron. When they do, we can represent the coupling term: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒)] + γ⃖ Since the photon is prime and so does the electron, they sum up to an even number as one previously covered in the thesis. Theorem (2) the photon is a net curvature of prime discrete amount, and the electron was proven the invariant three. [(𝑒)] + γ⃖ = 3 + 5 As reader probably knows by now that even amount of variations are correlated to zero, that is how we derived the existence of fermions, and how we derived the invariant three to be an Electron. [(𝑒)] + γ = 0 That is an indication that there will be a complete absorption of the photon onto the electron. We have introduced the superscript on the electron to sum the elements it contains. Manor O [(24 ∗5) + (𝑒)] + γ + γ + γ Which is exactly the structure of the Graviton, as in the 8T thesis we considered it to be the combination of three net variations, summing up to spin two. Which is exactly the structure of the Graviton, as in the 8T thesis we considered it to be the combination of three net variations, summing up to spin two. [(24 ∗5) + (𝑒)] + γ + γ + γ = 2𝑁2 + 2 If we can create a situation in which an electron will be emitting three photons at the same moment, than the result would be a Spin two particle, i.e. a Graviton which is composed of three photons. If we can create a situation in which an electron will be emitting three photons at the same moment, than the result would be a Spin two particle, i.e. a Graviton which is composed of three photons. However in order for the Electron to emit three photons it has to absorb three photons and retain them, the photons emission must be in sync to reach the desired higher spin to which we correlate the Graviton. The creation of the Graviton than is decreasing as time goes by as one theorized that the electron aspiring lowest index in the superscript over time, synonymous with the lowest state of energy, (subscript meant to index the electron itself, to classify which one absorbed and which one emitted). 𝑒 𝒩 𝒦⟶𝑒 𝒩 0 For some time parameter: 𝑡→∞ 𝑒 𝒩 𝒦⟶𝑒 𝒩 0 For some time parameter: 𝑡→∞ 𝑒 𝒩 𝒦⟶𝑒 𝒩 0 𝑡→∞ For some time parameter: That is to state that leptons has a varying probability of emission over time, and if we aspire to be more brave, according to the superscript prediction, the probability of emission should be lower, and aspire zero over time. Curvature Co-Products ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (2.1. A) Let us analyze the product of two primes, which are distinct higher primes. Odd Photon Absorptions and The Graviton 𝑒⟶𝑒𝒦 𝒦∈ ℝ For the electron that absorbed a photon, the new parameterization will be: For the electron that absorbed a photon, the new parameterization will be: 𝑒𝒦= 𝑒+1 For the electron that emitted the photon the probability in the new parameterization will be: For the electron that emitted the photon the probability in the new parameterization will be: 𝑒𝒦= 𝑒0 We need to introduce a sub-script to differentiate the two electrons, so overall: We need to introduce a sub-script to differentiate the two electrons, so overall: 𝑒𝒦= 𝑒+1 →𝑒 1 +1 𝑒𝒦= 𝑒0 →𝑒0 0 𝑒𝒦= 𝑒+1 →𝑒 1 +1 𝑒𝒦= 𝑒0 →𝑒0 0 𝑒𝒦= 𝑒+1 →𝑒 1 +1 𝑒𝒦= 𝑒0 →𝑒0 0 𝑒𝒦= 𝑒0 →𝑒0 0 Suppose the electron that absorbed now absorbed an additional two photons at the same time: [(𝑒)] + γ⃖ + γ⃖ = 3 + 5 + 5 They don’t sum to an even number, which indicate that the absorption of a photon cluster by a single electron could not exceed one photon at the time. Alternatively that the photon can absorb only odd number of photons together: [(𝑒)] + γ⃖ + γ⃖ + γ⃖ = 3 + 5 + 5 + 5 = 0 Notice that if the absorption of three net variations is possible and complete, so does the emission of three net variations, now for the Electric coupling term assume instead of absorption we would have an emission. [(𝑒)] + γ + γ + γ [(𝑒)] + γ + γ + γ 168 Manor O According to the author proof of the Riemann hypothesis, primes form a non-abelian group, the condition under addition is to have an odd amount of primes under addition. Similar to the idea of quadratic curvature, we can present the idea of co-products of net curvature under multiplication, not in any case of multiplication by two primes will yield a prime, it could yield an odd, suppose we consider the cases it will yield a prime. 𝑁𝑉1 ⟶ 𝑁𝑉1 × 𝑁𝑉2 ⟵𝑁𝑉2 𝑁𝑉1 × 𝑁𝑉2 ∈ ℙ 𝑁𝑉1 × 𝑁𝑉2 ≢𝑁𝑉1 ∪ 𝑁𝑉2 𝑁𝑉1 ⟶ 𝑁𝑉1 × 𝑁𝑉2 ⟵𝑁𝑉2 𝑁𝑉1 × 𝑁𝑉2 ∈ ℙ 𝑁𝑉1 × 𝑁𝑉2 ≢𝑁𝑉1 ∪ 𝑁𝑉2 𝑁𝑉1 × 𝑁𝑉2 ≢𝑁𝑉1 ∪ 𝑁𝑉2 In between each net variation element, we can define an automorphism arrow from itself to itself: 1𝑎∶ 𝑁𝑉1 ⟶𝑁𝑉1 1𝑎∶ 𝑁𝑉2 ⟶𝑁𝑉2 1𝑎∶ 𝑁𝑉1 ⟶𝑁𝑉1 1𝑎∶ 𝑁𝑉2 ⟶𝑁𝑉2 Since at high energies we can align the net variations elements on the same value, which in the case of the first three interactions lead to alignment at 𝑁𝑉 = +3, and thus unification at 24 + 2 variations. 8 + (1) + 2: [(8 ∗3) + (3)] + 3 ∶[(24 ∗5) + (3)] + 𝟑 The two real exchanges between the third and the first will modify the same middle element the exact same manner as presented in the 8T thesis. The only term we can vary is the left, as we want to ensure duality among the forces; we cannot touch the net variation, marked in black; [(8 ∗3) + (3)] + 𝟑. We cannot vary the invariant three; the modification will be to the left term in the coupling series; We cannot vary the invariant three; the modification will be to the left term in the coupling series; (8 ∗3) + 2 = 26 (8 ∗3) + 2 = 26 169 Manor O The physical meaning of such equivalence relation in high energy is a morphism between the Bosons. A gluon morphism the weak interaction 𝑊+, 𝑊−, 𝑍 Bosons, and photon morphism to the 𝑊+, 𝑊−,𝑍 Bosons. γ ⟶𝑊+/𝑊−/𝑍 [(24 ∗5) + (𝑒)] + 𝑊+ [8 + (𝔤+ 2)] →8 + 𝑊+/𝑊−/𝑍 [8 + (𝔤+ 2)] →8 + 𝑊+/𝑊−/𝑍 it is possible to build an additional arrow from one independent element to another. it is possible to build an additional arrow from one independent element to another. Δ1 ∶ 𝑁𝑉1 ⟶𝑁𝑉2 Δ2 ∶ 𝑁𝑉2 ⟶𝑁𝑉1 Δ1 ∶ 𝑁𝑉1 ⟶𝑁𝑉2 Δ2 ∶ 𝑁𝑉2 ⟶𝑁𝑉1 In general form: In general form: In general form: Δ ∶ 𝑁𝑉𝐾⟶𝑁𝑉𝑀 Δ ∶ 𝑁𝑉𝐾⟶𝑁𝑉𝑀 Before the morphism they were distinct: 𝑁𝑉𝐾≢ 𝑁𝑉𝑀 𝑁𝑉𝐾≢ 𝑁𝑉𝑀 𝑁𝑉𝐾≢ 𝑁𝑉𝑀 In the context of category theory, we have a category that has one class of objects, which vary in two ways, from themselves to themselves, from themselves to other objects in the class that are distinct, and are able to morph via addition and multiplication to other object in the class. The setting of category theory makes it simpler to understand the nature of Bosons, via arrows and morphisms, both are presented in the thesis in a mere numerical form. SEW unification in page twenty-nine, is a manifestation of the features of category theory and the morphisms among elements. 170 Variational Fermion Distributions & Dark matter ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) Variational Fermion Distributions & Dark Manor O Let us analyze the idea of dark matter. Author submitted two options in the 8T thesis. The first in pages (25-28) which has to do with the Quark masses series. Such a series was derived as a result of trying to eliminate the question of three families, and by simple rescale of the third and the first a pattern was found. The rate of convergence to zero however is rather fast and thus already in the fifth family we reach total mass aspiring zero. The rate of convergence to zero could indicate that it is not suffienct to be the sole cause of dark matter. The second explanation in thesis pages (123-124), used to universe packet representation, i.e. a gravitational effect from a distinct manifold. In this paper, author is going to expend on the second explantiaion, which is variational matter distributions of distinct manifold of the same class. Let us use the universe packet pairs, with opposite curvature orientations, which flatten each other. Let the arbitrary variations distributions be identical in amount but different across each area of extremum curvature such as galaxy. As an example one took a set of two distinct Lorentz manifolds, which invoked stationary: ℘= [𝑠1 , 𝑠2] 𝑠𝑛= (𝑀, 𝑔) 𝑠𝑛= (𝑀, 𝑔) To each associate a finite number of arbitrary variations which vanish into matter, isomorphic to each other. This cluster of arbitrary variations vanish into a galaxy. ∑𝛿𝑔𝑖 𝑀 1=1 ∈𝑠1 ∑𝛿𝑔𝑖 𝐾 1=1 ∈𝑠2 𝐾≡𝑀 ∑𝛿𝑔𝑖 𝐾 1=1 ~ ∑𝛿𝑔𝑖 𝑀 1=1 (1.62) ∑𝛿𝑔𝑖 𝑀 1=1 ∈𝑠1 ∑𝛿𝑔𝑖 𝐾 1=1 ∈𝑠2 𝐾≡𝑀 ∑𝛿𝑔𝑖 𝐾 1=1 ~ ∑𝛿𝑔𝑖 𝑀 1=1 (1.62) (1.62) This cluster has the same amount of matter. The key point is that despite the clusters are isomorphic the Fermionic distributions could be different. If one manifold has a star like earth at the matric, that fact does not mean that the complimentary manifold has a star at the exact place, or any star at all. The exactness condition using that framework does not include exact matter distributions but rather isomorphic number of arbitrary variations in the total cluster. Variational Fermion Distributions & Dark (2.1. A) Manor O different states, similar to QFT idea in which the phases are independent from the amplitudes, the phases are signaling to a degree of rotation, 8T equivalent would be certain degree of acceleration due to curvature on the manifold. We defined the manifold as the variable "s" so the isomorphism can be put in rigor: △: 𝑠→𝑠 𝑠= (𝑀, 𝑔) 𝑠= (𝑀, 𝑔) The notion of isomorphism's, of an object which vary but still stay as is, is the mathematical equivalence of physical co-variance, as one believes. In that context, another interesting analog is of the similarly of a mechanical system. In particular is that the equation of motion is left unaltered if multiplied by constant, which applies to cases in which the potential energy is an homogenous function of the co-ordinates. Suppose we would multiply the equations of manifold variation by some factor, ∑𝛿𝑔𝑖∗𝐾 𝑁 𝑖=1 Assuming manifold invoked stationary: ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (1.48) ∑𝛿𝑔𝑖∗𝐾 𝑁 𝑖=1 = 0 ∑𝛿𝑔𝑖∗𝐾 𝑁 𝑖=1 Assuming manifold invoked stationary: ∑𝛿𝑔𝑖= 0 𝑁 𝑖=1 (1.48) ∑𝛿𝑔𝑖∗𝐾 𝑁 𝑖=1 = 0 Assuming manifold invoked stationary: The stationarity condition is not altered by scalar multiplication; matter can be created in higher amounts while keeping the manifold stationary. From that argument, we can extract that the potential energy of the manifold is an homogenous function of the matric tensor. That is very different than the QFT formalism which require for each matter created, anti-matter creation keeping the S matrix unvaried. If that was in fact the case, anti-matter would be quite common in the universe which is not the case. That idea of matter anti-matter pairs vanishing to zero, led to the massive difference in estimating the source of dark energy. 8T allows creation of matter while keeping the manifold stationary, matter pairs in such way that no curvature is allowed, that is by the anti-commutation relation of fermions. The result of this construction is that energy is not conserved. That is because matter can morph into energy, and arbitrary variations of the manifold vanish into matter. Those ideas could be somewhat hard to accept, similar to how the discovery of Planck and Heisenberg principle were hard to accept at the time. In certain sense If the QFT idea was right, anti-matter would be as common as matter, as those pair to each other for each matter particle created, which is not the case. Manor O Define one fermion distribution at an interval: ∑𝛿𝑔𝑖 𝑀 1=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑖 𝐾 1=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 ∑𝛿𝑔𝑖 𝑀 1=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑖 𝐾 1=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 If one considers that variational distributions are not identical while the clusters of total variations are isomorphic, the result is invisible matter traces within each manifold. It is more reasonable to assume the distributions are different rather to assume they are identical at the a star of a scale. Is it possible to assume that two equal amounts of gas projectile at the same space would diffuse the exact same molecule distributions? We can only take the average in such cases and state that the total distributions must be identical overtime, i.e. spread over the limits of the space. Those invisible matter distributions are the role of dark energy according to this idea. Other universes is not a question anymore, 8T correlate the major features of our own universe to their existence. Dark energy is given by universe packet, as a result flatness as well. Those are proven, agreed upon measurable facts, i.e. the flatness and dark energy, which can not be solved assuming there exist only one universe. It could be even possible 171 Manor O to estimate the distance between each two universe assuming we know the added gravitational effect added by dark matter. to estimate the distance between each two universe assuming we know the added gravitational effect added by dark matter. 8T – Isomorphism's and Covariance ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) 8T – Isomorphism's and Covariance ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (2.1. A) Let us analyze the idea of co-variance in this framework. The paper will present two roots in which co- variance rise. The idea behind the co-variance is the following sentence – there is only one set of rules. In the context of variational manifolds, there exist only one manifold. The idea of co-variance can firstly be analyze via the notion of isomorphism. In this context, isomorphism would be to state that the manifold is the same manifold; the difference is that observer may watch the manifold in different configurations, i.e. 8T – Spin Chronicles ∑δgi M 1=1 > 0 (3.13) (3.13) We have presented the spin classification in the 8T thesis, page seventeen, while using the prime critical line: Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Let us analyze the idea of spin. As one can see, each Boson is firstly described by spin, which is not an integer, i.e. one half. That is an indication to it's particle like traits which are preserved. Than since each coupling term is containing an additional term on the prime critical line, which is the invariant three, the spin total reaches to one. Since the invariant three is always there, the minimal spin which Bosons from the second coupling term and above will have is one. Since the Higgs does not have this invariant three which is the sole generator of net variations as we now believe, it is represented by spin zero. Than if an additional photon is getting scattered onto the photon emitted the spin of the system than again varies, which is presented in the 8T thesis: [(24 ∗5) + (3)] + 5 + 5 →[2𝑁2 + 1 2] + 1 2 + 1 2 = [2𝑁2 + 3 2] The 8T emphasized the spin variation by half unit as a result of measurement, changing the photon from wave like to a particle like, at first used in the context of super symmetry. Such a view of analysis than indicate that the super-partners are not needed. That is because the same particle represents spectra of behaviors according to spin variations. It is the same particle. Later view of SUSY showed that using the SEW unification it is a possible to predict a morphism from the photon onto the electron, as it is isomorphic to the W boson of the weak interaction, same as for the Gluon. Therefore, at high energies Bosons can turn into matter, without any need for super partners. Additional important point, which were not mentioned before. First, since each photon contain energy of certain amount, it is impossible to measure with it without interfering with the experiment. That is the analog of QM principle of uncertainty with regards to time an energy. Even a measurement with a Higgs will interfere with the system as the Higgs contains quanta of energy given it has a positive mass. Manor O Manor O Let ℱ be the summation of net variations of all kind, which is a function of time. In the 8T framework: Let ℱ be the summation of net variations of all kind, which is a function of time. In the 8T framework: ℱ= ∑𝓅𝑖 Χ 𝑖=𝑁𝑣 𝑖=1 The subscript is the kind of net variation: The subscript is the kind of net variation: φ: 𝑁𝑉⟶𝓅𝑖 (2.4) φ: 𝑁𝑉⟶𝓅𝑖 (2.4) (2.4) The superscript is the element that absorbed The superscript is the element that absorbed The superscript is the element that absorbed Χ = 𝑒𝑖; 𝑖∈ℝ Χ = 𝑒𝑖; 𝑖∈ℝ Χ = 𝑒𝑖; 𝑖∈ℝ Χ = 𝑒𝑖; 𝑖∈ℝ Over time, the probability of violations increases as more matter is being created: ℱ𝑡< ℱ𝑡+∆𝑡 ℱ𝑡< ℱ𝑡+∆𝑡 Manor O The second quantity which is not conserved is the number of violations, as matter being created in larger amounts, there exist higher probability of violations which are prime (or one) amount of curvature to arise from it, to despite the manifold is isomorphic to itself, the number of elements in the subgroup of violations is ever increasing. 𝑠= (𝑀, 𝑔, ℱ) 172 Manor O 8T – Spin Chronicles Although the Higgs will not interfere with the spin, it will interfere with the environment of the experiment, making the energy of the system larger than before. 173 Manor O [(24 ∗5) + (𝑒)] + γ + 𝐻0 ⟶2𝑁2 + 1 𝐻0 ⟺𝐸𝑛 𝐸𝑛> 0 So despite the Higgs does not affect the system at the spin level, it does vary the energy. The invariance of total spin due to the Higgs can be put using an automorphism of the coupling term: 𝐻0: 2𝑁2 + 1 ⟶2𝑁2 + 1 The second important point is the following, at the heart of it all, each Boson (weak and above) start with spin one-half. Than due to additional element it receives total spin one. It is possible to classify the behavior of Bosons according to the number of elements in the coupling term. Odd number of elements in the coupling term would lead to a behavior of a particle, while even number would lead to an integer spin, a wave like behavior; we take into account the invariant three and the outer Bosons, manifested as prime outside the bracket. Define the summation of elements on the prime critical line: ∑𝒫𝑖= 𝒮 𝑁 𝑖=1 2 | 𝑁⟶𝑇𝑟𝑢𝑒 Than wavelike: Than wavelike: 𝒮⟶𝔚 Else, it would be particle like Else, it would be particle like Since the lepton is represented as the invariant three, which is also the element used to describe the Boson of the weak interaction, which could either behave like a particle or a wave, depends upon the total elements on the critical line, so does the electron. That is given by the previously mentioned equivalence relation: 𝑊−≡(𝑒−) (1.61) Interactions Separation & The Gravitational Class Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Interactions Separation & The Gravitational Class Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Let us analyze the idea of separation of the forces, i.e. Bosonic net variations. We have previously mentioned that the direction of the arrow is the direction of time. The strongest interactions appear at the first, the result is an endless process of clustering, with weaker and weaker interactions, given by ratio of net to total. 174 Manor O 1 > 3 30 > 5 128 > 7 850 … When the author derived the coupling series back in March 2021, each coupling term was analyzed by extracting the net variation element. In the following page, presented the original part of the construction of the pre-equation idea which yielded the primorial: Manor O 1 > 3 30 > 5 128 > 7 850 … Manor O When the author derived the coupling series back in March 2021, each coupling term was analyzed by extracting the net variation element. In the following page, presented the original part of the construction of the pre-equation idea which yielded the primorial: Analyze the (7,11) total variations pair with NV = (+1): Total variations sum is divisible by two: Total variations sum is divisible by two: 18/2 = 9 And then by three And then by three 9/3 = 3 We know that we have 𝑁𝑉 = (+1) so it can be extracted to yield: 𝐹1 = 8 + 1 However, even amounts of variations vanish so we can ignore the element 8 and write: 𝐹1 = 1 Notice that the first interaction can be represented the following way: [𝑊−+ γ + 𝔤 ] = 9 (1.62) (1.62) That is because each Boson is isomorphic to a prime, that was speculated on theorem two, pre-equation idea: That is because each Boson is isomorphic to a prime, that was speculated on theorem two, pre-equation idea: 𝑊−≡ +3 γ ≡+5 𝑊−≡ +3 γ ≡+5 so already in the first term we can represent the Bosons of the rest of the two interactions. Since we have taken it out from the total sum, we have separated it from the mixture of the three interactions. The Boson of the weak interaction can be either one of the three. We have taken 𝑊− as it has the same charge as the electron, fact that the author used for the SUSY construction. Manor O Electroweak combined term. In contrast to other theories that aspire to describe Gravity moment of separation, the 8T does not include Gravity directly, as it is a composite interaction that has infinite variations, assuming we use the spin representation of the main equation. [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≢ 𝑁𝑉𝐾4 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 𝑁𝑉𝐾3 ≢ 𝑁𝑉𝐾4 The structure of the gravity is invariant to the change of the element. Since the total spin is invariant, i.e. two, there is not a change in the nature of gravity; the structure is the same while the composite element could be different. [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾3 →2𝑁() + 2 [(2𝑁()) + (3)] + 𝑁𝑉𝐾1 + 𝑁𝑉𝐾2 + 𝑁𝑉𝐾4 →2𝑁() + 2 the last interesting twist, the fact that the three terms are combined in one equation without Gravity, is already means that Gravity has broken, as the term (1.62) contains three distinct amount of net curvature, put another way, those interactions are just different amounts of gravity, manifested in different primes. So the term already contain the separation of ''gravity'' if the author reasoned clearly enough. To put this confusing idea another way, Gravity is the class and the interactions are different objects in the class, so the fact that we have different objects in the class means that the class it is not unified, i.e. that the class is separated or "broken". (1.62) indicate that these are the interactions that were separated as they exactly match the Bosons of two next interactions. To express the idea of the Gravity class we can define a set of net variations: ∧:𝑇𝑜𝑝⟶𝑆𝑒𝑡 ∧:𝑇𝑜𝑝⟶𝑆𝑒𝑡 𝐺𝑐𝑙𝑎𝑠𝑠= {𝑁𝑉 |𝑁𝑉 ∈ ℙ ∪ (+1)} (1.63) (1.63) Interactions Separation & The Gravitational Class Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations So now after the Gluon was taken out: [𝑊−+ γ + 𝔤] ⟶[𝑊−+ γ] + 𝔤 [𝑊−+ γ + 𝔤] ⟶[𝑊−+ γ] + 𝔤 Next, the term in the parenthesis represents the Electroweak Bosonic combinations. This term create total sum of an even, and thus we took it to zero in the 8T thesis. However here we have two independent Bosons, so each can be set in his way. After the Gluon was separated, the electroweak combination is needed to be separated, to the electric Boson, the photon and the weak interaction Bosons. The combination of all the three also implies higher state of energy and as each interaction stands on its own, the energy is lower. So already we can reason why the forces should be separated. [𝑊−+ γ + 𝔤] ⟶𝐸𝑛+ 𝐸𝑛+1 + 𝐸𝑛+2 = 𝐸3𝑛+3 𝐸3𝑛+3 > 𝐸𝑛 ∩𝐸𝑛+1 ∩𝐸𝑛+2 [𝑊−+ γ + 𝔤] ⟶𝐸𝑛+ 𝐸𝑛+1 + 𝐸𝑛+2 = 𝐸3𝑛+3 𝐸3𝑛+3 > 𝐸𝑛 ∩𝐸𝑛+1 ∩𝐸𝑛+2 That is to state that the SEW combined term has a higher energy state and each of the Bosons in itself, which indicate nature aspires to "break the combination" of the SEW into separated elements. it is quite a remarkable fact that the Bosons of the first three interactions sum exactly to the term of the first interaction to equation (1.62). Therefore, the 8T predicts that first the strong is separated than the 175 Manor O γ + γ + 𝑊−= 13 13 ∈ ℙ γ + γ + 𝑊−= 13 That Boson is associated with the sixth coupling term: (120,120 + (3)) + 13 = 120,136 Assuming that Boson has a lifetime that is unknown but still present approximately stable behavior, using two Bosons of the weak interaction we can reach again to an higher Boson: 13 + γ + γ = 23 So the paper present two major ways in which those Bosons can be detected. First by colliding hadron formations to reach bigger formations. That lepton propagation and from there net variation from the Lepton. The second, once Bosons are at the same space they can morph into higher coupling Bosons in odd combinations, given by the primorial. Those new coupling Bosons are in somewhat of a Superposition of Bosons, they composed by combinations of distinct Bosons. That means that it is possible to detect them based on set of possible decays. As to test, the 8T author will provide a set of possible decays, which can be used to examine the 8T primorial, each Prime is isomorphic to a Boson: 𝑊−+ 𝑊−+ γ = 11 𝑊−+ 𝑊−+ 𝔤= 7 𝑊−+ 𝔤+ 𝔤= γ 𝑊−+ 𝑊−+ 𝑊−+ 𝔤+ 𝔤= 11 γ + γ + 𝑊−= 13 13 + 7 + 3 = 23 23 + 13 + 𝑊−= 49 49 + 𝑊−+ γ = 57 𝑊−+ 𝑊−+ 𝑊−+ 𝔤+ 𝔤= 11 γ + γ + 𝑊−= 13 13 + 7 + 3 = 23 23 + 13 + 𝑊−= 49 49 + 𝑊−+ γ = 57 49 + 𝑊−+ γ = 57 And on we go endlessly, since there exist an equivalence relation between the electron and the Boson of the second interaction, it is possible to replace them without changing the result. That equivalence relation was at the heart of SUSY variation of the 8T, which based upon aligning the net variations on the same term of 𝑁𝑉 . 𝑾−≡(𝒆) (𝟏. 𝟔𝟏) (𝟏. 𝟔𝟏) The morphisms ignore the fact that the gluon is "confined" within the hadron. However as one can see, the morphisms can occur without the gluon as a direct participator, we can regard the 𝑁𝑉 = +7 to be an independent element and not nested by lower magnitude primes (and one). High Energy Paradox Now we are left with the question of detection of those higher coupling terms that are getting weaker as the arrow develops. Reason might indicate that to detect in order to detect those weaker interactions, civilization ought to examine weaker and weaker energies. Author will argue that the opposite is the case. To detect those weaker interactions we need higher energies. That is the case, as there exist a clear pattern given by the primorial. The clusters of variations, excluding the invariant three and the 𝑁𝑉 element are getting larger from term to term: 2𝑁1 < 2𝑁2 < 2𝑁3 Put more elegantly 2𝑁𝐾⟶∞ K ∈ℝ 2𝑁1 < 2𝑁2 < 2𝑁3 Put more elegantly 2𝑁𝐾 are the cluster of total variations which vanish into matter. That means that the increase the chance of detecting the weaker coupling, we need to collide hadrons in such way that the cluster of total variations will grow accordingly. That implies we need to collide many hadrons rather than just a few. Since each hadron has energy, as it has mass, that is synonymous with high energy collusion. Than once the cluster has been created, there exist an unknown probability of emission of an electron, i.e. the destabilizer. According to the right multiplier of the 2𝑁𝐾, which belongs to the prime, the Boson than will appear. At high energies, suppose one of those Bosons was detected. It is immediately creating an option for a creation of an higher magnitude Boson. As an example: 176 Manor O Summing up, the larger the energy the higher the chance to observe those weaker coupling terms, the high energy collusions also creating a setting in which those higher coupling terms can morph into even higher coupling terms. Alone those terms are stronger, but as they are part of a much bigger cluster they get weaker from term to term. The strongest term has the smallest cluster, as previously mentioned in the 8T, as dictated by the arrow of time. 0.111 > 0.1 > 0.039 > 0.008 … 1 9 > 3 30 > 5 128 > 7 850 … 177 Manor O N-Tuples ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) Manor O N-Tuples (2.1. A) Let us analyze the idea of an N-tuple, which is an ordered set of elements. To start the analysis one will change the setting using functor, from a varying manifold onto a set. ∧:𝑇𝑜𝑝⟶𝑆𝑒𝑡 Each prime pair used in the original derivation now stand as a set of elements. One mentioned that the number of primes could be infinite, for simplicity sake, two was chosen. The criteria one was asking is the prime pair sum to be two and three devisable. So to put the idea mathematically: Ν = (𝒫1 … . 𝒫𝑛) Ν𝒮= (𝒫1 + 𝒫2+. . 𝒫𝑛) [2, 3] | Ν𝒮 Ν = (𝒫1 … . 𝒫𝑛) Ν𝒮= (𝒫1 + 𝒫2+. . 𝒫𝑛) [2, 3] | Ν𝒮 We have defined the Boson to be a prime amount of net variation, which arises from the N-Tuple, in an amount proportional to the average of the N-Tuple. Ν𝒮 𝑁 ∝𝑁𝑉 to expend the idea of a Boson using N Tuple, we can associate the Boson to be a product type of the N- Tuple. The Boson is the product of the N-Tuple that satisfy the devisors requirements. So to put it mathematically. NV = ∏δgϕ ϕ=NV ϕ=1 = δgϕ=1 × δgϕ=2 × δgϕ=3 (1.23) (1.23) δgϕ=1 ≡δgϕ=2 ≡δgϕ=3 Since in contrast to Fermions we have only one sign for Boson, positive summation of curvature, or negative if we consider: 178 178 Manor O ∂g ∂t = −2Ric (1.23.A) (1.23.A) So according to this idea, the subscript is mere index that counts the number of times the arbitrary curvature is chained. Manor O The difference among the Bosons is the number of chained arbitrary variations. As an example the difference among the photon with 𝑁𝑉= +5 and the 𝑊− with 𝑁𝑉= +3 would be: So according to this idea, the subscript is mere index that counts the number of times the arbitrary curvature is chained. The difference among the Bosons is the number of chained arbitrary variations As an example the difference among the photon with 𝑁𝑉= +5 and the 𝑊− with 𝑁𝑉= +3 would be: W−= δgϕ=1 × δgϕ=2 × δgϕ=3 γ = δgϕ=1 × δgϕ=2 × δgϕ=3 × δgϕ=4 × δgϕ=5 δgϕ=4 ≡δgϕ=5 ≡δgϕ=1 δgϕ=4 ≡δgϕ=5 ≡δgϕ=1 since all the terms are prime number multiples, if we consider (1.63) they are negative amount of curvature, the difference between each Boson is the number of elements in the term of (1.23). that is the analog of the original idea made in March. The prime pairs are N-tuples, which has products of prime type of the same element in different amount, which is not devisor of two and three as fermions. [2, 3] | ≠∏δgϕ ϕ=NV ϕ=1 (1.24) So using the idea of N-tuples it is easier to grasp the difference among Fermions, which arise in even numbers of two distinct elements which differ in sign and summed as zero, as (2.12) indicate. From those summations, product type may rise, which are proportional to the average size of the summation, those product type has one element which has a negative sign, and the different products differ in the number of times this element is multiplied, as (1.23) indicate. Since the number of repetition is prime, the more repetitions we have, the weaker the element, since the photon has five times versus three times for the W Boson, it yields, assuming 𝛿𝑔𝜙∈ℝ a bigger negative number, or if we assume that 𝛿𝑔𝜙< 1 and positive, a smaller positive number as the repetitions increase. Those assumptions rely of the idea that 𝛿𝑔𝜙 is a curvature spike which can be quantified. Observables -8T ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (2.1. A) Let us analyze the idea of observables. We know from Quantum mechanics that observables obey a certain operator relations: Let us analyze the idea of observables. (3,3) (3, 5) (3,7) (3,11), (3,13) … (5, 3) (5,5) (5,7) (5,11) (𝟓,𝟏𝟑) … (7, 3) (7,5) (7,7) (𝟕, 𝟏𝟏) (7,13) … … . (29, 19)(29,23), (29, 29), (𝟐𝟗,𝟑𝟏)… All the pairs, which are not marked in yellow, are in commutation relation, i.e. they vanish into zero. As an example: As an example: (7, 3) −(3,7) = 0 The pairs marked in yellow do not commute as they have a non-vanishing element, either prime or one propagating from them. That was the idea which yielded the primorial. (7,11) − (11, 7) = +1 (29,31) −(31,29) = +3 Those elements propagating from those pairs are violations of stationarity, which are discrete quanta of prime curvature on the matric tensor. Since those quanta's contain energy, and they obey seemingly no law in which we can predict their propagation, it is than impossible to measure the energy of a system at such scales, or even at all. Not only because of the lack of commutation of those pairs, but also because the manifold arbitrary variations vanish into matter, which has innate energy, given by its mass and mass and energy relation. Therefore, the picture is the following: matter is created by the requirement of a stationary manifold, it does not contradict this condition as it pairs in such way that no curvature is allowed. However, matter is potential curvature, as it is composed by quarks, and for that reason energy can not be conserved. QFT suggested that for each particle of matter created, an anti-particle is also created and so they annihilate each other. That idea implying that there exist equal amounts of matter and anti-matter in the universe, which contradict the experimental data, anti-matter, is much rarer. So that is quite paradoxical, matter creation while the manifold is stationary (due to the anti-commutation relation), and at the same time energy is not conserved, for two reasons: because anti-matter and matter asymmetry and because matter has a potential energy, assuming it has mass. Manor O We know from Quantum mechanics that observables obey a certain operator relations: [𝐴̂, 𝐵̂] = 𝐴̂𝐵̂ −𝐵̂𝐴̂ = 0 And in cases they do not obey the relation, the result is: And in cases they do not obey the relation, the result is: 𝐴̂𝐵̂ −𝐵̂𝐴̂ = 𝑖ℎ 𝐴̂𝐵̂ −𝐵̂𝐴̂ = 𝑖ℎ Since we did not use any operators nor we did not use the Planck constants in the theory we need an analog for the idea of observables. To do just than we can replace the bracket used in Quantum mechanics by a prime pair, if those two pairs are vanishing onto zero, i.e. they commute, if they don’t, they have a non-vanishing element, isomorphic to a prime or one. 179 String Theory - The Devil's Gift - 8T Let us analyze the idea that stands at the heart of String theory. If one understood correctly, each "mode of vibration" of the string is isomorphic to a particle of certain kind. The more volatile the vibration the higher the energy. The string has infinite potential geometrical combinations and knots. That core idea according to the 8T author is wrong. First, if it was correct there would not be a standard model, as slight variations of the string would account in a bound state of the proton and the neutron, they would not form an abelian group but rather a non-abelian group with infinite kind of particles. Second, it is impossible to derive the action or the Lagrangian of such a theory, as it is impossible to derive which state out of the infinite set of states of the string should have minimal energy or considered stationary string. It is not promised that the string will stay at the lowest state of vibration. String theory is impossible to work with, its core idea is flawed. It is impossible to make any sort of prediction 180 Manor O assuming that is the case, let alone any laws of nature or reasoning. If that theory was correct there would be infinite bound state of matter, not just nine combinations as found with the omega minus. It has been built almost 60 years ago, many average brained physicists worked on it and the result is no testable predictions were made at all. To map the scope of theory, one will have to map all over the combinations of the string and associate each geometrical pattern to a particle. How can that even be done? that idea is ridiculous as there are infinite variations, an suppose that was correct, the variation of the string is impossible to predict, so even if one can map N combinations, each combinations have infinite morphism options. They could be more volatile, such variations would than yield measureable variations in the coupling magnitude, or alternatively infinite coupling magnitudes. If it were correct, we would measure infinite bound states of the electron, photon, and hadrons. It is the opposite of a Lagrangian oriented, as it implies nature would generate infinite couplings with no reason behind it, in contrast to the 8T which gives a exact reason to the magnitudes of the couplings.. According to string theory each universe (as they have many solutions, called the 'landscape') could have a different set of laws, that is ''standard model of its own". How many laws nature would generate? Why bother to create so many set of distinct particles for each universe, it is a instead of just one?. In the 8T because of the invariance of the prime ring, the same magnitudes, and Bosons appear at all the manifolds at the same order, the dominating forces are depended upon the unique arrow of the manifold, as it gets older the weaker interactions are more common, as it close to singularity the strong forces are dominating creating the hadrons. All manifolds have Quarks in them in decreasing total mass order, all universes are obeying the same laws, nets are primes and interacting via areas of extremum curvatures flattening each other, and of the same kind. There exist one equation for all, one principle - a varying manifold in a packet. That is it and to prove it the primorial was yielded. Manor O Is there a stronger equations than (1.2) or (1) in string theory in terms of the spectra and accuracy of predictions? To put it another way, String theory is the devils gift. It is a dead end theory, which makes simple things complicated as build upon the implicit assumption that those infinitesimal things, i.e. particles are important. It shifts the center of attention to the object rather to the principles and to the ideas, it gives no testable predictions, it is long and hard to comprehend, it predicts no law, it is not Lagrangian oriented in any way, and many worked on it without aligning it with anything in the particle scales, nor in the cosmological scales. It should have no place in physics anymore, 8T is far superior by all means, predictions wise it has much more predictions, which are correct to date, and length wise it is much shorter, it is also easier to comprehend. 8T was built upon one subject of descripting which yielded all the other infinitesimal quantities and the equation which results in Dark energy and flatness, string theory is built upon describing an infinite set of states of a varying infinitesimal object that yielded not a single prediction in sixty plus years, in any scale infinitesimal nor cosmological and it uses measured values such as Planck and the speed of light to reach the idea of a string rather than deriving those parameters and ideas from pure thought, as presented in the 8T. assuming that is the case, let alone any laws of nature or reasoning. If that theory was correct there would be infinite bound state of matter, not just nine combinations as found with the omega minus. It has been built almost 60 years ago, many average brained physicists worked on it and the result is no testable predictions were made at all. To map the scope of theory, one will have to map all over the combinations of the string and associate each geometrical pattern to a particle. How can that even be done? that idea is ridiculous as there are infinite variations, an suppose that was correct, the variation of the string is impossible to predict, so even if one can map N combinations, each combinations have infinite morphism options. Manor O They could be more volatile, such variations would than yield measureable variations in the coupling magnitude, or alternatively infinite coupling magnitudes. If it were correct, we would measure infinite bound states of the electron, photon, and hadrons. It is the opposite of a Lagrangian oriented, as it implies nature would generate infinite couplings with no reason behind it, in contrast to the 8T which gives a exact reason to the magnitudes of the couplings.. According to string theory each universe (as they have many solutions, called the 'landscape') could have a different set of laws, that is ''standard model of its own". How many laws nature would generate? Why bother to create so many set of distinct particles for each universe, it is a instead of just one?. In the 8T because of the invariance of the prime ring, the same magnitudes, and Bosons appear at all the manifolds at the same order, the dominating forces are depended upon the unique arrow of the manifold, as it gets older the weaker interactions are more common, as it close to singularity the strong forces are dominating creating the hadrons. All manifolds have Quarks in them in decreasing total mass order, all universes are obeying the same laws, nets are primes and interacting via areas of extremum curvatures flattening each other, and of the same kind. There exist one equation for all, one principle - a varying manifold in a packet. That is it and to prove it the primorial was yielded. Is there a stronger equations than (1.2) or (1) in string theory in terms of the spectra and accuracy of predictions? To put it another way, String theory is the devils gift. It is a dead end theory, which makes simple things complicated as build upon the implicit assumption that those infinitesimal things, i.e. particles are important. It shifts the center of attention to the object rather to the principles and to the ideas, it gives no testable predictions, it is long and hard to comprehend, it predicts no law, it is not Lagrangian oriented in any way, and many worked on it without aligning it with anything in the particle scales, nor in the cosmological scales. Manor O 0 ⟶1 2 ⟶1 ⟶2 ⟶⋯ In a certain sense, if we consider the Boson as a separate entity, it has spin one-half, but the invariant present of the majestic three than ensures that the Boson will summed as spin one. In other words the electron fields and the Boson fields are entangled in such way that is ensuring the spin of the Boson to be one. Since the primorial is time invariant, given by the invariance of the coupling terms under temporal shifts, so does this relation. So now instead of the classification of spin we can represent each coupling term to be a mixture of spins of distinct kind. Define spin operator: 𝒷: [ℝ, ℚ] ⟶𝒮[ℝ,ℚ] (1.64) (1.64) That is a mapping between the integers and non-integer fields onto a spin operator in such way that the spin is matching the subscript: That is a mapping between the integers and non-integer fields onto a spin operator in such way that the spin is matching the subscript: 𝒷: 0 ⟶𝒮0 𝒷: 1 2 ⟶𝒮1/2 𝒷: 1 ⟶𝒮1 Now each coupling term can be represented in a way that reflect the idea of interacting fields, which are represented by spin operators. [(24 ∗5) + (𝑒)] + γ ⟶ 𝒮0 + 𝒮1 2 + 𝒮1 [(24 ∗5) + (𝑒)] + γ ⟶ 𝒮0 + 𝒮1 2 + 𝒮1 Manor O It should have no place in physics anymore, 8T is far superior by all means, predictions wise it has much more predictions, which are correct to date, and length wise it is much shorter, it is also easier to comprehend. 8T was built upon one subject of descripting which yielded all the other infinitesimal quantities and the equation which results in Dark energy and flatness, string theory is built upon describing an infinite set of states of a varying infinitesimal object that yielded not a single prediction in sixty plus years, in any scale infinitesimal nor cosmological and it uses measured values such as Planck and the speed of light to reach the idea of a string rather than deriving those parameters and ideas from pure thought, as presented in the 8T. 8T – Interacting Fields ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn k/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) (2.1. A) We have presented the spin classification in the 8T thesis, page seventeen, while using the prime critical line: We have presented the spin classification in the 8T thesis, page seventeen, while using the prime critical line: Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations pin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Let us analyze the idea of interacting fields. Instead of the spin classification presented above, we can consider the coupling terms as expressions that contain all the "fields" or the distinct kind of particles in the term itself. Those fields are not only interacting but generating each other in succession, in such way: 181 8T – Five-Vectors We have presented the spin classification in the 8T thesis, page seventeen, while using the prime critical line: Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations So now instead of the classification of spin we can represent each coupling term to be a mixture of spins of distinct kind. Define spin operator: 𝒷: [ℝ, ℚ] ⟶𝒮[ℝ,ℚ] (1.64) (1.64) That is a mapping between the integers and non-integer fields onto a spin operator in such way that the spin is matching the subscript: That is a mapping between the integers and non-integer fields onto a spin operator in such way that the spin is matching the subscript: 𝒷: 0 ⟶𝒮0 𝒷: 1 2 ⟶𝒮1/2 𝒷: 1 ⟶𝒮1 182 Manor O Manor O Now each coupling term can be represented in a way that reflect the idea of interacting fields, which are represented by spin operators. Now each coupling term can be represented in a way that reflect the idea of interacting fields, which are represented by spin operators. [(24 ∗5) + (𝑒)] + γ ⟶ 𝒮0 + 𝒮1/2 + 𝒮1 (1.44. 𝐴) Let us analyze the idea of four vectors, which are closely related to invariance. Since in our framework we have an even number of universes aspiring infinity in the universe packet (1.2A), the vector must specify which universe out of the packet the motion occurs in, which represents by the parameter 𝑠𝑛. 𝔏: [𝑥, 𝑦, 𝑧. 𝑡] ⟶[𝑥, 𝑦, 𝑧. 𝑡, 𝑠𝑛] [(𝑥, 𝑦, 𝑧. 𝑡) ∈𝑠𝑛] 𝔏: [𝑥, 𝑦, 𝑧. 𝑡] ⟶[𝑥, 𝑦, 𝑧. 𝑡, 𝑠𝑛] [(𝑥, 𝑦, 𝑧. 𝑡) ∈𝑠𝑛] So there as to be a clear specification between space and distance. When motion occurs an object coordinate varies in a certain space, the variation depends upon the frame of reference. So there as to be a clear specification between space and distance. When motion occurs an object coordinate varies in a certain space, the variation depends upon the frame of reference. Since the frame of reference also is effected by the distribution of matter on the universe, it has to be taken into account. So coordinate variation in fourth vectors has to do with relativistic motion, but surface variation is the new idea. 8T – Manifolds Heredity Let us analyze the idea of heredity. We have presented a manifold creation as a curvature spike departing from the original manifold, the curvature spike immediately gets flatten by equation (1.2A) as it is part of the packet. The process of flatting due to the packet is the reason for the acceleration at all stages. We have presented the variation of the Dirac Delta. So the Dirac delta in 8T describe the process in which arbitrary amount of curvature appear, and vanish into matter. However, there is no restriction with regard to Time. Arbitrary amount of curvature can appear at any time, so we must modify the idea of the Dirac in our framework. 𝛿g ≠0 𝑎𝑡 𝑡= 𝑄(𝑡) 𝛿𝑔= 0 𝑎𝑡 𝑡1 = 𝑄(𝑡+ ∆𝑡) At later continuation of time: 𝑡2 > 𝑡1 This condition is satisfied: 𝛿g ≠0 𝑎𝑡 𝑡2 = 𝑄(𝑡+ ∆𝑡+ ∆𝑡) At later continuation of time: This condition is satisfied: 𝛿g ≠0 𝑎𝑡 𝑡2 = 𝑄(𝑡+ ∆𝑡+ ∆𝑡) 𝛿g ≠0 𝑎𝑡 𝑡2 = 𝑄(𝑡+ ∆𝑡+ ∆𝑡) Moreover, the amount of variations is either prime or one: 𝛿𝑔= 2 (𝑉+ 1 2) ; 𝑉≥0 The process of manifold creation can be put as means of an arrow: The process of manifold creation can be put as means of an arrow: 𝜍: 𝑠𝑛⟶𝑠𝑛+1 The process of birth has an heredity condition, the new manifold must have the same number of dimensions as the original manifold and must possess the same traits, i.e. be a simply connected manifold and a complete manifold. In other words, we can add an additional superscript with the (3,1) signature to the newborn manifold, to ensure the heredity condition. 𝜍: 𝑠𝑛 (3,1) ⟶𝑠𝑛+1 (3,1) So that is in agreement with the idea of the multiverse as presented in the 8T thesis, i.e. infinite set of surfaces, each with a finite dimensions of it's own. The heredity condition prevents the theoretical scenario in which the newborn manifold will possess a higher number of dimensions or alternatively that the new manifold will not by complete or simply connected. Such is needed as for simplicity sake, if each newborn manifold is of a different class than the packet process of flattening could result in complications. Nature is satisfied with simplicity as the primorial is indicating. What is simpler than generating only one class of manifolds? Manor O Manor O 8T – Five-Vectors This idea imposes a new constraint, that in order to describe the motion we have to specify which universe it occurs. It is also possible to construct the jumps across the manifolds in the packet as a non-linear motion, where you travel in three dimensions, jump to an higher or lower surface and move on those dimensions with a distinct arrow. [(𝑥, 𝑦, 𝑧. 𝑡) ∈𝑠𝑛] ⟶[(𝑥, 𝑦, 𝑧. 𝑡) ∈𝑠𝑛+1] 𝑠𝑛≢𝑠𝑛+1 Since those surfaces have different matter distribution in jumping from surface to surface one must ensure that the trajectory chosen does not have any Fermionic obstacles which manifest themselves in the new surface and had no equivalent in the original surface which the jumped occurred. The idea of variational Fermion distributions was presented by arbitrary variations vanishing into matter, in identical amount but in different configuration, and is considered as one of the two explanations for dark matter: ∑𝛿𝑔𝑖 𝑀 1=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑖 𝐾 1=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 𝐾≡𝑀 ∑𝛿𝑔𝑖 𝑀 1=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑖 𝐾 1=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 𝐾≡𝑀 𝐾≡𝑀 Summing up, when we consider motion, the fourth vector must become a five-vector, which specify surface on which the motion occurs. 183 Dark Matter is a Must We can represent the second variation of the main equation as the following ∑𝛛𝓵 𝛛𝐬𝐦 𝐊/𝟐 𝐦=𝟏 𝛛𝐬𝐦 𝛛𝐌 𝛛𝐌 𝛛𝐠 𝛛𝐠 𝛛𝐭𝛅𝐠𝒎−∑𝛛𝓵 𝛛𝐬𝐧 𝐊/𝟐 𝐧=𝟏 𝛛𝐬𝐧 𝛛𝐌 𝛛𝐌 𝛛𝐠 𝛛𝐠 𝛛𝐭𝛅𝐠𝒏= 𝟎 (𝟐. 𝟏. 𝐁) That construction is the following. Each manifold has area of extremum curvatures on it. Those extremum curvatures are surrounded by arbitrary variations, which vanished into matter. Up to this point, it was covered. For each arbitrary amount of variation absorbed onto the ∂g/ ∂t there is a radiation emitted from the area of extremum curvature to ensure it does not vary over time. those galaxies than has areas of extremum curvatures, and arbitrary variations around them, that is matter clusters, spiraling around those areas. The galaxy as an area of extremum curvature is getting flattened by another galaxy of the same magnitude and different matter distribution. As was previously analyzed: ∑𝛿𝑔𝑚 𝐾/2 𝑚=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑛 𝐾/2 𝑛=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 ∑𝛿𝑔𝑚 𝐾/2 𝑚=1 ≡∑𝛿𝑔𝑛 𝐾/2 𝑛=1 ∑𝛿𝑔𝑚 𝐾/2 𝑚=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑛 𝐾/2 𝑛=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 ∑𝛿𝑔𝑚 𝐾/2 𝑚=1 ≡∑𝛿𝑔𝑛 𝐾/2 𝑛=1 ∑𝛿𝑔𝑚 𝐾/2 𝑚=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑛 𝐾/2 𝑛=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 ∑𝛿𝑔𝑚 𝐾/2 𝑚=1 ≡∑𝛿𝑔𝑛 𝐾/2 𝑛=1 The key point of this paper which was not presented before is the following – if we require the manifold to have areas of extremum curvatures and time invariant acceleration away from those areas, than by (2.1.B) we also require a set of arbitrary variations vanishing onto matter, around the matter of our own galaxy. The amount is the same, the distribution is different, invisible matter is also an immediate result of the main equation. Now take an infinite set of manifolds in the packet and the matter within one distinct manifold is now at the position of a minority as it is only belong to one manifold. Not only the main equation represent matter formation in our manifold, it also represent matter creation in other manifolds. While areas of extremum curvature flattening each other, matter is constantly being created in different distribuends across all manifolds, and as a result accounts for what is speculated as invisible matter. It could be explain that way, that the areas of extremum curvature alone with the matter distribution of one distinct manifold is not sufficient for holding the condition ∂g/ ∂t = 0 . Manor O Dark Matter is a Must Manor O Manor O 8T – Manifolds Heredity it is reasonable argument to claim that nature is Lagrangian oriented in the number of manifold classes it generates. 184 𝒳: 𝕄⟶𝔅𝕄 𝒳: 𝕄⟶𝔅𝕄 The Gluon, W and Z Bosons and the photon has the relation according to the above operator: 𝔅𝕄=0: 𝔅𝕄>0: 𝔅𝕄=0 The author will make a prediction that the fourth coupling term according to the primorial will possess a positive mass. That is the fourth coupling term will described by 𝔅𝕄>0. By the ratio of the net variations of the fourth and the second, the fourth term is describing bigger Quanta than the Quanta of the Weak interaction. In particular, the mass of the fourth coupling term should be 2.1333 higher than the Masses of the Bosons of the weak interaction. Dark Matter is a Must However, an infinite set of pairs forming the packet is sufficient. That means that in order for the stationarity condition of the manifold to hold, one must have at least two distributions of arbitrary variations vanishing onto matter, in different configurations. Dark matter than is a must and is just regular matter signature of a distinct manifolds. that matter is creating the additional gravitational effect ensuring the ∂g/ ∂t = 0 condition of stationarity. The formation of matter than can be described by the Quark masses series, which indicate nature is devising in increasing amount to eliminate those arbitrary variations. with the arrow of time, families with total mass which is lighter and lighter is formed but the structure of the families is the same, i.e. two distinct elements which differ in sign and create threefold combinations. Put another way the main equations describe an infinite distinct sets of matter creations, and the Quark masses series indicate their total mass direction, assumed same for all. 185 The Bosonic Mass Pattern Let us analyze the idea of the Bosonic mass pattern. The Boson of the first coupling term is described by one number, and we know its massless. The second coupling Bosons has a positive mass. The second coupling also differ by an additional term from the first coupling. The Boson of the third coupling, i.e. the photon, is again considered massless, even though it is a carrier of energy and can exhort pressure. The subject of this paper is the following question: can we predict a pattern in which the primorial coupling series generate mass ?. That is, is there a discrete jumps between massless to mass positive. Author is going to assume there is a mass pattern of that kind. Define the Boson mass by the parameter: Strong Interaction Paradox We have presented the idea of Quark confinement using that framework, stating that each net curvature is increasing the probability of arrival to its position. The end result is endless cluster of Gluons inside the hadron, which causes the Quark triplet to position on the lowest point on the curve. At high energy trying to break the triplet is synonymous with trying to roll the distinct arbitrary amount of curvature up hill, or to flatten the curve. The illustration below is the Quark triplet before collusion, as presented in the 8T thesis, page sixty-nine. At high energy, there exist a hadron collusion that leads to immense increase of the energy, which is synonymous with trying to roll the Quarks up-hill: As the number of Gluons in the cluster is infinite, the illustration is not accurate, as the Quarks seems to be separated and to reach the height of half of the curve. In reality the moments the triplet are separated in hadron collusion, they will aspire again to the lowest point on the curve, in other words, they will accelerate toward one another. That is the elements that differ sign toward one another. The At high energy, there exist a hadron collusion that leads to immense increase of the energy, which is synonymous with trying to roll the Quarks up-hill: At high energy, there exist a hadron collusion that leads to immense increase of the energy, which is synonymous with trying to roll the Quarks up-hill: As the number of Gluons in the cluster is infinite, the illustration is not accurate, as the Quarks seems to be separated and to reach the height of half of the curve. In reality the moments the triplet are separated in hadron collusion, they will aspire again to the lowest point on the curve, in other words, they will accelerate toward one another. That is the elements that differ sign toward one another. The As the number of Gluons in the cluster is infinite, the illustration is not accurate, as the Quarks seems to be separated and to reach the height of half of the curve. In reality the moments the triplet are separated in hadron collusion, they will aspire again to the lowest point on the curve, in other words, they will accelerate toward one another. That is the elements that differ sign toward one another. Manor O second illustration can be represented in a different manner, somewhat resembles relativity in different frames of reference. The Quark triplet is the same, it’s the curve itself that is getting flattened. The fact that the curve is flattened means that the quarks are less bounded that before, and at a stage where the curve is completely flat the Quarks are no longer confined. Such a construction can reason for the fact that the strong interaction is getting weaker at high energy, as the curve itself is effected by an additional amount of matter, assuming the curve is negative given by the Ricci flow: second illustration can be represented in a different manner, somewhat resembles relativity in different frames of reference. The Quark triplet is the same, it’s the curve itself that is getting flattened. The fact that the curve is flattened means that the quarks are less bounded that before, and at a stage where the curve is completely flat the Quarks are no longer confined. Such a construction can reason for the fact that the strong interaction is getting weaker at high energy, as the curve itself is effected by an additional amount of matter, assuming the curve is negative given by the Ricci flow: ∂g ∂t = −2𝑅𝑖𝑐 ∑δgi M 1=1 > 0 ⟶ ∑δgi M 1=1 < 0 (3.13) (3.13) That term (1.49.A) is effected by positive energy given by (2.12) which is the colliding hadron, so overall the negative amount of net curvature decreases, causes the curve to get flatten, after the collusion, the net curvature which retained, will causes the extra net curvature to reach its positon and the original curve will be retain, with the original Quark triplet locked to the minima. So at high energy the term representing the strong interaction is a weaker term, i.e. the 8 + (1) and not just the one. At high energies than→9: 30: 128 The idea of the curve flattening due the positive energy on the hadron and the negative sign of the curve is given by the below illustration: Strong Interaction Paradox The As the number of Gluons in the cluster is infinite, the illustration is not accurate, as the Quarks seems to be separated and to reach the height of half of the curve. In reality the moments the triplet are separated in hadron collusion, they will aspire again to the lowest point on the curve, in other words, they will accelerate toward one another. That is the elements that differ sign toward one another. The As the number of Gluons in the cluster is infinite, the illustration is not accurate, as the Quarks seems to be separated and to reach the height of half of the curve. In reality the moments the triplet are separated in hadron collusion, they will aspire again to the lowest point on the curve, in other words, they will accelerate toward one another. That is the elements that differ sign toward one another. The 186 Manor O 8T – Multiverse Uncertainties Now the subject of the paper is the following. What is the nature of nature in terms of certainty. Despite 8T is able to put under one equation some of the major questions of modern physics, such as dark energy, flatness and dark matter, which are direct results of the multiverse as presented in (2.1. A ) and (2.1. B), how much can we predict really ? 8T can predict the coupling magnitudes and all the numbers nature will ever generate, which is a significant step forward. At the same time, there are many uncertainties, which go beyond the conjugate relation of QM, which famously known between momenta and position or time and energy. The first uncertainty is the uncertainty of decays. Given higher term coupling, The Boson can be represented as a nested Boson, composed of lower primes. It is impossible to derive which combination is serving the actual decay, and the higher the coupling the larger the possible combinations of Bosonic decays. The idea is presented in page one hundred forty seven in the 8T thesis. The second uncertainty revolves around the arbitrary variation term (2.12) which vanish onto matter. It is impossible, as far as one can see, to derive the amount of matter being created at each moment, nor it is possible to derive where those arbitrary variations will appear, that can only be done in retrospective, where starts and galaxies are at. These are two major uncertainties. It is also impossible to estimate how fast those arbitrary variations form into matter, what is the time segment, is it same for all or varying according to the surface and the amount of matter that was already there? A third uncertainty is regarding the matter configuration on other universes, it is not possible to predict the configuration, the rate or the position of matter on other universes, other than stating that the matter on those manifolds is identical to matter on our own manifold, skeleton wise, i.e. Quarks. 187 Manor O ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t δg𝑚−∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t δg𝑛= 0 (2.1. 8T – Multiverse Uncertainties B) ∑𝛿𝑔𝑖 𝑚 𝑖=1 ⟶ℛ𝑠1 ∈[0,1] ∑𝛿𝑔𝑖 𝑛 𝑖=1 ⟶ℛ𝑠2 ∈[0,1] ℛ𝑠1 ≢ℛ𝑠2 ∑𝛿𝑔𝑖 𝐾/2 𝑚=1 ≡∑𝛿𝑔𝑖 𝐾/2 𝑛=1 Manor O ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t δg𝑚−∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t δg𝑛= 0 (2.1. B) (2.1. B) The last uncertainty is the uncertainty of class. We assumed that the heredity condition, which state newborn manifolds will belong to the same class but that is not proven. We also assumed that the families on other manifolds will have the exact same families in terms of total mass, given by the Quark masses pattern, which in contrast to the primorial, took the measured values and used them to define the pattern, where in the primorial the measured values were derived from a variational principle. The other universes have the same form of matter in terms of the kind, i.e. Quarks, but are the masses identical? Is there a principle involved which guarantees those numbers will form? To examine those assumptions a civilization could have two options: The first is to find the principals involved which the kind of manifolds arise, and the numbers of the masses arise, without measurement. The principles than must match the measurement, the second is to jump across the packet and measure the traits of matter on each distinct manifold. We assumed that nature is Lagrangian oriented, and generate the minimal number of laws, minimal kind of manifolds and minimal sets of distinct fermion masses, but it is just an educated guess after all. In contrast to the invariance of the prime ring that ensures that the same Bosons will appear at the same order, the Fermion and manifold classes is still requires work. That picture indicate that at least part of the laws of nature are identical across the multiverse, the same coupling constants will appear at all, matter and Quarks will appear at all, leptons will appear at all the same way. All universes will contain galaxies, and will be flat. However, are they all of the same kind? Are the Fermions masses identical for all? The last uncertainty is the uncertainty of class. We assumed that the heredity condition, which state newborn manifolds will belong to the same class but that is not proven. Primorial as a Wave Equation Let us analyze the idea of a positional primorial, which is constructed to provide a solution to the question of position. To solve the issue of position the author will use a five- vector on the subscript. The three spatial coordinates, one temporal and the index of the manifold on the packet. 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉⟶(8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇 Assuming the even sum has vanished into matter, which means it stands for the nuclei, from which the electron propagate, we than are required to insert the subscript to that term as well. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇 (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) The time is indexed as to represent the idea of unique arrow to each manifold. The additional subscript on the primorial allows us to expend the idea and include positional variables with temporal variables. Since the 8T was built upon an Euler Lagrange framework which yields an equation of motion that is invariant to changes of coordinate so does the primorial is invariant, and assumed relativistic. The invariance of the primorial is also due to the invariance of the prime ring. In no frame of reference does the primes change their order nor their innate values. Any observer clever enough will find that the primes are at the heart of the coupling magnitudes, does not vary if measured from that coordinate or another. That is because in all frames of reference arbitrary variations are vanishing onto matter, while primes are possessing a non-vanishing nature, which is a feature universal to all observers, i.e. all matter clusters of distinct amounts. We can go further and state a morphism similar to the term presented in the non-relativistic Schrodinger equation: 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉𝑠𝑛 (1.35) (1.35) Which means that instead of a wave propagation in space, we have net curvature ripple propagation in space. As the curvature ripple propagates over larger matric tensor surface, the ripple gets weaker and weaker. That is somewhat a more modern version of the Coulomb law and the Newton law, which regard force to be inversely proportional to the square of the distance. Manor O Manor O 8T – Multiverse Uncertainties We also assumed that the families on other manifolds will have the exact same families in terms of total mass, given by the Quark masses pattern, which in contrast to the primorial, took the measured values and used them to define the pattern, where in the primorial the measured values were derived from a variational principle. The other universes have the same form of matter in terms of the kind, i.e. Quarks, but are the masses identical? Is there a principle involved which guarantees those numbers will form? To examine those assumptions a civilization could have two options: The first is to find the principals involved which the kind of manifolds arise, and the numbers of the masses arise, without packet and measure the traits of matter on each distinct manifold. We assumed that nature is Lagrangian oriented, and generate the minimal number of laws, minimal kind of manifolds and minimal sets of distinct fermion masses, but it is just an educated guess after all. In contrast to the invariance of the prime ring that ensures that the same Bosons will appear at the same order, the Fermion and manifold classes is still requires work. That picture indicate that at least part of the laws of nature are identical across the multiverse, the same coupling constants will appear at all, matter and Quarks will appear at all, leptons will appear at all the same way. All universes will contain galaxies, and will be flat. However, are they all of the same kind? Are the Fermions masses identical for all? Finding out the principle in which the specific masses arise from a variational principle identical to primorial, and this specific class of manifold arise from all potential classes of manifolds, could be the biggest challenge facing modern theoretical physics, and the last pillar to reach the completion of the unified theory, 8T. At the current stage, the theory is able to explain the major what's and one of the two major whys, the why of the coupling magnitudes,. The "why" which still requires work is the "why" of those masses. What is the variational principle leading to those numbers? Is there one at all? 188 Primorial as a Wave Equation Alternatively, if we consider that the Laplace operator already contain the unique manifold, 𝑠𝑛 ∈∇2 than (1.32) takes the simpler form: 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 (1.36) ∂2g ∂tn2 = ∂2g ∂x2 + ∂2g ∂y2 + ∂2g ∂z2 ∂2g ∂x2 + ∂2g ∂y2 + ∂2g ∂z2 ∈M ∈ sn (1.36) ∂2g ∂x2 + ∂2g ∂y2 + ∂2g ∂z2 ∈M ∈ sn 189 Fermionic and Multiverse Superposition's Given the fundamentals, (2.12) and (1), it is possible to analyze the subject of linearity for Fermions. Suppose that in two distinct locations arbitrary variations are vanishing into matter. Both have an even amount of arbitrary variations. The combination of the two is also a solution, which keeps the manifold stationary, as it yields an even number that vanishes into matter. That is similar to the superposition of states, the idea of a linear differential equation, which serve a crucial role in Quantum mechanics. The following can be put mathematically: ∑δgi + ∑δgi = 0 𝑁2 i=1 𝑁1 i=1 𝑁1 ≢𝑁2 2 |𝑁1 ∩𝑁2 ∑δgi + ∑δgi = 0 𝑁2 i=1 𝑁1 i=1 𝑁1 ≢𝑁2 2 |𝑁1 ∩𝑁2 2 |𝑁1 ∩𝑁2 That is both are devisors of two, i.e. even number of arbitrary variations. than the described outcome is the summation of the two to be an even number which keeps the manifold stationary, i.e. no curvature is allowed: (𝑁1 + 𝑁2) = 𝑁1+2 2 |𝑁1+2 ⟶0 (𝑁1 + 𝑁2) = 𝑁1+2 2 |𝑁1+2 ⟶0 The idea of linearity can be expressed in another manner, that is by pair of two distinct manifolds flatting each other, combined with another pair of two distinct manifold is also a solution of (1.2.A) and how the packet is constructed. ∂ℓ 𝜕𝑠1 − ∂ℓ 𝜕𝑠2 = ℨ1 = 0 ∂ℓ 𝜕𝑠3 − ∂ℓ 𝜕𝑠4 = ℨ2 = 0 ℨ1 + ℨ2 = ℨ1+2 = 0 ℨ1 + ℨ2 = ℨ1+2 = 0 Which means that in the universe packet any even number of universe pairs with opposite curvature orientation is a valid solution. The even number can be of any magnitude, and be a result of lower magnitude solutions, which represent pairs of universes flattening each other via areas of extremum curvature, causing outward acceleration from those areas. The superposition concerning Bosons was analyzed in detail in previous papers and for the superposition to hold under addition operation, odd number of primes are required. 190 Manor O Manor O Therefore, despite matter can jump across the manifolds while keeping the manifold stationary, there is still a conservation of variation if we consider that matter can not escape the packet. Such idea is than revolutionary as it is imply it is impossible to know whether matter was originally created from variations of our own manifold, or it is matter which "jumped' or was transferred from a distinct manifold. That means that within one universe the conservation of energy does not hold, as matter has a potential energy, i.e. curvature in the 8T framework as was previously covered in the thesis. That idea however shades light on a conservation law, which indicates that while matter can jump from manifold to manifold, it can not escape the packet itself. So there is a conservation of entities within the packet. That does not indicate that there is a conservation of energy, as new entities are constantly being created as the manifold has arbitrary variations, all it indicates is the following: once those entities are created they must appear in some manifold within the packet. Nature does not impose a restriction on the index of the manifold that those entities should exist on. 8T – Arbitrary Variations Transfer We partitioned and discretized the arbitrary variation term of equation (1) and derived the existence of Fermions. In particular, we have shown that it must have an even amount of elements, which differ in sign and create nine threefold combination, and no more than two distinct elements. ∑δgi = 0 N i=1 (2.12) (2.12) In addition, with bosons, described by the term (1.49) as they were proven discrete amount of prime curvature on the matric tensor: ∑δgi M 1=1 > 0 (3.13. B) (3.13. B) Suppose that we have a finite number of pairs of distinct universes flatting each other via areas of extremum curvatures. ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t δg𝑚−∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t δg𝑛= 0 (2.1. B) (2.1. B) K = 2𝑘 Let the arbitrary variation terms, we also have a finite number of arbitrary variation vanishing into matter. Notice that if we change the amount in one universe and insert it to another universe, the stationarity condition will hold. 𝒷: δg𝑚⟶δg𝑚̌ 𝒷: δg𝑛⟶δg𝑛̇ δg𝑚< δg𝑚̌ δg𝑛̇ > δg𝑛 δg𝑚−δg𝑚̌ = Δ δg𝑛̇ −δg𝑛= Δ 𝒷: δg𝑚⟶δg𝑚̌ 𝒷: δg𝑛⟶δg𝑛̇ δg𝑚< δg𝑚̌ δg𝑛̇ > δg𝑛 δg𝑚−δg𝑚̌ = Δ δg𝑛̇ −δg𝑛= Δ 𝒷: δg𝑚⟶δg𝑚̌ 𝒷: δg𝑛⟶δg𝑛̇ δg𝑚< δg𝑚̌ δg𝑛̇ > δg𝑛 δg𝑚−δg𝑚̌ = Δ δg𝑛̇ −δg𝑛= Δ In other words, the term Δ is the amount of arbitrary variations that vanished into matter, and transferred from the first manifold into the second manifold. The only requirement is that that this amount would be an even amount of variations that will ensure that the manifold will stay at the condition of stationarity. 2|Δ →T𝑟𝑢𝑒 ∑𝛿𝑔𝑚 𝐾 2−Δ 𝑚=1 < ∑𝛿𝑔𝑛 𝐾 2+Δ 𝑛=1 ∑𝛿𝑔𝑚 𝐾 2−Δ 𝑚=1 = 0 ∩∑𝛿𝑔𝑛= 0 𝐾 2+Δ 𝑛=1 191 Manor O Manor O why the time is indexed in (1.33). The 8T framework is than allowing us to combine the nature of Quantum mechanics and discrete amount of energy, together with the setting of curvature and Riemannian geometry on continuous and smooth surfaces. It is also possible to correlate inverse relation between time and energy, and state that the longer the period of curvature diverging, the weaker the net curvature is, the more flat the ripple becomes. In physical theories that can correspond the redshifts. why the time is indexed in (1.33). The 8T framework is than allowing us to combine the nature of Quantum mechanics and discrete amount of energy, together with the setting of curvature and Riemannian geometry on continuous and smooth surfaces. It is also possible to correlate inverse relation between time and energy, and state that the longer the period of curvature diverging, the weaker the net curvature is, the more flat the ripple becomes. In physical theories that can correspond the redshifts. Discrete Curvature Ripples Therefore, the light quanta is a discrete ripple of net curvature, prime isomorphic which diverge to all directions of the unique manifold, that is 192 Manor O Discrete Curvature Ripples The wave equation of the Bosonic class is represented by the wave equation, which is net curvature diverging on the matric tensor. that is by a five-vector. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 (1.33) ∂2g ∂tn2 = ∂2g ∂x2 + ∂2g ∂y2 + ∂2g ∂z2 (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 (1.33) ∂2g ∂tn2 = ∂2g ∂x2 + ∂2g ∂y2 + ∂2g ∂z2 (1.33) The key point to take from this is the following. It is possible to settle the issue of particle wave duality if we consider the idea of discrete ripples of net curvature. The discrete is manifested in the fact that the Bosons are isomorphic to the class of primes, and each Bosons has a unique signature of a prime. The Boson itself is represented by spin one-half, but as it is entangled with the majestic three, only than it accumulated as spin one. That is important if one would like to settle the subject of what is the light Quanta. NV = ∏δgϕ ϕ=NV ϕ=1 = δgϕ=1 × δgϕ=2 × δgϕ=3 … (1.23) (1.23) δgϕ=1 ≡δgϕ=2 ≡δgϕ=3 … Therefore, the discrete nature is associative to the particle nature of the Bosonic class. In addition, the wave like nature is associative to the diverging nature of the ripple across the Lorentz manifold. It is possible to expend the idea using the spin representation as presented in the 8T thesis, in particular, it is possible to state the number of elements in the coupling term to a certain behavior, either particle or wave like. If the number of primes, including the majestic three is odd, than the overall behavior of the system would be particle like. If even it would be wave-like. Each prime added is considered a spin variant, which interfere with the overall system. As we can add prime together it means we can add those discrete ripples together to create the potentials itself, similar to summation of particles of a certain volume to get the potentials in classical field theories. 8T – The Fields Mixture 𝒬: 𝑇𝑜𝑝⟶𝑆𝑒𝑡 𝒬: 𝑇𝑜𝑝⟶𝑆𝑒𝑡 𝐺𝑐𝑙𝑎𝑠𝑠= {𝑁𝑉 |𝑁𝑉 ∈ ℙ ∪ (+1)} 𝒬: 𝑇𝑜𝑝⟶𝑆𝑒𝑡 8T – The Fields Mixture 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) (1.1) (1.2) Now, suppose we would like to represent the invariant multiplier by a combination of Bosons. Now, suppose we would like to represent the invariant multiplier by a combination of Bosons. 8 = 5 + 3 8 = γ + 𝒲 8 = 5 + 3 8 = γ + 𝒲 8 = γ + 𝒲 It is possible than to represent the multiplier by the following: It is possible than to represent the multiplier by the following: 𝐹𝑅# = ((γ + 𝒲) ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉 Let the forth coupling and above be represented by the parameters: 𝑁𝑉4⟶𝑉∞= ⋀𝑘 𝐾⟶[4, ℝ] ℝ→∞ 𝑁𝑉4⟶𝑉∞= ⋀𝑘 𝐾⟶[4, ℝ] ℝ→∞ So now it is possible to represent each coupling term in the following way, for the higher couplings as an example: So now it is possible to represent each coupling term in the following way, for the higher couplings as an example: (γ + 𝒲) × ⋀𝑘+ 3 + ⋀𝑘= γ⋀𝑘+ 𝒲⋀𝑘+ 3 + ⋀𝑘 (γ + 𝒲) × ⋀𝑘+ 3 + ⋀𝑘= γ⋀𝑘+ 𝒲⋀𝑘+ 3 + ⋀𝑘 γ⋀𝑘+ 𝒲⋀𝑘 (1.43) (1.43) equation (1.43) indicate that there is an interaction of the higher coupling terms, and therefore Bosons, with the electric and the weak interaction Bosons as each interaction contains elements from both. Since the invariant multiplier can be represented as a series of eight Gluons, the same applies there as well. 8 = ∑𝔤𝑖 𝑖=8 𝑖=1 In other words, replacing the invariant multiplier by the prime combinations which represent it allows us another glimpse into the valuable interaction of the higher couplings. That is than a source of a prediction: The higher coupling terms are in constant interaction with the Bosons of the strong, weak and electric. If we represent each in terms of a diverging ripples of net curvature, than (1.43) indicate that there is a ripple intersection between those Bosons: 193 Manor O 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 That is reasonable to assume as all of them are of the same class, the curvature class. That is reasonable to assume as all of them are of the same class, the curvature class. 8T – Indexed Hamiltonians The Hamilton idea could be presented in a rather simple way assuming we accept the notion of the multiverse as true, which is the case according to the 8T main equation in universe packet representation (2.1.A) and (2.1.B). In classical physics, the Hamilton is classified as summation of two terms, potential and kinetic. However, in the 8T it is the summation of two indexed terms, which indicate we have to sum over the kinetic and potential energies of all accelerating manifolds in the packet. That is in classical and Quantum mechanics: 𝐻̂ = T̂ + Û So in the 8T the Hamiltonian must be varied to represent the idea of distinct manifolds which composing the packet. For that purpose, the author will present an arrow and a morphism: 𝒢: T̂ ⟶T̂𝑖 T̂𝑖= 𝜕2g′ 𝑖 ∂t2 ; ∀ 𝑠1→𝑛 𝒢: T̂ ⟶T̂𝑖 For the potential energy, i.e. matter operator, the author will use the arbitrary variation term which vanish into zero, i.e. (1.48) over all the manifolds in the packet. ℴ: Û ⟶Û𝑖 Û𝑖= ∑δgi N i=1 ; ∀ 𝑠1→𝑛 an indication that the potential energy is zero, but rather that the potential energy means no That is by has no curvature, i.e. that matter pairs in a way that does not allow curvature to manifest itself, but it's still exit in a form of Quarks and the reason Quarks can not escape, to ensure the manifold stationarity condition. The new Hamiltonian is a summation of two indexed terms; each represents the kinetic energy of the manifold, and the potential energy of the manifold with a unique index and a unique arrow. The idea of indexed Hamiltonian is a result of the features of the 8T, acceleration outward from extremum curvatures, flatness, arbitrary variations vanishing into matter and manifold packets. This idea is clearly indicating that in order to understand one universe, it can only be done by analyzing the packet of universes itself. 𝐻̂ = T̂𝑖+ Û𝑖 (1.5) 𝐻̂ = T̂𝑖+ Û𝑖 (1.5) 194 Manor O Manor O Manor O 8T – Electron Propagation The equation of the Bosonic class is represented by the new form of the wave equation, which is net curvature diverging on the matric tensor. that is by a five-vector. We have proven the invariant three to be the electron by putting it in the fine structure formula. It is considered the electron to each of the higher coupling terms. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + 𝑁𝑉𝜇 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 (1.33) ∂2g ∂tn2 = ∂2g ∂x2 + ∂2g ∂y2 + ∂2g ∂z2 (1.33) The key point to is the following, to represent the relation of the electron to the strong interaction, it is possible to use the primorial in the following form: (8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + 𝑁𝑉𝜇→(𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + 𝑁𝑉𝜇 That is to state that the strong interaction even term is ever containing the electron. From the second term and above the electron is propagated out-ward. Since the electron is isomorphic to the Boson of the weak interaction, which can be either particle or a wave, so does the electron possess that particle duality. Given by the wave equation the electron is propagating all across the nuclei, as it is bounded by the bracket. That is important to clarify as it comes to an agreement with the insight of Quantum mechanics. That is to state that the strong interaction even term is ever containing the electron. From the second term and above the electron is propagated out-ward. Since the electron is isomorphic to the Boson of the weak interaction, which can be either particle or a wave, so does the electron possess that particle duality. Given by the wave equation the electron is propagating all across the nuclei, as it is bounded by the bracket. That is important to clarify as it comes to an agreement with the insight of Quantum mechanics. (𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + 𝑁𝑉𝜇⟶(𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + γ𝜇 8 ⟶𝟐𝒆− (1.25) (1.25) That new form of the primorial could be analyzed in the following form. The electron is not generated from the second term and above, but was already in existence inside the hadron. This instability, once That new form of the primorial could be analyzed in the following form. 8T – Inclusion Arrows The idea of inclusion arrows can fit the features of manifold creation. That is each new manifold is an embedded space of the original manifold, each manifold is the domain of a set of manifolds embedding's which rose from the original manifold. Define the inclusion arrow as: 𝜄: 𝑠0 ⟶𝑠𝑛+1 One will require the inclusion map of the 8T to possess the homomorphism trait. Such is required to ensure that the new sub-manifold will preserve the same features and structure of the original manifold. I.e. to be a complete manifold, which is simply connected and possess (3,1) signature. For that purpose the heredity condition was presented: 𝜍: 𝑠0 (3,1) ⟶𝑠𝑛+1 (3,1) In contrast to the idea of universe packet as presented in the 8T thesis, this idea emphasize of sub- structure within larger structures. Those substructures are than yielding from them sub-manifolds and the process than goes endlessly. The idea can be simply expressed using functors and sets. ℘:𝑇𝑜𝑝⟶𝑆𝑒𝑡 𝑠𝑝𝑎𝑛(𝑠0) = { ∑𝑠𝑛+1 𝐾 𝑛=1 | n ∈ℝ , K ∈ℝ } . ∑𝑠𝑛+1 𝐾 𝑛=1 ⊂ 𝑠0 Using that idea it is possible to provide a possible answer to the question of what was before the beginning. The answer using that idea, is that there was another manifold which had a unique time arrow. Time existed before singularity, it existed on the spanning space of our own manifold, when are own manifold was generated, it was than allocated to this structure the features of the original structure, and a unique arrow of time. Therefore, "the beginning of time" is only "the beginning of time" of this Three dimensional space, which serve as part of infinite spaces, embedded in one another, flattening each other via areas of extremum curvatures. In addition, our space itself serves as a generator for other spaces, with a distinct arrow and dimensions of their own. 195 8T – Electron Propagation The electron is not generated from the second term and above, but was already in existence inside the hadron. This instability, once there can not stay inside the hadron and must propagate outward, in all directions. This instability is bounded to the hadron, unlike the Bosons which are net curvature diverging unbound. Since it is possible to replace the positions of the elements using spin symmetry, there could be unbounded electrons in nature. there can not stay inside the hadron and must propagate outward, in all directions. This instability is bounded to the hadron, unlike the Bosons which are net curvature diverging unbound. Since it is possible to replace the positions of the elements using spin symmetry, there could be unbounded electrons in nature. (𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + γ𝜇⟶(𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + γ𝜇) + (𝒆−)𝜇 It is possible to express the motion of the electron inside the hadron before it gets propagated if the subscript is preserved. (𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + γ𝜇⟶(𝟐𝜇𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + 𝒆−𝜇) + γ𝜇 196 Manor O 8T – Curvature Ripples and Entanglement The wave equation of the Bosonic class is represented by the wave equation, which is net curvature diverging on the matric tensor. that is by a five-vector. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇) + 𝑁𝑉𝜇 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 (1.5) (1.5) Suppose we would like to execute an experiment in particle physics. The measured object can be either a Fermion such as the Electron or any Boson, which is net curvature diverging unbound. Since based on the second coupling term, the Electron is isomorphic to the Boson of the weak interaction as given by equation (1.61) in the Thesis, it is considered to possess the same effect during measurement. The following mathematical reasoning is needed. The measurer is arbitrary variation cluster that vanished into matter. The measurer is an infinite amount of matter. ∑δgi = 0 N i=1 ; 𝑁⟶∞ Assuming the measurer is a matter cluster of immense amount which stay as it is, it must have Bosons, which ensure it will retain its shape. Define the set of Bosons that are composing the arbitrary variation cluster, which is the observer, using functor: ℘:𝑇𝑜𝑝⟶𝑆𝑒𝑡 Τ = [𝑁𝑉𝑘| 𝐾∈ℝ ] (1.51) ℘:𝑇𝑜𝑝⟶𝑆𝑒𝑡 Τ = [𝑁𝑉𝑘| 𝐾∈ℝ ] (1.51) (1.51) The set (1.51) is representation of the type of Bosons that composes the observer. That is given by the mathematical relation: Τ ⊆∑δgi N i=1 A = {∑δgi N i=1 ; 𝑁⟶∞} Τ ⊂A Τ ⊂A Since the set T is the set of net curvature composing the observer, and the measured object is of the following class, i.e. net curvature diverging unbound, even before measurement or the experiment there exist a modification, those ripples from the object and from the matter cluster interest with each other. Just a observer itself is causing a major variation of manifold. Assuming again the measurer has inner curvature retaining its shape and preventing its decomposition. Therefore, the manifold has those infinitesimal quantities, either with spin one-half or spin one: 197 Manor O Manor O 𝐵= {𝒮1 ∪ 𝒮1 2 } With the existence of the measurer, the two sets are representing the system itself, no matter how far they are there exist now a collection of the two set under one new set. The measured object is 𝐵 and the matter cluster is 𝐴. 𝐴 ∪𝐵⟶𝐶 𝐶= 𝐴+ 𝐵 There exist a modification of the system due to the mere existence of the observer. The net curvature which are discrete ripples of curvature diverging are creating compositions with the net curvature diverging of the observer; they are now part on the single entity. In certain cases, the combinations of ripples creating new ripples of higher magnitude that are solutions of the primorial higher terms. The observer, which has Electrons in it, are getting modified from the wave-like nature of the measured particle, and vice versa. Moreover, once the sets are joint to a single entity, it is impossible to reverse the action of joining them. It is impossible to decompose which element were modified nor which ones came from the matter cluster and which belonged to the infinitesimal quantity. That idea of irreversibility can be represented by the existence of two or more possible decompositions with equal probabilities. 𝐶⟶𝐴1 + 𝐵1 ∈𝑃(𝑢1) 𝐶⟶𝐴2 + 𝐵2 ∈𝑃(𝑢2) 𝐴1 + 𝐵1 ≢𝐴2 + 𝐵2 𝑃(𝑢1) = 𝑃(𝑢2) 𝑃(𝑢1) = 𝑃(𝑢2) The idea of observers as separated entity from the measured object is leading us far astray as we can possibly get according to the 8T author. Not only that, the observer and the measured object are united in one system, which can not be decomposed nor it is stops at larger distances. The mere existence of the measurer is effecting the system, modifying it to a new joint set of elements, which interact with each other. Since no two observers are identical, no two joints sets are identical; each observer is causing a different joint set to be created, or a unique entanglement. That root is different than the one taken in the thesis, which used spin variation to derive the effect of measurement on the system. The spin variation is encompassed with additional unit of energy quanta, which causes the system energy to vary, making the experiment impossible to make without changing the measured object spin and energy. 198 Manor O The 𝑉̇ Operator The wave equation of the Bosonic class is represented by the wave equation, which is net curvature diverging on the matric tensor. that is by a five-vector. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝑒)𝜇) + 𝑁𝑉𝜇 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 (1.33) ∂2g ∂tn2 = 𝜕2𝑔 𝜕𝑥2 + 𝜕2𝑔 𝜕𝑦2 + 𝜕2𝑔 𝜕𝑧2 (1.33) The key point to take from this is the following – the curvature ripple is the momenta as presented in Quantum mechanics. The key point to take from this is the following – the curvature ripple is the momenta as presented in Quantum mechanics. Quantum mechanics. 𝑀𝑉̇ = −∇𝑈 𝑈= (𝑋, 𝑌, 𝑍) We can also instantly see that the mass is inversely proportional to the velocity derivative. In the 8T, the mass is considered curvature converging represented by the Quark masses series, that is by 8 −(1) variations, which allowed to the derive the pattern of total masses decreasing: 19,600 ⟶1400 ⟶56 176,400 ⟶1400 ⟶6.3 ⟶0.113 ⟶ By the primorial, the series of diverging net curvature unbound is represented by terms which are of the sort of 8 + (1) for the first, and scalar multiples of that 8 + (1) and additional prime for the higher coupling term. The key point is the following. The velocity operator is than represented by the relation of 𝑉̇ = 8 + (1) 8 −(1) = −∇𝑈 𝑀 𝑉̇ = 8 + (1) 8 −(1) = −∇𝑈 𝑀 In the 8T thesis, there is an analog to this relation by the Einstein equation between mass and energy. Energy is represented as the curvature diverging, and the mass as the curvature converging, the following ratio has a root which is the speed of light. In the 8T thesis, there is an analog to this relation by the Einstein equation between mass and energy. Energy is represented as the curvature diverging, and the mass as the curvature converging, the following ratio has a root which is the speed of light. 8 −(1) ⟶ℊ𝑐 8 + (1) ⟶ℊ𝑑 Curvature diverging ℊ𝑑 is equal to curvature converging, ℊ𝑐, times the square of speed of light. A new version of the Einstein equation, equation 5.1). Manor O Manor O 8T – Proof: Anti Matter The wave equation of the Bosonic class is represented by the wave equation, which is net curvature diverging on the matric tensor. that is by a five-vector. (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝑒−)𝜇) + 𝑁𝑉𝜇 8 ⟶𝟐𝒆− 8 = γ + 𝒲 𝐹𝑅# = ((γ + 𝒲) ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉 (γ + 𝒲) = 𝟐𝒆− (γ + 𝒲)2 = (𝟐𝒆−) 𝟐 (𝟐𝒆−)𝟐= 64 = 0 (γ + 𝒲)2 = γ2 + 2𝒲γ + 𝒲2 2𝒲γ = 0 γ2 + 𝒲2 = 0 [(γ, 𝒲) > 0 ] ∪[(γ, 𝒲) < 0] 𝓦≡𝒆− [(𝛄,𝒆−) > 𝟎 ] ∪[(𝛄,𝒆−) < 𝟎] 𝑬𝒏𝒅 𝒐𝒇 𝒑𝒓𝒐𝒐𝒇 (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)𝜇) + 𝑁𝑉𝜇⟶(8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝑒−)𝜇) + 𝑁𝑉𝜇 8 ⟶𝟐𝒆− 8 = γ + 𝒲 𝐹𝑅# = ((γ + 𝒲) ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉 (γ + 𝒲) = 𝟐𝒆− (γ + 𝒲)2 = (𝟐𝒆−) 𝟐 (𝟐𝒆−)𝟐= 64 = 0 (γ + 𝒲)2 = γ2 + 2𝒲γ + 𝒲2 2𝒲γ = 0 γ2 + 𝒲2 = 0 [(γ, 𝒲) > 0 ] ∪[(γ, 𝒲) < 0] 𝓦≡𝒆− [(𝛄,𝒆−) > 𝟎 ] ∪[(𝛄,𝒆−) < 𝟎] 𝑬𝒏𝒅 𝒐𝒇 𝒑𝒓𝒐𝒐𝒇 199 The Graviton Rise The Graviton in the 8T is represented by a combination of three Bosons and one Lepton, i.e. the electron, even though due to the EMT symmetry it can be the Muon of the Tao. The following form is the structure of the graviton: [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 + 𝑁𝑉3 = [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2] + 1 2 + 1 2 + 1 2 (2.2) (2.2) It recently became evident to one that there could be a several forms in which we can represent gravity which exceeds the invariance of spin due to replacing the Bosons. A more interesting form of Gravitons includes timed emission of two Bosons from two distinct Leptons which aspire to "stay away" from each other, or to obey the Pauli exclusion Principle. [2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦+ 1 2 + 1 2] + 1 2 + 1 2 = [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 (2.2. 𝐴) (2.2. 𝐴) To emphasize the idea of the Leptons to be different state it is possible to map them in different directions ensuring they will never vanish into an even number and ruin the coupling series: [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3) + (3)] + 𝑁𝑉1 + 𝑁𝑉2 →[(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3⃖ ) + (3 )] + 𝑁𝑉1 + 𝑁𝑉2 (2.2. 𝐵) So that is a much simpler version than the first which is represented in the 8T thesis. First, it balance out the inequality between the Lepton and the Bosons in the coupling term. Instead of having one Lepton to generate three Bosons, we require now two Leptons to generate one Boson each. Than the summation of spin accumulates to spin two. The actual type of the Boson is not relevant to this discussion, for simplicity sake it can be the photon. [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3⃖ ) + (3 )] + 𝛾+ +𝛾 ⟶ [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (𝑒− ⃖ ) + (𝑒− )] + 𝛾+ 𝛾 (2.2. 𝐶) (2.2. 𝐶) So the result of (2.2. 𝐶) than accumulate to an even number which in variation from can be ignored, as it vanishes. The result is one term in the coupling of Gravity similar to the coupling of the strong which contain only one term. The one term indicate that similar to the Boson of the Strong, the Graviton is massless and it is short range. The 𝑉̇ Operator The fact that these two are similar is indicating that the speed of light is the limit of the acceleration operator, which is in agreement with private relativity. ℊ𝑑= ℊ𝑐𝑐2 (5.1) (5.1) 200 Manor O The Graviton Rise [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (𝑒− ⃖ ) + (𝑒− )] + 𝛾+ 𝛾 ⟶(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + 𝐸𝑣𝑒𝑛 (2.4) (2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + 𝐸𝑣𝑒𝑛 ⟶(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) (2.41) (2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + 𝐸𝑣𝑒𝑛 ⟶(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) (2.41) (2.41) Another option of the rise of the Graviton, which is the last option, as far as one can see, is the result of three Leptons and one Boson, which will again lead to equation (2.4) and (2.41): [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3⃖ ) + (3 ) + (3⃛)] + 𝑁𝑉1 ⟶ [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (𝑒− ⃖ ) + (𝑒− ) + (𝑒−)] + 𝛾 These forms of gravity indicate that the Graviton is more likely to rise in elements with large number of Leptons, i.e. heavy elements, when we have the balanced form of Gravitons, those Bosons have to be timed, that is to be propagated in the same temporal segment for some arbitrary frame of reference. These forms of gravity indicate that the Graviton is more likely to rise in elements with large number of Leptons, i.e. heavy elements, when we have the balanced form of Gravitons, those Bosons have to be timed, that is to be propagated in the same temporal segment for some arbitrary frame of reference. [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (𝑒⃖) + (𝑒 )] + 𝛾+ 𝛾→[(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (𝑒⃖) + (𝑒 )] + 𝛾𝑡1 + 𝛾𝑡2 (2.42) (2.42) 𝑡1 = 𝑡2 𝑡1 = 𝑡2 201 Manor O 8T –The Photon Propagation The photon propagation is presented in two different ways in the 8T thesis. The first via a different form of the Feynman diagrams, using arrows and the framework of variational curvature vanishing in summation of even numbers: [(24 ∗5) + (3)] + 5 → [(24 ∗5) + (𝑒−)] + γ 𝑒−↘ ⟶ (𝛾) ⟶ 𝑒− ↗ (+3) ⟶(𝛾) ⟶ (+3) (+3) ⟶(+5) ⟶ (+3) (+3) + 𝑁𝑉= (+3) + 5 = 8 8 = 𝐸𝑣𝑒𝑛 𝐸𝑣𝑒𝑛= 0 d is via the photon absorption in a visual means: (𝑒) (𝑒) The question and the subject matter is why does the electron repeal its other. Why does the Electrons does not get in to the curve which is the photon but rather "escape" to different directions. There exist several whys to do just that. The first is mentioned in the 8T thesis, if the two Electrons would get into the curve, they will aspire the lowest point and meet each other. Such a scenario will lead to a vanishing of the two Electrons, making the primorial impossible to begin with, if we pre-condition the Electrons to propagate photons. That is the Pauli exclusion principle. For that reason, we presented the form of balanced Graviton with indexing the states of the electrons, to ensure they will not meet each other. The second is via the photon absorption in a visual means: (𝑒) (𝑒) The question and the subject matter is why does the electron repeal its other. Why does the Electrons does not get in to the curve which is the photon but rather "escape" to different directions. There exist several whys to do just that. The first is mentioned in the 8T thesis, if the two Electrons would get into the curve, they will aspire the lowest point and meet each other. Such a scenario will lead to a vanishing of the two Electrons, making the primorial impossible to begin with, if we pre-condition the Electrons to propagate photons. That is the Pauli exclusion principle. For that reason, we presented the form of balanced Graviton with indexing the states of the electrons, to ensure they will not meet each other. The question and the subject matter is why does the electron repeal its other. Why does the Electrons does not get in to the curve which is the photon but rather "escape" to different directions. 8T –The Photon Propagation There exist several whys to do just that. The first is mentioned in the 8T thesis, if the two Electrons would get into the curve, they will aspire the lowest point and meet each other. Such a scenario will lead to a vanishing of the two Electrons, making the primorial impossible to begin with, if we pre-condition the Electrons to propagate photons. That is the Pauli exclusion principle. For that reason, we presented the form of balanced Graviton with indexing the states of the electrons, to ensure they will not meet each other. [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (3⃖ ) + (3 )] + 𝛾+ 𝛾 ⟶ [(2𝑁𝑔𝑟𝑎𝑣𝑖𝑡𝑦) + (𝑒− ⃖ ) + (𝑒− )] + 𝛾+ 𝛾 (2.2. 𝐶) (2.2. 𝐶) There is a second way to explain this phenomena, the Electron which absorbed a photon has absorbed a net curvature of prime amount, it will pull the particle toward the curve putting it in a lower height, while the emitting electron just gave up a certain amount on net curvature which will elevate him to the higher direction. Those two heights will not cross, and thus the two electrons will not meet. This explanation can be put mathematical rigor. Define the absorbing Electron using the Quantum manifold setting, i.e. using subscripts for classifying the absorbing/emitting elements and superscripts for the number of elements within the Electron. 202 The Quest of Defining 𝐻̂ The subject matter of this paper is the following. Is it possible to expend the idea of energy using variational manifolds? The term energy has been used and still is used all over the modern spectra of physics. However, no theory has been able to explain what is the idea that stands behind it. Since the main equation describes the phenomena of 'dark energy" or time invariant acceleration outward from extremum curvatures, supported by the coupling constants series, it is possible using that equation, to expend and clarify the idea of Energy. Consider the idea of a certain mapping: 𝜑: g ⟶𝐸 𝐸= 𝐻̂ 𝜑: g ⟶𝐸 𝐸= 𝐻̂ This mapping is between the Ricci curvature and the operator of energy in Quantum mechanics. So that the main equation (1.2) now can look as the following: ∂ℓ ∂s ∂s ∂M ∂M ∂g 𝛛𝐄 𝛛𝐭−∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐄′ 𝛛𝐭𝟐= 0 ∂ℓ ∂s ∂s ∂M ∂M ∂g 𝛛𝐄 𝛛𝐭−∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐄′ 𝛛𝐭𝟐= 0 ∂ℓ ∂s ∂s ∂M ∂M ∂g 𝛛𝐄 𝛛𝐭−∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝝏𝟐𝐄′ 𝛛𝐭𝟐= 0 If we consider outward acceleration from extremum curvatures on the manifold to the term of "energy", than energy is isomorphic to the degree of curvature. The more curved the matric on the manifold the more energy it has and vice versa. The more flat the manifold is the less energy it contain. This idea than indicating that the manifold will aspire to reach the highest degree of flatness, indicating lowest degree of curvature overtime. That is in agreement with the idea of stationary manifold. "Energy" can also by analyzed in the context of the transformation between curvature to flatness, the manifold in the beginning was highly curved and as a result of being a part of the packet it was immediately flattened by the packet, which led to very high rate of change of curvature, which is isomorphic to energy, which manifested itself in the beginning. Once the manifold got flattened, the rate of change of curvature to time is significantly lower and aspire lower and lower value as the manifold expends. Each time net amount of curvature appearing matter is clustering toward it leading to formations of stars and galaxies. Using that construction, it is possible to construct the idea of potential energy. Manor O ℋ𝐴: ( 𝑒𝐾 −(0) ⟵γ) ⟶ 𝑒𝐾 −(1) ℋ𝐸:( 𝑒𝐾 −(1) ⟶γ) ⟶ 𝑒𝐾 −(0) ℋ𝐸:( 𝑒𝐾 −(1) ⟶γ) ⟶ 𝑒𝐾 −(0) Allocate the elevation parameter due to absorption by a term: Allocate the elevation parameter due to absorption by a term: 𝑒1 −(0) −𝑒1 −(1) = ∆↓ Allocate the opposite parameter due to emission: 𝑒2 −(1) −𝑒2 −(0) = ∆↑ Using the following idea, it is possible to imagine a new form of interaction between the Electrons. In such that they can pass on the same spatial coordinates but in different heights. The photon is pulling the absorbed particle to a lower elevation altitude, while its release leading to a higher elevation to the emitting Electron. The Feynman diagram now can be modified: 𝑒−↑ ⟶ (𝛾) ⟵ 𝑒− ↓ 𝑒−↑ ⟶ (𝛾) ⟵ 𝑒− ↓ 203 8T – Wave Functions and Spin 8T – Wave Functions and Spin 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30:128: 850:9254.. (1.2) 𝑁𝑉= 2 (𝑉+ 1 2); 𝑉≥0 𝑁𝑉 ∈ ℙ ∪ (+1) ; ℙ→𝑆𝑒𝑡 𝑜𝑓 𝑃𝑟𝑖𝑚𝑒𝑠 𝑁𝑉 = 𝑃𝑚𝑎𝑥 ∈[0, ℝ] ∪ (+1) ; 𝑃𝑚𝑎𝑥∈ ℙ The second form of the primorial is locating each prime on the prime critical strip. This construction leading to a new form of the original equation, which assumed to be describing the trait of spin. We can solidify that claim using the fact in Quantum mechanics, spin of systems can only change in discrete amounts, that is a positive indication as there is one critical strip in which the Bosons are arising from. Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 0: 2𝑁() variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations Spin 1 2 : 2𝑁() + 3 variations Spin 1: 2𝑁() + 3 + 𝑁𝑉 variations Spin 𝑁= 2𝑁() + 3 + 𝑁𝑉1 + 𝑁𝑉2 … variations That is the first point of intersection with Quantum mechanics. The second point of intersection, is the following. In the paper, "Net Versus Spin", page 140 in the 8T thesis, author argued that the state of the system depends upon the number of elements on the prime critical line, if it is even, the system will behave like a wave, if odd like a Fermion, or particle like. Same argument arisen without the author correlating the arguments at the time, in QM. In particular, symmetrical wave functions versus anti- symmetrical wave functions. The Quest of Defining 𝐻̂ Potential energy is value, which describe the amount of curvature within fermion cluster. This term also include the fact that each matter unit itself is composed by Quarks, which are vanishing curvature spikes. Therefore, the potential energy can be put in rigor as the sum of two equations (2.12) and (3.13.B): 𝑈= ∑δgi N i=1 + ∑δgi M 1=1 (1.55) ∑δgi M 1=1 ⊂ ∑δgi N i=1 (1.56) (1.55) (1.56) Equation (1.56) meant to express the idea of a matter cluster with Bosons that i hi i h b f i l h f Equation (1.56) meant to express the idea of a matter cluster with Bosons that propagate within it. As the subset of Bosons is larger, the more amount of matter being clustered making the potential energy higher. It is important to clarify that matter itself is only a part of the picture itself, as it has no curvature itself, but it is composed to arbitrary amounts of curvature that must vanish to keep the manifold stationary, which are two distinct elements which differ in sign, or Quarks. The total energy of the specific manifold is described by: 𝐻̂ = 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 + ∑δgi M 1=1 = T̂ + Û (1.57) (1.57) 204 Manor O 8T – Wave Functions and Spin Odd numbers obey the Fermi statistics, and Even numbers obey the Bose statistics, which is exactly what the primorial is indicating and the same formalism used to describe the Bosons which act like Fermions, or the Particle wave duality. The author would like to dive deeper into the subject of identical particles and symmetrical versus anti-symmetrical wave functions which serve as a significant part in QM. Suppose a given set of Bosons of a given coupling: 𝐾= {𝛾1 …𝛾𝑛} Since all the Bosons are isomorphic to a unique prime there index can change without any difference, i.e. it is impossible to distinguish between two photons. 𝐾→𝑁𝑉=2 ∀ 𝛾𝑛∈𝐾 The same apply to each Boson of each coupling in the series. The notion of equivalence can be solidified using the idea of class. All Bosons belong to the class of curvature on the manifold, page 176 in the thesis, and thus it is possible to expend the idea of indistinguishable to classes of Bosons. As an example, consider the arrow: Τ: (𝐾→𝑁𝑉=2 ) ⟶𝐾2 Alternatively, more generally: Τ: (𝐾→𝑁𝑉=𝑍 ) ⟶𝐾𝑍 205 Manor O Since all Bosons belong to the same class, we can create an higher class, summing the distinct classes of Bosons; 𝒯= {𝐾1 …𝐾𝑍} 𝐾1 ≡𝐾2 … ≡𝐾𝑍 The last point of intersection is the nature of Bosons versus the nature of Fermions, excluding the complication arises from the duality of the Electron and W Boson. Sin The last point of intersection is the nature of Bosons versus the nature of Fermions, The last point of intersection is the nature of Bosons versus the nature of Fermions, excluding the complication arises from the duality of the Electron and W Boson. Since Bosons has only one sign, that is they are net amount (assumed positive, although it makes no difference and considered negative) they are described under one sign in the thesis. And thus, if the Boson is interchanged it makes no difference to sign of the wave function. Consider the set of signs to the class of Bosons to be a subset of discrete amount: 𝔛𝐵⊂𝒯 𝔛𝐵= {+} While Fermions are vanishing curvature spikes, the set of elements for class for Fermions, excluding the Electron, would then be: ℱ= {𝛿g1, 𝛿g2} Consider the set of signs to the class of Fermions to be a subset of discrete amount: 𝔛𝐹⊂ℱ 𝔛𝐹= {+, −} Thus the interchange of Bosons does not change the sign of the wave function, while the interchange of Fermions does changes the sign of the wave function. The immediate result is that Bosons can be propagated as wave to long-range distances, while Fermions cannot. That is similar to stating that the Fermions will accelerate toward one another. Another way to explain it is to state that the opposite curvature ripples cancel each other out, yielding an higher level entity which has no curvature manifestation, what we call matter. A threefold combination must match another threefold combination to eliminate the curvature ripples, in agreement with stationary manifold. The exclusion of the Electron is due to the fact that it is isomorphic to the Boson of the weak interaction, which imposes a complication as it can theoretically belonged to either Fermions or Bosons. Thus the interchange of Bosons does not change the sign of the wave function, while the interchange of Fermions does changes the sign of the wave function. The immediate result is that Bosons can be propagated as wave to long-range distances, while Fermions cannot. 8T – 𝑇̂ as a Sum of Accelerations The main equation of the 8T is describing the variation of a Lorentz manifold, which according to the second equation (1.1) can be expressed as being part of a manifold packet. Such a theoretical construction manifested in one equation, is able to provide an answer to three major questions at the heart of major cosmology. The flatness puzzle, the "dark energy" puzzle and the "dark matter" puzzle by (1.2). ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ 𝜕2g′ ∂t2 = 0 (1) (1) ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K 2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1.A) ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t δg𝑚−∑∂ℓ ∂sn K 2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t δg𝑛= 0 (2.1.B ) (2.1.B ) The subject matter of this paper is the following question, can we represent the The subject matter of this paper is the following question, can we represent the Hamiltonian of the manifold as a sum of spikes arbitrary spikes all across the matric. The author will argue that the answer is positive. First, let us represent the Hamiltonian of the system as presented in pages 205-206 within the 8T thesis: 𝑈= ∑δgi N i=1 + ∑δgi M 1=1 ∑δgi M 1=1 ⊂ ∑δgi N i=1 𝐻̂ = 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 + ∑δgi M 1=1 = T̂ + Û (1.57) 𝑈= ∑δgi N i=1 + ∑δgi M 1=1 ∑δgi M 1=1 ⊂ ∑δgi N i=1 𝐻̂ = 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 + ∑δgi M 1=1 = T̂ + Û (1.57) (1.57) The potential energy is arbitrary variation clusters which vanish into matter, than Bosonic ripples propagating within them, that is vanishing spikes and non-vanishing curvature spikes. The kinetic energy is given by the Ricci flow equivalent to the acceleration in (1). Now to make the Hamiltonian more accurate it is possible to decompose the kinetic term: 𝛛𝐠 𝛛𝐭= ∑(𝛛𝐠 𝛛𝐭)𝝓 𝑲 𝝓=𝟏 (𝟏,𝟓𝟖) (𝟏,𝟓𝟖) Such an operation would allow us to regard the kinetic the term as the sum of acceleration of distinct galaxies. As new areas of extremum curvature are being created, 𝐾⟶∞ and energy, as previously mentioned is not conserved as the Hamiltonian term is a subject to constant variance, which is increase. Manor O That is similar to stating that the Fermions will accelerate toward one another. A h l i i i h h i i l l h h Summing up in three points. First point, the wave function of systems should be classified according to certain criteria. The first is the Bosonic or Fermionic classes, the second is the number of elements in the class. Second point, Fermion class must contain an even number of elements, Bosonic class can contain any number of element, if the number is even, the statistic obey the Bose rules, if it is odd than Fermi rules. Third, the number of elements dictates the spin summation using the prime critical line. The spin summation determines the behavior of the system, which can be either a smooth wave or particle like. The symmetry of wave function is due to subset of signs to each of the two distinct classes. The number of elements also dictating the Quanta's of energy in the system. 206 Manor O 8T – 𝑇̂ as a Sum of Accelerations The increase does not interfere with the stationarity of the manifold as matter is appearing in a way that does not allow curvature to manifest itself. There could be additional uses for the kinetic term decomposition such as a collusion between two galaxies, which now mean that there is new area of extremum curvature, with new rate of acceleration. The new rate of acceleration is the summation of the kinetic terms of each distinct galaxies: (∂g ∂t)𝜙=1 + (∂g ∂t)𝜙=2 = (∂g ∂t)𝜙=1+2 207 Manor O The Grand Field The Primorial equation of the 8T is describing the coupling magnitudes of all known interactions and the interactions which are not yet discovered. In the 8T thesis the primorial is has several forms which are correlated to different uses and ideas of this unique series of dimensionless numbers. In the beginning of the 8T thesis, the even terms of each coupling are correlated to fermions which are two and three divisible to vanish into matter. In this paper, the even terms will represent a direct product of fields, which add up to a one Grandfield, which may or may not have a physical meaning. 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) 𝐹𝑉=0 = 8 + (1) (1.1) 𝐹𝑅# = (8 ∗ ∏𝑁𝑉 𝑉=𝑅 𝑉=1 + (3)) + 𝑁𝑉= 30: 128:850: 9254.. (1.2) (1.1) (1.2) Since the first multiplier is representing the summation of Gluon type, it relates to the strong interaction. We have proven each Boson to be in a state of one to one correspondence with a unique prime. Since the primorial is taking each prime in the set of primes and multiplies it with the Gluon type, we get a term, which contains all the Bosons of the known interactions and the next interactions in line. The following form of the primorial is somewhat more "advanced" as it appears in the end of the thesis but it is identical to (1.1). (𝟐𝜇 𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + 𝒆− 𝜇) + 𝑁𝑉𝜇= (𝟐𝝁 𝒆−× 𝑵(𝑽=𝟏)𝝁× … 𝑵(𝑽=𝒌)𝝁+ 𝒆− 𝜇) + 𝑁𝑉𝜇 For the first three interactions, we get the term: 𝟐𝜇𝒆−× 𝒲𝜇× 𝛾𝜇= (𝟐𝒆−𝒲γ)𝝁 This idea is very different from the original idea that appeared in the beginning of the thesis and regard the even terms to vanish into nucleons. Rather it shows that there exist one term which contain all the Bosonic 'particles' within it. Since those Bosons are net curvature on the manifold, they belong to the same class, which solidifies that idea of one united field, over an idea of distinct fields for each Boson type. This new interoperation of the primorial many indicate that there is ripples intersection among the Bosons as the arrow of time develops. That makes sense as those higher Bosons come from the original first term 𝟐𝜇𝒆− as proven previously. Momenta and Wavelength (𝟐𝜇 𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + 𝒆− 𝜇) + γ𝜇= 30: 128: 850: 9254 … The subscript stands for a five vector that is given by: The subscript stands for a five vector that is given by: 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) 𝜕2 𝜕𝑡𝑛 2 𝑁𝑉= ∇2𝑁𝑉 ∂2g ∂tn 2 = 𝜕2𝑔 𝜕𝑥2 + 𝜕2𝑔 𝜕𝑦2 + 𝜕2𝑔 𝜕𝑧2 In the recent past 8T has defined "Energy" or the "Hamiltonian" of the system as the arrow: 𝜑:g ⟶𝐸 𝐸= 𝐻̂ 𝜑:g ⟶𝐸 𝐸= 𝐻̂ Given that introduction, it is possible to construct an analog for the relation that defined physics of the past century, the relation between momenta and wavelength. That can be done in several ways. Since each photon is a net curvature diverging unbound, and by the arrow above, that curvature is accounting for a certain energy, the more curved the element the more energy it contains. That is how one presented the EMT symmetry of equation . The setting of the theory is stationary manifold in which curvature is "not allowed", i.e. the manifold aspires to reach the lowest state of energy, or flatness, because of being of a packet of manifolds which flatten each other via areas of extremum curvatures. Therefore, for that we have momenta, which must be in inverse relation to the wavelength. The shorter the wavelength, the higher the diverging rate of the curvature across the matric, due to the stationarity condition. The longer the diverging process, the amount of energy is devised across larger areas so that the manifold can "get rid" of those elements which violate the stationarity condition first. What is new here is not the relation of the two terms, but rather the reasoning and the nature of photons when considered in a variational curvature framework. Than we can extrapolate the usual relation, curvature is inversely proportional to wavelength, and wavelength is inversely proportional to momenta. Which indicating that the momenta is directly proportional to curvature/energy. The shorter wavelength the higher the energy, the higher the momenta, and the faster the wave or ripple or curvature travels or diverge all across, so as a result this is the 8T explanation to the fact that the earthly sky are blue. 208 Manor O the class. In the 8T it is the mapping between Ricci flow to Energy which set to clarify the ambiguous term of "energy". The universe aspires reaching flatness by the stationarity condition, which is in one to one correspondence for lowest state of energy. The Grand Field In contrast to QFT which has many type of fields which are hard to grasp, 8T aspire to examine those Bosons as excitements or curvature spikes of one entity which is the manifold. The new form of the primorial is expressing that idea. The idea was also presented under the name "Gravity classes" which defined the Bosons to be distinct object of the same class, the curvature class. 𝐺𝑐𝑙𝑎𝑠𝑠= {𝑁𝑉 |𝑁𝑉 ∈ ℙ ∪ (+1)} One last point, In contrast to other theories of physics, the symmetry break of the strong electroweak is something that happens constantly given the Bosons of the three first interactions sums up to the first coupling term: 𝟐𝜇 𝒆−+ 1 = (𝒲𝜇+ 𝛾𝜇) + 𝔤𝜇 (𝒲𝜇+ 𝛾𝜇+ 𝔤𝜇) > 𝒲𝜇∪𝛾𝜇∪ 𝔤𝜇 The separation accrues as each element has energy of discrete amount, and all of them unified has higher energy than each as distinct, so first the strong depart, than the Electroweak depart from one another. Gravity is not there as it is the class, which is already broken given the first three distinct elements, which serve as representative of 209 SUSY and Invariant Three The alignment of the first three interactions was based upon aligning the net variations, which stand for Bosons according to theorem two, which is part of three theorems that yielded the primorial. The alignment was presented in two ways, for simplicity sake the author will present only the second as it is simpler. The alignment is due to two real net variations going from the photon to the Gluon. 8 + (1) + 2: [(8 ∗3) + (3)] + 3 ∶[(24 ∗5) + (3)] + 𝟑 8 + (1) + 2 ⟶8 + (3) 8 + 𝟑∶[(8 ∗3) + (3)] + 𝟑∶[(24 ∗5) + (3)] + 𝟑 [(24 ∗5) + (𝑒)−] + γ ⟶[(24 ∗5) + (𝑒)−] + 𝑊− [8 + (𝔤+ 2)] ⟶[8 + (𝑊 −)] We said that the modification could not affect the invariant three, which is the electron. The reason it cannot effect the electron is because it is isomorphic to the Bosons of the first interaction, as both are represented by the same number, and in particle physics one of the Bosons of the weak interaction carry the same charge as the electron. [(8 ∗3) + (3)] + 𝟑 3 ≡(3) (𝑒)−= 𝑊− Therefore, a modification on the electron is identical to modification on the alignment Boson which is in the case of the first three interactions is the Weak interaction. The only term in which the net variations unbound can modify on the third term than. Is the first term. It is important to emphasize, as reader may rightfully ask why the modification cannot affect the lepton. As a result of those exclusions on the Lepton and the Boson the only modification which is allowed will result in alignment at: (8 ∗3) + 2 = 26 210 8T –Quantum Variantics O Manor Index: Energy & Schrodinger Equation____________________________214 Orthogonality __________________________________________216 Amplitudes____________________________________________ 217 Universe at singularity___________________________________218 QM Axioms ___________________________________________220 Quantum tunneling ______________________________________221 The Atom ____________________________________________222 Immense spin formations ________________________________224 Gravity in Fermion cluster & Binders ______________________225 Curvature scattering by matter ____________________________226 The Road to Reality ____________________________________227 Manor O Manor O 211 Manor O 8T –Quantum Variantics O Manor October 14, 2021 Index: Energy & Schrodinger Equation____________________________214 Orthogonality __________________________________________216 Amplitudes____________________________________________ 217 Universe at singularity___________________________________218 QM Axioms ___________________________________________220 Quantum tunneling ______________________________________221 The Atom ____________________________________________222 Immense spin formations ________________________________224 Gravity in Fermion cluster & Binders ______________________225 Curvature scattering by matter ____________________________226 The Road to Reality ____________________________________227 Introduction Manor O Manor O 8T –Quantum Variantics O Manor October 14, 2021 Index: Energy & Schrodinger Equation____________________________214 Orthogonality __________________________________________216 Amplitudes____________________________________________ 217 Universe at singularity___________________________________218 QM Axioms ___________________________________________220 Quantum tunneling ______________________________________221 The Atom ____________________________________________222 Immense spin formations ________________________________224 Gravity in Fermion cluster & Binders ______________________225 Curvature scattering by matter ____________________________226 The Road to Reality ____________________________________227 ⟨𝐵|𝐿̂|𝐴⟩ Another dominating theme is the following transformations between two states by a linear operator. First of all, its too abstract, and it does not tell you anything about nature, despite being solvable. What's the worth of calculating without understanding? This is what computers do. These things are solvable but they are telling very little in intuitive fashion. The authors of QM are not explaining why QM is described by linear operators rather than non-linear operators, which is another flaw in this formulation. Let alone the fact that it goes from right to left, rather than all equations from left to right. One must clarify that it is not a case against QM but a case against how QM is described. There is a difference. The questions concerning the issues of "why" are just as important the Questions concerning "what", the entire extrapolation of the primorial coupling series was built on the notion of "why". In QM there exist very little to no explanations to why things are the way they are and that another major flaw in the current formulation of Quantum mechanics. So the challenge in hand is first of all, given the recent advancements in the field of Theoretical physics, and the unification of the interactions using manifolds is to re-build the flawed QM by using axioms and mappings. From here on out, QM is considered old formulation, and QV is the new formulation, which stands for Quantum Variantics. The new formulation is of variational curvature which will aspire to build more complete analogs for the dominating themes of QM. The most urgent mapping that is lacking in the old formulation of Quantum mechanics, is the following map between Ricci flow to Energy, which meant to provide a clear and solid definition to "Energy". 𝜑: g ⟶𝐸 The Schrodinger equation for the electron than is describing how does a non-vanishing curvature spike which has spin one half is moving across a nine-fold combination of two distinct elements which differ in sign. The electron has a superscript which describe the number of elements it contains. There is no need to use the Planck constant, which is a measureable constant. The problem with using this constant is that it is not a result of a variational principle. The equation with the Planck constant, has a flawed beauty for two reasons. First, it is not clear what it is. Introduction This paper is in depth analysis with the flaws of Quantum mechanics, by the author of the 8T, theory which unified the known interactions using variational manifolds, by doing so provided simple answer to several major unanswered questions, both in Quantum scale and in cosmological scales. Quantum mechanics is a set of ideas which derived by a set of experiments. While the experiments dictate the kind of equations and descriptions that are in the set of ideas. While one certainly cannot doubt what experiments indicate, it is possible to examine and improve the ideas which are describing reality, and in particular the methods, questions and equations. As Dirac once indicated that, there should be a more complete version of Quantum Mechanics. Let us begin with the flawed formulation. 212 Manor O First of all, the most obvious flaw in formulation is with the definition of "Energy". How can one solves equations in QM without a proper definition of "Energy"? The operator 𝐸 and 𝐻 does not mean anything, so it is possible to calculate with it, but understanding the meaning of energy is still not part of the current formulation of QM. 𝐸Ψ(𝑥, 𝑡) = 𝐻̂Ψ(𝑥, 𝑡) This equation may be solvable but it is not clear, it's too abstract and it does not tell you anything about nature. One can argue that we know the definition of energy as to the summation of kinetic and potential. If so one must also ask, how does mass form, as there exist both mass positive and massless particles. And mass plays a rule in the formulation of the kinetic and poetical energy. Below is another example of the flawed formulation of Quantum mechanics. Notice the interesting result and the major simplification. That is, in extremum energy, the equation reduces: Manor O 𝑒 (𝜕𝑔 𝜕𝑡)Ψ(𝑀𝜇,𝑡0) ⟶𝑒(0)Ψ(𝑀𝜇,𝑡0) 𝚿(𝑀𝜇,𝒕) = 𝚿(𝑀𝜇, 𝒕𝟎) 𝑒 (𝜕𝑔 𝜕𝑡)Ψ(𝑀𝜇,𝑡0) ⟶𝑒(0)Ψ(𝑀𝜇,𝑡0) If one is correct than the only parameters which will describe at extremum are the amplitudes and the matric itself. That is to say that the motion of the Electron would be an exclusive result of the space-time configuration. Assuming lowest energy, there exist only one electron in the system, and so the spin is half integer. At ground state, the electron will act as a particle. If the electron than at that stationary state would absorb a net amount of curvature, that would result in the change in its nature. 𝜕𝑔 𝜕𝑡= ∇2≠0 The new QV equation: 𝑒 (𝜕𝑔 𝜕𝑡)Ψ(𝑀𝜇,𝑡0) = 𝑒(∇2)Ψ(𝑀𝜇,𝑡0) Ψ(𝑀𝜇,𝑡) = 𝑒(∇2)Ψ(𝑀𝜇,𝑡0) Now the electron is not at stationary state, it absorbed a photon a net amount of curvature. Now the total spin of the system is an integer. That is given by the primorial spin formation. So at stationary state: (2𝑒−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝑒−)𝜇) ⟶(2𝑁2 + 1 2) With the insertion of the net curvature: With the insertion of the net curvature: (8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝑒−)𝜇) + 𝑁𝑉𝜇 ⃐ ⟶(2𝑁2 + 1 2) + 1 2 = 30: 128:850: 9254… (2𝑁2 + 1 2) + 1 2 = 128 (8 ∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝑒−)𝜇) + 𝑁𝑉𝜇 ⃐ ⟶(2𝑁2 + 1 2) + 1 2 = 30: 128:850: 9254… Now after the electron absorbed a photon it will diverge to all directions. Since it is bounded to the nuclei it will propagate across the nuclei in all directions, as a curvature ripple. The spatial dimensions are matching the Laplace. Now after the electron absorbed a photon it will diverge to all directions. Since it is bounded to the nuclei it will propagate across the nuclei in all directions, as a curvature ripple. The spatial dimensions are matching the Laplace. 𝜇= (∇2, 𝑡𝑛, 𝑠𝑛) The new equation can be described: The new equation can be described: ( ∂ ∂t)Ψ(𝑀𝜇, 𝑡) = ∇2Ψ(𝑀𝜇,𝑡0) That is much simpler an cleaner equations than the QM ones. The solving is not included yet. The most urgent issue is crafting equations which are free of arbitrary constants as such the Planck constant and in unclear undefined terms such as 𝑖ℏ and 𝑐. In here, they are not important, radical as it sounds. ⟨𝐵|𝐿̂|𝐴⟩ Second it is not clear why it has the value it has. Why that number and not another? Ψ(𝑀𝜇,𝑡) = 𝑒(𝜕𝑔/𝜕𝑡)Ψ(𝑀𝜇, 𝑡0) Where the term in the exponential stands for: Where the term in the exponential stands for: (𝜕𝑔 𝜕𝑡) = 𝑖𝐻̂∆𝑡/ℏ Notice the interesting result and the major simplification. That is, in extremum energy, the equation reduces: Notice the interesting result and the major simplification. That is, in extremum energy, the equation reduces: 213 Manor O One last modification is that this equation must specify which arrow it belongs, which universe is it contained in. ( ∂ ∂t𝑛 )Ψ(𝑀𝜇, 𝑡𝑛) = ∂2g ∂x𝑛2 + ∂2g ∂y𝑛2 + ∂2g ∂z𝑛2 Ψ(𝑀𝜇,𝑡0) + 𝑉 𝑡0 ∈𝑡𝑛;𝑛∈ℝ ( ∂2 𝜕𝑡2𝑛 )Ψ(𝑀𝜇, 𝑡𝑛) = ∂2g ∂x𝑛2 + ∂2g ∂y𝑛2 + ∂2g ∂z𝑛2 Ψ(𝑀𝜇, 𝑡0) 𝛛𝟐𝑴𝝁 𝝏𝒕𝟐𝒏 = 𝛛𝟐𝐠 𝛛𝐱𝒏𝟐+ 𝛛𝟐𝐠 𝛛𝐲𝒏𝟐+ 𝛛𝟐𝐠 𝛛𝐳𝒏𝟐 214 Manor O Matric via time is equal to rate of change of Ricci flow over spatial dimensions of a specific manifold. The quantum equations of Einstein general relativity: 𝑔𝑢𝑣= 8𝜋𝑇𝑢𝑣 Manor O Either way 8T is providing an answer to why QM operators must be flat. The areas of extremum curvature flattening each other, causing the manifold to accelerate outward, that is basic in the 8T thesis. There is a very good chance that some, and maybe all above examples do not have a computational applications, but computation is what computers do, and it does not require thinking. Still those equations are saying something as oppose to those terms in QM that are too abstract. Is it better to have a theory which is all-computational, not simple, impossible to imagine and full with vague terms and arbitrary numbers as the current form of QM? Alternatively, a theory which is highly simple and free of constants and un-important calculations such as 8T, which is the attempt for QM analogs using VC (Varying curvature) framework, a theory which in a sense deemed the highly important Planck constant and the speed of light as not necessary, as the coupling magnitudes are attainable without it, in a sense a theory which generate all the numbers of nature with zero measurements or effort using one algorithm. Orthogonality That is another dominating theme in QM old formulation which needed to be shaded light upon is the subject of orthogonal states. Is there a simpler way to explain why it is important, before diving into the unclear notation of QM? This idea is used in a sense of "distinguishable states". First of all, the uses of the word "state" is too abstract, state of what? In QV all we have is the manifold. So that is much clearer. Given two distict states of manifolds, there exist zero probability of joint union, they are disjoint. ⟨Mμ1|Mμ2⟩= 0 (Mμ)k ∈s; k ∈ℝ 𝑠= (M,g) Let us examine the following equation: 𝛿𝑖𝑗 = {0 → 𝑖≠𝑗 1 →𝑖= 𝑗 Such does not tell anything about nature. Again it's not incorrect just badly crafted. No matter how many "important calculations" can be made with equations of that sort. What we can tell about Fermions and Bosons using the Knocker delta? The idea of orthogonal states can be used in different manner and contexts. Below are several of those new ideas. First, orthogonality between Fermions and Bosons as an example. Alternatively, between Bosons and Bosons, that is to state that there could not be a Boson which Is the inner product/average of two Bosons, that is that distinguishable Bosons are orthogonal. ⟨𝑁𝑉=𝑘1|𝑁𝑉=𝑘2⟩= 0 It can also be used to describe the orthogonality of universes: ⟨𝑠𝑛|𝑠𝑛+1⟩= 0 ; 0 < n < 𝐾 ⟨𝑠𝑛|𝑠𝑛+1⟩= 0 ; 0 < n < 𝐾 But here is the interesting turn of events. Since the main equation of the 8T requires: But here is the interesting turn of events. Since the main equation of the 8T requires: ∑∂ℓ ∂sm K/2 m=1 ∂sm ∂M ∂M ∂g ∂g ∂t −∑∂ℓ ∂sn K/2 n=1 ∂sn ∂M ∂M ∂g ∂g ∂t = 0 (2.1. A) Given two distinct manifolds, which indistinguishable extremum curvatures: Given two distinct manifolds, which indistinguishable extremum curvatures: ⟨∂g ∂t𝑛| ∂g ∂t𝑛+1⟩≠0 ⟨∂g ∂t𝑛| ∂g ∂t𝑛+1⟩≠0 One must ask what is the physical implication of such an equation. It comes to an agreement with the Hamiltonian of the 8T. One must ask what is the physical implication of such an equation. It comes to an agreement with the Hamiltonian of the 8T. 𝐻̂ = 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 + ∑δgi M 1=1 𝛛𝐠 𝛛𝐭= ∑(𝛛𝐠 𝛛𝐭)𝝓 𝑲 𝝓=𝟏 215 𝑃(∆X, ∆Y, ∆𝑧) = 1 𝑃(∆X, ∆Y, ∆𝑧) = 1 And we would also expect different distributions at different times. such would exclude any preferred location for the particle. The particle will aspire to reach the lowest point on the Ricci curvature, but it is not possible to tell where this point will be. Since the particle is effected by curvature ripple, which is wavelike it will move as a smoothly as a wave. This new version of QM aspires to eliminate the use of the Planck constant from those equations. Manor O 𝑧𝐾−𝑧2 = ∆𝑧 Manor O 𝑧𝐾−𝑧2 = ∆𝑧 Manor O 𝑧𝐾−𝑧2 = ∆𝑧 In other words, the probability of finding a particle is correlated to Ricci curvature configuration at time segment. The curvature orientation is dictating the probability to find a particle at a certain location. The more flat the matric, the wider the variance, and the probability to find a particle at a certain location would be equalized. Not yet computational and maybe not at all, but still a clearer version than the formulation of QM, computational as it can be. The probability to find the particle somewhere along ∆𝑧 is one. 𝑃(X, Y, ∆𝑧) = 1 Since reality is much complicated those are oversimplifications. In real situation we would expect: Amplitudes Another major subject which appears in QM formulation is the subject of amplitudes. The square of the amplitude gives a certain probability of occurrence. In relativistic QM the phases are independent from the amplitudes. The latter can be consider an as overlap. The amplitudes in QM are synonymous with energy. Now given the arrow which takes Ricci curvature to "Energy": 𝜑: g ⟶𝐸 The probability of occurrence is proportional to the square of Ricci curvature, or the square of the new amplitude. assuming Ricci curvatures are constants, which will make the calculation easier to make. (∂g ∂t)𝜙=1 = 0 | (𝛛𝐠 𝛛𝐭)𝝓=𝟏 |𝟐= 𝑃ℝ | (𝛛𝐠 𝛛𝐭)𝝓=𝟏 |𝟐= 𝑃ℝ Shading light on the mechanism: Shading light on the mechanism: ∫ |Ψ(𝑀𝜇,𝑡𝑛)| 2𝑑𝑀 𝑀2 𝑀1 = ∇2g To make things more complete, in order to understand what is the probability to find a particle at 3D regions on the three dimensional matric 𝑀𝜇, one must compute the Ricci curvature of the system in three dimensional space, as the Ricci curvature defines the geometrical setting in which effecting the motion of the particle and thus it's potential location. As an example consider a situation where: ∂2g ∂x𝑛2 = 0, ∂2g ∂y𝑛2 = 0 , 𝛛𝟐𝐠 𝛛𝐳𝒏𝟐≠𝟎 ∂2g ∂x𝑛2 = 0, ∂2g ∂y𝑛2 = 0 , 𝛛𝟐𝐠 𝛛𝐳𝒏𝟐≠𝟎 Since the curvature does not vary in time segment on the first two spatial dimensions but only on the third, there is a major simplification and one now need to compute just one term instead of three. The particle will be found in between a range of a matric which two set of coordinate vary only in the third spatial dimension. Since the curvature does not vary in time segment on the first two spatial dimensions but only on the third, there is a major simplification and one now need to compute just one term instead of three. The particle will be found in between a range of a matric which two set of coordinate vary only in the third spatial dimension. In rigor, there exist a certain probability to find a particle in coordinate: [𝑥1, 𝑦1,𝑧2] ⟷[𝑥1,𝑦1, 𝑧𝐾] 216 The Universe at Singularity ( ∂ ∂t𝑛 )Ψ(𝑀𝜇, 𝑡𝑛) = ∂2g ∂x𝑛2 + ∂2g ∂y𝑛2 + ∂2g ∂z𝑛2 Ψ(𝑀𝜇,𝑡0) + 𝑉Ψ ( ∂ ∂t𝑛 )Ψ(𝑀𝜇, 𝑡𝑛) = ∇2 ∂2g ∂t𝑛2 Ψ(𝑀𝜇, 𝑡0) + 𝑉Ψ ( ∂ ∂t𝑛 )Ψ = ∇2 + 𝑉Ψ ( ∂ ∂t𝑛 )Ψ = ∇2 + 𝑉Ψ If before/at the moment of singularity, the universe had a finite size, i.e. not varying, we can require: If before/at the moment of singularity, the universe had a finite size, i.e. not varying, we can require: ∇2= 0 ( ∂ ∂t𝑛 )Ψ = 𝑉Ψ Since we require the potential energy to an entity which manifest itself as entity of the form: ∑δgi = 0 N i=1 ∑δgi = 0 N i=1 In addition, more recently, the potential energy of the universe was considered in the thesis: In addition, more recently, the potential energy of the universe was considered in the thesis: 𝑈= ∑δgi N i=1 + ∑δgi M 1=1 217 Manor O 𝐻̂ = 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 + ∑δgi M 1=1 = T̂ + Û Manor O 𝐻̂ = 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 + ∑δgi M 1=1 = T̂ + Û Manor O At finite size, or at the singularity arrow: At finite size, or at the singularity arrow: 𝕊:𝑡0 ⟶𝑡0 Such that Such that We are left with the term which is the Bosonic part of the arbitrary variation term which belong to: ∑𝛅𝐠𝐢 M 1=1 ⊂ ∑δgi N i=1 = 𝑉 ∑𝛅𝐠𝐢 M 1=1 > 0 ∑𝛅𝐠𝐢 M 1=1 ⊂ ∑δgi N i=1 = 𝑉 ∑𝛅𝐠𝐢 M 1=1 > 0 Such that at the moment of singularity for the manifold indexed by "𝑛" at the time 𝑡0 : ( ∂ ∂t0(𝑛) )Ψ = Ψ𝛅𝐠𝐢 Since there is not outward acceleration, as we required finite size at the moment of 𝑡0 to 𝑡0, and the manifold is still three dimensional, the arbitrary variation term on the right term is describing curvature arising from a fermion cluster and maintaining its shape. Here is term describing inner curvature composing a finite size manifold at 𝑡0. the manifold must appear flat, compressed. Now imagine that the term is finite, that will result in: ( ∂ ∂t0(𝑛) )Ψ = 0 ∪ Ψ𝛅𝐠𝐢= 0 Suppose it does vary Suppose it does vary ( ∂ ∂t0(𝑛) )Ψ = Ψ𝛅𝐠𝐢⟶Ψ𝛅𝐠𝐢+ 𝛛𝐠 𝛛𝐭+ ∑δgi N i=1 𝛛𝐠 𝛛𝐭≠0 The flattening moment. Manor O have existed before singularity, just not in a manner to allow time to manifest as we know it. For the arrow of time, we need ∂g/ ∂t ≠0 resulting in the primorial, net curvature appearing in distinct discrete amounts. The problem with the singularity can be solved using the idea of manifold spans, that the manifold was ignited by another manifold, in which time rose earlier. The moment the manifold departed it had potential curvature which broke, caused ∂g/ ∂t ≠0 and the acceleration. That was also the moment the manifold inherited a unique arrow and spatial dimensions. It is possible to imagine the manifold as a three-dimensional compressed sphere which had a non-varying state, curvature wise, for some continuous or discrete segment in time, the moment in which curvature varied ignited the acceleration and thereby the flattening moment. The Universe at Singularity That thing is rather abstract, as far as one can see, if one derived those equations correctly. The manifold has innate curvature at singularity that keeps it flat, finite in size, it does not accelerate thereby, and is at a state where it is only a potential curvature. At the moment of singularity the potential term varies, yielding a varying curvature on the manifold, agreeing with a varying time as it is possible to differentiate the states. Now the result of the innate curvature term varying is igniting that acceleration term to take a positive value, and the flatting moment comes with it. That is of course just an idea. Another interesting question is concerning time. Assuming the manifold does not vary until the term of potential term breaks, the manifold still exists, but not in such way that "time could pass". Same term that appear on black holes, which we associate freeze in time:∂g/ ∂t = 0. So the manifold could 218 Quantum Tunneling What is the analog of tunneling using varying manifolds? Since photons can travel via matter, in a solid compact formation, and Bosons are belonging to the same class as Fermions, that is also given by the Electron and it's duality to the Boson of the weak interaction, there result of this construction is the following: matter and in particular electrons can travel via matter. A statement which is synonymous with the idea of tunneling. Now since 8T is relativity new, it is less evident to date, on how to perform the calculations on the probability of tunneling, as this framework does not have constants such as the Planck, which is used in almost all calculations In QM. Instead the author will follow reason in trying to predict. It is possible to assume that there are several factors which effect the probability of tunneling. The first is the amount of matter which the tunneled particle should cross. The bigger it is the smaller the probability of crossing. That is assumed correct as the larger the matter count in volume, the bigger the chance the particle would be 'trapped' or pulled onto one on the nucleons. The second element is the tunneled particle energy, which is proportional to momenta. The higher the energy, the bigger the momenta, the particle diverges faster and thus will go via the matter cluster in faster pace, which will decrease the chance of being trapped in the matter liar. There exist another option concerning tunneling, which involves the idea of identical particles. If one had an electron in region, which was destroyed by pairing it with the positron and in the mean time, the liar of nucleons propagated from within it a distinct electron, which now is outside the region, since the two Electrons are manifested by the same number, it is impossible to distinguish them. One can conclude that the Electron crossed the barrier. (𝟐𝜇 𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + 𝒆− 𝜇) ∈[𝐴, 𝐵] 𝒆+ 𝜇∈[𝐴,𝐵] 𝒆− 𝜇+ 𝒆+ 𝜇∈[𝐴, 𝐵] = 0 𝟐𝜇 𝒆−⟶𝒆− 𝜇1 ∈[𝐶, 𝐷] 𝒆− 𝜇1 ≡𝒆− 𝜇 [𝐶, 𝐷] ∉[𝐴,𝐵] (𝟐𝜇 𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + 𝒆− 𝜇) ∈[𝐴, 𝐵] Because of the Electron duality to the Boson of the weak interaction, any feature which is related to the class of Boson, should be inherited by the Lepton. QM Axioms What really stands at the heart of the 8T that is so "hard to grasp"? The author does not think there exist such a thing. What is hard to grasp is the methods and techniques that are used to describe the reality of QM. That is the methods are the source of the problem and in particular the dominant part of LA, which is horrendous as means of trying to imagine what is happening. At the heart of it QM is composed of few simple Axioms which are: (1) The spectrum of 'Energy' is discrete (2) A physical system has a set of potential arrows leading to different results. (3) there exist a chance to each arrow. The sum of all arrows is one. (4) Objects are randomly generated. (5) Physical systems has objects, which are disjoint, joint, and semi-disjoint. That is orthogonal, identical or entangled. Distinct arrows are orthogonal. (6) Time variance of objects has an Iso-arrow to 3D spatial variance. 219 Manor O 𝔽𝐶𝑙𝑎𝑠𝑠= {2𝑛 ∪ 𝒆−| (2𝑛 ∩ 𝒆−) ∈𝑠} 𝔹𝐶𝑙𝑎𝑠𝑠 = {ℙ ∪(+𝟏)| (ℙ ∪(+𝟏) ∈𝑠} 𝑠= (𝑀𝐸,𝑔) ⟵(3,1) 𝔽𝐶𝑙𝑎𝑠𝑠∩𝔹𝐶𝑙𝑎𝑠𝑠 = 𝒆− 𝒆−≡(3) ∈ℙ The unique term that belong to both classes must exhibit the features of both classes. That insight was known long before 8T was crafted. However, the primorial validates and shade light on way those things are correct. It does so in such a simple fashion, compared to QFT which has to go via SUSY to reach the insight of alignment at 26 variations. The Atom (δg1δg1δg1) (δg2δg2δg2) (δg2δg2δg1) (δg1δg1δg2) (δg2δg1δg1) (δg1δg2δg2) (δg2δg1δg2) (δg1δg2δg1) (δg3δg3δg3) The pairing of atoms is such that inverse threefold combinations pairs. Quantum Tunneling This version of Boson Fermion duality is by the primorial second term, which is the most symmetrical. 8 + (1):(24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … [(8 ∗3) + (3)] + 𝟑 3 ≡(3) (𝑒)−= 𝑊− We can define the inheritance conduction by equalizing the classes. The Fermion class are of vanishing curvature spikes and the Electron. The Bosonic class are discrete prime amount of curvature, which are non- vanishing, and the Electron. The Electron is the unique term that belong to both classes. 220 The Atom (δg1δg1δg1) (δg2δg2δg2) (δg2δg2δg1) (δg1δg1δg2) (δg2δg1δg1) (δg1δg2δg2) (δg2δg1δg2) (δg1δg2δg1) (δg3δg3δg3) The pairing of atoms is such that inverse threefold combinations pairs. The pairing of atoms is such that inverse threefold combinations pairs. The pairing of atoms is such that inverse threefold combinations pairs. (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) That is due to the conditions of stationarity on the Lorentz manifold. The order of the pairing is not evident in the 8T, as the author did not consider it relevant at that point. The physics is was important in that context back in the day, while the 8T was still in early stages of construction. Now emphasis will be made on the formation of atoms. Since the pairing of each threefold is incomplete in a sense that not all threefold combination bring an element to itself: (δg1δg2δg2) That element will pair to another threefold combination that is imperfect and include the inverse signs. That is synonymous with the process of just two arbitrary variations vanishing to zero, given by the stationarity condition. It is possible to examine the threefold combinations in terms of edges. Those edges will pull another threefold 221 Manor O Manor O Manor O Manor O combinations, thereby creating elements with heavier Hadrons, i.e. large number of Hadron composites. (δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) ⟷(δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) And all of those formations are due to the condition of stationary manifold. δg1 + δg2 … = ∑𝛿g𝑖 𝑁 𝑖=1 ∑𝛿g𝑖= 0 𝑁 𝑖=1 Another important note is the following. The most stable state in the set of the threefold combinations must be the lowest with the lowest energy. Since we had the mapping from Ricci curvature to energy, the most stable threefold combination must be most flat. Now another point which is important is the following. It is impossible to associate to each threefold combination different degrees of curvature, i.e. energy. The threefold combination does not tell how volatile are the arbitrary amount of curvature. To demonstrate: (δg1δg2δg1) ⟶𝐸121 ℒ0 (δg1δg2δg1) ⟶𝐸121 ℒ1 𝐸121 ℒ0 ≠𝐸121 ℒ1 Such a construction will allow us to indicate that the direction of development would be as such that threefold combination will aspire lowest energy state. That is in agreement with 8T idea of emission of the electron. And it is also similar to the Quantum formulation of subscript and superscript on the electron, to indicate his aspiration to reach lowest energy state. We can make the transformation between energy state of threefold combinations. 𝐸121 ℒ1 = 𝐸121 ℒ0 + 𝒆− Using that idea it is also possible to solidify the direction of the families formation. As the manifold develops, i.e. accelerates due to being a part of the packet, flatter and flatter combinations should rise. That means lower and lower masses. In the 8T, the Quark masses pattern indicate to that direction, i.e. third family is the first and first family is the third. 19,600 ⟶1400 ⟶56 176,400 ⟶1400 ⟶6.3 ⟶0.113 ⟶ Using those theoretical insights it is possible to reason for the similarity of the generations, they are the class of arbitrary variations, which differ in their level intensity. The latter is proportional to the arrow of time. 222 (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) Gravity within Fermion Clusters (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) At any given time an heavy element which has two "clouds of probability", i.e. electrons, and those electron propagate on a common segment, there exist a chance those the Electrons will emit a discrete amount of net curvature at the same time. That spatial alignment and temporal alignment will result in the Graviton. That nature of the composition given by the spin two trait, is what makes the Graviton short range. (𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇+ (𝒆−)𝜇) + 𝑁𝑉1𝜇+ 𝑁𝑉2𝜇= 2𝑁() + 2 (𝑁𝑉2𝜇≡𝑁𝑉1𝜇) ∪(𝑁𝑉2𝜇≢𝑁𝑉1𝜇) (𝟐𝒆−∗ ∏𝑁𝑉𝜇 𝑉=𝑅 𝑉=1 + (𝒆−)𝜇+ (𝒆−)𝜇) + 𝑁𝑉1𝜇+ 𝑁𝑉2𝜇= 2𝑁() + 2 (𝑁𝑉2𝜇≡𝑁𝑉1𝜇) ∪(𝑁𝑉2𝜇≢𝑁𝑉1𝜇) The only condition one is requiring is the five vector to be aligned, that the ripples will interest, both on the spatial and temporal. Since both contain the Laplacian in the five vector, and time as well, it is important to align all the elements on the five vector. Than the decay of the Graviton can be put as two Photons/or any two Bosons and two Electrons. (𝒆−)𝜇+ (𝒆−)𝜇+ 𝑁𝑉1𝜇+ 𝑁𝑉2𝜇⟷G where there Thus the Graviton may likely be rising at large scale Fermion formations, exit immense amount of Leptons which may emit together, yielding an higher spin particle, such as the Graviton. In the 8T, for those reasons has infinite combinations. primorial are Gravity, i.e. curvature is the class, where different objects, given by the rising. The interaction among starts than is taking place by long range meditator such . as light, which is represented by one independent term of prime Immense Spin Formations p (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) (δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) ⟷(δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) (δg1δg1δg1) ⟷(δg2δg2δg2) ⟷(δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2𝛅𝐠𝟐) ⟷(δg2δg1𝛅𝐠𝟏) ⟷(δg1δg2𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg2) (δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) ⟷(δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) (δg1δg1δg1) ⟷(δg2δg2δg2) ⟷(δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2𝛅𝐠𝟐) ⟷(δg2δg1𝛅𝐠𝟏) ⟷(δg1δg2𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg2) Using the endless clustering of matter, i.e. construction of the periodic time table, which will eventually yield a proportional number of electrons as each of those threefold combination will and can emit an electron given by the primorial. (δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) ⟷(δg2δg1𝛅𝐠𝟐) ⟷(𝛅𝐠𝟏δg2δg1) ⟶K𝒆− K ∈ℝ Since those electron has spin, given the second form of the primorial, the immense cluster of electrons and therefore the Bosons they emit contain spin as well. The summation of spin must be valid in large-scale formation of matter, i.e. arbitrary variations of the manifold, which takes the form of threefold combinations of two distinct elements, which differ in sign and summed as zero, or thanks to the contributions of physicists, Quarks. The large formation of matter must posses spin, or angular momenta around a self-Axis. The spin summation is due to the contribution of Electrons and Bosons in the cluster. 𝒮= ∑𝒆− 𝑁 𝑖=0 + ∑𝑍𝑘 𝑁 𝑘=1 ∑𝑁𝑉𝑘 𝑁 𝑘=1 𝒮∈∑𝛿g𝑖= 0 𝑁 𝑖=1 The two-fold summation reason is the following. One must sum across the kind of Bosons in play inside the matter cluster. Such a construction allow one to make a prediction: (1) All mega scale Fermion formations, stars and galaxies must have spin. The matter spirals of galaxies should spin around the axis of the center of galaxy and stars should spin around a self-axis taken from pole to pole. 223 (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) Manor O Manor O Manor O Imagine that a threefold combination could break to a certain spatial coordinate, its matched pair is breaking to another spatial coordinate, that is a violation of stationarity on the manifold. ∑𝛿g𝑖≠0 𝑁 𝑖=1 ∑𝛿g𝑖≠0 𝑁 𝑖=1 Within each threefold combination, there must be than at least by reason a threefold binder. The nature of the threefold binder is related to the size of the prime paring, given by theorems two and three of the 8T. for the threefold binder the element is represented by 𝑁𝑉1𝜇= +1. Those threefold binders are net curvature on the manifold. Each net is increasing the chance of another net to arrive to its positon. The result an endless succession of net curvature converging toward the threefold combination, or a "sea of Gluons", which ensures the trapped nature of the two distinct elements. (δg1 (↔)δg1 (↔)δg1) ⟷(δg2 (↔)δg2 (↔)δg2) (⟷) = ∑g𝑖= +1 𝐾 𝑖=1 + 1 + 1 … (⟷) = ∑g𝑖= +1 𝐾 𝑖=1 + 1 + 1 … Curvature scattering by matter Within a threefold combination we assumed there exist a sea of threefold binders. That is given by the term: (δg1 (↔)δg1 (↔)δg1) ⟷(δg2 (↔)δg2 (↔)δg2) Imagine the following scenario in which higher coupling Boson is propagating directly into matter. Since the threefold combination is emitting the Lepton and the Lepton is responsible for the Boson emissions, it is possible to assume that for the higher coupling Boson will not penetrate directly but rather by scattered by the threefold combination. Imagine the following scenario in which higher coupling Boson is propagating directly into matter. Since the threefold combination is emitting the Lepton and the Lepton is responsible for the Boson emissions, it is possible to assume that for the higher coupling Boson will not penetrate directly but rather by scattered by the threefold combination. 𝑁𝑉1𝜇⟶(δg1 (↔)δg1 (↔)δg1) ⟷(δg2 (↔)δg2 (↔)δg2) ↘𝑁𝑉1𝜇 It is getting interesting, it is possible to assume that the higher term Boson did get in to the threefold cluster, and the scattered Boson is composed by the same amount of distinct elements. such that the photon which came in, is not the photon which came out, but five net curvature which were in the original Fermion. ↘𝑁𝑉𝜇= ∑g𝑖= +1 5 𝑖=1 + 1 + 1 + 1 + 1 Threefold Binders (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) 224 Epilogue ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ ∂2g′ ∂t2 = 0 (1) Epilogue ∂ℓ ∂s ∂s ∂M ∂M ∂g ∂g ∂t −∂ℓ ∂s′ ∂s′ ∂M ∂M ∂g′ ∂2g′ ∂t2 = 0 (1) Curvature Subsets The subset condition, for any fermion cluster we have, which Bosons propagate within it, by the three critical theorems of the 8T, we have that the Bosonic class is a subset of the Fermion cluster. That is the: ∑𝑍𝑘 𝑁 𝑘=1 ∑𝑁𝑉𝑘 𝑁 𝑘=1 ⊂(∑𝛿g𝑖= 0 𝑁 𝑖=1 ) 225 Manor O The Road to Reality (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (𝛅𝐠𝟏δg2δg1) ⟷ ⏞ (𝒆−)𝜇 (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (𝛅𝐠𝟏δg2δg1) (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇 (δg2δg2δg2) ⟷ ⏞ (𝒆−)𝜇 (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇 (δg2δg2δg2) (δg1δg2𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (δg2δg1𝛅𝐠𝟏) ⟷ ⏞ (𝒆−)𝜇 (δg1δg2𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (𝛅𝐠𝟏δg2δg2) The Road to Reality (δg3δg3δg3) (δg1δg1δg1) ⟷(δg2δg2δg2) (δg1δg2δg2) ⟷(δg2δg1δg1) (δg1δg2δg1) ⟷(δg2δg1δg2) (δg2δg2δg1) ⟷(δg1δg1δg2) (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (𝛅𝐠𝟏δg2δg1) ⟷ ⏞ (𝒆−)𝜇 (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (𝛅𝐠𝟏δg2δg1) (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇 (δg2δg2δg2) ⟷ ⏞ (𝒆−)𝜇 (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇 (δg2δg2δg2) (δg1δg2𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (δg2δg1𝛅𝐠𝟏) ⟷ ⏞ (𝒆−)𝜇 (δg1δg2𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇 (𝛅𝐠𝟏δg2δg2) (δg2δg2δg1) ⟷ ⏞ (𝒆)𝜇+𝑁𝑉𝜇 (δg1δg1δg2) (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (𝛅𝐠𝟏δg2δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (𝛅𝐠𝟏δg2δg1) (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (δg2δg2δg2) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (δg2δg2δg2) (𝒆−)𝜇+𝑁𝑉𝜇 (𝒆−)𝜇+𝑁𝑉𝜇 (𝒆−)𝜇+𝑁𝑉𝜇 (δg2δg2δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇 (δg1δg1δg2) (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (𝛅𝐠𝟏δg2δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (δg2δg1𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (𝛅𝐠𝟏δg2δg1) (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (δg2δg2δg2) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (δg1δg1δg1) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (δg2δg2δg2) (δg1δg2𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (δg2δg1𝛅𝐠𝟏) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (δg1δg2𝛅𝐠𝟐) ⟷ ⏞ (𝒆−)𝜇+𝑁𝑉𝜇⟶ (𝛅𝐠𝟏δg2δg2) 226 Manor O Manor O Epilogue (1) In contrast to other theories, 8T is a result of individual effort, as such there could be inaccuracies in certain parts of particle physics, such as the graph in beginning, as one is a not a particle physicist but a Theorist. One would like to commend the reader for taking the time to analyze this thesis. There are still some open questions that are currently beyond 8T reach, such as the variational principle of the Fermion masses and mixing angles. Despite that, 8T showed that nature is governed by be reason, that things are surprisingly simple and beautiful at the most fundamental level. As a giant of history and an inspiration, Steven Weinberg stated in 1990: "it’s a faith that we have, that the final answer has to be a simple one". Simple it is, indeed. beautiful at the most fundamental level. As a giant of history and an inspiration, Steven Weinberg stated in 1990: "it’s a faith that we have, that the final answer has to be a simple one". Simple it is, indeed. 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … 8 + (1): (24 + (3)) + 3: (120 + (3)) + 5: (840 + (3)) + 7 … Type text here Manor O Digitally signed by Manor O Date: 2021.10.15 04:43:20 +03'00' Type text here Manor O Digitally signed by Manor O Date: 2021.10.15 04:43:20 +03'00' Manor O Digitally signed by Manor O Date: 2021.10.15 04:43:20 +03'00' 227

https://openalex.org/W4320484778

https://link.springer.com/content/pdf/10.1007/s11207-023-02183-w.pdf

English

The SWAP Filter: A Simple Azimuthally-Varying Radial Filter for Wide-Field EUV Solar Images

Research Square (Research Square)

cc-by

Solar Physics (2023) 298:92 https://doi.org/10.1007/s11207-023-02183-w Solar Physics (2023) 298:92 https://doi.org/10.1007/s11207-023-02183-w RESEARCH RESEARCH Abstract We present the SWAP Filter: an azimuthally varying, radial normalizing filter specifically developed for EUV images of the solar corona, named for the Sun Watcher with Active Pixels and Image Processing (SWAP) instrument on the Project for On-Board Autonomy 2 (PROBA2) spacecraft. We discuss the origins of our technique, its implementation and key user-configurable parameters, and highlight its effects on data via a series of examples. We discuss the filter’s strengths in a data environment in which wide field-of-view observa- tions that specifically target the low signal-to-noise middle corona are newly available and expected to grow in the coming years. Keywords Corona · Instrumental effects · Instrumentation and data management · Image processing The SWAP Filter: A Simple Azimuthally Varying Radial Filter for Wide-Field EUV Solar Images Daniel B. Seaton1 ·David Berghmans2 ·D. Shaun Bloomfield3 · Anik De Groof4 ·Elke D’Huys2 ·Bogdan Nicula2 ·Laurel A. Rachmeler5 · Matthew J. West1 Received: 9 February 2023 / Accepted: 16 June 2023 / Published online: 18 July 2023 © The Author(s) 2023  D.B. Seaton [email protected] 1 Southwest Research Institute, Boulder, CO, USA  D.B. Seaton [email protected] 1 Southwest Research Institute, Boulder, CO, USA 2 Solar-Terrestrial Centre of Excellence–SIDC, Royal Observatory of Belgium, Brussels, Belgium 3 Department of Mathematics, Physics and Electrical Engineering, Northumbria University, Newcastle Upon Tyne, UK 4 European Space Agency, European Space Astronomy Centre, Madrid, Spain 5 National Centers for Environmental Information, National Oceanic and Atmospheric Administration, Boulder, CO, USA 1. Introduction A problem of understated difficulty and importance for observational coronal physics is simply finding ways to display and analyze observations so they are actually useful. The need for such techniques stems from a number of sources, but it generally boils down to a few key facts: the corona has a tremendous dynamic range in brightness, many interesting features and events are inherently faint and hard to detect, and various sources of noise can interfere with both imaging itself and can be amplified by image-processing techniques.  D.B. Seaton [email protected] 92 Page 2 of 17 92 Page 2 of 17 D.B. Seaton et al. 92 Pa Noise sources include instrumental effects and photon shot noise embedded in the signal itself. Noise sources include instrumental effects and photon shot noise embedded in the signal itself. This problem is not unique to observations in the extreme ultraviolet (EUV), but it has emerged as an especially important one in this spectral band, particularly in light of the de- velopment of new EUV imagers – and new ways of using these imagers – that can observe the corona to heights as large as ≈6 R⊙. Such imagers include the Sun Watcher with Active Pixels and Image Processing (SWAP) on the Project for On-Board Autonomy 2 (PROBA2) spacecraft (Seaton et al., 2013b; Halain et al., 2013), the Extreme-Ultraviolet Imager on Solar Orbiter (Rochus et al., 2020; Auchère et al., 2023a; see also the observations of a prominence eruption to >6R⊙reported by Mierla et al., 2022), and the Solar Ultraviolet Imager (SUVI) on NOAA’s GOES-R line of spacecraft (Seaton et al., 2020; Darnel et al., 2022). Both SWAP and SUVI have fields of view (FOV) that can reveal features in the EUV corona to heights greater than 2 R⊙, and both have been used in off-pointed campaigns that revealed EUV corona structures to substantially larger heights, as great as ≈5 R⊙(e.g. Goryaev et al., 2014; O’Hara et al., 2019; Tadikonda et al., 2019; Seaton et al., 2021). Ob- serving the region between 1.5 – 6 R⊙– the middle corona (West et al., 2023) – is important for a wide variety of scientific questions, including coronal-mass-ejection (CME) initiation, solar-wind acceleration, and characterizing the processes that give the corona its large-scale structure. 1. Introduction Upcoming and proposed instruments, including the EUV CME and Coronal Connectivity Observatory (ECCCO: Golub et al., 2020) and the Sun’s Coronal Eruption Tracker CubeSat (SunCET: Mason et al., 2021, 2022), should reveal the EUV middle corona in dramatically more detail than has been possible, even with SWAP and SUVI. All the new observations that these instruments yield point to the great need for image-processing techniques that can help improve the accessibility of interesting features and events buried in these datasets. In fact, this need is also shared by other data sets, including both those from the Asso- ciation of Spacecraft for Polarimetric and Imaging Investigation of the Corona of the Sun (ASPIICS) coronagraph (Shestov and Zhukov, 2018; Shestov, Zhukov, and Seaton, 2019) on the formation-flying PROBA3 mission and ground-based eclipse observations (Habbal et al., 2007, 2011) This and other new instruments will also observe the corona in largely the same region, and they will likely encounter many of the same challenges, albeit in other wavelength regimes. In this article, we briefly describe our approach for processing these images to man- age dynamic range and reveal features that would otherwise be invisible, which we refer to as the SWAP Filter, thanks to its origins. These techniques have been demonstrated exten- sively on EUV coronal observations, including observations of large-scale EUV structure and evolution (Seaton et al., 2013a), eruptive flares (Seaton and Darnel, 2018), and off- pointed observations (Seaton et al., 2021), but they have not yet been described in detail in the literature. In addition to the technique described here, there are many other resources for simi- lar or related image processing available, most of which are included in common soft- ware packages such as SolarSoft IDL, the SunPy package Sunkit-Image, jHelioviewer, or other open-source repositories. These include Normalizing-Radial-Graded Filtering (NRGF: Morgan, Habbal, and Woo, 2006), Fourier-normalizing Radial Gradient Filtering (FNRGF; Druckmüllerová, Morgan, and Habbal, 2011), Noise Adaptive Fuzzy Equalization (NAFE: Druckmüller, 2013), Multiscale Gaussian Normalizaton (MGN: Morgan and Druckmüller, 2014), Wavelet-Optimized Whitening (WOW: Auchère et al., 2023b), jHelioviewer’s r3 ra- dial brightness scaling (Müller et al., 2017), and numerous other simpler techniques (e.g. Patel et al., 2022). The SWAP Filter for EUV Solar Images Page 3 of 17 92 Although today this problem is typically addressed using software, techniques to manage dynamic range in coronal imaging have their origins many decades ago, during the era of photographic observations. 1. Introduction One of the earliest examples was developed by Gordon Newkirk, who found that photographic images lacked sufficient dynamic range to capture the detail in the corona visible by eye during a total solar eclipse. In response, Newkirk devised an optical, radially graded neutral-density filter that successfully compensated for the steep brightness gradient of the corona (Eddy, 1989). Newkirk first deployed his filter at an eclipse in 1966 in Bolivia, generating a stunning image of the corona that arguably revealed more detail than was visible to eye (Newkirk, 1967). Subsequently, Newkirk-style filters were widely used for eclipse photography before digital imaging made it possible to assemble high-dynamic range (HDR) composites using multiple exposure times instead of radial- normalizing filters. Although the origins of our technique are firmly rooted in the era of space-based elec- tronic imaging, the fundamental motivation is the same as Newkirk’s: during our work with SWAP images, we realized that there was more to see in our images than we could display using simple techniques. In this article, we describe the problem and the simple observation that led to the development of our approach (Section 2), describe the approach itself and some of its configurable parameters (Section 3), and finally highlight a few examples of its applications (Section 4) before making a few concluding remarks (Section 6). 2. The Origins of Our Filter Technique Figure 1 The same SWAP image, displayed on a linear scale (left), log scale (center), and 1/4-power (rig Figure 1 The same SWAP image, displayed on a linear scale (left), log scale (center), and 1/4-power (right). function of height above the disk. Typically one generates the filter by sampling the coronal brightness in successive rings at greater heights, and then dividing the coronal image by the resulting array to eliminate the gradient with height, leaving only local variations to be displayed.1 Although filters like this have proven highly successful for visible-light coronal imaging, we found that this simple approach was not satisfactory for large-scale EUV images due to the large variation in brightness between polar coronal holes and equatorial streamers and other bright features. We sought another approach to “normalize” – that is divide – the data that could be adapted to adjust the degree of data scaling as a function of azimuth around the Sun. (Note that throughout this article we adopt the term normalize to refer to a process in which data are rescaled by division by a function or array.) Because at the time we were working with observations that spanned complete Carrington rotations, we had a large amount of data that revealed the evolution of the Sun and corona over a full solar rotation – or even longer – readily available to work with. As an initial experiment, we adopted a concept similar to the monthly minimum images (e.g. Section 10 in Morrill et al., 2006), which are used to subtract background (stray light and F-corona) from coronagraph observations, to see whether a similar type of stacking might yield an image that we could use to normalize our data. This approach turned out to be very effective, because long-term averages smooth out most (but not all) local variations and leave a largely smooth – but azimuthally variable – fall-off in intensity that can serve as the basis to normalize the height gradient of an image. Because the large-scale features in the off-disk corona often persist for many months (Seaton et al., 2013a), long-term image stacks preserve some of the variation of coronal brightness as a function of latitude, which leaves a normalizing filter that acts differently at equatorial latitudes, where bright active regions are often present; at mid-latitudes, where extended coronal fans are often observed; and at the poles, where coronal holes are most common. 1An implementation of this type of filter is available in the SunPy Examples Gallery: docs.sunpy.org/en/ stable/generated/gallery/computer_vision_techniques/off_limb_enhance.html#sphx-glr-generated-gallery- computer-vision-techniques-off-limb-enhance-py. 2. The Origins of Our Filter Technique Early in the SWAP mission, we realized that SWAP’s large FOV might reveal features and evolution in the solar corona that had not been well observed previously in these wave- lengths. We began to produce movies composed of deep-exposure images spanning full Carrington rotations, using median stacking – that is, computing the median of each image pixel in the temporal direction – of many individual SWAP observations to suppress noise and enhance the visibility of faint features in the middle corona (see Section 2.2 in West et al., 2022). We quickly realized, however, that without a technique to equalize the dynamic range in the corona, there was no path to take full advantage of these new observations, since we could not clearly visualize the large-scale features in the data. Figure 1 highlights this dilemma, showing an example of a deep-stacked SWAP im- age from 28 October 2014 at 13:21:13 UT, displayed using three different data-scaling ap- proaches. The left panel uses a linear scale, which results in an image with too much contrast to be shown on typical displays. A common approach to address this problem is to use a logarithmic scaling (center), but this tends to suppress fine variations too much to visualize anything other than the very large-scale features. A third approach is to raise the image val- ues to a power or other nonlinear function, (e.g. Lupton et al., 2004), which reduces dynamic range enough to reasonably display the image without washing out fine features. We found a good value for SWAP observations was the 1/4-power, but found it was still not adequate to provide a detailed view of the entire corona within the FOV. Thus, while these various data scaling approaches are used widely for narrow-field EUV image data, we found that none were adequate to capture both large-scale structure, which can vary in brightness by up to five orders of magnitude over the FOV, and fine features, which often represent variations of < 1%. A straightforward approach to overcoming this challenge is to rescale the data using a symmetrical radial-gradient normalizing filter. Much like Newkirk’s analog implementa- tion, these filters are used to neutralize the steep radial falloff of coronal brightness as a 92 Page 4 of 17 D.B. Seaton et al. Figure 1 The same SWAP image, displayed on a linear scale (left), log scale (center), and 1/4-power (right). 2. The Origins of Our Filter Technique In our case, we found median-stacking the data to be most effective approach to generating these filters, since the median suppresses both temporal noise and transient events that might otherwise skew the behavior of the filter. g Figure 2 shows an example of a long-term median-stack image generated using all SWAP observations during Carrington Rotation 2156, and a sample of an image (the same image Page 5 of 17 92 Page 5 of 17 92 The SWAP Filter for EUV Solar Images Figure 2 (left) Running median of all SWAP images during Carrington Rotation 2156, displayed at 1/4- power and (right) the result of using this running median as an image filter to normalize the same test image as in Figure 1 (28 October 2014 13:21:13 UT). An accompanying animation in the Electronic Supplementary Material shows the results of applying this type of filter to a full solar rotation’s worth of observations. Figure 2 (left) Running median of all SWAP images during Carrington Rotation 2156, displayed at 1/4- power and (right) the result of using this running median as an image filter to normalize the same test image as in Figure 1 (28 October 2014 13:21:13 UT). An accompanying animation in the Electronic Supplementary Material shows the results of applying this type of filter to a full solar rotation’s worth of observations. shown in Figure 1) processed using this image as the basis for a normalizing filter. In addi- tion to their value in equalizing the brightness in the off-disk region, an important feature of these filters is the bright bands left by active regions as they transit the solar disk. When used to normalize a full-Sun image, these bands suppress a bit of brightness at latitudes where the most prominent active regions appear, and comparatively amplify quiet-Sun fea- tures. The net result is a disk image that is normalized so both bright active regions and dark coronal holes and filaments are well scaled and overall contrast is better than what can be achieved with simpler image-normalization techniques. It is important to note that the filtered images should not be used to determine absolute brightness or density on their own, but they are excellent for determining where signal is coming from, and characterizing coronal morphology. 3. Implementation and Use Our strategy uses the three key features that we identified in filters based on Carrington- rotation image stacks discussed above: it captures the fall-off in brightness above solar limb, it captures the slow variation of this gradient as a function of azimuth, and captures the lati- tudinal distribution of brightness on the solar disk. The method to generate filter arrays is im- plemented in a SolarSoft IDL function, called p2sw_image_filter.pro in the PROBA2/SWAP software package. The generation of the filter is mostly implemented in two straightforward steps, both of which leverage transformations of coordinates that facilitate data smoothing and averaging in directions that are not practical in image coordinates. We generate the off-disk part of the filter by reprojecting the part of the image above the limb into polar coordinates, with azimuthal angle along the horizontal axis and radial height on the vertical. By smoothing the image in the azimuthal direction with a spatial median filter, which does not affect the radial direction, we generate an array that varies slowly in azimuth, but captures the remaining local gradient with good fidelity. We pad the edges of the transformed image with copies of itself to ensure that the processing wraps smoothly around the Sun at the discontinuity in the azimuthal angle (i.e. 0◦/360◦). After the smoothing process, we transform the processed array back from radial coordinates to Cartesian image coordinates. We generate the on-disk part of the filter by transforming the disk image to a simple cylin- drical map projection, which accounts for the observatory’s viewing angle in heliographic latitude (that is, the B0-angle). We then compute the median value for each latitude (that is, across the longitudinal direction) and replace all the pixels in the map with their appropriate median. Again, we transform this back to image coordinates. These two processed parts of the filter are merged into a single image and modulated using a set of user-configurable parameters (described below), yielding a basis image to perform radial filtering that shares its broad characteristics with the Carrington-rotation- stack images that gave us the idea for this approach. Figure 3 shows the resulting filter and processed image, displayed using the same param- eters as Figure 2. 2. The Origins of Our Filter Technique Because our experimental strategy leveraged all of the data for a full solar rotation, it turned out that these filters were extremely useful in processing movies of the evolution of the extended EUV corona over one or more solar rotations. (See the animation accom- panying Figure 2 for an example.) However, although this experimental approach yielded dramatic results, it suffered from an important drawback: because the method requires a full Carrington rotation of SWAP observations to generate the filter, it was only practical for processing SWAP’s long-duration movies, when the data were already available. Thus we sought an alternative approach that could be more easily deployed for short-duration events such as CMEs and solar flares, but would share most of the benefits that we realized from this technique. The Carrington-rotation-filter method yielded filters with several specific, helpful fea- tures: First, they capture the bright limb and the gradual falloff in brightness with increas- ing height, which largely determines the overall dynamic range of the image. Second, they capture gradual changes as a function of azimuthal angle, allowing the filter to adapt to ap- propriately normalize both bright extended features at lower latitudes and fainter features in the coronal-hole regions near the poles appropriately. Finally, they achieve good nor- malization across the disk and off-limb regions, specifically improving contrast on the disk because they include band-like features that result from the passage of either active regions 92 Page 6 of 17 D.B. Seaton et al. 92 Page 6 of 1 or darker features, such as coronal holes and filaments, at different latitudes. In Section 3 we describe how we translated these naturally derived features into a filtering approach that requires only a single input image and subsequently applied this filter technique to a variety of observations from SWAP and other instruments. or darker features, such as coronal holes and filaments, at different latitudes. In Section 3 we describe how we translated these naturally derived features into a filtering approach that requires only a single input image and subsequently applied this filter technique to a variety of observations from SWAP and other instruments. 3. Implementation and Use The method that we describe here is more sensitive to the specific features visible in the image used to manufacture the filter than the long-term stacking approach is, and thus it more strongly modulates the characteristics of these specific features. The result- ing image has overall less contrast than the processed image in Figure 2, but in each case both the overall global structure of the corona and its local-scale variations, both on- and off-disk, are far more visible than in any of the images in Figure 1. The filter’s overall sensitivity to local features can be adjusted somewhat by altering the width of the median filter used off-disk, which is one of several user-configurable parameters in the code. Here we used a value of ±15◦, which we found matches the scale of many typ- ical off-disk coronal features, but which can be adjusted to emphasize different scales. Very large values diminish the filter’s ability to adapt azimuthally to changing coronal brightness, while very small values make the filter too sensitive to local variations and suppress all but the most fine-scale features. The filter can also be tuned by an overall smoothing parameter, Page 7 of 17 92 The SWAP Filter for EUV Solar Images Figure 3 Filter image generated by our method displayed at 1/4-power (left) and the result of using this filter on our test image from 28 October 2014 13:21:13 UT (right). Though specific features differ somewhat between this and the processed image in Figure 2, the overall results are comparable without the requirement of large quantities of data. Figure 3 Filter image generated by our method displayed at 1/4-power (left) and the result of using this filter on our test image from 28 October 2014 13:21:13 UT (right). Though specific features differ somewhat between this and the processed image in Figure 2, the overall results are comparable without the requirement of large quantities of data. which is useful both for setting the scale of features to emphasize and suppressing small artifacts that sometimes develop during the filter generation. Here we used a 2D Gaussian blur with a σ-value of four pixels on the final filter in Cartesian space, but both the blurring method and width are adjustable depending on the user’s needs. 3. Implementation and Use also enhance noise in faint regions as well, thus there is always a trade-off to make between how aggressively to filter regions where the signal-to-noise ratio is low. For the SWAP images in both Figures 2 and 3, we set the constants in Equation 1 to be t0 = 1.5 and c0 = 0.75. In our experience, the distribution of brightness in typical wide-FOV EUV images is such that the value of t0 should be roughly the median of all pixels in the filter image, but manual fine-tuning is often required to obtain the user’s desired appearance for the image. Note that the value of t0 differs between the two techniques that we illustrate in this article, so we select t0 = 1.5, which is larger than the median of the Carrington- rotation-stack and smaller than the median of our method’s filter, for both cases to ensure the overall image is treated roughly equally by both methods. Figure 4 illustrates why it is necessary to rescale the filter using c0 before processing an image using some cross-sectional cuts through the data in our sample image. If we do not compress (or “crush”) the filter dynamic range, dividing by the filter will completely flatten all variation in brightness with height above limb, which results in an image that looks significantly modified from the original observations to many viewers. Crushing this function using c0 = 0.75 has little effect on how the filter treats faint features, but reduces somewhat how much it flattens bright features, resulting in an image that retains some – but much less – of the gradient from limb to the large heights and is both appropriate for display and more natural in appearance. Figure 5 illustrates the results of neglecting these two parameters with two processed im- ages. This yields images that either lack contrast (the c0 = 1.0 case; left) or have excessive enhancement of the background (the c0 = 0.75, t0 = 0. case; right). The overall result of appropriate selection of these parameters, as shown in Figure 3, yields a final product that achieves balance between contrast, dynamic range, and resemblance to the original obser- vation without excessively amplifying noise in dark regions. 3. Implementation and Use There are two other very important parameters that must be set to obtain a high-quality, low-noise, processed image that retains the overall natural appearance of typical solar EUV images: a bright limb, bright active regions, and a gradual decrease in brightness with height. (Such aesthetic choices are not necessarily requirements of a generic image filter; they help meet one of the objectives of our specific filter: namely, to generate images and movies for public-outreach materials.) The filter, as derived, acts uniformly across the entire image. If applied with no adjustments, the brightest features will be re-normalized to have the same brightness as the faintest features, while the faintest features, which are also the noisiest in most images, will be enhanced to match the brightness of all other features. The result is an image that both lacks contrast – or is excessively “flat” in appearance – and excessively noisy in its faintest regions. We therefore must tune the filter slightly to avoid both of these outcomes. This is straightforwardly achieved by modulating the overall filter array [F] to yield a final filter  F′ , in the following way, F′ = (F + t0)c0 . (1) (1) Here, t0 is a constant offset, which we refer to as the tapering parameter, that depends on the image background brightness (or can be adjusted by the user) and reduces the amount by which the filter enhances the faintest pixels – and specifically the noise in these regions. c0 is a factor that we refer to as the crush factor and is defined such that 0 < c0 < 1. This factor reduces the overall normalization effect of the filter, particularly on bright values. Careful tuning of these parameters can significantly improve the overall image result, but any filter that amplifies the faintest part of the image without also applying noise reduction is likely to 92 Page 8 of 17 D.B. Seaton et al. 92 Page 8 of 17 Figure 4 Brightness of a cross-section of our test image from South Pole to the edge of the FOV compared to the unprocessed filter in Figure 3 (green) and the same rescaled using c0 = 0.75 (gold). Figure 4 Brightness of a cross-section of our test image from South Pole to the edge of the FOV compared to the unprocessed filter in Figure 3 (green) and the same rescaled using c0 = 0.75 (gold). 4. Applications Although we have refined these filtering techniques somewhat since first developing them, largely to make them more robust to anomalies in image data, they have been used effec- tively on a variety of EUV image data and applications, including tracking the long-term evolution of large-scale features in SWAP data (Seaton et al., 2013a), observations of erup- tive solar flares in AIA (Seaton, Bartz, and Darnel, 2017) and SUVI (Seaton and Darnel, 2018; Veronig et al., 2018), as well as the EUV waves that often accompany these events (de Koning, Pizzo, and Seaton, 2022), and even on simulated observations of eruptions (Ma- son et al., 2022) from planned instruments (Mason et al., 2021). These techniques have proven especially valuable for ultra-wide-field observations of the EUV corona (Tadikonda et al., 2019; Seaton et al., 2021), where the dynamic range from limb to 3 R⊙can reach almost five orders of magnitude. Figure 6 contrasts the detailed struc- ture visible in wide-field EUV images from SUVI when processed with our filter with the minimal structure visible when these images are displayed using log-scaling. Enhanced ob- servations such as these make it possible to characterize the very faint dynamics associated with the origins of the solar wind’s embedded structure (Chitta et al., 2022). Figure 7 and the accompanying animation show how processing with our filter makes visible multi-scale dynamics in wide-field EUV observations. The movie makes visible both small, ubiquitous, jet-like features and several large eruptions near the east limb. These faint variations would not be detectable compared to the steep radial gradient in the brightness without this processing. See Seaton et al. (2021) for a complete discussion of the processing required for these SUVI mosaic images and the origins of artifacts in the resulting images and movies. 3. Implementation and Use where, I0 is the unprocessed, calibrated image and F′ is the filter generated in Equation 1. I′ is re-scaled as a byte-array using the values indicated. 3. Implementation and Use Two final choices remain before the processed image can be displayed: how to scale the final filtered image for display, if at all, and how to set the appropriate limits to clip the displayed image’s dynamic range. We found that our filtered images still have enough variation with height that it is useful to adjust the image scaling, and further that scaling The SWAP Filter for EUV Solar Images Page 9 of 17 92 Page 9 of 17 92 The SWAP Filter for EUV Solar Images Figure 5 Comparison of a filtered image with c0 = 1.0 and t0 = 1.0 (top left; i.e. no rescaling, but including low-end tapering), resulting in a “flat” looking result and with c0 = 0.75, but t0 = 0.0 (top right; rescaling, but no low-end tapering), which leads to an over-enhancement of background brightness and low-level temporal noise, and (bottom) the optimally processed image from Figure 3 (28 October 2014 13:21:13 UT). Figure 5 Comparison of a filtered image with c0 = 1.0 and t0 = 1.0 (top left; i.e. no rescaling, but including low-end tapering), resulting in a “flat” looking result and with c0 = 0.75, but t0 = 0.0 (top right; rescaling, but no low-end tapering), which leads to an over-enhancement of background brightness and low-level temporal noise, and (bottom) the optimally processed image from Figure 3 (28 October 2014 13:21:13 UT). 92 Page 10 of 17 D.B. Seaton et al. 92 Page 10 of 17 92 the image to the 1/4-power yields a subjectively aesthetically appealing result – although, as we discuss in Section 2, a variety of nonlinear functions can work depending on the specific need. For many of the processed images we have generated, we find that a good display range for the filtered and scaled image is between 0.5 – 2.0 renormalized counts. We used this combination for the images in this article, but users are encouraged to experiment and find display approaches that suit their particular needs and data. Some examples in the p2sw_image_filter.pro code itself illustrate how to implement this and display an image in IDL. In the example above, the fully renormalized image  I′ is then described by I′ =  I0/F′1/42.0 0.5 , (2) (2) where, I0 is the unprocessed, calibrated image and F′ is the filter generated in Equation 1. I′ is re-scaled as a byte-array using the values indicated. 5. Comparison to Other Techniques In this sense, the SWAP Filter has much in common 92 Page 12 of 17 D.B. Seaton et al. Figure 7 SUVI image of extended EUV corona from a campaign during April 2021, showing an erupting prominence cavity, indicated by arrow near the northeast limb. See also the accompanying animation in the Electronic Supplementary Material. Figure 7 SUVI image of extended EUV corona from a campaign during April 2021, showing an erupting prominence cavity, indicated by arrow near the northeast limb. See also the accompanying animation in the Electronic Supplementary Material. with other radial normalizing filters such as the NRGF and FNRGF, which are optimized specifically for off-disk image processing. The SWAP Filter marries its off-disk processing technique to a related on-disk filtering technique, but is nonetheless fundamentally similar in approach to these methods. Figure 8 presents examples of the same SWAP image shown elsewhere in this article pro- cessed with the MGN and NRGF filters, implemented in the sunkit-image Python package.2 While the MGN-processed image clearly reveals coherent structures on multiple scales, the overall appearance of the image – particularly off of the solar disk – is somewhat differ- ent from the SWAP Filter. The spatial filtering of the MGN suppresses the low-frequency background brightness in the image, strongly sharpening features on smaller scales and ren- dering them with largely uniform intensity across the full FOV. Like the SWAP Filter, the processing of the disk image improves the visibility of loops and large-scale features such as the filament in the northwestern part of the Sun, but the MGN makes small-scale variations within these features much more prominent. Furthermore, the MGN can be flexibly tuned to both highlight features on specific scales, by adjusting the specific spatial scales isolated by the filter, and remix the relative importance of each frequency in the image, by adjusting weights for each individual spatial component in the final image. Here we set the MGN to isolate a range of spatial scales, ranging from just over a single pixel to 40 pixels, with the strongest weights for the large-scale features, but alternate scalings are easily achieved via configurable parameters. The NRGF image, on the other hand, more closely resembles the processed image in Figure 3, because the approach used by the NRGF filter is similar to the off-disk approach used in the SWAP Filter. 5. Comparison to Other Techniques Since the development of the SWAP Filter, early in PROBA2’s mission, a number of power- ful new image-processing techniques have also been published (e.g. Druckmüllerová, Mor- gan, and Habbal, 2011; Druckmüller, 2013; Morgan and Druckmüller, 2014; Auchère et al., 2023b). The fundamental idea underpinning our approach was very simple: to adapt well- known radial-normalizing filter approaches – dating all the way back to Newkirk (1967) Page 11 of 17 92 The SWAP Filter for EUV Solar Images Figure 6 Comparison between log-scaled and filtered data from a SUVI offpoint campaign in 2018. Concen- tric dotted circles indicate heights of 2 and 3 R⊙. Solar North is indicated by the tick mark near the top of the solar disk. Adapted from Seaton et al. (2021), used with permission. Figure 6 Comparison between log-scaled and filtered data from a SUVI offpoint campaign in 2018. Concen- tric dotted circles indicate heights of 2 and 3 R⊙. Solar North is indicated by the tick mark near the top of the Figure 6 Comparison between log-scaled and filtered data from a SUVI offpoint campaign in 2018. Concen- tric dotted circles indicate heights of 2 and 3 R⊙. Solar North is indicated by the tick mark near the top of the solar disk. Adapted from Seaton et al. (2021), used with permission. – to be more responsive to azimuthal variations, which are particularly pronounced in the EUV corona. The filter’s primary purpose is thus to reduce dynamic range – as a function of altitude within localized features off-disk, and as a function of latitude on-disk – to improve display of data that span many orders of magnitude in radiance. Other methods filter specific spatial frequencies to emphasize and sharpen features with specific characteristics, using, for example, wavelets (e.g. Stenborg and Cobelli, 2003 and Auchère et al., 2023b) or multi-scale Gaussian normalization (Morgan and Druckmüller, 2014). Although these techniques can be tuned to match features on many scales, they are particularly powerful for enhancing the visibility and clarity of low-contrast features on fine scales; see, for example, Figure 3 in Auchère et al. (2023b) or Figure 6 in Morgan and Druckmüller (2014). In contrast, the SWAP Filter’s strength lies in its ability to make visible large-scale co- herent features without trading off localized contrast – as with logarithmic or power scaling (compare Figure 1 and Figure 3). 2See docs.sunpy.org/projects/sunkit-image/en/latest/ for additional info and documentation. 92 Page 14 of 17 92 Page 14 of 17 92 Page 14 of 17 92 Pa D.B. Seaton et al. The MGN enhances structures on relatively small scales, and thus has the most power in this range, but it fully suppresses the largest-scale structures, and has the least power for these scales. The 1/4-power, log-scaled, and SWAP-Filtered images all similarly behave, with the SWAP Filter emphasizing small and mid-range scales, and thus having the most power in this range. The log-scaled image suppresses the mid-range frequencies, but equalizes brightness across the whole image, and thus has the most power at large scales. Surprisingly, the log-scaled and SWAP-filtered images behave very similarly at high frequencies. In fact, this appears to be because they enhance off-disk noise – a very high-frequency feature – compared to the 1/4-power filter, which attenuates the off-disk brightness much more than these other methods. Overall, these power spectra show more quantitatively what we have, up until now, mainly illustrated qualitatively: the SWAP Filter results in images that more closely re- semble scaled, but otherwise unprocessed, solar images, while also improving the image contrast over these simpler methods for features across a wide range of scales. 5. Comparison to Other Techniques A notable difference is that the NRGF processing – which is de- scribed by Morgan, Habbal, and Woo (2006) as being specifically developed to address the steep brightness gradient in coronagraph images – is not optimized for use on disk. This Page 13 of 17 92 The SWAP Filter for EUV Solar Images Figure 8 (left) MGN-filtered version and (right) NRGF-filtered version of the image shown in Figure 3, illustrating similarities and differences in the same image when processed with different techniques. Figure 8 (left) MGN-filtered version and (right) NRGF-filtered version of the image shown in Figure 3, illustrating similarities and differences in the same image when processed with different techniques. Figure 9 Azimuthally averaged power spectra of images processed with the SWAP Filter compared to log- scaling, 1/4-power image scaling, and MGN processing. Figure 9 Azimuthally averaged power spectra of images processed with the SWAP Filter compared to log- scaling, 1/4-power image scaling, and MGN processing. region is often omitted from NRGF-processed images (or, in the case of coronagraphic im- ages, is simply unavailable). Nonetheless, because both the SWAP Filter and NRGF were both developed to address the radial gradient in coronal images, it is not surprising that they emphasize many of the same large-scale coronal features. To quantitatively highlight differences in the SWAP Filter’s performance compared to other image scaling and filtering techniques, we compute the azimuthally averaged power spectrum for the images processed with the SWAP Filter (see Figure 3) and compare the resulting spectrum to those of the log and 1/4-power scaling (see Figure 1), as well as the MGN filter (see Figure 8); the results are shown in Figure 9. Such power spectrum analyzes have been used before to show similarities and differences between solar images where the spatial frequencies are known to differ, such as by Rachmeler et al. (2019), Figure 10. References Auchère, F., Berghmans, D., Dumesnil, C., Halain, J.-P., Rochus, P., Delmotte, F., François, S., Hermans, A., Hervier, V., Kraaikamp, E., Meltchakov, E., Morinaud, G., Philippon, A., Smith, P.J., Stegen, K., Verbeeck, C., Zhang, X., Andretta, V., Abbo, L., Buchlin, E., Frassati, F., Gissot, S., Gyo, M., Harra, L., Jersell, G., Landini, F., Mierla, M., Nicula, B., Parenti, S., Renotte, E., Romoli, M., Russano, G., Sasso, C., Schühle, U., Schmutz, W., Soubrié, E., Susino, R., Teriaca, L., West, M., Zhukov, A.N.: 2023a, Beyond the disk: EUV coronagraphic observations of the Extreme Ultraviolet Imager on board Solar Orbiter. Astron. Astrophys. DOI. É Auchère, F., Soubrié, E., Pelouze, G., Buchlin, É.: 2023b, Image enhancement with wavelet-optimi whitening. Astron. Astrophys. 670, A66. DOI. ADS. Chitta, L.P., Seaton, D.B., Downs, C., DeForest, C.E., Higginson, A.K.: 2022, Direct observations of a com- plex coronal web driving highly structured slow solar wind. Nat. Astron. DOI. ADS. D l J M S D B B h C R h l L J i A Hill S M P k C L H h J M Darnel, J.M., Seaton, D.B., Bethge, C., Rachmeler, L., Jarvis, A., Hill, S.M., Peck, C.L., Hughes, J.M., Shapiro, J., Riley, A., Vasudevan, G., Shing, L., Koener, G., Edwards, C., Mathur, D., Timothy, S.: Darnel, J.M., Seaton, D.B., Bethge, C., Rachmeler, L., Jarvis, A., Hill, S.M., Peck, C.L., Hughes, J.M., Shapiro, J., Riley, A., Vasudevan, G., Shing, L., Koener, G., Edwards, C., Mathur, D., Timothy, S.: 2022, The GOES-R Solar UltraViolet Imager. Space Weather 20, e2022SW003044. DOI. ADS. de Koning, C.A., Pizzo, V.J., Seaton, D.B.: 2022, The solar eruption of 2017 September 10: wavy with a chance of protons. Astrophys. J. 924, 106. DOI. ADS. Druckmüller, M.: 2013, A noise adaptive fuzzy equalization method for processing solar extreme ultraviolet images. Astrophys. J. Suppl. 207, 25. DOI. ADS. g p y pp Druckmüllerová, H., Morgan, H., Habbal, S.R.: 2011, Enhancing coronal structures with the Fourier normalizing-radial-graded filter. Astrophys. J. 737, 88. DOI. ADS. Eddy, J.A.: 1989, Gordon Newkirk’s Contributions to Coronal Studies, In: McNally, D. (ed.) Highlights of Astronomy 8, Springer, Dordrecht, 503. ISBN 978-94-009-0977-9. DOI. Golub, L., Cheimets, P., DeLuca, E.E., Madsen, C.A., Reeves, K.K., Samra, J., Savage, S., Winebarger, A., Bruccoleri, A.R.: 2020, EUV imaging and spectroscopy for improved space weather forecasting. J. Space Weather Space Clim. 10, 37. DOI. ADS. Declarations Competing interests The authors declare no competing interests. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 6. Conclusions Here we have described the SWAP Filter: a newly available azimuthally varying radial nor- malizing filter, specifically designed for wide-field EUV images of the solar corona. Al- though simple, the filter is highly effective for producing displayable EUV images that preserve both large-scale EUV features and small-scale variations within the global coro- nal structure. The filter has been demonstrated on a large variety of EUV image data, and is now publicly available via the PROBA2/SWAP SolarSoft IDL software package. Perfor- mance tests using an 2.40 GHz Intel Core i9-9980HK demonstrate that filter generation with default parameters typically takes less than 0.3 seconds for a 1024 × 1024 SWAP image, so the technique is fast and computationally light compared to other similar techniques, making this a useful approach when computational resources are limited or for large datasets. Image-processing tools such as this one are especially important for studies of the mid- dle corona, as observations that connect this region to the solar disk inherently contain very large brightness gradients that must be reduced in order to effectively display the entire corona. Numerous approaches are available, each of which offers various advantages for specific applications. Our filter is fast to apply and it yields images that do not appear ex- tensively processed, and is well suited for the display of data that emphasizes global-scale features, and for generating outreach images and movies that resemble the appearance of inner coronal images that have been available for decades, and thus feel familiar to viewers. Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1007/s11207-023-02183-w. Author contributions All authors contributed to both the development of the image-processing techniques described in this article and to the writing and review of the manuscript. Funding SWAP is a project of the Centre Spatial de Liège and the Royal Observatory of Belgium funded by the Belgian Federal Science Policy Office (BELSPO). D.B. Seaton and M.J. West acknowledge support from NASA’s Heliophysics Guest Investigator program, grant 80NSSC22K0523. Materials Availability SWAP observations appearing in this article are available via the Virtual Solar Obser- vatory or the PROBA2 website via proba2.sidc.be/. SWAP software tools, including the SWAP Filter, are available via SolarSoft IDL. SUVI observations shown in this article are available via NOAA NCEI’s Space Weather data website, accessible via doi.org/10.25921/D60Q-G238. Page 15 of 17 92 The SWAP Filter for EUV Solar Images References Morgan, H., Habbal, S.R., Woo, R.: 2006, The depiction of coronal structure in white-light images. Solar Phys. 236, 263. DOI. ADS. Morrill, J.S., Korendyke, C.M., Brueckner, G.E., Giovane, F., Howard, R.A., Koomen, M., Moses, D., Plun- kett, S.P., Vourlidas, A., Esfandiari, E., Rich, N., Wang, D., Thernisien, A.F., Lamy, P., Llebaria, A., Biesecker, D., Michels, D., Gong, Q., Andrews, M.: 2006, Calibration of the Soho/Lasco C3 white light coronagraph. Solar Phys. 233, 331. DOI. ADS. Müller, D., Nicula, B., Felix, S., Verstringe, F., Bourgoignie, B., Csillaghy, A., Berghmans, D., Jiggens, P., García-Ortiz, J.P., Ireland, J., Zahniy, S., Fleck, B.: 2017, JHelioviewer. Time-dependent 3D visualisa- tion of solar and heliospheric data. Astron. Astrophys. 606, A10. DOI. ADS. Newkirk, G. Jr.: 1967, Structure of the solar corona. Annu. Rev. Astron. Astrophys. 5, 213. DOI. ADS. Newkirk, G. Jr.: 1967, Structure of the solar corona. Annu. Rev. Astron. Astrophys. 5, 213. DOI. ADS. O’Hara, J.P., Mierla, M., Podladchikova, O., D’Huys, E., West, M.J.: 2019, Exceptional extended field-of- i b i b PROBA2/SWAP 2017 A il 1 d 3 A h J 883 59 DOI ADS O’Hara, J.P., Mierla, M., Podladchikova, O., D’Huys, E., West, M.J.: 2019, Exceptional extended field view observations by PROBA2/SWAP on 2017 April 1 and 3. Astrophys. J. 883, 59. DOI. ADS. y p view observations by PROBA2/SWAP on 2017 April 1 and 3. Astrophys. J. 883, 59. DOI. ADS. Patel, R., Majumdar, S., Pant, V., Banerjee, D.: 2022, A simple radial gradient filter for batch-processing of coronagraph images. Solar Phys. 297, 27. DOI. ADS. Patel, R., Majumdar, S., Pant, V., Banerjee, D.: 2022, A simple radial gradient filter for batch-processing of coronagraph images. Solar Phys. 297, 27. DOI. ADS. Rachmeler, L.A., Winebarger, A.R., Savage, S.L., Golub, L., Kobayashi, K., Vigil, G.D., Brooks, D.H., Cir- tain, J.W., De Pontieu, B., McKenzie, D.E., Morton, R.J., Peter, H., Testa, P., Tiwari, S.K., Walsh, R.W., Warren, H.P., Alexander, C., Ansell, D., Beabout, B.L., Beabout, D.L., Bethge, C.W., Champey, P.R., Cheimets, P.N., Cooper, M.A., Creel, H.K., Gates, R., Gomez, C., Guillory, A., Haight, H., Hogue, W.D., Holloway, T., Hyde, D.W., Kenyon, R., Marshall, J.N., McCracken, J.E., McCracken, K., Mitchell, K.O., Ordway, M., Owen, T., Ranganathan, J., Robertson, B.A., Payne, M.J., Podgorski, W., Pryor, J., Samra, J., Sloan, M.D., Soohoo, H.A., Steele, D.B., Thompson, F.V., Thornton, G.S., Watkinson, B., Windt, D.: 2019, The High-Resolution Coronal Imager, Flight 2.1. Solar Phys. 294, 174. DOI. ADS. References Goryaev, F., Slemzin, V., Vainshtein, L., Williams, D.R.: 2014, Study of extreme-ultraviolet emission and properties of a coronal streamer from PROBA2/SWAP, Hinode/EIS and Mauna Loa Mk4 observations. Astrophys. J. 781, 100. DOI. ADS. Habbal, S.R., Morgan, H., Johnson, J., Arndt, M.B., Daw, A., Jaeggli, S., Kuhn, J., Mickey, D.: 2007, Lo- calized enhancements of Fe+10 density in the corona as observed in Fe XI 789.2 nm during the 2006 March 29 total solar eclipse. Astrophys. J. 663, 598. DOI. ADS. p Habbal, S.R., Druckmüller, M., Morgan, H., Ding, A., Johnson, J., Druckmüllerová, H., Daw, A., Arndt, M.B., Dietzel, M., Saken, J.: 2011, Thermodynamics of the solar corona and evolution of the solar magnetic field as inferred from the total solar eclipse observations of 2010 July 11. Astrophys. J. 734, 120. DOI. ADS. Halain, J.-P., Berghmans, D., Seaton, D.B., Nicula, B., De Groof, A., Mierla, M., Mazzoli, A., Defise, J.-M., Rochus, P.: 2013, The SWAP EUV imaging telescope. Part II: in-flight performance and calibration. Solar Phys. 286, 67. DOI. ADS. y Lupton, R., Blanton, M.R., Fekete, G., Hogg, D.W., O’Mullane, W., Szalay, A., Wherry, N.: 2004, Preparing red-green-blue images from CCD data. Publ. Astron. Soc. Pac. 116, 133. DOI. ADS. Mason, J.P., Chamberlin, P.C., Seaton, D., Burkepile, J., Colaninno, R., Dissauer, K., Eparvier, F.G., Fan, Y., Gibson, S., Jones, A.R., Kay, C., Kirk, M., Kohnert, R., Pesnell, W.D., Thompson, B.J., Veronig, A.M., West, M.J., Windt, D., Woods, T.N.: 2021, SunCET: the Sun coronal ejection tracker concept. J. Space Weather Space Clim. 11, 20. DOI. ADS. 92 Page 16 of 17 92 Page 16 of 17 D.B. Seaton et al. 92 Mason, J.P., Seaton, D.B., Jones, A.R., Jin, M., Chamberlin, P.C., Sims, A., Woods, T.N.: 2022, Simultaneous high dynamic range algorithm, testing, and instrument simulation. Astrophys. J. 924, 63. DOI. ADS. Mierla, M., Zhukov, A.N., Berghmans, D., Parenti, S., Auchère, F., Heinzel, P., Seaton, D.B., Palmerio, E., Jejˇciˇc, S., Janssens, J., Kraaikamp, E., Nicula, B., Long, D.M., Hayes, L.A., Jebaraj, I.C., Talpeanu, D.- C., D’Huys, E., Dolla, L., Gissot, S., Magdaleni´c, J., Rodriguez, L., Shestov, S., Stegen, K., Verbeeck, C., Sasso, C., Romoli, M., Andretta, V.: 2022, Prominence eruption observed in He II 304 Å up to > 6 R⊙by EUI/FSI aboard Solar Orbiter. Astron. Astrophys. 662, L5. DOI. ADS. Morgan, H., Druckmüller, M.: 2014, Multi-scale Gaussian normalization for solar image processing. Solar Phys. 289, 2945. DOI. ADS. References Rochus, P., Auchère, F., Berghmans, D., Harra, L., Schmutz, W., Schühle, U., Addison, P., Appourchaux, T., Aznar Cuadrado, R., Baker, D., Barbay, J., Bates, D., BenMoussa, A., Bergmann, M., Beurthe, C., Borgo, B., Bonte, K., Bouzit, M., Bradley, L., Büchel, V., Buchlin, E., Büchner, J., Cabé, F., Cadiergues, L., Chaigneau, M., Chares, B., Choque Cortez, C., Coker, P., Condamin, M., Coumar, S., Curdt, W., Cutler, J., Davies, D., Davison, G., Defise, J.-M., Del Zanna, G., Delmotte, F., Delouille, V., Dolla, L., Dumesnil, C., Dürig, F., Enge, R., François, S., Fourmond, J.-J., Gillis, J.-M., Giordanengo, B., Gissot, S., Green, L.M., Guerreiro, N., Guilbaud, A., Gyo, M., Haberreiter, M., Hafiz, A., Hailey, M., Halain, J.-P., Hansotte, J., Hecquet, C., Heerlein, K., Hellin, M.-L., Hemsley, S., Hermans, A., Hervier, V., Hochedez, J.-F., Houbrechts, Y., Ihsan, K., Jacques, L., Jérôme, A., Jones, J., Kahle, M., Kennedy, T., Klaproth, M., Kolleck, M., Koller, S., Kotsialos, E., Kraaikamp, E., Langer, P., Lawrenson, A., Le Clech’, J.-C., Lenaerts, C., Liebecq, S., Linder, D., Long, D.M., Mampaey, B., Markiewicz-Innes, D., Marquet, B., Marsch, E., Matthews, S., Mazy, E., Mazzoli, A., Meining, S., Meltchakov, E., Mercier, R., Meyer, S., Monecke, M., Monfort, F., Morinaud, G., Moron, F., Mountney, L., Müller, R., Nicula, B., Parenti, S., Peter, H., Pfiffner, D., Philippon, A., Phillips, I., Plesseria, J.-Y., Pylyser, E., Rabecki, F., Ravet-Krill, M.-F., Rebellato, J., Renotte, E., Rodriguez, L., Roose, S., Rosin, J., Rossi, L., Roth, P., Rouesnel, F., Roulliay, M., Rousseau, A., Ruane, K., Scanlan, J., Schlatter, P., Seaton, D.B., Silliman, K., Smit, S., Smith, P.J., Solanki, S.K., Spescha, M., Spencer, A., Stegen, K., Stockman, Y., Szwec, N., Tamiatto, C., Tandy, J., Teriaca, L., Theobald, C., Tychon, I., van Driel-Gesztelyi, L., Verbeeck, C., Vial, J.-C., Werner, S., West, M.J., Westwood, D., Wiegelmann, T., Willis, G., Winter, B., Zerr, A., Zhang, X., Zhukov, A.N.: 2020, The Solar Orbiter EUI instrument: the Extreme Ultraviolet Imager. Astron. Astrophys. 642, A8. DOI. ADS. References V., Hochedez, J.-F., Houbrechts, Y., Ihsan, K., Jacques, L., Jérôme, A., Jones, J., Kahle, M., Kennedy, T., Klaproth, M., Kolleck, M., Koller, S., Kotsialos, E., Kraaikamp, E., Langer, P., Lawrenson, A., Le Clech’, J.-C., Lenaerts, C., Liebecq, S., Linder, D., Long, D.M., Mampaey, B., Markiewicz-Innes, D., Marquet, B., Marsch, E., Matthews, S., Mazy, E., Mazzoli, A., Meining, S., Meltchakov, E., Mercier, R Meyer S Monecke M Monfort F Morinaud G Moron F Mountney L Müller R Nicula Marquet, B., Marsch, E., Matthews, S., Mazy, E., Mazzoli, A., Meining, S., Meltchakov, E., Mercier, R., Meyer, S., Monecke, M., Monfort, F., Morinaud, G., Moron, F., Mountney, L., Müller, R., Nicula, B., Parenti, S., Peter, H., Pfiffner, D., Philippon, A., Phillips, I., Plesseria, J.-Y., Pylyser, E., Rabecki, , Ravet-Krill, M.-F., Rebellato, J., Renotte, E., Rodriguez, L., Roose, S., Rosin, J., Rossi, L., Roth, P., K., Smit, S., Smith, P.J., Solanki, S.K., Spescha, M., Spencer, A., Stegen, K., Stockman, Y., Szwec, N., Tamiatto, C., Tandy, J., Teriaca, L., Theobald, C., Tychon, I., van Driel-Gesztelyi, L., Verbeeck, C., Vial, J.-C., Werner, S., West, M.J., Westwood, D., Wiegelmann, T., Willis, G., Winter, B., Zerr, A., Zhang, X., Zhukov, A.N.: 2020, The Solar Orbiter EUI instrument: the Extreme Ultraviolet Imager. Astron. Astrophys. 642, A8. DOI. ADS. Seaton, D.B., Bartz, A.E., Darnel, J.M.: 2017, Observations of the formation, development, and structure of a current sheet in an eruptive solar flare. Astrophys. J. 835, 139. DOI. ADS. p p y Seaton, D.B., Darnel, J.M.: 2018, Observations of an eruptive solar flare in the extended EUV solar corona. Astrophys. J. Lett. 852, L9. DOI. ADS. Seaton, D.B., Berghmans, D., Nicula, B., Halain, J.-P., De Groof, A., Thibert, T., Bloomfield, D.S., Raftery, C.L., Gallagher, P.T., Auchère, F., Defise, J.-M., D’Huys, E., Lecat, J.-H., Mazy, E., Rochus, P., Rossi, L., Schühle, U., Slemzin, V., Yalim, M.S., Zender, J.: 2013b, The SWAP EUV Imaging Telescope part I: instrument overview and pre-flight testing. Solar Phys. 286, 43. DOI. ADS. The SWAP Filter for EUV Solar Images Page 17 of 17 92 Seaton, D.B., De Groof, A., Shearer, P., Berghmans, D., Nicula, B.: 2013a, SWAP observations of the long- term, large-scale evolution of the extreme-ultraviolet solar corona. Astrophys. J. 777, 72. DOI. ADS. Seaton, D.B., Darnel, J.M., Hsu, V., Hughes, J.M.: 2020, Chapter 18 – GOES-R series solar dynamics. In: Goodman, S.J., Schmit, T.J., Daniels, J., Redmon, R.J. (eds.) The GOES-R Series, Elsevier, Amsterdam, 219. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References ISBN 978-0-12-814327-8. DOI. Seaton, D.B., Hughes, J.M., Tadikonda, S.K., Caspi, A., DeForest, C.E., Krimchansky, A., Hurlburt, N.E., Seguin, R., Slater, G.: 2021, The Sun’s dynamic extended corona observed in extreme ultraviolet. Nat. Astron. 5, 1029. DOI. ADS. g , , , , Astron. 5, 1029. DOI. ADS. Shestov, S.V., Zhukov, A.N.: 2018, Influence of misalignments on the performance of externally occulted solar coronagraphs. Application to PROBA-3/ASPIICS. Astron. Astrophys. 612, A82. DOI. ADS. Shestov, S.V., Zhukov, A.N., Seaton, D.B.: 2019, Modeling and removal of optical ghosts in the PROB 3/ASPIICS externally occulted solar coronagraph. Astron. Astrophys. 622, A101. DOI. ADS. Stenborg, G., Cobelli, P.J.: 2003, A wavelet packets equalization technique to reveal the multiple spatial-sc nature of coronal structures. Astron. Astrophys. 398, 1185. DOI. ADS. p y Tadikonda, S.K., Freesland, D.C., Minor, R.R., Seaton, D.B., Comeyne, G.J., Krimchansky, A.: 2019, Coronal imaging with the Solar UltraViolet Imager. Solar Phys. 294, 28. DOI. Veronig, A.M., Podladchikova, T., Dissauer, K., Temmer, M., Seaton, D.B., Long, D., Guo, J., Vršnak, B., Harra, L., Kliem, B.: 2018, Genesis and impulsive evolution of the 2017 September 10 coronal mass ejection. Astrophys. J. 868, 107. DOI. ADS. West, M.J., Seaton, D.B., D’Huys, E., Mierla, M., Laurenza, M., Meyer, K.A., Berghmans, D., Rachmeler, L.R., Rodriguez, L., Stegen, K.: 2022, A review of the extended EUV corona observed by the Sun Watcher with Active Pixels and image Processing (SWAP) Instrument. Solar Phys. 297, 136. DOI. ADS. West, M.J., Seaton, D.B., Wexler, D.B., Raymond, J.C., Del Zanna, G., Rivera, Y.J., Kobelski, A.R., DeForest, C., Golub, L., Caspi, A., Gilly, C.R., Kooi, J.E., Alterman, B.L., Alzate, N., Banerjee, D., Berghmans, D., Chen, B., Chitta, L.P., Downs, C., Giordano, S., Higginson, A., Howard, R.A., Mason, E., Mason, J.P., Meyer, K.A., Nykyri, K., Rachmeler, L., Reardon, K.P., Reeves, K.K., Savage, S., Thompson, B.J., Van Kooten, S.J., Viall, N.M., Vourlidas, A.: 2023, Defining the middle corona. Solar Phys. 298, 78. DOI. ADS.

W2896912960.txt

https://hts.org.za/index.php/hts/article/download/4031/8019

fr

Die kerk en die staatkundig-kulturele vlak van die volks-lewe

HTS teologiese studies

cc-by

DIE KERK EN DIE STAATKUNDIGKULTURELE VLAK VAN DIE VOLKSLEWE ’n Oorsig van die Afgelope Dekade DR. J. P. OBERHOLZER Die dekade w at hier te sprake is, is reeds m eer as ’n halfeeu ver verw yder van die negentiende-eeuse verbondenheid tussen die Ned. Hervormde Kerk en die ou Zuid-Afrikaanse Republiek, toe die stryd om kerklike en nasionale onafhanklikheid tw ee sye van dieselfde saak was. Die poging to t kerkvereniging en die drie­ jarige oorlog, gevolg deur ’n tydperk van skrynende armoede op materiële vlak, oorstrom ing vanuit die suide en gevolglike minderheidstatus, het daartoe gelei dat die eerste helfte van die 20ste eeu vir die Hervormde Kerk grotendeels nog ’n stryd om bestaan was. Die heropbou van die kerk het geskied in ’n gees van isolement, bevorder deur die mindere m aatskaplike posisie van sy lidmate, die tek o rt aan predikante, die im perialistiese strew e van susterkerke, en ander faktore. Vir rustige en om­ vattende form ulerings van die verhouding tussen kerk en maatskappy w as daar weinig geleentheid. Probleme op hierdie vlak is die hoof gebied deur optrede in casu eerder as deur vir die toekoms steekhoudende uitsprake. Woelinge op politieke en kulturele gebied het egter nie by die Kerk verbygegaan nie. Die opkoms van die nasionale beweging vanaf 1912, die rebellie van 1914, die m ynstakings van 1922 en 1942, die tw ee wêreldoorloë en die bittere stryd van die dertiger en veertiger jare het op een cf ander w yse w eerklank gevind in die vergaderinge van die Kerk. Die nasionale oorwinning van 1948 het ’n tydperk van groter eensgesindheid op Afrikaans-politieke en kulturele vlak ingelui. Terselfdertyd w as die materiële opbouwerk van die Kerk in so ’n m ate gevorder dat die geleentheid vir rustiger besinning oor die taak van die kerk, binne bereik voorgekom het. Ten spyte van m eer as ’n halfeeu van verdringing en oordondering, het die Herv. Kerk sy solidariteit m et die volk behou, en hierdie solidariteit het die K erk betrokke gem aak by die nasionale stryd teen internasionale bedreiging in die 50er en 60er jare. Die solidariteit tussen kerk en volk blyk byvoorbeeld uit die besluit van die Algemene Kerkvergadering van 1957, toe by kerkrade en kerklike kommissies aanbeveel is dat voorkeur gegee sal w ord aan Afrikaanse handelsinstellings. Deelname van die kerk in die bedrywighede van Afrikaanse kulturele organisasies, w aar deelname hom gegun is — en dít is helaas nie so vanself219 sprekend as w at verwag sou kon word nie — het steeds geskied sonder voorbehoud. Dit het vir die Kerk so gehoort. Sy lidmate is aangemoedig om deel te neem aan volkefeeste en dankfeeste by geleentheid van nasionale fees- en roudae. In sy besluit oor die A frikaner Broederbond het die Kerk nie alleen ingegaan op die teologiese implikasies nie, m aar ook kennis geneem van „die onbaatsugtige, „toegewyde en lofwaardige w yse” w aarop die bond „ons volk gedien het en nog dien.” 1). Hierdie solidariteit met die volk het ook uiting gevind in die w eerstand teen ideologiese en politieke bedreiging van die Protestants-Christelike erfenis van volk en kerk. Die deelname van die kerk aan die stryd teen die kommunisme en liberalisme het van die begin af geskied m et oortuiging — getuie die u it­ sprake van verskillende vergaderinge, die oprigting van die Antikom munistiese Aksiekommissie, die steun aan die Volkskongres oor Kommunisme en die Nasionale Raad teen Kom­ munisme.2) Die tradisionele beleid van afsonderlike naasbestaan was egter dié aspek w at die meeste aandag ontvang het, nie omdat die kerk dit so verkies het nie, m aar om dat ’n volgehoue en telkens hem ude druk van buite hom daartoe genoodsaak het. Vir die kerk was en is die afsonderlike bestaan van die verskil­ lende volksgroepe in Suid-Afrika, elk na eie aard en moontlikheid, van die vroegste jare af ’n gegewene w aarm ee hy rekening wou hou. Hy het uitdrukking gegee aan sy oortuiginge in hierdie verband by die formulering van die vereistes vir lidm aatskap en van ’n sendingbeleid (Artikel III). Saam met die ander Afrikaanse Kerke het die Hervormde Kerk die Volkskongres oor die Toekoms van die Bantoe gereël en verantwoordelikheid aanvaar vir cue besluite daar geneem. Hierdie kongres is gesien as die bekroning van jarelange getuienis teenoor die owerheid oor hierdie saak. Sindsdien w as die Kerk genoodsaak om telkens sy standpunt te stel en om te antwoord op beskuldiginge in dié verband. In kronologiese volgorde was dit veral die Naturellewysigingswetsontw erp van 1957, die H oogverraadsaak van 1958, die W et op Groepsgebiede en die onluste by Sharpville en Langa in 1960 w at w eerklank gevind het in die kerklike vergaderinge. Midde-in ’n hewige openbare stryd oor die sogenaamde kerkklousule in die Naturellew ysigingsw etsontw erp van 1957 het die Kommissie van die Algemene Kerkvergadering besluit dat 1) N otule 1964 A lgem ene Kerkvergadering, bl. 184. 2) N otule 1964 A lgem ene Kerkvergadering, bis. 166 vv. 220 „die standpunt van ons kerk is dat die kerk ten volle die beginsel van apartheid ook in die kerke onderskryf . . . ” en die betrokke m inster so verwittig. In ’n antw oord aan die British Council of Churches is apartheid verder om skryf as die enigste Christelike oplossing. Onrus binne die kerk oor hierdie opvatting is beantw oord m et ’n Herderlike skryw e w aarin ampsdraers en lede gem aan is om nie te handel in stryd m et A rtikel III en verw ante artikels nie. ’n Skrywe aan am psdraers h et gew aarsku teen die verkondiging van beskouings in stryd m et kerklike besluite, ens.3) Die H oogverraadsaak van 1957-’58 het weereens die aandag van die buiteland sterk op Suid-Afrika gevestig. Dr. J. J. Buskes, w at Suid-Afrika in 1956 besoek het en daarna ’n boek gepubliseer het „Suid Afrikas Rassenbeleid — O naanvaarbaar”, het in H er­ vormd Nederland ’n oproep geplaas om hulp vir die beskuldigdes en hulle afhanklikes. Hierteen het die Algemene Kommissie geprotesteer in ’n skrywe aan die Nederlands Hervormde Kerk. In dieselfde jaar het die Pretoriase Indiërgemeenskap die steun van die Kerk gevra vir ’n protes teen die proklam ering van groepsgebiede. Die Algemene Kommissie het sodanige steun geweier. Die onluste by Sharpville en Langa in 1960 het hier en in die buiteland ’n histerie ontketen w at ongekende gevolge sou hê. Die Algemene Kommissie het op 29 M aart 1960 ’n verklaring uitgereik w aarin lidmate gemaan is to t kalmte en vertroue, ’n beroep op alle blankes gedoen is om onderlinge geskille by te lc, en die owerhede vir voorbidding aanbeveel is. Skrywes van die Presbiteriaanse Allianse, Presbiteriaanse kerke in Australië, Engeland en Ierland, die Nederlands Hervormde Kerk, die Re­ formed Church in America en die Free Church Federal Council van Brittanje is uitvoerig beantwoord, hoofsaaklik deur die voorsitter, ds. A. J. G. Oosthuizen. Hierdie antw oorde het bestaan in ’n uiteensetting van die historiese en praktiese omstandighede, ’n beoordeling van die premisses w aarop die onderskeie skrywes berus het, en ’n verdediging van die ow erheid en beleid van Suid-Afrika. Na aanleiding van die onluste en bloedvergieting in die Kongo is dieselfde kerke gewys op die diskrepansie van hulle swye in daardie geval. Binne die Hervormde Kerk was daar steeds eenstemmigheid met betrekking to t die beleid op afsonderlike ontwikkeling, m et die uitsondering van ’n klein aantal am psdraers onder aanvoering van prof. A. S. Geyser. Hy het reeds in die laat 50er jare na 3) Bylaag tot DIE HERVORMER van 1.5.1957. 221 vore gekom m et ’n steeds onbuigsam er houding teenoor die Kerk ten opsigte van Artikel III en die konsekwensies daarvan. In die histerie na Sharpville het dan ook die boek „VERTRAAG­ DE AKSIE” verskyn, waarin die beleid van afsonderlike ont­ wikkeling op allerlei maniere veroordeel is. Die Algemene Kom­ missie h et die Hervormde m edewerkers verm aan vanweë die beroering w at hul optrede in die kerk veroorsaak het. Die belangrikste uitloper van die Sharpville-gebeure was egter die Cottesloe-konferensie van die Suid-Afrikaanse ledekerke van die W êreldraad van Kerke in Desember 1960. Dié konferensie is gereël na ’n protes van aartsbiskop Joost de Blank by die W êreldraad van Kerke oor die houding van die Afrikaanse kerke ten opsigte van die rassebeleid. Die houding van die Her­ vormde Kerk is uitvoerig uiteengesit in die gesprekke met dr. R. Bilheimer op 20 April 1960 en 28 April 1960 w at gehou is ter voorbereiding van die konferensie, in die memorandum van die kerk w at op die konferensie gedien het en in die besluite van die Algemene Kerkvergadering in 1961 toe die lidm aatskap van die W êreldraad van Kerke laat vaar is. Uit hierdie stukke haal ek die volgende te r illustrasie aan: 1. Die kerk stel hom in sy profetiese taak teenoor die ower­ heid nie byvoorbaat op die standpunt van ’n opposisie nie. 2. Die kerk beskou die apartheidsbeleid nie as ’n beginsel nie m aar as ’n werkmetode, w at wrywing en diskriminasie moet voorkom. 3. Die kerk sien in die agitasie teen apartheid die aanval van ideologiese magte w at die evangelie self wil aantas en die kerk wil vernietig. 4. Die kerk glo dat dit die evangelie self is w at die verhoudinge binne die m aatskppy kersten en dat dit nie op die weg van die kerk lê om die verandering van sam elew ingstrukture as sodanig as primêre taak te sien nie. 5. Die kommunikasiemedia is verantw oordelik vir ’n groot deel van die wanbegrip en onrus op staatkundig-kulturele vlak. 6. A partheid of eiesoortige ontwikkeling is die enigste regverdige oplossing vir die rasseproblem e van ons land. 7. Op Cottesloe het ’n verskil in teologiese aksent aan die lig gekom, naamlik ten opsigte van die beskouing van Christus as sosiale Hervormer en as Verlosser. 8. Die Cottesloe-besluite is onregverdige en ongeoorloofde in­ menging in staatsake van ons land w at m et Gods Koninkryk geen verband hou nie. 222 Die standpunt van die Hervormde Kerk rondom Cottesloe het by die volk as geheel aanklank gevind, en die kerk miskien meer as w at gesond is in die openbare beeld vereenselwig laat raak m et die beleid van afsonderlike ontwikkeling soos voorge­ stel en uitgevoer deur die regering van die dag. Dit moet egter beklemtoon word, selfs ná Cottesloe, d at die kerk in die eerste plek by wyse van A rtikel III betrokke is by hierdie beleid en dat hy hom steeds uitgespreek het vir die groot beginsel van skeiding ten opsigte van diè" volksgroepe omdat hy daarin ’n Christelike oplossing vir ’n probleem sien, langs dié weg ’n praktiese metode vir die uitvoering van sy evangeliese roeping sien en om dat hy in die yw er vir integrasie die kerkvyandelike magte van die wéreld aan die w erk sien. Dit is duidelik gestel in die rapport oor A rtikel III w at deur die Algemene K erkvergadering van 1961 aanvaar is. In hierdie rapport is die historiese en kerkregtelike agtergrond van Artikel III gestel, die artikel self ontleed en beoordeel w at betref die standpunt teen gelykstelling ( = ver­ menging) en diskriminasie, die wese en taak van volkskerke, die eenheid, gem eenskap en verkyningsvorm van die kerk en die goeie reg van ’n w erkverdeling soos deur A rtikel III gestel. Die verantw oording met betrekking to t die Kerk se houding voor en na Cottesloe was ’n eerste stap in die rigting van form ulering van ’n standpunt w at vroeër wel bestaan het en telkens uiting gevind het, m aar nooit prinsipieël onder woorde gebring is nie. Op hierdie eerste stappe is voortgebou in die publikasie „Gelyke Geleenthede” , ’n antw oord van die kerk op die herderlike skrywe van die Nederlands Hervormde Kerk, „Het R assenvraagstuk”. Op hierdie stuk is van die Nederlandse kerk geen antw oord ontvang nie. Trouens, die onder woorde bring van die Hervormde Kerk se standpunt, sou steeds stu it op die verskynsel dat hy as’t w are in ’n ledige ruim te in praat, sonder w eerklank. Dit w as die geval m et ’n memorandum w at aan die W êreldraad van Kerke voorgelê is na die Evanstonvergadering in 1954 w aar die W êreldraad van Kerke vir die eerste keer skeiding op grond van taal, kleur en etniese oorsprong verw erp het; m et die mem orandum van die Kerk by die Cottesloevergadering; die stuk „Gelyke Geleenthede”, die skrywe aan die W êreldraad van Kerke na aanleiding van die Rochester-verklaring en aan die Presbiteriaanse Alliansie na aanleiding van die Frankfortvergadering. Ook die afsonderlike kerke met wie hieroor gekorrespondeer is, het m et enkele uitsonderings geen reaksie op ons skrywes getoon nie. Die m islukking van die W êreldraad van Kerke se poging by Cottesloe om die Suid-Afrikaanse beleid deur middel van ’n 223 veroordeling deur die Afrikaanse kerke aan te tas, het gelei tot ’n veranderde strategie na 1960. Die eerste tekens hiervan het gekom by die Nieu Dehli-byeenkoms w aar die teenstanders van die apartheidsbeleid openlike aangemoedig is in ’n „boodskap aan Christene in Suid-Afrika”, die beleid van afsonderlike ontw ikke­ ling veroordeel is en die uitsprake van mense soos sir Francis Ibiam met instemming aangehoor is, naam lik d at Africans nie as mense nie m aar as slawe beskou word in Suid-Afrika en die Rhodesiës. Dit is opgevolg deur die Rochester-verklaring van die Sentrale Komitee in September 1963 w aarin aktiew e hulp aan die sogenaamde „slagoffers van diskrim inasie” in Suid-Afrika vir die eerste keer sy verskyning gem aak het. Die Algemene Kerkvergadering van 1964 het hierop skerp gereageer in ’n skrywe aan die W êreldraad van Kerke. Dit was die tyd toe organisasies soos Poqo en Umkonto we Sizwe aan bande gelê is deur wetgewing. Die optrede van die Division of Interchurch Aid, Refugee and World Service van die W êreldraad van Kerke het nou nuwe betekenis in Afrika gekry soos blyk uit Africa Survey van M atthews en Foot, en uit die jaarlikse begroting. Vlugtelinge uit Suid-Afrika w at la ter sou terugkeer as terroriste kon van nou af reken op hulp ook van die W êreldraad van Kerke. Die kerk het hom na 1964 nie w eer to t die W êreldraad van Kerke gerig nie. Hy het op hoogte gebly van dié liggaam se optrede, onder andere by die Kitwe-vergadering, die vergaderings van die Sentrale Komitee en by die Geneva-vergadering van Church and Society. In die Kroniek in DIE HERVORMER en by ander geleenthede is aandag geskenk aan die uitsprake van hierdie vergaderings, en so het die lidmate van die kerk op hoogte gebly van die diep kloof tussen die Hervormde Kerk en die W êreldraad van Kerke-gees. Die talle publikasies hier en oorsee w aarin ons land en kerk gekarikaturiseer w ord as ’n middeleeuse folterkerker vir alle Christene, sing so ’n eentonige lied dat ons aan die toon daarvan gewoond geraak het, en dit slegs terloops nog opmerk. Hierdie toestand is mede in die hand gewerk deur interne probleme, soos in verband m et die Geysersaak van 1962 en die bedanking uit die kerk van ’n aantal predikante en oud-predikante — gebeure w aarin die werklik prinsipiële verskilpunte oorskadu is deur onw aaragtigheid en valse gesindhede. Die reeds gespanne verhouding m et die Nederlands Hervorm­ de Kerk het to t breekpunt gekom in 1965 toe die Nederlandse en Deense regerings besluit het op skenkings aan die Defence and Aid Fund w at oorspronklik opgerig is m et die oog op die 224 H oogverraadsaak van 1958 en weer op die voorgrond gekom het m et die Rivonia-saak. Die Algemene Kommissie het in kabelgramme aan die Nederlands Hervormde Kerk en Deense Lutherse Kerk sy teleurstelling hieroor uitgespreek. Die Moderamen van die Nederlandse Kerk het op sy b eurt teleurstelling uitgespreek oor die „aggresiewe toon” van die kabelgram en ons verwys na die Nederlandse Ambassade in Suid-Afrika om dat dit ’n saak is w at die Nederlandse regering raak. ’n Teregwysing aan die Nederlandse Kerk omdat hy weier om verantw oording af te lê oor ’n saak w aarin hy dit eens is m et die Nederlandse regering, is na aandrang op ’n antwoord, gevolg deur ’n versoek tot sam espreking m et die Afrikaanse kerke. Uit so ’n samespreking sou dan blyk of ons en hulle dieselfde evangelie bely. Na korrespondensie tussen die drie Afrikaanse kerke en die twee Nederlandse kerke het die voorgenome sam esprekings nie deurgegaan nie, en is die posisie prakties een van geen kontak tussen Nederland en onsself. Die posisie is dus d at die solidariteit van kerk en volk en die feit dat die kerk op eie front dieselfde stryd moes stry as die owerheid op internasionaal-politieke front, gelei Tiet totT ’n toenemende isolasie en die afbreek van kontak m et kerke anders as die drie Afrikaanse kerke. Vir die Hervormde Kerk het dit in hierdie stryd eerstens gegaan om ’n lewenskwessie. Leer het alleen te r sprake gekom deurdat rasse-integrasie deur die ekumenies-kerklike wéreld to t ’n teologiese leerstuk verhef is as die toetssteen vir die w are Christelike geloof. In Suid-Afrika het die ontstaan van die blaadjie Pro Veritate en die Christelike Insituut die simptome geword van hierdie nuwe leerstuk. Die Algemene Kerkvergadering het hom in 1967 teen hierdie instansies uitge­ spreek en verklaar dat lidm aatskap van die Instituut en die Kerk nie versoenbaar is nie. Die korrespondensie met buitelandse kerke en kerklike liggame in die afgelope paar jaar het dieselfde patroon gevolg as voorheen. Van w eerskante is by die ingenome standpunte oor rasseverhoudinge volhard. Sam evattend kan ons dus stel dat die Hervormde Kerk in sy am ptelike uitsprake en optrede van die afgelope dekade sy solidariteit m et die Afrikanervolk op staatkundig-kulturele vlak gehandhaaf het. W aar hy onder die ou Zuid-Afrikaanse Republiek ’n m edevegter was vir staatkundige onafhanklikheid, het hy in die laaste jare op beskeie w yse deelgeneem aan ’n stryd om die staatkundige en kultrele voortbestaan van die A frikanervolk in ooreenstemming met sy historiese weg tot hiertoe. Die vordering 295 van die kerk op die gebied van die sending en sy deelname aan immigrasie-arbeid het hiertoe indirek bygedra. Die proses van die here Geyser en Naude teen prof. A. D. Pont en die afloop daarvan het ’n diepgaande indruk gemaak op talie lidmate van die Kerk en onder andere bygedra to t ’n be­ paalde terughoudenheid ten opsigte van politiek-kulturele aktiwiteite. Die metodes van bestryding van ideologiese en politieke sowel as teoJogiese gevare het hierdeur sterk in geding gekom. Gepaard daarm ee het gegaan ’n aksentverskuiw ing en aksentverskille binne die heersende politieke party, hoofsaaklik aangevuur deur sektore van die openbare pers. Tekens van politieke verdeeldheid en die bew eerde betrokkenheid van sommige am ps­ draers daarby het die vraag oor die verhouding van die Kerk en sy am psdraers to t die politieke aktiw iteite van die dag op die voorgrond gestel. Omstrede uitsprake en optrede van lede en am psdraers van die kerk w at in die pers so vertolk is dat die kerk vereenselwig is met partypolitieke groeperinge, het ’n teregwysing aan die pers en die verw erping van sodanige vereenselwiging van die kant van die Algemene Kommissie meegebring. Hierin het die Kommissie aangesluit by vroeëre uitsprake, onder andere soos gestel in ’n rapport van ds .J. G. M. Dreyer aan die Algemene Kommissie in 1957.4) Daarin stel hy: „Dit is w aar dat die Kerk aan die owerheid geen politieke oplossing kan voorskryf nie, m aar dis ook w aar, dat die Kerk geroepe is om die dade van die owerheid en van die volke te toets aan die Woord van God, w ant die beloftes en gebooie van die Evangelie geld ook vir die beslissinge op die terrein van die nasionale en intem asionale politiek. Die Kerk w at dit nie meer wil doen nie is ongehoorsaam aan Christus. As ons sê dat die Kerk hom nie afsydig van die politiek mag hou nie, dan word daarm ee nie bedoel dat die Kerk hom mag vereenselwig met een of ander politieke party nie. Daarcieur sou die Kerk immers: (1) sy kritiese funksie teenoor die verskillende politieke partye verloor; (2) Die Kerk sou daardeur sy reg verbeur om regter tussen en oor die partye te wees; (3) Die Kerk sou onvermydelik sy vryheid daardeur inboet; (4) Die Kerk bring daardeur oor hom die gevolge van die politiek van die betrokke party en hierdie politiek kan opportunisties korrup en sleg wees. Vgl. 1 Kon. 12:26-33, •*) D ie Hervormer, Junie 1957. 220 w aar die godsdiens in die diens van die politiek te staan gekom het. Tog kan dit gebeur dat die Kerk horn direk ook met die partypolitiek moet inlaat, w anneer naam lik ’n bepaalde party hom skuldig m aak aan magsmisbruik, aan verkragting van die reg, aan ’n antichristelike ideologie. Dan is die kerk geroepe om te getuig byvoorbeeld in die vorm van ’n protes, vir die heerskappy van Jesus Christus op al die terreine van die lewe, al is dit ook in die vorm van m artelaarskap. M aar w atter houding moet die am psdraers, leraars en kerkraadslede inneem ten opsigte van die partypolitiek? Mag hulle lede van partye wees en aktief daaraan deelneem? Hierop w ord deur sommige ,ja’ geantwoord. In verskillende Europese lande is dit die geval. In Engeland is al die biskoppe van die Anglikaanse Kerk ipso facto lid van die Engelse Hoërhuis, die Senaat. In Nederland was en is leidende Kerkfigure dikwels leidende politieke figure. Vgl. dr. Abr. Kuyper, prof. dr. H. Visscher, dr. De Visser, prof. dr. Slotemaker de Bruïne, prof. dr. G. van der Leeuw, terw yl verskillende bekende nog lewende teoloë op die hoofbesture van verskillende partye sit, sowel as lede is van die eerste of tweede kamer. In ons eie land word hierdie saak ietw at anders aangevoel. Eerstens dra dit die goedkeuring van ’n groot deel van die volk nie w eg nie dat ’n predikant hom aktief met die politiek inlaat. Tweedens laat die Sinodes nie toe dat ’n predikant of teologiese professor sy amp kan verenig met die lidm aat van die Volks­ raad of provinsiale Raad nie. D aar is ander w at ,nee’ antwoord op die vraag of ’n leraar en ’n kerkraadslid hulle aktief mag inlaat m et die partypolitiek. Dit het ook die Algemene Kerkvergadering gedoen soos bo aangetoon. Ons wens hiervan te verskil. Ons het hoegenaamd geen besw aar dat kerkraadslede hulle aktief bemoei m et die partypolitiek nie. Ons beskou dit selfs as wenslik met dien verstande egter d at hulle op ’n waardige wyse daaraan deel­ neem in ooreenstemming met die waardigheid van hulle ampte. Hierdie vryheid gun ons die predikante nie. Hulle posisie en taak, as die m ondstukke en eerste verteenw oordigers van die Kerk, m aak dit nie wenslik nie. Dit kan immers ’n babiloniese spraakverw arring op die kansels bring as die een leraar spreek in die belang van een party en miskien sy m edeleraar in die guns van ’n ander party optree. Die besw are w at ons bo geopper het, waarom die Kerk as sodanig hom nie moet 227 inlaat m et die partypolitiek nie, geld in die geval van leraars ook. Die Kerk kan natuurlik nie die vryheid van sy predikante tot so ’n m ate aan bande lê dat hulle geen partypolitieke opinie daarop na kan hou nie, m aar ons insiens kan die kerk wel van hulle verw ag om nie aktief daaraan deel te neem nie. Ons insiens m oet die band tussen hulle en ’n politieke party nie ’n te noue band wees nie. Dit neem egter nie weg nie, dat ’n predikant in sy gewete hom verplig kan ag om te getuig teen die optrede van een of ander party w at in stryd mag handel met God en Sy gebod. Dit mag egter nie oorhaastig geskied nie en liefs nie sonder raadpleging vooraf van sy kerklike instansies nie.” Met hierdie standpunt wil ek my graag vereenselwig. Die vraag kan gestel word of die solidariteit tussen kerk en volk nie in sekere tye verloop het to t solidariteit tussen kerk en ’n be­ paalde politieke party nie en die kerk daardeur sy kritiese funksie ingeboet het nie, of aan die ander k an t dat die kerk prinsipiëel-teologiese spreke laat vaar het te r wille van suiwer politieke spreke. W ant dit moet meteen duidelik wees, dat die kerk in elk geval eerstens vanuit die Heilige Skrif en in gehoorsaamheid aan die evangelie sal spreek. Lewe en leer staan immers nie los van m ekaar nie. Nogtans sal die Hervormde Kerk homself alleen t rou bly as hy volkskerk bly. mët die oog öp die kerstening van die hele volk en die getuienis teenoor volk en owerheid op elke terrein van die lewe. 228 

https://openalex.org/W4280492912

https://pure.aber.ac.uk/portal/files/51169491/plants-11-01321.pdf

English

Phenotypic and Genotypic Characterization and Correlation Analysis of Pea (Pisum sativum L.) Diversity Panel

Plants

cc-by

Citation for published version (APA): Uhlarik, A., Ćeran, M., Živanov , D., Grumeza, R., Skot, L., Sizer Coverdale, E., & Lloyd, D. (2022). Phenotypic and Genotypic Characterization and Correlation Analysis of Pea ( Pisum sativum L.) Diversity Panel. Plants, 11(10), Article 1321. https://doi.org/10.3390/plants11101321 Phenotypic and Genotypic Characterization and Correlation Analysis of Pea ( Pisum sativum L.) Diversity Panel Uhlarik, Ana; Ćeran, Marinia; Živanov , Dalibor; Grumeza, Radu; Skot, Leif; Sizer Coverdale, Ellen; Lloyd, David Citation for published version (APA): Uhlarik, A., Ćeran, M., Živanov , D., Grumeza, R., Skot, L., Sizer Coverdale, E., & Lloyd, D. (2022). Phenotypic and Genotypic Characterization and Correlation Analysis of Pea ( Pisum sativum L.) Diversity Panel. Plants, 11(10), Article 1321. https://doi.org/10.3390/plants11101321 Document License CC BY General rights i h d General rights Copyright and moral rights for the publications made accessible in the Aberystwyth Research Portal (the Institutional Repository) are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the Aberystwyth Research Portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain Y f l di ib h URL id if i h bli i i h Ab h R h P l • Users may download and print one copy of any publication from the Aberystwyth Research Portal for the purpose of private study or research. • Users may download and print one copy of any publication from the Aberystwyth Research Portal for the purp research search. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the Aberystwyth Research Portal arch. • You may not further distribute the material or use it for any profit-making activity or commercial ga Y f l di t ib t th URL id tif i th bli ti i th Ab t th R h P t l • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the Aberystwyth Research Portal You may not further distribute the material or use it for any profit making activity or commercial gain • You may freely distribute the URL identifying the publication in the Aberystwyth Research Portal Aberystwyth University Phenotypic and Genotypic Characterization and Correlation Analysis of Pea ( Pisum sativum L.) Diversity Panel Uhlarik, Ana; Ćeran, Marinia; Živanov , Dalibor; Grumeza, Radu; Skot, Leif; Sizer Coverdale, Ellen; Lloyd, David Citation: Uhlarik, A.; ´Ceran, M.; Živanov, D.; Grumeza, R.; Skøt, L.; Sizer-Coverdale, E.; Lloyd, D. Phenotypic and Genotypic Characterization and Correlation Analysis of Pea (Pisum sativum L.) Diversity Panel. Plants 2022, 11, 1321. https://doi.org/10.3390/ plants11101321 Keywords: pea; seed yield; protein; yield components; correlation; heritability Academic Editors: Fabio Palumbo, Gianni Barcaccia, Alessandro Vannozzi and Othmane Merah plants plants Article Ana Uhlarik 1,*, Marina ´Ceran 1, Dalibor Živanov 1 , Radu Grumeza 2 , Leif Skøt 3, Ellen Sizer-Coverdale 3 and David Lloyd 3 Marina ´Ceran 1, Dalibor Živanov 1 , Radu Grumeza 2 , Leif Skøt 3, Ellen Sizer-Coverdale 3 d 3 Ana Uhlarik 1,*, Marina ´Ceran 1, Dalibor Živanov 1 , Radu Grumeza 2 , Leif Skøt 3, Ellen Sizer-Coverdale 3 and David Lloyd 3 2 Agro Seed Research, Nijverheidslaan 1506, 3660 Oudsbergen, Belgium; [email protected] 3 Institute of Biological, Environmental & Rural Sciences (IBERS), Aberystwyth University, Penglais, Ceredigion, Aberystwyth SY23 3DA, UK; [email protected] (L.S.); [email protected] (E.S.-C.); [email protected] (D.L.) y g * Correspondence: [email protected] Abstract: Phenotypic and genotypic characterization were performed to assess heritability, variability, and seed yield stability of pea genotypes used in breeding to increase the pea production area. A European pea diversity panel, including genotypes from North America, Asia, and Australia consisting of varieties, breeding lines, pea, and landraces was examined in 2019 and 2020 in Serbia and Belgium using augmented block design. The highest heritability was for thousand seed weight; the highest coefficient of variation was for seed yield. The highest positive correlation was between number of seeds per plant and number of pods per plant; the highest negative correlation was between seed yield and protein content. Hierarchical clustering separated pea germplasm based on use and type. Different Principal component analysis grouping of landraces, breeding lines, and varieties, as well as forage types and garden and dry peas, confirms that there was an apparent decrease in similarity between the genotypes, which can be explained by their different purposes. Pea breeding should be focused on traits with consistent heritability and a positive effect on seed yield when selecting high-yielding genotypes, and on allowing for more widespread use of pea in various agricultural production systems. Take down policy Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. tel: +44 1970 62 2400 email: [email protected] tel: +44 1970 62 2400 email: [email protected] tel: +44 1970 62 2400 email: [email protected] Download date: 24. Oct. 2024 plants 1. Introduction Received: 10 March 2022 Accepted: 1 May 2022 Published: 16 May 2022 Pea (Pisum sativum L.) is considered to be the second most important protein crop after soybean, with a total production area of 2,436,344 ha in Europe in 2020 [1]. Despite the fact that the production area under peas has increased by 3.6% in the past five years [1], the demand for new sources of protein continues to grow [2], as total protein consumption in Europe, including plant-derived proteins, is about 70% higher than recommended [3]. To meet growing demand for quality protein while reducing reliance on imported plant protein from outside of Europe, improvement of the yield of protein crops produced in the region is desirable [4,5]. Protein content in pea as quoted in literature ranges from 16% to 30.9% [6,7]. In organic production systems in the European Union, peas are considered an important plant species because they are a source of biologically-fixed nitrogen and provide high-quality animal feed that is rich in crude protein and minerals [8]. Although soybean is one of the primary sources of plant-based protein, the advantages of growing peas over soybeans are its wider geographical area of cultivation and its ability to thrive in colder climates. Soybean, unlike pea, is one of the significant food allergens [9]; moreover, the fact that pea grain, unlike soybeans, can be used directly in the diet without prior heat treatment is essential, as this simplifies use and reduces processing costs. Combined sowing of peas with cereals is increasingly used in practice to improve the efficiency of land and water Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/plants Plants 2022, 11, 1321. https://doi.org/10.3390/plants11101321 Plants 2022, 11, 1321 2 of 17 2 of 17 resources, nutrients, and solar radiation [8]. Furthermore, peas can also be grown as green manure [10] and as a forage crop, especially in regions where climatic conditions are not favorable for a good seed yield [11]. 1. Introduction The fundamental goal of pea breeders is to increase seed yield to maximize plant productivity and allow for more widespread use of pea in various agricultural production systems. Seed yield is a complex trait that is quantitatively inherited, and its expression depends on genetic factors, environment, and interaction (GxE) [13]. Many studies have shown a significant influence of environment and genotype-by-environment (GxE) inter- action on seed yield and the yield components of their phenotypic performance [14–16]. Genetic variability is a prerequisite in increasing the seed yield, since it helps peas adapt to changing environmental conditions and is the source of variation and raw material for yield improvement. In this type of study, seed availability can be a limiting factor, as variability testing requires a large number of genotypes (lines or populations). In many of those, there is generally little seed material. Furthermore, the logistics of fully replicated trials can be prohibitive in terms of space. This can be solved by using an augmented trial design [17], which requires a significantly smaller number of seeds, giving access to a broader choice for populations in germplasm collections than when using traditional experiment designs. Assessment of genetic variability is essential for efficient parent selection in breeding pro- The objectives of this study were to examine the phenotypic and genotypic variability in a set of diverse pea genotypes using morphological and agronomic traits, assess heri- tability, discover the relationships among seed yield components of pea genotypes, identify germplasm structure using correlation analysis, and determine sensitivity to environmental conditions and possible GxE interaction for yield and yield components. This research aimed to assess the phenotypic diversity of studied pea germplasm in order to enrich the breeding collection of the pea used in Western and Central Europe. 1. Introduction Their adaptability to a broad range of environmental conditions makes them an excellent cover crop due to the short vegetation season [12] and can improve productivity in successor crops. Therefore, there are great needs and benefits for increasing the production area under peas. g p p The fundamental goal of pea breeders is to increase seed yield to maximize plant productivity and allow for more widespread use of pea in various agricultural production systems. Seed yield is a complex trait that is quantitatively inherited, and its expression depends on genetic factors, environment, and interaction (GxE) [13]. Many studies have shown a significant influence of environment and genotype-by-environment (GxE) inter- action on seed yield and the yield components of their phenotypic performance [14–16]. Genetic variability is a prerequisite in increasing the seed yield, since it helps peas adapt to changing environmental conditions and is the source of variation and raw material for yield improvement. In this type of study, seed availability can be a limiting factor, as variability testing requires a large number of genotypes (lines or populations). In many of those, there is generally little seed material. Furthermore, the logistics of fully replicated trials can be prohibitive in terms of space. This can be solved by using an augmented trial design [17], which requires a significantly smaller number of seeds, giving access to a broader choice for populations in germplasm collections than when using traditional experiment designs. Assessment of genetic variability is essential for efficient parent selection in breeding pro- grams and long-term selection gain [18]. The broad cultivation range of pea and reliable pea yield stability [19] is associated with a broad genetic variation observed at the phenotypic and molecular levels [13]. Before initiating any effective selection program, it is necessary to know the association of various traits with yield and with each other, as a negative association between the desired attributes under selection may result in genetic slippage. Consequently, this could limit the genetic advance of the yield, which represents the end product of many correlated characters. An understanding of the influence of individual contributors on seed yield informs the breeder on how effective selection can take place. It is considered that yield selection can be more effective when it is based on component traits that are highly heritable and positively correlated [20]. 2.1. Phenotypic Evaluation Descriptive statistics for eleven traits evaluated in pea genotypes at Agro Seed, Belgium. 2019 2020 Min Max Mean SD CV (%) SE He2B Min Max Mean SD CV (%) SE He2B p Values (t Test) GPP 1.43 13.29 3.72 1.02 27 0.11 69.55 1.09 13.5 3.7 1.05 28 0.08 70.59 0.12 PPP 4.81 8.47 6.07 0.57 9 0.00 27.58 7.43 18.2 9.98 1.49 15 0.12 42.03 ** SPP 13.21 38.20 22.13 4.18 19 0.32 45.01 26.5 76.9 41.97 6.74 16 42 42.72 ** FD 9.58 18.27 13.69 1.37 10 0.11 55.10 11.1 26 16.18 2.79 17 0.22 65.86 ** PH 48.52 150 84.78 22.20 26 85 87.11 55.5 154 92.25 19.2 21 1.51 73.13 ** PL 0.17 0.77 0.41 0.15 37 0.00 67.39 0.33 0.77 0.49 0.07 14 0.01 30.95 ** PoL 1.06 7.32 4.96 0.75 15 0.13 79.79 3.67 7.72 5.39 0.66 12 0.05 79.10 ** TSW 58.91 341 201.8 50.50 25 3.9 89.50 63.2 284 172.64 47.2 27 3.67 83.35 ** SWPP 3.05 6.74 4.61 0.65 14 0.11 33.65 4.52 13.7 7.79 1.64 21 0.13 49.52 ** SY 418 2180 1277 399 31 31 67.74 672 3349 1713.9 592 35 46.1 70.20 ** PC 23.1 30.4 26.94 1.22 5 0.12 84.24 21.7 32.3 27.48 1.74 6 0.14 88.55 ** GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). The broad-sense heritability estimates were moderate to high for all traits, indicating that genetic constituents are the primary source of these traits. The highest heritability of 89.5% was estimated for TSW in 2019 and 88.5% for PC in 2020, respectively. Heritability Table 1. Descriptive statistics for eleven traits evaluated in pea genotypes ats IFVCNS, Serbia. Table 1. Descriptive statistics for eleven traits evaluated in pea genotypes ats IFVCNS, Serbia. 2.1. Phenotypic Evaluation Descriptive statistics for eleven traits evaluated in pea genotypes at Agro Seed, Belgium. 2019 2020 Min Max Mean SD CV (%) SE He2B Min Max Mean SD CV (%) SE He2B p Values (t Test) GPP 1.43 13.29 3.72 1.02 27 0.11 69.55 1.09 13.5 3.7 1.05 28 0.08 70.59 0.12 PPP 4.81 8.47 6.07 0.57 9 0.00 27.58 7.43 18.2 9.98 1.49 15 0.12 42.03 ** SPP 13.21 38.20 22.13 4.18 19 0.32 45.01 26.5 76.9 41.97 6.74 16 42 42.72 ** FD 9.58 18.27 13.69 1.37 10 0.11 55.10 11.1 26 16.18 2.79 17 0.22 65.86 ** PH 48.52 150 84.78 22.20 26 85 87.11 55.5 154 92.25 19.2 21 1.51 73.13 ** PL 0.17 0.77 0.41 0.15 37 0.00 67.39 0.33 0.77 0.49 0.07 14 0.01 30.95 ** PoL 1.06 7.32 4.96 0.75 15 0.13 79.79 3.67 7.72 5.39 0.66 12 0.05 79.10 ** TSW 58.91 341 201.8 50.50 25 3.9 89.50 63.2 284 172.64 47.2 27 3.67 83.35 ** SWPP 3.05 6.74 4.61 0.65 14 0.11 33.65 4.52 13.7 7.79 1.64 21 0.13 49.52 ** SY 418 2180 1277 399 31 31 67.74 672 3349 1713.9 592 35 46.1 70.20 ** PC 23.1 30.4 26.94 1.22 5 0.12 84.24 21.7 32.3 27.48 1.74 6 0.14 88.55 ** GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). The broad-sense heritability estimates were moderate to high for all traits, indicating that genetic constituents are the primary source of these traits. The highest heritability of 89.5% was estimated for TSW in 2019 and 88.5% for PC in 2020, respectively. Heritability Results recorded at Agro Seed, Belgium, had mean SPP, PH and SY significantly lower in 2019 than in 2020. In contrast, TSW was higher in 2019 than in 2020, which is possibly explicable by higher temperatures observed during the pod filling period (Table S2). For all characteristics, statistically, significant differences were observed for mean values between two seasons for all traits, except for GPP. 2.1. Phenotypic Evaluation The highest coefficients of variation (CV) were observed for PL and SY in 2019 (37 and 31%) and for SY and GPP (31 and 28%) in 2020 (Table 2). High variation was also evident in PH and TSW in both years (16–31%). The coefficient of variability for all traits was generally higher in 2019. The lowest CV in 2019 was observed for PC (5%) and in 2020 for PC (6%). Table 2. Descriptive statistics for eleven traits evaluated in pea genotypes at Agro Seed, Belgium. 2019 2020 Min Max Mean SD CV (%) SE He2B Min Max Mean SD CV (%) SE He2B p Values (t Test) GPP 1.43 13.29 3.72 1.02 27 0.11 69.55 1.09 13.5 3.7 1.05 28 0.08 70.59 0.12 PPP 4.81 8.47 6.07 0.57 9 0.00 27.58 7.43 18.2 9.98 1.49 15 0.12 42.03 ** SPP 13.21 38.20 22.13 4.18 19 0.32 45.01 26.5 76.9 41.97 6.74 16 42 42.72 ** FD 9.58 18.27 13.69 1.37 10 0.11 55.10 11.1 26 16.18 2.79 17 0.22 65.86 ** PH 48.52 150 84.78 22.20 26 85 87.11 55.5 154 92.25 19.2 21 1.51 73.13 ** PL 0.17 0.77 0.41 0.15 37 0.00 67.39 0.33 0.77 0.49 0.07 14 0.01 30.95 ** PoL 1.06 7.32 4.96 0.75 15 0.13 79.79 3.67 7.72 5.39 0.66 12 0.05 79.10 ** TSW 58.91 341 201.8 50.50 25 3.9 89.50 63.2 284 172.64 47.2 27 3.67 83.35 ** SWPP 3.05 6.74 4.61 0.65 14 0.11 33.65 4.52 13.7 7.79 1.64 21 0.13 49.52 ** SY 418 2180 1277 399 31 31 67.74 672 3349 1713.9 592 35 46.1 70.20 ** PC 23.1 30.4 26.94 1.22 5 0.12 84.24 21.7 32.3 27.48 1.74 6 0.14 88.55 ** GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). The broad-sense heritability estimates were moderate to high for all traits, indicating that genetic constituents are the primary source of these traits. The highest heritability of 89.5% was estimated for TSW in 2019 and 88.5% for PC in 2020, respectively. Heritability Table 2. 2.1. Phenotypic Evaluation Trait 2019 2020 Min Max Mean SD CV (%) SE He2B Min Max Mean SD CV (%) SE He2B p Values (t Test) GPP 3.38 5.49 4.34 0.42 10 0.03 47.65 2.17 7.62 4.66 0.85 18 0.07 74.22 ** PPP 8.78 21.4 13.36 1.78 13 0.14 43.57 6.41 13.9 9.4 1.56 17 0.12 55.34 ** SPP 35.4 93.91 57.86 10.27 18 0.8 52.66 25.58 68.7 43.5 8.35 19 0.65 54.24 ** FD 12 38.21 22.85 4.78 21 0.37 87.52 9.40 41.7 22.06 4.64 21 0.36 79.32 0.06 PH 57.6 145.6 99.8 15.58 16 1.22 80.66 27.21 103 71.04 15.71 22 1.22 85.19 ** PL 0.42 0.88 0.69 0.10 15 0.01 54.56 0.32 0.93 0.68 0.15 22 0.01 82.03 0.11 PoL 3.16 8.93 5.76 0.90 16 0.07 91.44 4.04 9.91 5.97 0.84 14 0.07 93.12 ** TSW 37.2 315.8 174.7 53.48 31 4.18 96.60 40.54 307 187.5 54.92 30 4.31 97.70 ** SWPP 47 169.8 95.18 20.91 22 1.63 54.50 37.63 122 70.87 16.01 23 1.25 59.53 ** SY 305 2957 1555 0.42 42 51.1 86.89 2.17 3439 1813 0.85 36 50.2 86.88 ** PC 23.9 30.91 26.51 1.78 5 0.11 82.59 6.41 31.9 27.97 1.56 5 0.12 85.12 ** GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). Table 1. Descriptive statistics for eleven traits evaluated in pea genotypes ats IFVCNS, Serbia. 2.1. Phenotypic Evaluation Trait 2019 2020 Min Max Mean SD CV (%) SE He2B Min Max Mean SD CV (%) SE He2B p Values (t Test) GPP 3.38 5.49 4.34 0.42 10 0.03 47.65 2.17 7.62 4.66 0.85 18 0.07 74.22 ** PPP 8.78 21.4 13.36 1.78 13 0.14 43.57 6.41 13.9 9.4 1.56 17 0.12 55.34 ** SPP 35.4 93.91 57.86 10.27 18 0.8 52.66 25.58 68.7 43.5 8.35 19 0.65 54.24 ** FD 12 38.21 22.85 4.78 21 0.37 87.52 9.40 41.7 22.06 4.64 21 0.36 79.32 0.06 PH 57.6 145.6 99.8 15.58 16 1.22 80.66 27.21 103 71.04 15.71 22 1.22 85.19 ** PL 0.42 0.88 0.69 0.10 15 0.01 54.56 0.32 0.93 0.68 0.15 22 0.01 82.03 0.11 PoL 3.16 8.93 5.76 0.90 16 0.07 91.44 4.04 9.91 5.97 0.84 14 0.07 93.12 ** TSW 37.2 315.8 174.7 53.48 31 4.18 96.60 40.54 307 187.5 54.92 30 4.31 97.70 ** SWPP 47 169.8 95.18 20.91 22 1.63 54.50 37.63 122 70.87 16.01 23 1.25 59.53 ** SY 305 2957 1555 0.42 42 51.1 86.89 2.17 3439 1813 0.85 36 50.2 86.88 ** PC 23.9 30.91 26.51 1.78 5 0.11 82.59 6.41 31.9 27.97 1.56 5 0.12 85.12 ** GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). Results recorded at Agro Seed, Belgium, had mean SPP, PH and SY significantly lower in 2019 than in 2020. In contrast, TSW was higher in 2019 than in 2020, which is possibly explicable by higher temperatures observed during the pod filling period (Table S2). For all characteristics, statistically, significant differences were observed for mean values between two seasons for all traits, except for GPP. The highest coefficients of variation (CV) were observed for PL and SY in 2019 (37 and 31%) and for SY and GPP (31 and 28%) in 2020 (Table 2). High variation was also evident in PH and TSW in both years (16–31%). The coefficient of variability for all traits was generally higher in 2019. The lowest CV in 2019 was observed for PC (5%) and in 2020 for PC (6%). Table 2. 2.1. Phenotypic Evaluation 2.1. Phenotypic Evaluation Descriptive statistics and Pearson’s Correlation were conducted on morphological traits of analyzed pea genotypes. The studied features were number of grains per pod (GPP), number of pods per plant (PPP), number of seeds per plant (SPP), flowering duration (FD), plant height (PH), plot lodging (PL), pod length (PoL), thousand seed weight (TSW), seed weight per plant (SWPP), seed yield (SY), and protein content (PC). g p p y p According to the results of experiments at IFVCNS, Serbia, extensive phenotypic variation was observed in the analyzed pea panel. Means for PPP, SPP, SWPP, and PH were significantly lower in 2020 than in 2019, which is likely to be the consequence of drought in the flowering period in May 2020 (Table S1). In contrast, GPP, PoL, TSW, and SY were higher in 2020 than 2019. For all characteristics, statistically significant differences were observed for mean values between two seasons, except for PL and FD. The highest coefficients of variation (CV) were observed for SY and TSW in both years (42 and 36%, and 31 and 30%, respectively). High variation was also evident in SWPP and FD in both years (21–23%), Plants 2022, 11, 1321 3 of 17 3 of 17 while for PH and PL, higher values were recorded in the 2020 season (Table 1). The lowest CV was observed for PC (5%) in both years. The broad-sense heritability estimates were moderate for GPP, PPP, SPP, and SWPP and high for FD, PH, PoL, TSW, SY, and PC, which indicates that genetic constituents are the primary source of these traits. An exception was PL, whose heritability was significantly higher in 2020. The highest heritability of 96.6% and 97.7% were estimated for TSW in 2019 and 2020, respectively. Heritability was consistent between years for SPP, PH, PoL, TSW, SY, and PC and inconsistent for GPP, PPP, and PL (Table 1). Inconsistency between years for given traits could indicate their higher sensitivity to environmental conditions. and PL (Table 1). Inconsistency between years for given traits could indicate their higher sensitivity to environmental conditions. Table 1. Descriptive statistics for eleven traits evaluated in pea genotypes ats IFVCNS, Serbia. 2.1. Phenotypic Evaluation Trait 2019 2020 Min Max Mean SD CV (%) SE He2B Min Max Mean SD CV (%) SE He2B p Values (t Test) GPP 3.38 5.49 4.34 0.42 10 0.03 47.65 2.17 7.62 4.66 0.85 18 0.07 74.22 ** PPP 8.78 21.4 13.36 1.78 13 0.14 43.57 6.41 13.9 9.4 1.56 17 0.12 55.34 ** SPP 35.4 93.91 57.86 10.27 18 0.8 52.66 25.58 68.7 43.5 8.35 19 0.65 54.24 ** FD 12 38.21 22.85 4.78 21 0.37 87.52 9.40 41.7 22.06 4.64 21 0.36 79.32 0.06 PH 57.6 145.6 99.8 15.58 16 1.22 80.66 27.21 103 71.04 15.71 22 1.22 85.19 ** PL 0.42 0.88 0.69 0.10 15 0.01 54.56 0.32 0.93 0.68 0.15 22 0.01 82.03 0.11 PoL 3.16 8.93 5.76 0.90 16 0.07 91.44 4.04 9.91 5.97 0.84 14 0.07 93.12 ** TSW 37.2 315.8 174.7 53.48 31 4.18 96.60 40.54 307 187.5 54.92 30 4.31 97.70 ** SWPP 47 169.8 95.18 20.91 22 1.63 54.50 37.63 122 70.87 16.01 23 1.25 59.53 ** SY 305 2957 1555 0.42 42 51.1 86.89 2.17 3439 1813 0.85 36 50.2 86.88 ** PC 23.9 30.91 26.51 1.78 5 0.11 82.59 6.41 31.9 27.97 1.56 5 0.12 85.12 ** GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). Results recorded at Agro Seed, Belgium, had mean SPP, PH and SY significantly lower in 2019 than in 2020. In contrast, TSW was higher in 2019 than in 2020, which is possibly explicable by higher temperatures observed during the pod filling period (Table S2). For all characteristics, statistically, significant differences were observed for mean values between two seasons for all traits, except for GPP. The highest coefficients of variation (CV) were observed for PL and SY in 2019 (37 and 31%) and for SY and GPP (31 and 28%) in 2020 (Table 2). High variation was also evident in PH and TSW in both years (16–31%). The coefficient of variability for all traits was generally higher in 2019. The lowest CV in 2019 was observed for PC (5%) and in 2020 for PC (6%). Table 2. 2.2. Correlation Analysis between Traits Variables GPP PPP SPP FD PH PL PoL TSW SWPP SY PC GPP PPP −0.05 SPP 0.21 ** 0.75 ** FD 0 0.41 ** 0.29 ** PH 0.1 0.46 ** 0.30 ** 0.44 ** PL 0.03 −0.09 −0.06 −0.01 −0.12 PoL −0.02 −0.20 * −0.06 −0.20 * −0.12 0.04 TSW −0.17 * −0.26 ** −0.37 ** −0.08 0.08 0.29 ** 0.43 ** SWPP 0.05 0.36 ** 0.40 ** 0.05 0.21 ** 0.18 * 0.45 ** 0.51 ** SY 0.01 −0.04 0.07 −0.05 −0.12 0.5 0.23 ** 0.52 ** 0.64 ** PC −0.1 0.1 0.23 ** 0.01 −0.1 0.02 0.08 −0.2 0.07 0.14 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). The differences between the two locations regarding correlations between traits were GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). The correlation coefficients for the Agro Seed trial are shown in Table 4. Trait SWPP had positive correlation with all the traits (insignificant for GPP, FD, and PC). Trait SPP had significant positive correlation with FD and PH. Traits SY and PC were positively correlated (0,14). PC was also highly positively correlated with SPP. Table 4. Correlation coefficients between pea yield components at Agro Seed, Belgium. Table 4. Correlation coefficients between pea yield components at Agro Seed, Belgium. Variables GPP PPP SPP FD PH PL PoL TSW SWPP SY PC GPP PPP −0.05 SPP 0.21 ** 0.75 ** FD 0 0.41 ** 0.29 ** PH 0.1 0.46 ** 0.30 ** 0.44 ** PL 0.03 −0.09 −0.06 −0.01 −0.12 PoL −0.02 −0.20 * −0.06 −0.20 * −0.12 0.04 TSW −0.17 * −0.26 ** −0.37 ** −0.08 0.08 0.29 ** 0.43 ** SWPP 0.05 0.36 ** 0.40 ** 0.05 0.21 ** 0.18 * 0.45 ** 0.51 ** SY 0.01 −0.04 0.07 −0.05 −0.12 0.5 0.23 ** 0.52 ** 0.64 ** PC −0.1 0.1 0.23 ** 0.01 −0.1 0.02 0.08 −0.2 0.07 0.14 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. 2.2. Correlation Analysis between Traits Significant at p ≤0.05 (*), p ≤0.01 (**). 4. Correlation coefficients between pea yield components at Agro Seed, Belgium. GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). The differences between the two locations, regarding correlations between traits, were no significant correlation of GPP with FD, PH, PL, PoL, and SWPP in the Agro Seed trial, while the IFVCNS trial showed high correlations for these traits. Furthermore, in the Agro Seed trial there was no correlation between SY and traits FD and PL, while in the IFVCNS trial, there was no correlation between PC and SPP. The main difference between locations was in the correlation between PC and SY; IFVCNS trial showed negative correlation, while Agro Seed trial showed positive correlation. 2.2. Correlation Analysis between Traits Variables GPP PPP SPP FD PH PL PoL TSW SWPP SY PC GPP PPP −0.05 SPP 0.21 ** 0.75 ** FD 0 0.41 ** 0.29 ** PH 0.1 0.46 ** 0.30 ** 0.44 ** PL 0.03 −0.09 −0.06 −0.01 −0.12 PoL −0.02 −0.20 * −0.06 −0.20 * −0.12 0.04 TSW −0.17 * −0.26 ** −0.37 ** −0.08 0.08 0.29 ** 0.43 ** SWPP 0.05 0.36 ** 0.40 ** 0.05 0.21 ** 0.18 * 0.45 ** 0.51 ** SY 0.01 −0.04 0.07 −0.05 −0.12 0.5 0.23 ** 0.52 ** 0.64 ** PC −0.1 0.1 0.23 ** 0.01 −0.1 0.02 0.08 −0.2 0.07 0.14 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). The differences between the two locations, regarding correlations between traits, were no significant correlation of GPP with FD, PH, PL, PoL, and SWPP in the Agro Seed trial, while the IFVCNS trial showed high correlations for these traits. Furthermore, in the Agro Seed trial there was no correlation between SY and traits FD and PL, while in the IFVCNS trial, there was no correlation between PC and SPP. The main difference between locations was in the correlation between PC and SY; IFVCNS trial showed negative correlation, while Agro Seed trial showed positive correlation. 2.3. Correlation of Traits between Two Years Correlations between trait values observed in 2019 and 2020 for IFVCNS and Agro Seed were graphically represented using scatter diagrams (Figures 1 and 2). For all exam- ined traits, linear regressions in pairwise comparisons between years revealed positive correlations (Table S3). Table 3. Correlation coefficients between pea yield components at IFVCNS, Serbia. Table 3. Correlation coefficients between pea yield components at IFVCNS, Serbia. 2.2. Correlation Analysis between Traits The correlation coefficients for the IFVCNS trial between different traits are shown in Table 3. SWPP had positive and significant correlation with a great number of traits (GPP, PPP, SPP, PoL, TSW, and SY), but was highly negatively correlated with PC (−0.46); PC was also negatively correlated with SY and TSW. Furthermore, TSW was negatively correlated with GPP, PPP, and SPP, and positively with SY and PoL. Correlation between the remaining pairs of traits was mainly low and non-significant in both years. Table 3. Correlation coefficients between pea yield components at IFVCNS, Serbia. Variables GPP PPP SPP FD PH PL PoL TSW SWPP SY PC GPP PPP −0.12 SPP 0.41 ** 0.76 ** FD −0.18 * 0.16 * 0.02 PH −0.33 ** 0.23 ** 0.07 0.13 * PL −0.41 ** 0.14 −0.21 ** −0.06 −0.07 PoL 0.41 ** −0.42 ** −0.18 * −0.09 −0.25 ** −0.21 ** TSW −0.19 ** −0.42 ** −0.6 * 0.08 −0.04 0.16 ** 0.56 ** SWPP 0.22 ** 0.17 * 0.23 ** 0.13 −0.08 −0.03 0.56 ** 0.55 ** SY −0.02 −0.1 −0.13 0.18 * −0.2 0.21 * 0.29 ** 0.58 ** 0.64 ** PC 0 0.11 0.14 −0.1 0.14 −0.16 −0.22 −0.44 * −0.46 ** −0.7 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). The correlation coefficients for the Agro Seed trial are shown in Table 4. Trait SWPP had positive correlation with all the traits (insignificant for GPP, FD, and PC). Trait SPP had significant positive correlation with FD and PH. Traits SY and PC were positively correlated (0,14). PC was also highly positively correlated with SPP. Table 4. Correlation coefficients between pea yield components at Agro Seed, Belgium. 2.1. Phenotypic Evaluation Descriptive statistics for eleven traits evaluated in pea genotypes at Agro Seed, Belgium. GPP—grains per pod (#); PPP—pods per plant (#); SPP—seeds per plant (#); FD—flowering duration (days); PH—plant height (cm); PL—plot lodging; PoL—pod length (cm); TSW—thousand seed weight (g); SWPP—seed weight per plant (g); SY—seed yield (g/plot); PC—protein content (%); SD—standard deviation; CV—coefficient of variation (%); SE—standard error; He2B—broad-sense heritability (%); p-value (t test done for two-year performance measurement results). Significant at p ≤0.01 (**). The broad-sense heritability estimates were moderate to high for all traits, indicating that genetic constituents are the primary source of these traits. The highest heritability of 89.5% was estimated for TSW in 2019 and 88.5% for PC in 2020, respectively. Heritability Plants 2022, 11, 1321 4 of 17 was consistent between years for GPP, SPP, PoL and TSW, SY, and PC, and inconsistent for PPP, FL, PH, PL, and SWPP. was consistent between years for GPP, SPP, PoL and TSW, SY, and PC, and inconsistent for PPP, FL, PH, PL, and SWPP. 2.2. Correlation Analysis between Traits Variables GPP PPP SPP FD PH PL PoL TSW SWPP SY PC GPP PPP −0.12 SPP 0.41 ** 0.76 ** FD −0.18 * 0.16 * 0.02 PH −0.33 ** 0.23 ** 0.07 0.13 * PL −0.41 ** 0.14 −0.21 ** −0.06 −0.07 PoL 0.41 ** −0.42 ** −0.18 * −0.09 −0.25 ** −0.21 ** TSW −0.19 ** −0.42 ** −0.6 * 0.08 −0.04 0.16 ** 0.56 ** SWPP 0.22 ** 0.17 * 0.23 ** 0.13 −0.08 −0.03 0.56 ** 0.55 ** SY −0.02 −0.1 −0.13 0.18 * −0.2 0.21 * 0.29 ** 0.58 ** 0.64 ** PC 0 0.11 0.14 −0.1 0.14 −0.16 −0.22 −0.44 * −0.46 ** −0.7 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). Table 3. Correlation coefficients between pea yield components at IFVCNS, Serbia. Variables GPP PPP SPP FD PH PL PoL TSW SWPP SY PC GPP PPP −0.12 SPP 0.41 ** 0.76 ** FD −0.18 * 0.16 * 0.02 PH −0.33 ** 0.23 ** 0.07 0.13 * PL −0.41 ** 0.14 −0.21 ** −0.06 −0.07 PoL 0.41 ** −0.42 ** −0.18 * −0.09 −0.25 ** −0.21 ** TSW −0.19 ** −0.42 ** −0.6 * 0.08 −0.04 0.16 ** 0.56 ** SWPP 0.22 ** 0.17 * 0.23 ** 0.13 −0.08 −0.03 0.56 ** 0.55 ** SY −0.02 −0.1 −0.13 0.18 * −0.2 0.21 * 0.29 ** 0.58 ** 0.64 ** PC 0 0.11 0.14 −0.1 0.14 −0.16 −0.22 −0.44 * −0.46 ** −0.7 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). The correlation coefficients for the Agro Seed trial are shown in Table 4. Trait SWPP had positive correlation with all the traits (insignificant for GPP, FD, and PC). Trait SPP had significant positive correlation with FD and PH. Traits SY and PC were positively correlated (0,14). PC was also highly positively correlated with SPP. Table 4. Correlation coefficients between pea yield components at Agro Seed, Belgium. 2.4. Multivariate Analysis 2.4. Multivariate Analysis Multivariate analysis was performed based on all examined traits for both trial sites using mean values of both years to explore the population structure of the pea genotypes of different plant types (Figure 3), as well as of the variety type by use (Figure 4). The PCA gave a graphical representation of the broad phenotypic diversity of the investigated pea panel. Based on the results from IFVCNS, the first principal component accounted for 30.3% of the variance and showed no apparent clustering of the genotypes based on population type. The second principal component accounted for 20.1% of the variance and showed clustering of landrace types and overlapping of varieties and most breeding lines. Based on the results from Agro Seed, the first principal component accounted for 30% of the variance and showed some clustering of the breeding lines. In comparison, the second principal component accounted for 26% of the variance and showed a less clear subdivision of varieties. A large number of the breeding lines and varieties originated from North America and Serbia, while most of the landraces originated from Sweden (Table S4). y Multivariate analysis was performed based on all examined traits for both trial sites using mean values of both years to explore the population structure of the pea genotypes of different plant types (Figure 3), as well as of the variety type by use (Figure 4). The PCA gave a graphical representation of the broad phenotypic diversity of the investigated pea panel. Based on the results from IFVCNS, the first principal component accounted for 30.3% of the variance and showed no apparent clustering of the genotypes based on pop- ulation type. The second principal component accounted for 20.1% of the variance and showed clustering of landrace types and overlapping of varieties and most breeding lines. Based on the results from Agro Seed, the first principal component accounted for 30% of the variance and showed some clustering of the breeding lines. In comparison, the second principal component accounted for 26% of the variance and showed a less clear subdivi- sion of varieties. A large number of the breeding lines and varieties originated from North America and Serbia, while most of the landraces originated from Sweden (Table S4). Figure 3. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by plant type. 2.3. Correlation of Traits between Two Years Correlations between trait values observed in 2019 and 2020 for IFVCNS and Agro Seed were graphically represented using scatter diagrams (Figures 1 and 2). For all exam- ined traits, linear regressions in pairwise comparisons between years revealed positive correlations (Table S3). Plants 2022, 11, 1321 5 of 17 5 of 17 Figure 1. Correlations and linear regressions between data collected on the sa (165) grown in two years (2019 and 2020) at IFVCNS, Serbia. Figure 1. Correlations and linear regressions between data collected on the same pea genotypes (165) grown in two years (2019 and 2020) at IFVCNS, Serbia. Figure 1. Correlations and linear regressions between data collected on the sa (165) grown in two years (2019 and 2020) at IFVCNS, Serbia. Figure 1. Correlations and linear regressions between data collected on the sam (165) grown in two years (2019 and 2020) at IFVCNS, Serbia. Figure 1. Correlations and linear regressions between data collected on the same pea genotypes (165) grown in two years (2019 and 2020) at IFVCNS, Serbia. Figure 1. Correlations and linear regressions between data collected on the sam (165) grown in two years (2019 and 2020) at IFVCNS, Serbia. Figure 2. Correlations and linear regressions between data collected on the same pea genotypes (165) grown in two years (2019 and 2020) at Agro Seed, Belgium. Figure 2. Correlations and linear regressions between data collected on the same pea genotypes (165) grown in two years (2019 and 2020) at Agro Seed, Belgium. Plants 2022, 11, 1321 6 of 17 6 of 17 For the IFVCNS trial, correlations between seasons were mostly highly significant, except for flowering duration (FD). The strongest correlation was noticed for TSW (0.92), PH (0.77), and PoL (0.75), followed by GPP (0.60), SWPP (0.48), PL (0.46), SPP (0.45), and PPP (0.44), which had lower Pearson’s correlation coefficients (r). FD and PC had the lowest correlation between years (0.13 and 0.07, respectively). IEW 7 of 18 For the Agro Seed trial, the strongest correlation was noticed for GPP (0.96), PH (0.67), TSW (0.64), and the lowest correlation was for SPP (0.07) due to the high virus presence in 2020 (data not shown), FD (0.01), and PC (0.03). PC had a low correlation between localities (0.27 at IFVCNS and 0.17 at Agro Seed). IEW 7 of 18 Figure 2. 2.3. Correlation of Traits between Two Years Correlations and linear regressions between data collected on the same pea genotypes (165) grown in two years (2019 and 2020) at Agro Seed, Belgium. 2.4. Multivariate Analysis 2.4. Multivariate Analysis y p g yp IFVCNS, Serbia Agro Seed, Belgium Trait 2019 2020 2019 2020 F p-Value F p-Value F p-Value F p-Value GPP 33.171 0.0122 * 53.086 0.0005 ** 0.45 0.772 0.55 0.696 PPP 17.281 0.1463 0.6242 0.6459 1.10 0.357 1.44 0.222 SPP 30.741 0.018 * 12.562 0.2895 2.01 0.096 0.16 0.957 FD 14.681 0.2134 0.3419 0.8494 0.27 0.894 1.58 0.183 PH 79.438 0.000 ** 11.588 0.000 * 14.28 0.000 ** 14.32 0.000 ** PL 0.6091 0.6566 17.638 0.1387 3.42 0.010 2.05 0.089 * PoL 49.446 0.0009 ** 31.052 0.0171 * 1.66 0.162 3.16 0.016 * TSW 37.281 0.0063 * 3.375 0.0111 * 3.85 0.005 ** 2.57 0.040 * SWPP 30.949 0.017 * 16.254 0.170 1.51 0.201 0.47 0.761 SY 10.22 0.000 ** 4 0.009 * 1.97 0.101 3.26 0.013 * PC 4.87 0.001 ** 9.62 0.000 ** 2.50 0.045 6.1 0.000 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). y p g yp IFVCNS, Serbia Agro Seed, Belgium Trait 2019 2020 2019 2020 F p-Value F p-Value F p-Value F p-Value GPP 33.171 0.0122 * 53.086 0.0005 ** 0.45 0.772 0.55 0.696 PPP 17.281 0.1463 0.6242 0.6459 1.10 0.357 1.44 0.222 SPP 30.741 0.018 * 12.562 0.2895 2.01 0.096 0.16 0.957 FD 14.681 0.2134 0.3419 0.8494 0.27 0.894 1.58 0.183 PH 79.438 0.000 ** 11.588 0.000 * 14.28 0.000 ** 14.32 0.000 ** PL 0.6091 0.6566 17.638 0.1387 3.42 0.010 2.05 0.089 * PoL 49.446 0.0009 ** 31.052 0.0171 * 1.66 0.162 3.16 0.016 * TSW 37.281 0.0063 * 3.375 0.0111 * 3.85 0.005 ** 2.57 0.040 * SWPP 30.949 0.017 * 16.254 0.170 1.51 0.201 0.47 0.761 SY 10.22 0.000 ** 4 0.009 * 1.97 0.101 3.26 0.013 * PC 4.87 0.001 ** 9.62 0.000 ** 2.50 0.045 6.1 0.000 ** GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). 2.5. Genetic Distance and Hierarchical Clustering This work was part of a more extensive study on genetic diversity and phenotypic variation in yield and protein content of pea accessions in Europe. 2.4. Multivariate Analysis 2.4. Multivariate Analysis Analysis of variance for results at Agro Seed indicated a statistically significant difference Plants 2022, 11, 1321 7 of 17 ted site 7 of 17 ted site among pea genotypes for the following traits: PH and TSW in both years, and PC, PoL, and PC in 2020. Morpho-phenological data in both trial sites did not allow completely clear subdivi- sion between or within gene pools. among pea genotypes for the following traits: PH and TSW in both years, and PC, PoL, and PC in 2020. Morpho-phenological data in both trial sites did not allow completely clear subdivi- sion between or within gene pools. 3 2 1 0 -1 -2 -3 -4 -5 5 4 3 2 1 0 -1 -2 -3 -4 1st Comp 30.3% 2nd Comp. 20.1% dry forage landrace vegetable wild By use PCA Type by use Figure 4. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by variety type by usage. (Left): IFVCNS, Serbia site results; (Right): Agro Seed, Belgium site results. 3 2 1 0 -1 -2 -3 -4 -5 5 4 3 2 1 0 -1 -2 -3 -4 1st Comp 30.3% 2nd Comp. 20.1% dry forage landrace vegetable wild By use PCA Type by use Figure 4. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by variety type by usage. (Left): IFVCNS, Serbia site results; (Right): Agro Seed, Belgium site results. Table 5. Analysis of variance for data collected on pea genotypes tested on both sites. Table 5. Analysis of variance for data collected on pea genotypes tested on both sites. 2.4. Multivariate Analysis 2.4. Multivariate Analysis (Left): IFVCNS, Serbia site results; (Right): Agro Seed, Belgium site results. 3 2 1 0 -1 -2 -3 -4 -5 5 4 3 2 1 0 -1 -2 -3 -4 1st Comp 30.3% 2nd Comp. 20.1% Breeding Line Landrace Semi Wild Variety Wild Population type PCA Population type Figure 3. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by plant type. (Left): IFVCNS, Serbia site results; (Right): Agro Seed, Belgium site results. 3 2 1 0 -1 -2 -3 -4 -5 5 4 3 2 1 0 -1 -2 -3 -4 1st Comp 30.3% 2nd Comp. 20.1% Breeding Line Landrace Semi Wild Variety Wild Population type PCA Population type Figure 3. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by plant type. (Left): IFVCNS, Serbia site results; (Right): Agro Seed, Belgium site results. Figure 3. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by plant type. (Left): IFVCNS, Serbia site results; (Right): Agro Seed, Belgium site results. Morpho-phenological data in both trial sites did not allow completely clear subdivi- sion between or within gene pools. Morpho-phenological data in both trial sites did not allow completely clear subdivision between or within gene pools. dry forage landrace By use CA Type by use g p PCA analysis based on the variety type on both locations showed a higher prevalence of dry pea type and showed that accessions of dry pea type were generally clustered according to their use. However, it did not provide a clearer subdivision of pea genotypes by use. landrace vegetable wild y Although PCA analysis did not show a clear grouping of genotypes based on plant type, an analysis of variance was conducted to determine possible differences within the germplasm. Analysis of variance for results at IFVCNS indicated a statistically significant difference among pea genotypes for the following traits: GPP, PH, PoL, and TSW in both years (Table 5). Furthermore, differences were observed for SPP and SWPP only in 2019. 2.4. Multivariate Analysis 2.4. Multivariate Analysis This group consists of modern dry v are related and grouped together. dry types of use. This group consists of modern dry varieties, which are related and grouped together. mostly varieties and dry types of use. This group consists of modern dry v are related and grouped together. dry types of use. This group consists of modern dry varieties, which are related and grouped together. mostly varieties and dry types of use. This group consists of modern dry v are related and grouped together. Figure 5. Hierarchical clustering of 156 pea accessions used in the experiment. Th classified into four groups based on the clustering. Figure 5. Hierarchical clustering of 156 pea accessions used in the experiment. The accessions were classified into four groups based on the clustering. Figure 5. Hierarchical clustering of 156 pea accessions used in the experiment. The classified into four groups based on the clustering. Figure 5. Hierarchical clustering of 156 pea accessions used in the experiment. The accessions were classified into four groups based on the clustering. A Mantel test was performed to test the correlations between the g matrix and the phenotypic distance matrices at the two sites and years ( were significant correlations at both sites and in both years However, t A Mantel test was performed to test the correlations between the genetic distance matrix and the phenotypic distance matrices at the two sites and years (Table 6). There were significant correlations at both sites and in both years. However, the correlations were highest for the two seasons at IFVCNS, Serbia. g y , were highest for the two seasons at IFVCNS, Serbia. Table 6. Mantel test results for correlations between the genotype distance matrix and the phenotype distance matrices at the IFVCNS and ASR sites in 2019 and 2020. were highest for the two seasons at IFVCNS, Serbia. Table 6. Mantel test results for correlations between the genotype distance matrix and the phenotype distance matrices at the IFVCNS and ASR sites in 2019 and 2020. were highest for the two seasons at IFVCNS, Serbia. Table 6. Mantel test results for correlations between the genotype distance matrix and the phenotype distance matrices at the IFVCNS and ASR sites in 2019 and 2020. Table 6. Mantel test results for correlations between the genotype distance matrix type distance matrices at the IFVCNS and ASR sites in 2019 and 2020. 2.4. Multivariate Analysis 2.4. Multivariate Analysis After filtering, markers with a Gentrain score (Illumina quality measure) below 0.7 and a minimum allele frequency of 0.05–a total of 11,693 polymorphic markers—were left in the panel of 165 accessions used here. The genetic distance matrix was based on all the 16,693 markers and the phenotypic distance matrices were based on all the phenotypic traits measured in this study. The hierarchical grouping chart showed four groups (Figure 5). The first group (green) consists of nine genotypes, exclusively of the breeding line population type and dry type of use, originating from Serbia The second group (red) consists of 36 genotypes mainly of the GPP—grains per pod; PPP—pods per plant; SPP—seeds per plant; FD—flowering duration; PH—plant height; PL—plot lodging; PoL—pod length; TSW—thousand seed weight; SWPP—seed weight per plant; SY—seed yield; PC—protein content. Significant at p ≤0.05 (*), p ≤0.01 (**). This work was part of a more extensive study on genetic diversity and phenotypic variation in yield and protein content of pea accessions in Europe. After filtering, markers with a Gentrain score (Illumina quality measure) below 0.7 and a minimum allele frequency of 0.05–a total of 11,693 polymorphic markers—were left in the panel of 165 accessions used here. The genetic distance matrix was based on all the 16,693 markers and the phenotypic distance matrices were based on all the phenotypic traits measured in this study. The hierarchical grouping chart showed four groups (Figure 5). The first group (green) consists of nine genotypes, exclusively of the breeding line population type and dry type of use, originating from Serbia. The second group (red) consists of 36 genotypes, mainly of the breeding line population type and dry type of use, with some exceptions of the vegetable type. In this group, vegetable pea lines, as expected, are clustered very closely together (mainly the 00 lines, Table S4); however, a set of 00 lines shows introgressions from other types of peas. A number of forage peas are also part of this genetic group (Flex, H3) The third group (blue) consists of 57 genotypes, mainly landraces, forage, wild/semi-wild population, and dry types of use. Clustering showed that some of the landraces are very closely related. The fourth group (black) consists of 63 genotypes, mostly varieties and Plants 2022, 11, 1321 8 of 17 dry types of use. This group consists of modern dry varieties, which are related and grouped together. mostly varieties and dry types of use. 2.4. Multivariate Analysis 2.4. Multivariate Analysis BL– breeding line; L–landrace; SW–semi-wild; V–variety; W–wild. (Left): plant (population) type; (Right): variety type by usage. ure 6. Principal component analysis (PCA) of the broad phenotypic diversity of the inv panel composed by analyzed genotypes, classified by using the DNA extraction data. B line; L–landrace; SW–semi-wild; V–variety; W–wild. (Left): plant (population) type; (Ri The seed yield and protein content was projected on the first two principal compo- nents from the PCA of the genotype data to further analyze the phenotypic variance and G*E interaction using 3D graphs (Figures 7 and 8). The interactions between the genotypes of the accessions, the two sites, years, and usage or plant type showed that for seed yield, the dry and vegetable types tended to have the widest range of yields irrespective of year PCA analysis in relation to plant usage shows and reinforces the structure in the PCA graph by plant type, successfully grouping modern bred varieties (dry and vegetable) whose accessions dominate on the left, while the landrace, forage, and wild types, being clearly distinct phenotypically, are on the right-hand side. y type by usage. The seed yield and protein content was projected on the first two principal nts from the PCA of the genotype data to further analyze the phenotypic varia the dry and vegetable types tended to have the widest range of yields irrespective of year and site, while the forage, landraces, and wild types tended to be intermediary (Figure 7). For protein content, the forage, landraces and wild types were highest, particularly in 2020, at both sites. The seed yield and protein content was projected on the first two principal components from the PCA of the genotype data to further analyze the phenotypic variance and G*E interaction using 3D graphs (Figures 7 and 8). The interactions between the genotypes of the accessions, the two sites, years, and usage or plant type showed that for seed yield, the dry and vegetable types tended to have the widest range of yields irrespective of year and site, while the forage, landraces, and wild types tended to be intermediary (Figure 7). For protein content, the forage, landraces and wild types were highest, particularly in 2020, at both sites. g yp y p yp E interaction using 3D graphs (Figures 7 and 8). 2.4. Multivariate Analysis 2.4. Multivariate Analysis Mantel Test IFVCNS19 IFVCNS20 ASR19 Mantel Test IFVCNS19 IFVCNS20 ASR19 ASR20 Correlation (r) 0.238 0.240 0.120 0.110 Probability 1 × 10−4 1 × 10−4 2 × 10−4 6 × 10−4 Correlation (r) 0.238 0.240 0.120 Probability 1 × 10−4 1 × 10−4 2 × 10−4 A Principal Component Analysis (PCA) was also performed on the and the data for the first two components were related to population type and plant type (Figure 6, right). In the PCA graph grouped by plant type, th t l d d i ild d ild t i l t t d th A Principal Component Analysis (PCA) was also performed on the genotype data and the data for the first two components were related to population type (Figure 6, left) and plant type (Figure 6, right). In the PCA graph grouped by plant type, it is noticeable that landraces and semi-wild and wild materials were concentrated on the cluster’s right side, while breeding lines and varieties were mainly concentrated on the left. This shows a clear morphological group formed from landraces (longer internodes), semi-wild and wild, and the modern bred dry pea varieties. 9 of 17 f 18 Plants 2022, 11, 1321 Plants 2022, 11, x F Figure 6. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by using the DNA extraction data. BL–breed- ing line; L–landrace; SW–semi-wild; V–variety; W–wild. (Left): plant (population) type; (Right): va- riety type by usage. Figure 6. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by using the DNA extraction data. BL– breeding line; L–landrace; SW–semi-wild; V–variety; W–wild. (Left): plant (population) type; (Right): variety type by usage. Figure 6. Principal component analysis (PCA) of the broad phenotypic diversity of the inv pea panel composed by analyzed genotypes, classified by using the DNA extraction data. B ing line; L–landrace; SW–semi-wild; V–variety; W–wild. (Left): plant (population) type; (R Figure 6. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by using the DNA extraction data. BL–breed- ing line; L–landrace; SW–semi-wild; V–variety; W–wild. (Left): plant (population) type; (Right): va- riety type by usage. Figure 6. Principal component analysis (PCA) of the broad phenotypic diversity of the investigated pea panel composed by analyzed genotypes, classified by using the DNA extraction data. arly in 2020. 3. Discussion Discussion High variability in germplasm collection is desirable when it comes to introd w sources of variation in breeding programs. Evaluation and characterization of d agronomic traits can improve the identification of the phenotypic clusters w oad geographic groups [21]. In order to assess the phenotypic variability of rmplasm collection, the partially-replicated experimental design used in this wo wed successful estimation of broad-sense heritability (He2B) using limited seed q s, significantly reducing the cost of phenotyping [22] a diversity panel contain rge number of lines. Extensive phenotypic variation was present in the diversity both trial sites, illustrated particularly by high coefficients of variation for S VCNS and Agro Seed), for TSW (at IFVCNS), and for GPP (at Agro Seed), as well WPP and PL (at IFVCNS) in 2020 and for PH in 2019 and TSW in 2020 (at Agro S hile for PH, higher values were recorded in the 2020 season for both trial sites. ritability of TSW was observed at both IFVCNS and Agro Seed, similar to pre ndings [15,19,23,24], showing that this trait is highly genetically controlled. The l ritability was recorded for PPP in 2019 and SPP in 2020 for IFVCNS, and for PPP in d PL in 2020 for Agro Seed, similar to the findings of [19]. Lower heritability for aits could be due to the significant error variance caused by high environmental ce (particularly with regards to rainfall) and not the narrow genetic variance. Mea High variability in germplasm collection is desirable when it comes to introducing new sources of variation in breeding programs. Evaluation and characterization of inherited agronomic traits can improve the identification of the phenotypic clusters within broad geographic groups [21]. In order to assess the phenotypic variability of a pea germplasm collection, the partially-replicated experimental design used in this work allowed successful estimation of broad-sense heritability (He2B) using limited seed quantities, significantly reducing the cost of phenotyping [22] a diversity panel containing a large number of lines. 2.4. Multivariate Analysis 2.4. Multivariate Analysis The interactions between the ge the accessions, the two sites, years, and usage or plant type showed that for see dry and vegetable types tended to have the widest range of yields irrespective d site, while the forage, landraces, and wild types tended to be intermediary (F r protein content, the forage, landraces and wild types were highest, particu 20, at both sites. Figure 7. 3D graph of the pea seed yield and pea protein content projected on the first two principal components from the PCA of the genotype data for the type of use. (Left): seed yield; (Right): pro- tein content. Figure 7. 3D graph of the pea seed yield and pea protein content projected on the first two p components from the PCA of the genotype data for the type of use. (Left): seed yield; (Rig tein content. Figure 7. 3D graph of the pea seed yield and pea protein content projected on the first two principal com- ponents from the PCA of the genotype data for the type of use. (Left): seed yield; (Right): protein content. Figure 7. 3D graph of the pea s components from the PCA of t tein content. eed yield and pea protein content projected on the first two principa he genotype data for the type of use. (Left): seed yield; (Right): pro- Figu comp tein projected on the first two principa se. (Left): seed yield; (Right): pro gure 7. 3D graph of the pea s mponents from the PCA of t n content. eed yield and pea protein content projec he genotype data for the type of use. (L ure 7. 3D graph of the pea seed yield and pea protein content projected on the first two p mponents from the PCA of the genotype data for the type of use. (Left): seed yield; (Rig n content. Figure 7. 3D graph of the pea seed yield and pea protein content projected on the first two principal com- ponents from the PCA of the genotype data for the type of use. (Left): seed yield; (Right): protein content. Plants 2022, 11, 1321 2022, 11, x FOR PEER 10 of 171 Figure 8. 3D graph of the pea seed yield and pea protein content (the first two principal comp from the PCA of the genotype data) for the plant type. (Left): seed yield; (Right): protein c B–breeding Line; V–variety; SW–semi-wild; W–wild; L–landrace. Figure 8. 2.4. Multivariate Analysis 2.4. Multivariate Analysis 3D graph of the pea seed yield and pea protein content (the first two principal components from the PCA of the genotype data) for the plant type. (Left): seed yield; (Right): protein content. B–breeding Line; V–variety; SW–semi-wild; W–wild; L–landrace. gure 8. 3D graph of the pea seed yield and pea protein content (the first two principal comp m the PCA of the genotype data) for the plant type. (Left): seed yield; (Right): protein co breeding Line; V–variety; SW–semi-wild; W–wild; L–landrace. Figure 8. 3D graph of the pea seed yield and pea protein content (the first two principal components from the PCA of the genotype data) for the plant type. (Left): seed yield; (Right): protein content. B–breeding Line; V–variety; SW–semi-wild; W–wild; L–landrace. In terms of plant type, the varieties and breeding lines had the accessions wi ghest seed yield, but also the widest range of yield. Again, the patterns were s ithin the same year at the two sites (Figure 8). For protein content, landraces and nd semi-wild accessions were in the high category, together with a few varieties, p l i 2020 In terms of plant type, the varieties and breeding lines had the accessions with the highest seed yield, but also the widest range of yield. Again, the patterns were similar within the same year at the two sites (Figure 8). For protein content, landraces and wild and semi-wild accessions were in the high category, together with a few varieties, particularly in 2020. arly in 2020. 3. Discussion If there is a correlation between two traits, direct selection of one will cause a change in the other. Before initiating any effective selection program, it is necessary to know the association of various traits with each other [27]. The relationship between various traits of peas has been studied previously, but the results were found to vary significantly according to varietal differences and environmental conditions [28–31]. This study used a very diverse group of pea genotypes and many varieties have a low number of seeds per pod or small grains, therefore diluting the correlation. However, in both trial sites, a positive and significant correlation was found between the traits that represent plant yield (SWPP) with PoL, PPP, SPP, and TSW, similar to previous findings [32,33], as well as a high positive correlation of SY with TSW and SWPP. The high positive correlation between PPP and SPP is similar to the results of [34]. These results indicate that number of pods per plant has a significant influence on yield. The positive correlation of PPP with PH and the negative correlation of GPP with PH, both observed at IFVCNS, is similar to [35], suggesting that the height of the plant could affect the number and size of seeds. Furthermore, the negative correlation between PH and PoL and the positive correlation between PoL and GPP observed at IFVCNS may indicate an indirect negative correlation between height and grains per pod. A significant negative correlation was expressed between TSW with GPP, PPP, and SPP for both trial sites, and GPP with PL (in the IFVCNS trial). SPP and FL were positively correlated only in the Agro Seed trial. PC was significantly negatively correlated with TSW only in the IFVCNS trial, unlike the results of [36], who noted positive correlations between these two traits. Our results for the IFVCNS trial, similar to [37], showed that in pea, the relationship between seed yield and seed weight per plant is always positive, regardless of the environment and the genetic background, which implies that the relationship between seed protein content and seed yield is almost always negative. Correlations between remaining pairs of traits mainly were low and non-significant in both years, which might indicate the presence of nonlinear interaction between traits [38]. Correlation of a single trait between years was analyzed. A low correlation between years might suggest a higher genotype by environment (GxE) interaction [15]. arly in 2020. 3. Discussion Extensive phenotypic variation was present in the diversity panel at both trial sites, illustrated particularly by high coefficients of variation for SY (at IFVCNS and Agro Seed), for TSW (at IFVCNS), and for GPP (at Agro Seed), as well as for SWPP and PL (at IFVCNS) in 2020 and for PH in 2019 and TSW in 2020 (at Agro Seed), while for PH, higher values were recorded in the 2020 season for both trial sites. High heritability of TSW was observed at both IFVCNS and Agro Seed, similar to previous findings [15,19,23,24], showing that this trait is highly genetically controlled. The lowest heritability was recorded for PPP in 2019 and SPP in 2020 for IFVCNS, and for PPP in 2019 and PL in 2020 for Agro Seed, similar to the findings of [19]. Lower heritability for these traits could be due to the significant error variance caused by high environmental influence (particularly with regards to rainfall) and not the narrow genetic variance. Means for PPP, SPP, SWPP, and PH were significantly lower in 2020 than in 2019, which might be the consequence of drought during the flowering period in May 2020 in Serbia (Table S1). The mean for SY was higher in 2020; similar findings were given by [6], where a high amount of precipitation, especially at the flowering stage, positively affected the grain yield of field pea. Means for Plants 2022, 11, 1321 11 of 17 11 of 17 PPP, SPP, PH, and SY were lower in 2019 in Belgium. In contrast, TSW was higher in 2019 compared to 2020, which is similar to the findings of [25], where lack of moisture in the pod-filling period led to lower number of pods and seeds per plant. This causes the plants to concentrate their nutrients and energy in the weight of the seeds, giving a higher mass of 1000 seeds, but generally lower yields. The coefficient of variation for all traits in the Serbia trial was generally higher in 2020, which is most likely due to uneven rain distribution during the pea growing season (Table S1), which can affect all the elements of the yielding structure, the length of flowering, and the height and lodging of plants. The phenotypic correlation is conditioned by the relationship among individual traits and the influence of environmental factors [26]. arly in 2020. 3. Discussion Thus, it can be concluded that traits with lower correlation between 2019 and 2020, such as PPP, SPP, PL, and SWPP, were influenced by environmental factors, while TSW, PoL, GPP, and PH were less affected. Similar findings were observed by [15] and [19], which also detected lower GxE interactions for TSW and higher interactions for SPP. The lowest correlation was observed for FL, which might indicate genotype reaction to agro-ecological conditions (high GxE). This might be influenced by the fact that the varieties, wild accessions, and landraces included in the assessed germplasm collection originated from different climatic conditions. Since the correlations of all examined traits in two years were positive, it can be concluded that the studied set of genotypes did not have strong reactions to different agro-ecological factors. g g Despite the absence of clearly separate groups among pea genotypes based on plant type, comparison among groups by different subtypes (forage, dry, and vegetable) indi- cated their differences in phenotypic traits based on PCA, so deviations in morphological characteristics between groups comparing trial sites can still be visible. Based on the IFVCNS trials, where the grouping of most landraces is visible on the PCA diagram (Figure 5, left), grouping could result from landrace characteristics to have a higher plant Plants 2022, 11, 1321 12 of 17 12 of 17 height and smaller grains. Clustering can also be seen for some of the varieties and breeding lines. However, both these groups consist of genotypes from different climates and origins and different uses. Further studies, similar to [15] and [39], performed at the molecular level, could show a more detailed grouping of these genotypes. Comparing type by use, in the IFVCNS trial, clustering can be seen mostly for the dry pea subtype, followed by the forage subtype (Figure 6, left). Dry pea varieties are characterized by small number of seeds per pod, large grains, and about 20–25% protein content [40,41], while forage peas are mainly characterized by small grains, indeterminate type, and 17–24% protein content [41,42]. Similar clustering can be seen in the Agro Seed trial, with a less visible structure. Wild and landrace types are clustered together (on the right). There were marked phenotypic differences between the wild peas (climbing, low lodging resistance, indeterminate growth) and commercial pea varieties (dry pea, forage). However, there was significant variation in the traits themselves, which was confirmed by analysis of variance. arly in 2020. 3. Discussion In contrast, when the genotypes were analyzed by the population type, no apparent structure of phenotypic data was present because, within groups (such as breeding lines, varieties, landraces), there were variations resulting from different subtype representations (such as dry pea, vegetable pea, forage pea) within each of them. Based on the hierarchical clustering, there was evidently a difference between geno- types. This can be explained by the fact that both pea varieties and pea breeding lines can be classified into vegetable peas, dry peas for feed and food, and forage peas (grown primarily for forage) [42]. Landraces have distinctive characteristics arising from development and adaptation over time to conditions of a localized geographic region, and typically display greater genetic diversity than types subjected to formal breeding practices [43]. Higher correlations for the IFVCNS trials, compared to the Agro Seed trials, as cal- culated by the Mantel test, may partially be because the analyzed pea panel consisted of 64 accessions with a Serbian provenance. If they were adapted to conditions in Serbia more than other accessions, this may be reflected in the overall genetic makeup, meaning that these accessions would likely be genetically more similar due to adaptation to conditions in this specific region. In contrast, accessions from a broader range of geographic and ecologically distinct regions will be more genetically diverse. g y g g y The fact that landraces and semi-wild, and wild material were separated from most breeding lines and varieties (Figure 6, Left), and that forage types were separated from garden and dry peas (Figure 6, Right) confirms that there is a decrease in similarity between the genotypes, which can be explained by their different purposes. The 3D graphs (Figures 7 and 8) show that the patterns of variation in seed yield and protein content were remarkably similar at the two sites and relatively consistent between years. This was the case regardless of whether the accessories were grouped according to use or plant type; this confirms the adaptability of peas to different agroclimatic factors. 4.3. Meteorological Conditions The Novi Sad area is characterized by a moderate continental climate, with an average annual temperature of 11 ◦C and 122 sunny days per year [51]. The meteorological con- ditions in Novi Sad during the vegetation season of pea in 2019 and 2020 and long-term averages are presented in Table S1. The average temperature in 2019 was higher by 1.5 ◦C compared to the long-term average, while in 2020, it was higher by 1.3 ◦C. The precipitation in 2019 was higher by 5.7 mm, whereas the precipitation in 2020 was 14.3 mm higher than the long-term average, with uneven distribution of precipitation per month. The Kessenich area has a more continental climate compared to the rest of Belgium because it is less influenced by the Atlantic Ocean. Precipitation is frequent, but not particularly abundant. The annual average temperature is 10 ◦C [52]. The meteorological conditions in Kessenich during the vegetation season of pea in 2019 and 2020 and the long-term average are presented in Table S2. The average temperature in 2019 was higher by 1.06 ◦C compared to the long-term average, while in 2020, it was higher by 1.15 ◦C. The precipitation in both years was slightly higher than the long-term average (0.6 and 0.8 mm, respectively), with even distribution per month. 4.2. Experimental Design and Phenotyping 4.2. Experimental Design and Phenotyping The experiment was set up according to an experimental plan with partial repetitions similar to [44–47] in a row-column system with four experimental blocks. The trial was set up using augmented block design to minimize land and labor costs while still controlling for sources of variation. It has been shown that augmented designs are especially useful for assessing genotype effects practically and efficiently [17]. The size of an individual plot was 5 m2. The distance between the rows in the plot was 0.2 m so each plot had five rows, with a distance between the plots of 0.9 m. Of the total 165 genotypes, 146 were sown in one replicate, while 19 genotypes were presented in four replicates, so the total number of plots was 222. The usual pre-sowing preparation was done and the sowing depth was 2 cm, with a plant density of 80 plants/m2. After sowing, standard field practices were applied. Sowing was done in early March and the harvest was in late July in both years and at both trial sites. Seed yield components data were collected from ten randomly selected plants, avoid- ing marginal rows. The analyzed traits were grains per pod (GPP), pods per plant (PPP), seeds per plant (SPP), flowering duration (FD), plant height (PH), plot lodging (PL), pod length (PoL), thousand seed weight (TSW), seed weight per plant (SWPP), seed yield (SY), and protein content (PC). Plot lodging was determined by measuring the height of the plot in full flowering and again before harvest. The thousand seed weight were determined according to a method in which 100 grains from each subplot were weighed using an ana- lytical scale and the obtained result multiplied by 10 [48]. Protein content was determined using Fourier-transformed near infra-red spectroscopy (FT-NIRS) via an FT-NIR analyzer (Antaris Thermo Fisher Scientific, Waltham, MA, USA) [49,50]. 4. Materials and Methods 4.1. Plant Material The study was conducted on 165 pea genotypes (Table S4) during two vegetation seasons—2019 and 2020—In the experimental field of the Institute of Field and Vegetable Crops (IFVCNS), Novi Sad, Rimski šanˇcevi, Serbia (45◦20′ N, 19◦51′ E), and in the experi- mental field of the Agro Seed Research company, Kessenich, Belgium (51◦08′ N, 5◦48′ E). Studied genotypes encompassed 78 breeding lines, 59 varieties, 25 landraces and two wild and one semi-wild accession. The analyzed panel was set up to represent relevant European pea diversity and also included genotypes from Asia, Australia, and North America (USA). Genotypes used in the experiment were grouped based on variety by type into several groups: dry, forage, and vegetable. All material was kindly provided from germplasm collections from four institutions: Institute of Field and Vegetable Crops, Novi Sad, Serbia (IFVCNS), Agro Seed Research, Kessenich, Belgium (ASR), Institute of Biological, Environmental and Rural Sciences, Aberystwyth, United Kingdom (IBERS), and NorthGen Genetic Resource Center, Alnarp, Sweden (NorthGen). Plants 2022, 11, 1321 13 of 17 4.4. Statistical Analysis The spatial arrangement of plots in the field was used to assign row and column coor- dinates in each trial. For each trait, the best linear unbiased predictor (BLUP) was obtained for each genotype and year, using the spatial model as described by [53]. The variance components, estimated following [20], were used to assess the broad-sense heritability of each trait: (1) He2B = (σ2g/σ2p) × 100 (1) where: 2 σ2p = σ2e + σ2g: Phenotypic variance; 2 σ2g: Genotypic variance; g σ2e: Error variance. where: 2 σ2p = σ2e + σ2g: Phenotypic variance; 2 σ2g: Genotypic variance; 4.5. DNA Extraction and Genotyping DNA was extracted from 50 mg leaf sections of each accession using the QIAGEN DNAEasy 96 plant kit (QIAGEN, Manchester, UK). DNA samples were genotyped by NEOGEN EUROPE LTD (GENESEEK EUROPE), Auchincruive, Ayr, Scotland. The 13.2K GenoPea array (Illumina), was first described by [58]. The data were imported into the R program “argyle”, where initial assessment and quality control took place [59]. A total of 586 markers were removed due to low signal (1) and more than 20 no-calls in a sample (585). Based on the DNA extraction results, the hierarchical clustering, principal component analysis (PCA), and Mantel tests were performed using R (R Core Team 2020). For the Mantel Test, the R package “ade4” was used. The distance matrices were generated using the “Euclidian” method. The fan plot of the hierarchical clustering was done using the “ape” package. The PCA analysis was carried out using the prcomp command in base R [60]. g σ2e: Error variance. A principal component analysis (PCA) was conducted on normalized data using Minitab 17 software Trial version (Minitab Inc., Pennsylvania State University), taking into Plants 2022, 11, 1321 14 of 17 account sample plant type and variety type by usage. PCA results were graphically summa- rized in a biplot. Mixed model analysis, described by [54], was performed using Progeno 3.6.24 software (Progeno BV Company, Ghent University, Ghent, Belgium) [55]. Descriptive statistics and Pearson’s correlation coefficients between BLUPs for all quantitative traits were analyzed in XLStat according to [56], and analysis of variance was performed using Minitab 17 software Trial version (Minitab Inc., Pennsylvania State University, State College, PA, USA) [57]. 5. Conclusions Examination of phenotypic diversity of pea genotypes of different origins indicated considerable variation for a range of traits. The broad-sense heritability estimates were moderate to high for all examined traits, indicating the major influence of genetic factors. Inconsistent heritability between years and between sites for certain traits indicated their higher sensitivity to environmental conditions. Based on this work, it could be concluded that selection of traits with high and consistent heritability would be most valuable to breeders, including stable traits that effect seed weight per plant such as pod length and thousand seed weight. Moderate, consistent heritability with a long-lasting effect on seed weight per plant was observed for the number of seeds per plant, so direct selection of this trait could also be effective. Breeding for increased seed protein content is hampered by the negative correlation between protein content and yield. Therefore, increasing protein production can be done by increasing the area under protein crops or by improving protein quality over protein content, given that difference in protein content can be the result of many different environmental factors. The seed yield is a less challenging breeding target than protein content; therefore, high yields have been achieved among some of the cultivated accessions. Given the lack of the strong influence of GxE, confirmed by positive correlations between two years, this gene pool provides numerous possibilities as a starting point for future pea varieties adaptable to different agroclimatic conditions. Traits such as the number of grains per pod, plant height, pod length, thousand seed weight, seed yield, and protein content showed highly significant variations among the tested groups of genotypes. Knowledge of these variations could be used in further plant selection programs. p g Given the proven and evident differences between the analyzed genotypes based on type and use, and their great adaptability to tested conditions, it can be concluded that pea is a very adaptable plant species with untapped production potential. The results of this study should contribute to a better knowledge of variability and seed yield stability of pea genotypes used in Europe for future production and breeding. Obtained phenotypic results could improve pea breeding by developing new cultivars carrying favorable traits to attain more sustainable production and higher yields. References 1. Faostat Statistics. Crop and Livestock Products-Statistics on Area Harvested of Dry Pea in European Region. Available online: https://www.fao.org/faostat/en/#data/QCL (accessed on 1 March 2022). 1. Faostat Statistics. Crop and Livestock Products-Statistics on Area Harvested of Dry Pea in European Region. Available online: https://www.fao.org/faostat/en/#data/QCL (accessed on 1 March 2022). p g 2. Ismail, B.P.; Senaratne-Lenagala, L.; Stube, A.; Brackenridge, A. Protein demand: Review of plant and animal proteins used in alternative protein product development and production. Anim. Front. 2020, 10, 53–63. [CrossRef] [PubMed] 2. Ismail, B.P.; Senaratne-Lenagala, L.; Stube, A.; Brackenridge, A. Protein demand: Review of plant and animal proteins used in alternative protein product development and production. Anim. Front. 2020, 10, 53–63. [CrossRef] [PubMed] p p p p 3. Westhoek, H.; Rood, T.; Van Den Berg, M.; Janse, J.; Nijdam, D.; Reudink, M.; Stehfest, E.; Lesschen, J.P.; Oenema, O.; Woltjer, G.B. The Protein Puzzle: The Consumption and Production of Meat, Dairy and Fish in the European; PBL Netherlands Environmental Assessment Agency: Den Haag, The Netherlands, 2011; pp. 123–144. g y g pp 4. Lindsay, D.L.; Jha, A.B.; Arganosa, G.; Glahn, R.; Warkentin, T.D. Agronomic Performance in Low Phytic Acid Field Peas. Plants 2021, 10, 1589. [CrossRef] [PubMed] 5. Jha, A.B.; Warkentin, T.D. Biofortification of Pulse Crops: Status and Future Perspectives. Plants 2020, 9, 73. [CrossRef] [PubMed] 6. Harmankaya, M.; Özcan, M.M.; Karada¸S, S.; Ceyhan, E. Protein and mineral contents of pea (Pisum sativum L.) genotypes grown in central Anatolian region of Turkey. South West. J. Hortic. Biol. Environ. 2010, 1, 159–165. g y 7. Lam, A.C.Y.; Can Karaca, A.; Tyler, R.T.; Nickerson, M.T. Pea protein isolates: Structure, extraction, and functionality. Food Rev. Int. 2018, 34, 126–147. [CrossRef] 7. Lam, A.C.Y.; Can Karaca, A.; Tyler, R.T.; Nickerson, M.T. Pea protein isolates: Structure, extraction, and functionality. Food Rev. Int. 2018, 34, 126–147. [CrossRef] 8. Živanov, D.; Savi´c, A.; Katanski, S.; Karagi´c, Ð.; Miloševi´c, B.; Mili´c, D.; Ðor ¯devi´c, V.; Vuji´c, S.; Krsti´c, Ð.; ´Cupina, B. Intercropping of field pea with annual legumes for increasing grain yield production. Zemdirb. Agric. 2018, 105, 235–242. [CrossRef] 9 H E Miti ti f S b All b D l t f L All C t t S d ACS S S 2008 1001 431 445 , g y gy y p g y p , , 10. Acar, R.; Özel, A.; Sumiahadi, A.; Koç, N. The Importance of Legume Forage Crops As Green Manure. 5. Conclusions In conclusion, this work should promote the broader use of pea as a grain legume within Plants 2022, 11, 1321 15 of 17 15 of 17 diverse agricultural systems to provide multiple beneficial advantages, in line with the principles of sustainability. diverse agricultural systems to provide multiple beneficial advantages, in line with the principles of sustainability. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/plants11101321/s1, Table S1: Meteorological conditions in Novi Sad, Serbia during the vegetation seasons of pea in 2019 and 2020, and long-term average. Table S2: Meteorological conditions in Kessenich, Belgium during the vegetation seasons of pea in 2019 and 2020, and long-term average. Table S3: Pearson’s correlation coefficient ®for analyzed traits between two years (2019 and 2020). Table S4: The list of pea genotypes analyzed in the study. Author Contributions: Conceptualization and methodology, M. ´C. and D.Ž.; software, A.U., M. ´C., and L.S.; validation, R.G.; formal analysis, E.S.-C., D.L., and L.S.; investigation, A.U., D.Ž., and M. ´C.; writing—original draft preparation, A.U.; writing—review and editing, M. ´C., R.G., and L.S.; supervi- sion, M. ´C. and D.Ž. All authors have read and agreed to the published version of the manuscript. Author Contributions: Conceptualization and methodology, M. ´C. and D.Ž.; software, A.U., M. ´C., and L.S.; validation, R.G.; formal analysis, E.S.-C., D.L., and L.S.; investigation, A.U., D.Ž., and M. ´C.; writing—original draft preparation, A.U.; writing—review and editing, M. ´C., R.G., and L.S.; supervi- sion, M. ´C. and D.Ž. All authors have read and agreed to the published version of the manuscript. Funding: This project has received funding from the European Union’s Horizon 2020 Program for Research and Innovation under grant agreement no. 727312 (project EUCLEG) and the Ministry of Education, Science and Technological Development of the Republic of Serbia, grant number: 451-03-68/2022-14/200032. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to the ongoing project. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to the ongoing project. 5. Conclusions Acknowledgments: We thank the seed companies, institutes, and university that provided us with seeds of their accessions: Institute of Field and Vegetable Crops, Novi Sad, Serbia (IFVCNS), Agro Seed Research, Kessenich, Belgium (ASR), Institute of Biological, Environmental and Rural Sci- ences, Aberystwyth, United Kingdom (IBERS), and NorthGen Genetic Resource Center, Alnarp, Sweden (NorthGen). Conflicts of Interest: The authors declare no conflict of interest. References In-depth characterization of common bean diversity discloses its breeding potential for sustainable agriculture. Sustainability 2019, 11, 5443. [CrossRef] g tance and Association of Quantitative Characteristics in Syraian Landraces of Garden peas (Pisum sativum L.) 2, 198–203. [CrossRef] 23. Al-Aysh, F.M. Inheritance and Association of Quantitative Characteristics in Syraian Landraces of Garde Int. J. Life Sci. 2013, 2, 198–203. [CrossRef] 24. Kumar, B.; Kumar, A.; Kumar Singh, A.; Lavanya, G.R. Selection strategy for seed yield and maturity in field pea (Pisum sativum L. arvense). Afr. J. Agric. Res. 2013, 8, 5411–5415. ) f J g 25. Candráková, E.; Polláková, N.; Macák, M. The influence of agro-ecological conditions on common pea yield. Res. J. Agric. Sci. 2011, 43, 25–30. 26. Vasileva, V.; Kosev, V. Evaluation of nodule related components and forage productivity in Pea (Pisum sativum L.) genotypes. Int. J. Pharm. Life Sci. 2015, 6, 4230–4237. 27. Naeem, M.; Awais Ghani, M.; Muhammad Atif, R.; Amjad, M.; Ahmad, T.; Batool, A.; Muzammil Jahangir, M.; Ali, B.; Mehran Abbas, M.; Nadeem Akram, M. Estimation of correlation and path coefficient for morphological and quality related traits in pea (Pisum sativum L.). Bangladesh J. Bot. 2020, 49, 549–555. [CrossRef] g 28. Singh, B.K.; Sutradhar, M.; Singh, A.K.; Singh, S.K. Evaluation of genetic variability, correlation and path coefficients analysis for yield attributing traits in field pea [Pisum sativum (L.) var. arvense]. Res. Crops. 2017, 18, 316–321. [CrossRef] 29. Gautam, K.K.; Syamal, M.M.; Singh, A.K.; Gupta, N. Variability, character association and path coefficient analysis of green pod yield and its related traits in pea (Pisum sativum L.). Legume Res. 2017, 40, 818–823. [CrossRef] 30. Kumar, T.N.V.; Alloli, T.B.; Hadimani, H.P.; Ajjappalavar, P.S.; Satish, D.; Abdul, K.; Hanchinamani, C.N. Studies on Correlation and Path Coefficient Analysis in Garden Pea (Pisum sativum L.) varieties. Int. J. Curr. Microbiol. Appl. Sci. 2019, 8, 3024–3031. [CrossRef] 31. Mohanty, A.T.; Verma, A.; Pandey, B. Evaluation of Correlation and Path Coefficients Analysis for Yield Attributing Traits in Garden Pea (Pisum sativum L.) under Tarai tract of Uttarakhand. Int. J. Curr. Microbiol. Appl. Sci. 2020, 11, 1315–1322. , A.T.; Verma, A.; Pandey, B. Evaluation of Correlation and Path Coefficients Analysis for Yield Attributin ea (Pisum sativum L.) under Tarai tract of Uttarakhand. Int. J. Curr. Microbiol. Appl. Sci. 2020, 11, 1315–1322. 32. Rasaei, A.; Ghobadi, M.E.; Ghobadi, M.; Abdiniya, K. References In Proceedings of the International Soil Congress 2019, Successful Transformation toward Land Degradation Neutrality: Future Perspective, Ankara, Turkey, 17–19 June 2019. 11. Stone, J.L. Peas as a field crop. Small Farmer’s J. 2018, 42. 12. Krsti´c, Ð.; Vuji´c, S.; Ja´cimovi´c, G.; D’Ottavio, P.; Radanovi´c, Z.; Eri´c, P.; ´Cupina, B. The Effect of Cover Crops on Soil Water Balance in Rain-Fed Conditions. Atmosphere 2018, 9, 492. [CrossRef] 13. Tan, M.; Koc, A.; Dumlu Gul, Z. Morphological characteristics and seed yield of east Anatolian local forage pea (Pisum sativum ssp. arvense L.) ecotypes. Turk. J. Field Crops 2012, 17, 24–30. Plants 2022, 11, 1321 16 of 17 14. Tolessa, T.T.; Keneni, G.; Sefera, T.; Jarso, M.; Bekele, Y. Genotype 9 environment interaction and performance stability for grain yield in field pea (Pisum sativum L.) genotypes. Int. J. Plant Breed. 2013, 7, 116–123. y p g yp 15. Burstin, J.; Salloignon, P.; Chabert-Martinello, M.; Jean-Bernard Magnin-Robert, J.; Siol, M.; Jacquin, F.; Chauveau, A.; Pont, C.; Gregoire Aubert, G.; Delaitre, C.; et al. Genetic diversity and trait genomic prediction in a pea diversity panel. BMC Genom. 2015, 16, 105. [CrossRef] [PubMed] [ ] [ ] 16. Bocianowski, J.; Ksiezak, J.; Nowosad, K. Genotype by environment interaction for seeds yield in pea (Pisum sativum L.) using additive main effects and multiplicative interaction model. Euphytica 2019, 215, 191. [CrossRef] p p y 17. Federer, W.T.; Crossa, J. Screening experimental designs for quantitative trait loci, association mapping, genotype-by environment interaction, and other investigations. Front. Physiol. 2012, 3, 1–8. [CrossRef] [PubMed] 18. Reckling, M.; Döring, F.T.; Bergkvist, G.; Stoddard, L.F.; Watson, A.C.; Seddig, S.; Chmielewski, M.F.; Bachinger, J. Grain legume yields are as stable as other spring crops in long-term experiments across northern Europe. Agron. Sustain. Dev. 2018, 38, 191. [CrossRef] [ ] 19. Khan, R.A.; Mahmud, F.; Reza, M.A.; Mahbub, M.; Shirazy, J.B.; Rahman, M. Genetic diversity, correlation and path analysis for yield and yield components of pea (Pisum sativum L.). World J. Agric. Sci. 2017, 13, 11–16. y y p p g 20. Georgieva, N.; Nikolova, I.; Kosev, V. Association study of yield and its components in pea (Pisum sativum L.). Int. J. Pharmacogn. IJP 2015, 2, 536–542. 21. Aboagye, L.M. Strategies for sustainable conservation and use of legume genetic resources in Ghana. Ghana J. Sci. 2007, 47, 59–67. [CrossRef] 22. Caproni, L.; Raggi, L.; Ceccarelli, S.; Negri, V.; Carboni, A. References The study of traits correlation and path analysis of the grain yield of the peas in semi-dry conditions in Kermanshah. In Proceedings of the International Conference on Food Engineering and Biotechnology, Singapore, 7–9 May 2011. g p y 33. Goa, Y.; Ashamo, M. Evaluation of field pea (Pisum sativum L.) genotypes performance for yield and yield components at five growing environments of Southern Ethiopia. Curr. Res. Agric. Sci. 2014, 1, 65–76. 34. Togay, N.; Togay, Y.; Yildirim, B.; Dogan, Y. Relationships between yield and some yield components in Pea (Pisum sativum ssp arvense L.) genotypes by using correlation and path analysis. Afr. J. Biotechnol. 2008, 7, 4285–4287. 35. Jaiswal, K.N.; Lavanya, G.R. Correlation coefficient analysis in field pea (Pisum sativum L.). Int. J. Agric. Sci. Res. (IJASR) 2014, 4, 211–214. A.; Houtin, H.; Rond-Coissieux, C.; Naudet-Huart, M.; Touratier, M.; Marget, P.; Burstin, J. Meta-analysis o enetic control of yield-related traits and seed protein content in pea. Sci. Rep. 2020, 10, 15925. [CrossRef] [Pu 37. Krajewski, P.; Bocianowski, J.; Gawłowska, M.; Kaczmarek, Z.; Pniewski, T.; Swiecicki, W.; Wolko, B. QTL for yield components and protein content: A multienvironment study of two pea (Pisum sativum L.) populations. Euphytica 2012, 183, 323–336. [CrossRef] 38. Tsenov, N.; Kostov, K.; Gubatov, T.; Peeva, V. Study on the genotype × environment interaction in winte quality. Field Crops Stud. 2004, 1, 20–29. K.; Gubatov, T.; Peeva, V. Study on the genotype × environment interaction in winter wheat varieties I. Grain tud. 2004, 1, 20–29. q y p 39. Siol, M.; Jacquin, F.; Chabert-Martinello, M.; Smýkal, P.; Le Paslier, M.; Aubert, G.; Burstin, J. Patterns of genetic structure and linkage disequilibrium in a large collection of pea germplasm. G3 Genes Genomes Genet. 2017, 7, 2461–2471. [CrossRef] 17 of 17 17 of 17 Plants 2022, 11, 1321 40. USA Pulses. Processing Information and Technical manual, Dry Pea and Forage Pea. Available online: https://www.usapulses. org/ (accessed on 4 March 2022). 41. Abd El-Salam, M.H.; El-Shibiny, S. Natural biopolymers as nanocarriers for bioactive ingredients used in food industries. Encapsulations Nanotechnol. Agri-Food Ind. 2016, 2, 793–829. p g 42. Feedipedia, Pea Forage. Tables of Chemical Composition and Nutritional Value. Available online: https://feedipedia.org/ (accessed on 28 November 2021). 43. “Landrace.” Merriam-Webster.com Dictionary, Merriam-Webster. Available online: https://www.merriam-webster.com/ dictionary/landrace (accessed on 1 March 2022). 44. Cullis, B.R.; Smith, A.B.; Coombes, N.E. On the design of early generation variety trials with correlated data. J. Agric. Biol. References Environ. Stat. 2006, 11, 381–393. [CrossRef] bias-Pazaran, G.E. 3. Methods to measure heritability. In Optimizing Breeding Schemes Manual, Heritability: M ion; Wimmer, V., Ziemke, E., Martini, J., Storr, S., Eds.; Exellence in Breeding Platform: Montpellier, France; B 45. Covarrubias-Pazaran, G.E. 3. Methods to measure heritability. In Optimizing Breeding Schemes Manual, Heritability: Meaning and Computation; Wimmer, V., Ziemke, E., Martini, J., Storr, S., Eds.; Exellence in Breeding Platform: Montpellier, France; Bil&Melida Gates Platform: London, UK; International Maize and Wheat Improvement Center: Veracruz, Mexico, 2019; pp. 8–16. Computation; Wimmer, V., Ziemke, E., Martini, J., Storr, S., Eds.; Exellence in Breeding Platform: Montpellier, France; Bil&Melida Gates Platform: London, UK; International Maize and Wheat Improvement Center: Veracruz, Mexico, 2019; pp. 8–16. p g p Gates Platform: London, UK; International Maize and Wheat Improvement Center: Veracruz, Mexico, 2019; pp. 8–16. p pp 46. Lamichaney, A.; Parihar, A.K.; Hazra, K.K.; Dixit, G.P.; Katiyar, P.K.; Singh, D.; Singh, A.K.; Kumar, N.; Singh, N.P. Untangling the Influence of Heat Stress on Crop Phenology, Seed Set, Seed Weight, and Germination in Field Pea (Pisum sativum L.). Front. Plant. Sci. 2021, 12, 635868. [CrossRef] 47. Pandit, E.; Tasleem, S.; Barik, R.S.; Mohanty, D.P.; Nayak, D.K.; Mohanty, P.S.; Das, S.; Pradhan, S.K. Genome-Wide Association Mapping Reveals Multiple QTLs Governing Tolerance Response for Seedling Stage Chilling Stress in Indica Rice. Front. Plant Sci. 2017, 8, 552. [CrossRef] 48. Agrawal, P.K.; Singh, D.P. 8.Determination of Weight. Available online: https://seednet.gov.in/PDFFILES/Chapter%208.pdf (accessed on 4 March 2021). .; Zangirolami, S.M.; Silva, O.N.; Valderrama, P.; Março, P.H. Rapid non-invasive assessment of quality para 49. Santos, R.d.L.; Zangirolami, S.M.; Silva, O.N.; Valderrama, P.; Março, P.H. Rapid non-invasive assessment of quality parameters in ground soybean using near-infrared spectroscopy. Pesqui. Agropecuária Bras. 2018, 53, 97–104. [CrossRef] 49. Santos, R.d.L.; Zangirolami, S.M.; Silva, O.N.; Valderrama, P.; Março, P.H. Rapid non-invasive assessment of quality parameters in ground soybean using near-infrared spectroscopy. Pesqui. Agropecuária Bras. 2018, 53, 97–104. [CrossRef] 50. Osborne, B.G. Near-infrared Spectroscopy in Food Analysis. Encyclopedia of Analytical Chemistry; Online John Wiley and Sons Ltd.: H b k NJ USA 2006 in ground soybean using near-infrared spectroscopy. Pesqui. Agropecuária Bras. 2018, 53, 97–104. [CrossRef] 50. Osborne, B.G. Near-infrared Spectroscopy in Food Analysis. Encyclopedia of Analytical Chemistry; Online John Wiley and Sons Ltd.: Hoboken, NJ, USA, 2006. 51. Republic Hydrometeorological Institute of Serbia. Available online: http://www.hidmet.gov.rs/index_eng.php 52. Belgian Climate, Kessenich. Available online: www.climatestotravel.com (accessed on 22 January 2022 53. References Singh, M.; Ceccarelli, S.; Grando, S. Precision of the genotypic correlation estimated from variety trials block designs. TAG Theor. Appl. Genet. 1997, 95, 1044–1048. [CrossRef] g pp 54. Demidenko, E.; Mandel, I. Yield Analysis and Mixed Model. ASA Sect. Nonparametric Stat. 2005, 1, 1636–1643. 54. Demidenko, E.; Mandel, I. Yield Analysis and Mixed Model. ASA Sect. Nonparametric Stat. 2005, 1, 1636–1643. 55. Maenhout, S. Progeno; Ghent University: Gent, Belgium, 2018. Available online: https://progeno.net/ (accessed on g y g p p g 56. Steel, R.C.D.; Torrie, J.H. Principles and Procedures of Statistics; McGraw-Hill Book Company, Inc.: New York, NY, USA, 1981; pp. 34–74. 57. Minitab 17 Statistical Software [Computer Software]; Minitab, Inc.: State College, PA, USA, 2010. Available online: www.minitab.com (accessed on 22 November 2021). 58. Tayeh, N.; Aluome, C.; Falque, M.; Jacquin, F.; Klein, A.; Chauveau, A.; Bérard, A.; Houtin, H.; Rond, C.; Kreplak, J.; et al. Development of two major resources for pea genomics: The GenoPea 13.2K SNP Array and a high-density, high-resolution consensus genetic map. Plant J. 2015, 84, 1257–1273. [CrossRef] [PubMed] g p 59. Morgan, A.P. argyle: An R Package for Analysis of Illumina Genotyping Arrays. G3: Genes Genomes Genet. 2016, 6, 281–286. [CrossRef] [PubMed] 60. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2020. Available online: https://www.R-project.org (accessed on 8 March 2022).

https://openalex.org/W2159649486

https://hal-univ-paris.archives-ouvertes.fr/hal-03664884/file/eji3075.pdf

English

Acute myeloid leukemia impairs natural killer cells through the formation of a deficient cytotoxic immunological synapse

European Journal of Immunology

cc-by

Acute myeloid leukemia impairs natural killer cells through the formation of a deficient cytotoxic immunological synapse Zena Khaznadar, Guylaine Henry, Niclas Setterblad, Sophie Agaugue, Emmanuel Raffoux, Nicolas Boissel, Hervé Dombret, Antoine Toubert, Nicolas Dulphy Zena Khaznadar, Guylaine Henry, Niclas Setterblad, Sophie Agaugue, Emmanuel Raffoux, Nicolas Boissel, Hervé Dombret, Antoine Toubert, Nicolas Dulphy To cite this version: Zena Khaznadar, Guylaine Henry, Niclas Setterblad, Sophie Agaugue, Emmanuel Raffoux, et al.. Acute myeloid leukemia impairs natural killer cells through the formation of a deficient cy- totoxic immunological synapse. European Journal of Immunology, 2014, 43 (5), pp.1383-1388. ￿10.1002/eji.201444500￿. ￿hal-03664884￿ Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-03664884 https://u-paris.hal.science/hal-03664884v1 Submitted on 2 Jan 2023 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Acute myeloid leukemia impairs natural killer cells through the formation of a deficient cytotoxic immunological synapse Zena Khaznadar,1,2 Guylaine Henry,3 Niclas Setterblad,4 Sophie Agaugue,1 Emmanuel Raffoux,5,6 Nicolas Boissel,5,6 Hervé Dombret,5,6 Antoine Toubert,1-3 and Nicolas Dulphy1-3 1Institut National de la Santé et de la Recherche Médicale (INSERM UMRS 1160), Paris, France 2Univ Paris Diderot, Sorbonne Paris Cité, Institut Universitaire d'Hématologie, Paris, France 3Assistance Publique–Hôpitaux de Paris (AP–HP), Hôpital Saint-Louis, Laboratoire d’Immunologie et Histocompatibilité, Paris, France 4Univ Paris Diderot, Sorbonne Paris Cité, Institut Universitaire d'Hématologie, Département d’Imagerie, Paris, France 5AP–HP, Hôpital Saint-Louis, Laboratoire d’Hématologie Adulte, Paris, France 6Univ Paris Diderot, Sorbonne Paris Cité, EA3518, Paris, France 4Univ Paris Diderot, Sorbonne Paris Cité, Institut Universitaire d'Hématologie, Département d’Imagerie, Paris, France 5AP–HP, Hôpital Saint-Louis, Laboratoire d’Hématologie Adulte, Paris, France 6Univ Paris Diderot, Sorbonne Paris Cité, EA3518, Paris, France 5AP–HP, Hôpital Saint-Louis, Laboratoire d’Hématologie Adulte, Paris, France 6Univ Paris Diderot, Sorbonne Paris Cité, EA3518, Paris, France Key words: NK cells; Immunological synapse; Acute myeloid leukemia; Cancer immunity Corresponding author: Dr. Nicolas Dulphy, Hôpital Saint-Louis, 1, avenue Claude Vellefaux, 75475 Paris Cedex 10, France. Tel: +33-142494894. Fax: +33-142494641. [email protected] Abbreviations: AML acute myeloid leukemia. NCR natural cytotoxicity receptors. Received: 24-Jan-2014; Revised: 30-Jun-2014; Accepted: 10-Jul-2014 This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/eji.201444500. NKIS NK immunological synapse. ROI region of interest. RRI relative recruitment index This article is protected by copyright. All rights reserved. 1 1 This article is protected by copyright. All rights reserved. Abstract: Acute myeloid leukemia (AML) cells are killed by allogeneic NK cells. However, autologous NK cells from AML patients express decreased levels of activating receptors, and show reduced cytotoxicity. Here, we investigated how interactions between NK and AML cells might cause loss of NK cell activity in patients. Our results show that AML cell lines and primary blasts alter the NK cell phenotype, reducing their cytotoxic potential upon prolonged contact. Down-regulation of NK-cell-activating receptors was contact-dependent and correlated with conjugate formation. Time-lapse imaging of HL60 AML cell line and NK- cell interactions showed a high proportion of non-cytolytic contacts. Studies of NK-cell immunological synapses revealed a defect in lytic synapse formation. Namely, despite correct F-actin and LFA-1 recruitment, polarization of lytic granules toward primary blasts or AML cell lines was reduced. The NK-AML cell line synapses showed impairment of CD3ζ recruitment. Attempts to correct these synapse defects by cytokine stimulation of NK cells improved conjugate formation, but not granule polarization. Pre-treatment of AML cell lines with the immuno-modulating molecule lenalidomide significantly enhanced granule polarization. We speculate that combining immunomodulatory drugs and cytokines could increase AML cell sensitivity to autologous NK cells and reinforce the activity of allogeneic NK cells in adoptive immunotherapy. 2 2 This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Introduction: Natural killer (NK) cells are part of the innate immune system. They are regulated by a balance between signals from activating and inhibitory receptors [1]. NK cells mount highly effective anticancer immune responses, and thus represent major candidates for cancer immunotherapy [2]. Their ability to kill acute myeloid leukemia (AML) cells has been demonstrated by the absence of relapse after transplantation of allogeneic hematopoietic stem cells containing alloreactive NK cells [3]. Infusion of purified donor NK lymphocytes to consolidate engraftment after haplo-identical transplantation led to complete remission in some AML patients [4, 5]. Thus, NK cells are a potential tool for immunotherapy against AML [6, 7]. AML cells express ligands interacting with NK-cell activating receptors, making them susceptible to killing by NK cells [8, 9]. However, autologous NK cells do not effectively control AML growth in patients. Some mechanisms have been suggested for AML escape, including down-regulation of the ligands for NK-cell activating receptors, secretion of soluble forms of these ligands, or up-regulation of ligands for NK-cell inhibitory receptors [10]. Reduced toxicity has also been linked to down-regulation of activating receptors on NK cells, particularly NKG2D and the natural cytotoxicity receptors (NCR) NKp46 and NKp30 [11–13]. Upon remission, the NK-cell phenotype and function normalize, suggesting that the presence of AML blasts is responsible for the defects observed [12]. Interestingly, once out of the AML environment, NK cells that had previously expressed low levels of activating receptors regain their activity, becoming cytotoxic toward autologous leukemic blasts in mice [13]. This further implicates contact with AML cells in NK-cell activity down-modulation. 3 3 Interaction between NK cells and a cancerous cell involves a series of steps resulting in the formation of an immunological synapse (NKIS) [14]. Upon first contact, the NK-cell’s cytoskeleton is reorganized in the zone of contact with the target cell [15]. Lytic granules are then polarized and their content released into the intercellular space, causing target cell death [16]. In this article, we hypothesized that AML-NK-cell interactions might alter the NK-cell phenotype, and reduce their cytotoxic potential upon prolonged contact. We therefore studied the results of interaction with AML cell lines or primary blasts on NK cells using in vitro functional assays complemented by imaging experiments focusing on the lytic NKIS. Interaction between NK cells and a cancerous cell involves a series of steps resulting in the formation of an immunological synapse (NKIS) [14]. Introduction: Upon first contact, the NK-cell’s Interaction between NK cells and a cancerous cell involves a series of steps resulting in the formation of an immunological synapse (NKIS) [14]. Upon first contact, the NK-cell’s cytoskeleton is reorganized in the zone of contact with the target cell [15]. Lytic granules are then polarized and their content released into the intercellular space, causing target cell death [16]. In this article, we hypothesized that AML-NK-cell interactions might alter the NK-cell phenotype, and reduce their cytotoxic potential upon prolonged contact. We therefore studied the results of interaction with AML cell lines or primary blasts on NK cells using in vitro functional assays complemented by imaging experiments focusing on the lytic NKIS. 4 4 This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. AML cells have a heterogeneous capacity to induce NK-cell degranulation Only expression of the co-stimulatory pair LFA-1/ICAM-1 (CD11a/CD54) differed between THP1 and HL60 cells, with THP1 cells expressing significantly higher levels of CD54 and HL60 cells expressing CD11a, as expected for an acute promyelocytic leukemia cell line (Fig. 1C) [22]. This article is protected by copyright. All rights reserved. AML cells have a heterogeneous capacity to induce NK-cell degranulation NK-cell activation is mainly induced by interaction with ligands present on a target cell, such as leukemic blasts from AML patients. These blasts showed very variable ligand expression levels between patients (Fig. 1A), although all tested blasts expressed at least one ligand for NK-cell activating receptors. When allogeneic NK-cell degranulation was tested against leukemic cells from 24 AML patients, the level of degranulation induced was variable between patients, and did not correlate with the expression level of the ligands (Fig. 1B and data not shown). This suggested that the presence of activating receptors ligands is necessary but the relationship between their level of expression and capacity to activate NK-cell cytolytic function is not quantitative. To reflect this variability in controlled experiments, we used two cell lines - THP1 and HL60 - expressing different NK-cell receptor ligands (Fig. 1A). Similar to the heterogeneity seen in NK-cell response to primary AML cells, NK cells from healthy donors degranulated more extensively when challenged with THP1 than with HL60 cells (Fig. 1B). No major differences were found to correlate with NK-cell activation for the ligands of NKG2D (MICA/B, ULBP1-3) or DNAM-1 (CD155 “PVR”), which were differentially expressed between the two cell lines. Although the expression patterns for these molecules was slightly different, the total ligand expression measured at the cell surface, using Fc recombinant proteins for each receptor, showed that this was not significant (Fig. 1A, 1C) [17]. Ligands of NKp30 and NKp46 were below the detection threshold with this method; while expression levels for HLA class I molecules, ligands for NK-cell inhibitory receptors, were identical on both cell lines (Fig. 1A). The THP1 and HL60 cells profiles were completed by measuring the expression levels of several AML phenotypic markers (CD33, CD34, CD45, CD117, 5 5 CD11b, CD11c) [18] and some suggested inhibitory molecules (CD137L, CD200, CD273, CD274, and CD276) [19–21]. This revealed no major differences (Supporting Information Fig. 1). Only expression of the co-stimulatory pair LFA-1/ICAM-1 (CD11a/CD54) differed between THP1 and HL60 cells, with THP1 cells expressing significantly higher levels of CD54 and HL60 cells expressing CD11a, as expected for an acute promyelocytic leukemia cell line (Fig. 1C) [22]. CD11b, CD11c) [18] and some suggested inhibitory molecules (CD137L, CD200, CD273, CD274, and CD276) [19–21]. This revealed no major differences (Supporting Information Fig. 1). Contact with AML cells alters NK-cell phenotype and function With the hypothesis that AML cells could promote NK-cell deficiencies observed in patients [10–13], we proceeded to test whether these defects could be reproduced in vitro by co- culture of healthy donor NK cells with AML cell lines. First, we compared expression of activating receptors (NKG2D, DNAM-1 and NKp46) after cultivating PBMCs alone or with THP1 or HL60 cells for 48 h. DNAM-1 expression decreased with both cell lines, whereas NKG2D expression was reduced by 45% with THP1 cells only. Interestingly, for NKp46 we noted a trend with THP1 cells (p-value 0.05) but the decrease was not significant with HL60 cells (Fig. 2A). A greater fold-decrease was observed with higher AML/PBMC ratios (data not shown). Co-culture experiments performed as control with the non-myeloid, lymphoma cell line RAJI showed no effect on DNAM-1 or NKG2D expression on NK cells and a down- regulation of NKp46 (Fig. 2A). We next assessed NK-cell degranulation capacity after co-culture with AML cell lines. To do this, CD107a-mobilization assays with a HLA class-I-negative target (K562) were performed (Fig. 2B). Exposure to HL60 cells very slightly reduced NK-cell degranulation, whereas contact with THP1 cells severely decreased degranulation (16.8% vs. 43.4% for PBMC cultured alone, p=0.03). 6 When performing degranulation tests with primary AML blasts (Fig. 1B) the NKG2D level on NK cells decreased, and the degree of degranulation correlated with the extent of this down-regulation (Fig. 2C). Interestingly, NKG2D expression was reduced on CD107a- NK cells, suggesting that down-regulation of this receptor requires contact, but not necessarily degranulation. Taken together, these experiments demonstrated that AML cells can dampen the expression of activating receptors while also reducing the cytolytic activity of fresh allogeneic NK cells. In line with differences in conjugate formation assays - where NK cells formed twice as conjugates with THP1 as with HL60 cells (Fig. 2D) - the two cell lines differed in their effects on NK cells overall. THP1 cells generally had a more severe impact than HL60 cells. These data thus suggest that NK-cell inhibition could be the result of the number of NK/AML cells interactions. t n W c w s v r c c w We therefore tested whether contact with primary AML blasts had the same adverse effect as cell lines on healthy NK cells. PBMCs from three unrelated healthy donors were cultivated with AML blasts isolated from four untreated patients. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Contact with AML cells alters NK-cell phenotype and function The NK-cell phenotype after 48 h showed a slight decrease in NKp46 in all cases. AML cells from patients showed the same variability in their effect on NK cells with some patients (Pt3 and Pt4) causing a significant reduction in DNAM-1 (34%) and NKG2D (16%) expression (Fig. 2E). When tested for conjugate formation, blasts from these same patients formed a higher percentage of conjugates with NK cells than blasts from the other two patients (Fig. 2E). After co-culture with patient blasts, higher expression of both DNAM1 and NKG2D on NK cells correlated negatively with the number of conjugates formed by those cells (Fig. 2F). Expression levels of NKG2D and DNAM-1 ligands on blasts from the 4 patients did not relate to the receptors down-regulation. And phenotyping for the inhibitory molecules did not show a difference between the different AML cells (Supporting Information Table 1). 7 7 Soluble factors secreted by AML cells have been shown to inhibit NK-cell proliferation and are thought to be responsible for NK-cell receptor down-regulation in patients [12, 23, 24]. However, in our hands, exposing PBMCs to cell lines or primary blasts supernatants had no effect on NK-cell phenotype (data not shown). This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Low killing rate upon NK cell-AML cell line interaction Since contact was needed to induce changes in NK cells, we investigated whether AML cell lines could affect the dynamics and outcome of NK-cell interaction [25]. Target cells (K562 or HL60) were seeded into a multi-chamber culture dish in medium supplemented with a nucleic-acid stain. Due to their high motility, THP1 cells were not suitable for analysis in these experiments as they would be difficult to track on successive images. NK cells were pre-activated with IL-15, labeled and added to the chamber (NK/AML ratio 2:1), and interactions with target cells were scored. HL60 cells were engaged in a lower number of interactions with NK cells than K562 cells, with 72% of tracked HL60 cells taking part in interactions with NK cells versus 93% of K562 cells (Fig. 3A). This was consistent with the flow cytometry results for conjugate formation (Fig. 3B). To assess the efficacy of NK-cell killing, we measured the Conjugate-Kill time, i.e., the time between conjugate formation and fluorescence appearing in target cells as explained in the Materials and methods section (Fig. 3C, and Supporting Information Video 1). NK cells required an average of 5 min 6 sec to kill K562 cells, and 4 min 54 sec to kill HL60 cells (Fig. 3D). The difference was not significant. Some target cells were more resistant, requiring multiple lytic hits before dying. Analysis of the population dying only after contact with two NK cells (Supporting Information Fig. 2A) also showed no difference in time to death based 8 on the identity of the target: average times were 12 min for K562 and 11 min 42 sec for HL60 for the first contact, and 2 min 54 sec and 3 min 42 sec, respectively, for the second contact (Fig. 3E, and Supporting Information Video 2). Thus, the time needed for NK cells to kill target cells was no different for AML cells than for the “gold standard” of killing efficiency, K562 cells. Ineffective interactions were defined as contacts that detached after more than 5 min or lasted up to 30 min without inducing apoptotic changes in the target cell (Supporting Information Fig. 2B, and Supporting Information Video3). NK interactions with HL60 were considerably less effective, with 26% of HL60 cells resisting lysis, than interactions with K562 cells, where only 15% of cells were not killed after interaction (Fig. 3F, 3G). Low killing rate upon NK cell-AML cell line interaction Thus, the overall outcome of NK cells-AML cell lines interaction was reduced cell death, with a smaller number of HL60 cells lysed after contact with NK cells. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. NK-AML immunological synapse shows defective lytic granule polarization The low killing rate for NK-AML interactions described above suggests that AML cells engage NK cells in lytic interactions which are never completed. To investigate how this occurs, we analyzed the lytic NKIS formed in this context [26]. Depending on how the balance between activating and inhibitory signals tips, NK cells form two types of NKIS: inhibitory synapses, where actin reorganization is blocked, lytic granules are not polarized to the synapse, and the targeted cell is not killed [27]; and lytic synapses, where the NK-cell actin cytoskeleton reorganizes at the synapse and lytic granules become polarized for the directed secretion of their contents onto the target cell [14]. Indeed the use of MHCI transfected K562 cells reduced NK-cell degranulation in a CD107a test constantly with K562-E cell line and to a lesser point with K562-Cw*03:04 (Supporting Information Fig. 3). However when we examined synapses showing actin reorganization in the contact area i.e., lytic synapses, lytic granule polarization was not significantly different from MHCI- Depending on how the balance between activating and inhibitory signals tips, NK cells form two types of NKIS: inhibitory synapses, where actin reorganization is blocked, lytic granules are not polarized to the synapse, and the targeted cell is not killed [27]; and lytic synapses, where the NK-cell actin cytoskeleton reorganizes at the synapse and lytic granules become polarized for the directed secretion of their contents onto the target cell [14]. Indeed the use of MHCI transfected K562 cells reduced NK-cell degranulation in a CD107a test constantly with K562-E cell line and to a lesser point with K562-Cw*03:04 (Supporting Information Fig. 3). However when we examined synapses showing actin reorganization in the contact area i.e., lytic synapses, lytic granule polarization was not significantly different from MHCI- 9 9 negative K562 cells (Fig. 4A, 4B), thus ruling out a straightforward inhibition due to MHCI interaction with inhibitory receptors. NKIS formed with AML cell lines were then compared with those formed with K562 cells. With AML cell lines, a reduced percentage of NK-cell conjugates contained polarized CD107a+ lytic granules (Fig. 4A, 4B). A similar effect on polarization was found with primary leukemic cells bound to healthy donor NK cells (Fig. 4B, 4C). In lytic synapses, recruitment of LFA-1 integrin to the synapse provides an activation signal, and is necessary for lytic granule polarization [28, 29]. Inhibitory signals disrupt NK cell adhesion and granule polarization via LFA-1 [30, 31]. This article is protected by copyright. All rights reserved. NK-AML immunological synapse shows defective lytic granule polarization To test CD11a (LFA-1α subunit) recruitment to the NK immunological synapse, a relative recruitment index (RRI) was calculated for conjugates. This showed that CD11a was correctly recruited to NK-AML cells synapses (Fig. 5A,B). Due to the higher expression of CD11a on HL60 cells (Fig. 1), NK- HL60 cells interactions had a higher RRI than NK-THP1 cells and NK-K562 cells interactions. Since signaling through NCRs is partially responsible for granule polarization [32], we next examined CD3ζ polarization as an indicator of NCR (NKp46, NKp30) recruitment to the synapse area [33, 34]. NK-AML cell lines synapses are associated with little or no recruitment of CD3ζ compared with NK-K562 synapses (Fig. 5C, 5D). Co-staining for CD107a and CD3ζ showed that when CD3ζ was recruited lytic granule polarized to the NK- AML synapse, whereas with NKIS lacking CD3ζ recruitment (RRI<2) almost no lytic granule polarization was observed (Fig. 6). This reveals a direct relationship between defective NK-cell activation via CD3ζ-associated receptors and the uncompleted NKIS. 10 10 This article is protected by copyright. All rights reserved. NKIS If allogeneic NK cells are to be used in adoptive immunotherapy strategies, the resistance of AML cells to NK-mediated lysis must be overcome [6, 7]. We therefore tested whether pre- activating NK cells with IL-2 or IL-15 overnight improved lytic granule polarization in this context. Pre-activation significantly increased the number of NK cells interacting with AML cell lines, with a higher percentage of conjugates formed with both THP1 and HL60 cells, and as a result increased degranulation (Fig. 7A, 7B). However, the NKIS remained defective as lytic granule polarization was not significantly restored (Fig. 7C). We next tried treating AML cell lines with the immuno-modulating drug lenalidomide Lenalidomide is an interesting candidate because it has been effective in some AML cases [35, 36]. A proposed mechanism of this molecule is increasing the expression of NK- activating receptor ligands on cancer cells [37–39]. In a very similar context, lenalidomide was able to correct defects in the formation of T-cell lytic synapses with leukemic cells [20]. With NK cells, with a concentration corresponding to the dose used in trials for treating refractory AML [35, 36], lenalidomide treatment significantly increased polarization of lytic granules in NKIS, both with THP1 and HL60 cells (Fig. 7D). However NKG2D and DNAM- 1 ligands levels of expression did not change with lenalidomide treatment nor did the level of any of the studied inhibitory molecules: CD137L, CD200, CD273, CD274, and CD276 (Supporting Information Fig. 4). 11 This article is protected by copyright. All rights reserved. 11 This article is protected by copyright. All rights reserved. Discussion: When the interaction did proceed to completion, the time required by NK cells to recruit lytic granules, release their contents and lyse the leukemic cell was unchanged compared with other target cells A similar result was expresses higher levels of the adhesion molecule CD54 (ICAM-1), which binds to LFA-1. This pairing mediates NK-cell conjugate formation [44]. In our assays, THP1 effectively formed more conjugates with NK cells, leading to more severe perturbations. However, on patient cells, although considerable variations in CD54 expression levels were noted, this did not correlate with an impaired NK-cell degranulation capacity upon contact (data not shown). expresses higher levels of the adhesion molecule CD54 (ICAM-1), which binds to LFA-1. This pairing mediates NK-cell conjugate formation [44]. In our assays, THP1 effectively formed more conjugates with NK cells, leading to more severe perturbations. However, on patient cells, although considerable variations in CD54 expression levels were noted, this did not correlate with an impaired NK-cell degranulation capacity upon contact (data not shown). Cytotoxicity is the final outcome of NK-target cell interactions, but dynamic results showed a low killing rate for NK-HL60 cell conjugates, this stemmed from a high rate of uncompleted interactions between NK cells and AML cells. When the interaction did proceed to Cytotoxicity is the final outcome of NK-target cell interactions, but dynamic results showed a low killing rate for NK-HL60 cell conjugates, this stemmed from a high rate of uncompleted interactions between NK cells and AML cells. When the interaction did proceed to completion, the time required by NK cells to recruit lytic granules, release their contents and lyse the leukemic cell was unchanged compared with other target cells. A similar result was 5 interactions between NK cells and AML cells. When the interaction did proceed to completion, the time required by NK cells to recruit lytic granules, release their contents and lyse the leukemic cell was unchanged compared with other target cells. A similar result was recently reported for target lysis via antibody-dependent cellular cytotoxicity [45]. To kill, NK cells must direct their lytic granules toward the target cell, where their contents can be released to induce it effect [14]. NKIS with AML cells showed a normal accumulation of F-actin, the hallmark of lytic synapses, but a low rate of lytic granule polarization. Similar immature cytotoxic synapses have been described with anti-tumor T cells [46, 47]. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Discussion: AML cells express ligands for NK cell-activating receptors [8, 9, 40]. This makes them potential targets for NK-lysis [13]. However, NK cells do not kill autologous leukemic blasts or prevent leukemia development. In most patients this is due to low-level surface expression of NK-activating receptors [10–13]. Normalization of NK cells upon remission and the return of NK cells' ability to kill autologous AML cells in mice suggest that the defect is not due to a problem with NK cells themselves, but rather results from contact with AML cells [12, 13]. Interestingly, AML cells from patients with low NCR expression are highly susceptible to lysis by NK cells expressing normal levels of NCRs [11], thus indicating that AML cells are not simply resistant to NK-cell cytotoxicity. Combining these observations, it appears that a point may occur during AML development when normal cytotoxic NK cells become overwhelmed by the number of target AML cells, and start losing their functionality. p In this study, through observations from an integrated system of NK-cell functions and imaging, we propose a model of escape where, instead of avoiding contact, AML cells actively engage NK cells in ineffective interactions. In support of this, NK cells from healthy donors partially lost activity after interactions with AML cells (Supporting Information Video 4) [41–43]. In addition, a significant percentage of NK- HL60, an AML cell line, interactions did not result in lysis, thus interaction can decrease NK cell activity without killing the target cell. Our results indicate that inefficient killing was the result of qualitative impairment of the immunological synapse. Direct cell-cell contact is necessary to produce NK-cell defects, with conjugate formation causing down-regulation of receptors activating NK cells, such as NKG2D, DNAM-1 and NKp46, and subsequently impairing cytotoxicity. Among the cell lines tested, THP1 12 12 12 expresses higher levels of the adhesion molecule CD54 (ICAM-1), which binds to LFA-1. This pairing mediates NK-cell conjugate formation [44]. In our assays, THP1 effectively formed more conjugates with NK cells, leading to more severe perturbations. However, on patient cells, although considerable variations in CD54 expression levels were noted, this did not correlate with an impaired NK-cell degranulation capacity upon contact (data not shown). Cytotoxicity is the final outcome of NK-target cell interactions, but dynamic results showed a low killing rate for NK-HL60 cell conjugates, this stemmed from a high rate of uncompleted interactions between NK cells and AML cells. Discussion: It was previously known that polarization and exocytotic signals can be uncoupled in NK cells [29]; our findings show that cytoskeleton polarization and lytic granule polarization can also be uncoupled. Different signaling proteins are involved in these two steps of NKIS formation. Inhibition of a MAP kinase, JNK-1, or of (WASp)-interacting protein (WIP) has been shown to inhibit polarization of lytic granules to the immune synapse, but had no effect on conjugate formation or localization of actin at the synapse [48, 49]. Most studies on NKIS have been conducted using coated surfaces or cell lines expressing ligands for only one or two NK- activating or inhibitory receptors, and mostly with NK leukemic-like cell lines, such as NK92 or NKL, as effectors. Using AML cell lines expressing a wide range of activating and 13 13 13 inhibitory ligands and primary polyclonal NK cells better reflects the disease conditions. inhibitory ligands and primary polyclonal NK cells better reflects the disease conditions. Thereby the dissociation seen in our experiments suggests that each step in NKIS formation requires a particular level of activation, determined by the number and type of inhibitory and stimulatory ligands engaged with their receptors. inhibitory ligands and primary polyclonal NK cells better reflects the disease conditions. Thereby the dissociation seen in our experiments suggests that each step in NKIS formation requires a particular level of activation, determined by the number and type of inhibitory and stimulatory ligands engaged with their receptors. The activation signal necessary for lytic granule polarization is provided by recruitment of the integrin LFA-1 to the synapse [28, 50]. KIR-mediated inhibitory signals can disrupt NK- cell adhesion and granule polarization via LFA-1 [30, 31]. However, these appear not to be responsible for the defective NK-AML synapse described here, as LFA-1 is correctly recruited. Staining for inhibitory molecules reported to impair T-cell cytotoxic synapse formation (CD200, CD273, CD274, and CD276) [19, 20], revealed the presence of CD200 on the AML cell lines and the primary blasts studied (Supporting Information Fig. 1 and Supporting Information Table 1). In AML, CD200 expression is associated with poor outcome, with CD200high patients showing reduced NK-cell activity and NCR expression [19]. Lytic granule polarization is triggered by CD3ζ recruitment, this signaling adaptor combines with NKp46 to transmit its signal to the MAP kinase JNK1 [32]. NK-AML cell lines synapses generally recruited very little or no CD3ζ, another indication of their non-activated state. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Discussion: In NK-AML synapses where CD3ζ was not recruited, NK cells exhibited no polarization of lytic granules, revealing a direct relationship between defective NK-cell activation and uncompleted NKIS, highlighting the importance of CD3ζ in killing AML cells. Chronic NKG2D stimulation due to disease-related ligand expression has been shown to trigger CD3ζ degradation [51]. This could contribute to impairing NK-cell activity upon long term interactions. Thereby NKIS defects can be the result of higher inhibition with molecules like CD200 and/or lower activation through CD3ζ. 14 14 Some NK-cell defects could be overcome by pre-activation with cytokines (IL-2, IL-15), which increased conjugate formation and degranulation, but did not improve synapse formation. More promisingly, pre-treatment of AML cell lines with the immuno-modulating molecule lenalidomide improved the polarization rate significantly. A possible mechanism is a decrease in The TNF family member receptor activator for NF-κB ligand (RANKL). Lenalidomide has been shown to decrease RANKL in bone marrow [52]. RANK signaling into NK cells directly impairs NK-cell anti-leukemia reactivity [53]. Combining the correcting effect of lenalidomide on NKIS with the qualitative effect of cytokines on conjugate formation could increase AML cell sensitivity to autologous NK and reinforce the activity of allogeneic NK cells in adoptive immunotherapy. ac 15 15 Cell isolation and cell lines PBMCs were isolated from healthy donor samples obtained from the French blood service (EFS). NK cells were isolated by negative selection using a NK-cell Isolation Kit (Miltenyi Biotec). Purified NK cells were used without any prior culture or exposure to growth factors for all experiments except when preactivation with IL-2 or IL-15 is mentioned. PBMCs were isolated from healthy donor samples obtained from the French blood service (EFS). NK cells were isolated by negative selection using a NK-cell Isolation Kit (Miltenyi Biotec). Purified NK cells were used without any prior culture or exposure to growth factors for all experiments except when preactivation with IL-2 or IL-15 is mentioned. All cell lines (THP1: human acute monocytic leukemia; HL60: human acute promyelocytic leukemia cell line; controls: K562, K562-Cw*0304 transfected to express HLA-C, and K562- E2B4 transfected to express HLA-E [54]) were grown in RPMI supplemented with 10% FBS, and L-glutamine, sodium pyruvate, hepes buffer (1% each), and with the corresponding selection agent for transfected cells. The expression of HLA molecules on transfected cell lines was verified by flow cytometry. All cell lines (THP1: human acute monocytic leukemia; HL60: human acute promyelocytic leukemia cell line; controls: K562, K562-Cw*0304 transfected to express HLA-C, and K562- E2B4 transfected to express HLA-E [54]) were grown in RPMI supplemented with 10% FBS, and L-glutamine, sodium pyruvate, hepes buffer (1% each), and with the corresponding selection agent for transfected cells. The expression of HLA molecules on transfected cell lines was verified by flow cytometry. All patients gave written informed consent to participate in this study, the protocol for which was approved by the local review board in line with the Declaration of Helsinki. Samples from patients with AML were collected at diagnosis in the Hematology Department of Saint- Louis Hospital, Paris, France, before treatment. CD33+ blasts were isolated by positive sorting from PBMCs using the StemSep System (StemCell-Technology). This article is protected by copyright. All rights reserved. Flow cytometry Ligands for NKG2D, DNAM-1, NKp30 and NKp46 were stained with Fc recombinant proteins for each receptor (R&D system) and revealed using a Biotin-conjugated F(ab')2 goat anti-human IgG (Jackson ImmunoResearch) followed by streptavidin R PE (Molecular probes). Monoclonal antibodies (mAbs) used to stain AML and NK cells were: ULBP1, ULBP2, ULBP3, MICA, MICB, CD155 “PVR” from R&D system; CD112 “Nectin2”, 16 16 CD276 “B7H3”, CD137L, CD117, CD200, CD11c, HLA-E, DNAM-1 from Biolegend; CD11a, CD273 “PDL-2”, CD34, CD11b, CD54, HLA-A,B,C, CD3, CD33, CD56, CD16, IFN-γ, an isotype control from BD biosciences; CD45 (Invitrogen); CD33, NKp46, NKG2D from Beckman Coulter; and CD274 “PD-L1”(ebioscience). CD276 “B7H3”, CD137L, CD117, CD200, CD11c, HLA-E, DNAM-1 from Biolegend; CD11a, CD273 “PDL-2”, CD34, CD11b, CD54, HLA-A,B,C, CD3, CD33, CD56, CD16, IFN-γ, an isotype control from BD biosciences; CD45 (Invitrogen); CD33, NKp46, NKG2D from Beckman Coulter; and CD274 “PD-L1”(ebioscience). Flow cytometry analysis was performed on a FACSDiva (BD), and results were analyzed using Diva v6.1.3 or FlowJo vX.0.6 software. Conjugation and NK-cell degranulation assays by flow cytometry Conjugation tests were performed with purified NK cells stained with CellTracker Green (CMFDA) and AML cells stained with CellTracker Orange (CMRA) (Molecular Probes) according to the manufacturer's instructions. Cells were mixed 1:1 and incubated for 1 h at 37 °C. Percentage of bicolor cell-conjugates among CMFDA positive cells was measured. NK cells were incubated overnight with target cells. Degranulation was then estimated by measuring the percentage of CD107a-positive cells [55]. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Time-lapse microscopy Microscopy was performed in culture conditions at 37 °C under 5% CO2 on a on a NIKON Biostation timelapse microscope equipped with a 40× N.A 0.8 using the Biostation IM software (Nikon). Target cells, 1.5x105 K562 or HL60 cells, were seeded onto two different segments of a Hi-Q4 Culture Dish (Ibidi), previously coated with 1% collagen (Sigma Aldrich)/2% fibronectin (Gibco), and left to adhere for at least 2 hours. Sytox Blue Nucleic- Acid Stain (Molecular Probes) was added to the medium at 5 µM. After choosing two fields of interest per segment at 40× magnification, 3x105 CMRA-labeled NK cells were added to each segment (NK:Target ratio 2:1) and imaging was started simultaneously for the two target cells. Images were acquired on three channels (Phase-contrast, DAPI and TRITC filter cubes) every 45 sec for 3 hours. Images were analyzed using BioStation software and the Fiji 17 17 package [56]. Target cells were manually tracked, and each interaction with an NK cell was scored. Conjugates were defined as physical contact between NK cells and target cells lasting longer than 2 min 15 sec (3 imaging cycles). Target cell death was detected by the gradual appearance of Sytox Blue fluorescence, and by visual signs of death in the phase-contrast image (blebbing). This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. Cytokines and lenalidomide treatment Where indicated, purified NK cells were incubated overnight in medium supplemented with rhIL-2 1000 IU/mL or rhIL-15 10 ng/mL (ImmunoTools), prior to performing degranulation, conjugate and polarization tests as described. In other assays, lenalidomide (LC Laboratories) was added to AML cell cultures at a final concentration of 10 µM 18-24 hours before assessing lytic granule polarization as described. Granule polarization was counted on blinded slides. This article is protected by copyright. All rights reserved. Fluorescence microscopy and image analysis Slides of cell conjugates (NK:AML ratio 2:1) were mounted as follows: purified NK cells and AML cells were incubated together for 1 hour at 37 °C, resuspended gently, seeded on slides previously coated with poly-L-lysine (Sigma Aldrich) and left to adhere for 30 min at 37 °C. Cells were fixed in 4% paraformaldehyde at room temperature, permeabilized with 0.1% saponin, and non-specific sites were blocked with TBS-T 5% BSA. Cells were labeled for 30 min to reveal CD107a (H4A3-FITC, BD or H4A3-Brilliant Violet421, Biolegend), LFA1 (HI111-FITC, BD Pharmingen), and F-actin (Phalloidin Alexa Fluor568, Molecular Probes), or for 1 hour to reveal CD3ζ (6B10.2-FITC, Biolegend) and with the isotype control. Slides were washed with TBS-T 1% BSA, mounted with hard-set Vectashield with or without DAPI (Vector Laboratories). Polarization of lytic granules was scored under an Axiovert 200M microscope (Zeiss) for at least 50 conjugates per slide in randomly selected fields. Conjugates were defined based on morphology - when opposing NK cell and target cell membranes appeared flat, with accumulated F-actin. Conjugates were considered polarized when CD107a-positive lytic granules were located in the NK cell quarter nearest the target cell [32, 48]. Results were validated by a second examiner who re-counted randomly selected slides. Representative images were captured with Axiovert 200M microscope equipped with a Plan Apochromat 63X/N.A.1.4 oil-immersion objective, an optavar 1.6× lens (resulting magnification 100.8×) 18 and an Axiocam MRM camera (Zeiss) using the Axiovision v4.5.0.0 software (Zeiss), or with LSM 510 confocal microscope (Zeiss) equipped with a 64× oil-immersion objective and optavar 4× lens using the associated software. and an Axiocam MRM camera (Zeiss) using the Axiovision v4.5.0.0 software (Zeiss), or with LSM 510 confocal microscope (Zeiss) equipped with a 64× oil-immersion objective and optavar 4× lens using the associated software. The Fiji package was used to analyze fluorescence microscopy images [56]. On unprocessed images for actin, LFA1 and CD3ζ, a region of interest (ROI).1 was defined on the NK cell membrane touching the target cell. ROI.2 was located on the opposing NK-cell membrane, and ROI.3 was in the background. The mean gray value (MGV) was measured for each ROI. A relative recruitment index (RRI) was calculated based on: RRI= (MGV ROI.1 – MGV ROI.3) / (MGV ROI.2 – MGV ROI.3) Statistics Statistics and graphical presentations were produced using GraphPad Prism5 software (GraphPad). Wilcoxon's signed rank test was used to compare NK cells in microscopy experiments or after co-culture and to compare results after cells treatments (cytokines/lenalidomide). Other data were compared using the Mann-Whitney test. 19 This article is protected by copyright. All rights reserved. 19 Acknowledgements: This work was supported by research grants from the Association Laurette Fugain, Association pour la Recherche sur le Cancer (Z.K, #DOC20100600956), Fondation pour la Recherche Médicale (S.A, #SPF20100518400), Fondation de France (S.A, #2012-26957), Institut National du Cancer (grants #R8S09081HHA, #RPT12008HHA), and Assistance Publique–Hopitaux de Paris (A.T, Translational Research grant in Biology 2010, #RTB10002). This article is protected by copyright. All rights reserved. Conflict of interest The authors declare no commercial or financial conflict of interest. 20 20 This article is protected by copyright. All rights reserved. References: 1. Caligiuri MA. Human natural killer cells. Blood. 2008. 112:461–469. 2. Ljunggren H-G, Malmberg K-J. Prospects for the use of NK cells in immunotherapy of human cancer. Nat Rev Immunol. 2007. 7:329–339. 3. Ruggeri L, Mancusi A, Burchielli E, Perruccio K, Aversa F, Martelli MF, Velardi A. Natural killer cell recognition of missing self and haploidentical hematopoietic transplantation. Semin. Cancer Biol. 2006. 16:404–411. 4. Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch SK, McKenna D, et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood. 2005. 105:3051–3057. 5. Passweg JR, Tichelli A, Meyer-Monard S, Heim D, Stern M, Kühne T, Favre G, et al. Purified donor NK-lymphocyte infusion to consolidate engraftment after haploidentical stem cell transplantation. Leukemia. 2004. 18:1835–1838. 6. Farag SS, Caligiuri MA. Immunologic Approaches to Acute Leukemia in the Elderly. Seminars in Hematology. 2006. 43:118–125. 7. Vey N, Bourhis J-H, Boissel N, Bordessoule D, Prebet T, Charbonnier A, Etienne A, et al. A phase 1 trial of the anti-inhibitory KIR mAb IPH2101 for AML in complete remission. Blood. 2012. 120:4317–4323. 8. Salih HR, Antropius H, Gieseke F, Lutz SZ, Kanz L, Rammensee H-G, Steinle A. Functional expression and release of ligands for the activating immunoreceptor NKG2D in leukemia. Blood. 2003. 102:1389–1396. 9. Pende D, Spaggiari GM, Marcenaro S, Martini S, Rivera P, Capobianco A, Falco M, et al. Analysis of the receptor-ligand interactions in the natural killer-mediated lysis of freshly isolated myeloid or lymphoblastic leukemias: evidence for the involvement of the Poliovirus receptor (CD155) and Nectin-2 (CD112). Blood. 2005. 105:2066–2073. 10. Lion E, Willemen Y, Berneman ZN, Van Tendeloo VFI, Smits ELJ. Natural killer cell immune escape in acute myeloid leukemia. Leukemia. 2012. 26:2019–2026. 11. Costello RT, Sivori S, Marcenaro E, Lafage-Pochitaloff M, Mozziconacci M-J, Reviron D, Gastaut J-A, et al. Defective expression and function of natural killer cell- triggering receptors in patients with acute myeloid leukemia. Blood. 2002; 99:3661–3667. 12. Fauriat C, Just-Landi S, Mallet F, Arnoulet C, Sainty D, Olive D, Costello RT. Deficient expression of NCR in NK cells from acute myeloid leukemia: Evolution during leukemia treatment and impact of leukemia cells in NCRdull phenotype induction. Blood. 2007. 109:323–330. 21 This article is protected by copyright. All rights reserved. 21 13. Siegler U, Kalberer CP, Nowbakht P, Sendelov S, Meyer-Monard S, Wodnar- Filipowicz A. References: Activated natural killer cells from patients with acute myeloid leukemia are cytotoxic against autologous leukemic blasts in NOD/SCID mice. Leukemia. 2005. 19:2215– 2222. 14. Orange JS. Formation and function of the lytic NK-cell immunological synapse. Nat. Rev. Immunol. 2008. 8:713–725. 15. Wulfing C, Purtic B, Klem J, Schatzle JD. Stepwise cytoskeletal polarization as a series of checkpoints in innate but not adaptive cytolytic killing. PNAS. 2003. 100:7767– 7772. 16. Mace EM, Wu WW, Ho T, Mann SS, Hsu H-T, Orange JS. NK Cell Lytic Granules Are Highly Motile at the Immunological Synapse and Require F-Actin for Post- Degranulation Persistence. J. Immunol. 2012. 189:4870–4880. 17. Mamessier E, Sylvain A, Bertucci F, Castellano R, Finetti P, Houvenaeghel G, Charaffe-Jaufret E, et al. Human Breast Tumor Cells Induce Self-Tolerance Mechanisms to Avoid NKG2D-Mediated and DNAM-Mediated NK Cell Recognition. Cancer Research. 2011. 71:6621 –6632. 18. i e i ri es e - er es -Mediavilla J, Gutiérrez N, Cañizo C, Ramos F, et al. Early immunophenotypical evaluation of minimal residual disease in acute myeloid leukemia identifies different patient risk groups and may contribute to postinduction treatment stratification. Blood. 2001. 98:1746–1751. 19. Coles SJ, Wang ECY, Man S, Hills RK, Burnett AK, Tonks A, Darley RL. CD200 expression suppresses natural killer cell function and directly inhibits patient anti-tumor response in acute myeloid leukemia. Leukemia. 2011. 25:792–799. 20. Ramsay AG, Clear AJ, Fatah R, Gribben JG. Multiple Inhibitory Ligands Induce Impaired T Cell Immunological Synapse Function in Chronic Lymphocytic Leukemia That Can Be Blocked with Lenalidomide. Blood. 2012. 120: 1412–1421. 21. Baessler T, Charton JE, Schmiedel BJ, Grünebach F, Krusch M, Wacker A, Rammensee H-G, et al. CD137 ligand mediates opposite effects in human and mouse NK cells and impairs NK-cell reactivity against human acute myeloid leukemia cells. Blood. 2010. 115:3058–3069. 22. Zhou Y, Jorgensen JL, Wang SA, Ravandi F, Cortes J, Kantarjian HM, Medeiros LJ, et al. Usefulness of CD11a and CD18 in Flow Cytometric Immunophenotypic Analysis for Diagnosis of Acute Promyelocytic Leukemia. AJCP. 2012. 138:744–750. 23. Bergmann L, Schui DK, Brieger J, Weidmann E, Mitrou PS, Hoelzer D. The inhibition of lymphokine-activated killer cells in acute myeloblastic leukemia is mediated by transforming growth factor-beta 1. Exp. Hematol. 1995. 23:1574–1580. 24. Orleans-Lindsay JK, Barber LD, Prentice HG, Lowdell MW. Acute myeloid leukaemia cells secrete a soluble factor that inhibits T and NK cell proliferation but not 22 This article is protected by copyright. References: All rights reserved. 22 cytolytic function--implications for the adoptive immunotherapy of leukaemia. Clin. Exp. Immunol. 2001. 126:403–411. cytolytic function--implications for the adoptive immunotherapy of leukaemia. Clin. Exp. Immunol. 2001. 126:403–411. 25. Vanherberghen B, Olofsson PE, Forslund E, Sternberg-Simon M, Khorshidi MA, Pacouret S, Guldevall K, et al. Classification of human natural killer cells based on migration behavior and cytotoxic response. Blood. 2013. 121:1326–1334. 26. Orange JS, Harris KE, Andzelm MM, Valter MM, Geha RS, Strominger JL. The mature activating natural killer cell immunologic synapse is formed in distinct stages. PNAS. 2003. 100:14151–14156. 27. Eissmann P, Davis DM. Inhibitory and regulatory immune synapses. Curr. Top. Microbiol. Immunol. 2010. 340:63–79. 28. Barber DF, Faure M, Long EO. LFA-1 Contributes an Early Signal for NK Cell Cytotoxicity. J Immunol. 2004. 173:3653–3659. 29. Bryceson YT, March ME, Barber DF, Ljunggren H-G, Long EO. Cytolytic granule polarization and degranulation controlled by different receptors in resting NK cells. J. Exp. Med. 2005. 202:1001–1012. 30. Burshtyn DN, Shin J, Stebbins C, Long EO. Adhesion to target cells is disrupted by the killer cell inhibitory receptor. Current Biology. 2000. 10:777–780. 31. Das A, Long EO. Lytic Granule Polarization, Rather than Degranulation, Is the Preferred Target of Inhibitory Receptors in NK Cells. The Journal of Immunology. 2010. 185:4698 – 4704. 32. Chen X, Trivedi PP, Ge B, Krzewski K, Strominger JL. Many NK cell receptors activate ERK2 and JNK1 to trigger microtubule organizing center and granule polarization and cytotoxicity. PNAS. 2007. 104:6329 –6334. 33. Pessino A, Sivori S, Bottino C, Malaspina A, Morelli L, Moretta L, Biassoni R, et al. Molecular cloning of NKp46: a novel member of the immunoglobulin superfamily involved in triggering of natural cytotoxicity. J. Exp. Med. 1998. 188:953–960. 34. Pende D, Parolini S, Pessino A, Sivori S, Augugliaro R, Morelli L, Marcenaro E, et al. Identification and molecular characterization of NKp30, a novel triggering receptor involved in natural cytotoxicity mediated by human natural killer cells. J. Exp. Med. 1999. 190:1505–1516. 35. Fehniger TA, Byrd JC, Marcucci G, Abboud CN, Kefauver C, Payton JE, Vij R, et al. Single-agent lenalidomide induces complete remission of acute myeloid leukemia in patients with isolated trisomy 13. Blood. 2009. 113:1002–1005. 36. Blum W, Klisovic RB, Becker H, Yang X, Rozewski DM, Phelps MA, Garzon R, et al. Dose Escalation of Lenalidomide in Relapsed or Refractory Acute Leukemias. JCO. 2010. 28:4919–4925. 23 This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. References: 23 37. Wolschke C, Stübig T, Hegenbart U, Schönland S, Heinzelmann M, Hildebrandt Y, Ayuk F, et al. Postallograft lenalidomide induces strong NK cell-mediated antimyeloma activity and risk for T cell-mediated GvHD: results from a phase I/II dose-finding study. Exp. Hematol. 2012. 41:134–142. 38. Kotla V, Goel S, Nischal S, Heuck C, Vivek K, Das B, Verma A. Mechanism of action of lenalidomide in hematological malignancies. J Hematol Oncol. 2009; 2:36. 39. Benson DM Jr, Bakan CE, Zhang S, Collins SM, Liang J, Srivastava S, Hofmeister CC, et al. IPH2101, a novel anti-inhibitory KIR antibody, and lenalidomide combine to enhance the natural killer cell versus multiple myeloma effect. Blood. 2011. 118:6387–6391. 40. Nowbakht P, Ionescu M-CS, Rohner A, Kalberer CP, Rossy E, Mori L, Cosman D, et al. Ligands for natural killer cell-activating receptors are expressed upon the maturation of normal myelomonocytic cells but at low levels in acute myeloid leukemias. Blood. 2005. 105:3615–3622. 41. Abrams SI, Brahmi Z. Target cell directed NK inactivation. Concomitant loss of NK and antibody-dependent cellular cytotoxicity activities. J Immunol. 1988. 140:2090–2095. 42. Jewett A, Bonavida B. Target-induced anergy of natural killer cytotoxic function is restricted to the NK-target conjugate subset. Cell. Immunol. 1995; 160:91–97. 43. Gill S, Vasey AE, De Souza A, Baker J, Smith AT, Kohrt HE, Florek M, et al. Rapid development of exhaustion and down-regulation of eomesodermin limit the antitumor activity of adoptively transferred murine natural killer cells. Blood. 2012. 119:5758–5768. 44. Gross CC, Brzostowski JA, Liu D, Long EO. Tethering of intercellular adhesion molecule on target cells is required for LFA-1-dependent NK cell adhesion and granule polarization. J. Immunol. 2010. 185:2918–2926. 45. Rudnicka D, Oszmiana A, Finch DK, Strickland I, Schofield DJ, Lowe DC, Sleeman MA, et al. Rituximab causes a polarisation of B cells which augments its therapeutic function in NK cell-mediated antibody-dependent cellular cytotoxicity. Blood. 2013. 121:4694–4702. 46. oc’h A i A er non I h ns c r ism th G ho ib et al. αEβ7 integrin interaction with E-cadherin promotes antitumor CTL activity by triggering lytic granule polarization and exocytosis. J Exp Med. 2007. 204:559–570. 47. r ncis kiewic K oc’h A o tet er non I chmitt A mi-Chouaib F. CD103 or LFA-1 Engagement at the Immune Synapse between Cytotoxic T Cells and Tumor Cells Promotes Maturation and Regulates T-cell Effector Functions. Cancer Res. 2013. 73:617–628. 48. This article is protected by copyright. All rights reserved. References: Li C, Ge B, Nicotra M, Stern JNH, Kopcow HD, Chen X, Strominger JL. JNK MAP kinase activation is required for MTOC and granule polarization in NKG2D-mediated NK cell cytotoxicity. PNAS. 2008. 105:3017–3022. 49. Krzewski K, Chen X, Strominger JL. WIP is essential for lytic granule polarization and NK cell cytotoxicity. PNAS. 2008. 105:2568–2573. 24 This article is protected by copyright. All rights reserved. 24 50. Liu D, Bryceson YT, Meckel T, Vasiliver-Shamis G, Dustin ML, Long EO. Integrin- dependent organization and bidirectional vesicular traffic at cytotoxic immune synapses. Immunity. 2009. 31:99–109. 51. Hanaoka N, Jabri B, Dai Z, Ciszewski C, Stevens AM, Yee C, Nakakuma H, et al. NKG2D initiates caspase-mediated CD3ζ degradation and lymphocyte receptor impairments associated with human cancer and autoimmune disease. J Immunol. 2010. 185:5732–5742. 52. Breitkreutz I, Raab MS, Vallet S, Hideshima T, Raje N, Mitsiades C, Chauhan D, et al. Lenalidomide inhibits osteoclastogenesis, survival factors and bone-remodeling markers in multiple myeloma. Leukemia. 2008. 22:1925–1932. 53. Schmiedel BJ, Nuebling T, Steinbacher J, Malinovska A, Wende CM, Azuma M, Schneider P, et al. Receptor Activator for NF-κB Ligand in Acute Myeloid Leukemia: Expression, Function, and Modulation of NK Cell Immunosurveillance. J Immunol. 2012. 190: 821–831. 54. Schleypen JS, Von Geldern M, Weiss EH, Kotzias N, Rohrmann K, Schendel DJ, Falk CS, et al. Renal cell carcinoma-infiltrating natural killer cells express differential repertoires of activating and inhibitory receptors and are inhibited by specific HLA class I allotypes. Int. J. Cancer. 2003. 106:905–912. 55. Alter G, Malenfant JM, Altfeld M. CD107a as a functional marker for the identification of natural killer cell activity. J. Immunol. Methods. 2004. 294:15–22. o 5 P M 56. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, et al. Fiji: an open-source platform for biological-image analysis. Nature Methods. 2012. 9:676–682. 25 This article is protected by copyright. All rights reserved. 25 FIGURE 1. Characterizing primary AML blasts (dots), THP1 (blue line) and HL60 (red line) cell lines. (A) NK-cell ligand expression levels, expressed as the ratio of the mean fluorescence intensity (MFI) for each marker divided by the MFI for the negative controls, was determined on the indicated cell types by flow cytometry. For primary AML blasts, gating was done on CD33+ cells in bone marrow or peripheral blood samples, each dot represents an individual sample out of 24. This article is protected by copyright. All rights reserved. References: For THP1 and HL60 cells, lines represent the mean of three independent experiments. (B) Percentage of healthy donor NK cells degranulating in response to AML cells was determined by flow cytometry. After over-night co-incubation of PBMCs with AML cells in the presence of CD107a antibodies, percentage of CD107a+ (i.e. degranulated) NK cells was assessed on the (CD3-CD56+) population in the life lymphocyte’s gate defined morphologically. Each symbol represents an individual experiment out of 24 for primary AML blasts and 5 for THP1 and HL60 cells. Bars represent means, *p < 0.05; **p < 0.01 (Mann-Whitney test). (C) Expression levels of NKG2D and DNAM-1 ligands, CD54 and CD11a were determined on THP1 and HL60 cell lines by flow cytometry. Representative histograms from one out of three independent experiments are shown. histograms from one out of three independent experiments are shown. 26 26 FIGURE 2. Effects of AML cells on NK-cell phenotype and function. PBMCs were co- cultured in round-bottomed 96-well plates alone or with irradiated AML cells (ratio 1:1). After 48 hours, PBMCs were phenotyped and tested for NK cell degranulation. (A) Receptor down-regulation on NK cells was assessed by flow cytometry, gating was done as indicated in Figure 1. Level of expression on NK cells cultured alone was compared with that on NK cells co-cultured with THP1 or HL60 myeloid cell lines, or with the control lymphoma cell line RAJI. Each symbol represents an individual experiment out of six, and mean ± SD are shown. (B) PBMCs retrieved from cultures as described above were tested for degranulation against K562 as described in figure 1. Representative dot plot of one out of six are shown; for cells retrieved from culture alone (upper left), with THP1 cells (upper right) or with HL60 cells (lower left). In the graph (lower right) percentage of CD107a+ NK cells after co-culture with THP1 or HL60 cells is presented. Each symbol represents an individual experiment. Bars represent mean ± SD. *p < 0.05; **p < 0.01 (Wilcoxon's signed rank test). (C) Degranulation levels for IL-2-activated NK cells against primary AML blasts was measured as described in Figure 1 and correlated (Pearson correlation coefficient) with NKG2D down- regulation (MFI before/MFI after degranulation) on NK cells. MFI was assessed by flow cytometry as described in (A). This article is protected by copyright. All rights reserved. References: (D) Conjugation tests were performed with purified NK cells stained with CellTracker Green (CMFDA) and THP1 or HL60 cells stained with CellTracker Orange (CMRA). Cells were mixed 1:1 and incubated for 1 h at 37°C. Percentage of bicolor cell-conjugates among CMFDA positive cells was measured by flow cytometry. Each symbol represents an individual experiment and bars represent mean ± SD. **p < 0.01 (Wilcoxon's signed rank test). (E and F) Experiments were performed as in (A and D), with PBMCs from three donors co-cultured with AML blasts isolated from four patients (Pt1 – Pt4). (E) DNAM-1, NKG2D and NKp46 levels were assessed after 48 hours of co-culture (left). Percentage of conjugates was also assessed (right). Plotted are means + SEM of the three PBMCs. *p < 0.05; **p < 0.01 (Mann-Whitney test). (F) Correlation between receptor down- regulation and the percentage of conjugates (Pearson correlation coefficient). t D P P r 27 This article is protected by copyright. All rights reserved. 27 This article is protected by copyright. All rights reserved. 28 28 FIGURE 3. Time-lapse imaging of HL60 and K562 interactions with NK cells. In 3 independent experiments imaging was performed on a BioStation IM-Q (Nikon) at 40× magnification. Target cells (TCs) were manually tracked and interactions with NK cells (red) were scored for 137 K562 cells and 133 HL60 cells pooled from the 3 experiments. The time between conjugate formation and TC’s killing, visualized by the appearance of intracellular fluorescence in TCs, was calculated. Timing (hh:mm:ss) started when NK cells were added to the plate. (A) Percentage of K562 or HL60 cells forming conjugates with NK cells was manually counted in images. ***p < 0.0001 (Fisher's exact test). (B) For comparison, percentage of NK-cell conjugates was also estimated by flow cytometry as described in Figure 2D. Each symbol represents an individual experiment and bars represent mean ± SD. **p < 0.01 (Mann-Whitney test). (C) Representative example of NK-cell interaction with HL60 cells, taken from the time-lapse experiences performed as described above: overlay (upper row), starting before the encounter (first frame to the left). In this example Conjugate- Kill time was 2 min 15 sec. Lower row presents blue channel fluorescence intensity in 16 colors. The trace represents the intensity profile in TC. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. References: (D) Conjugate-Kill time was determined for K562 and HL60 cells after contact with a single NK cell (mean 5 min 7 sec and 4 min 58 sec, respectively). NS: no significant difference (Mann-Whitney test). (E) Some TCs were hit by two NK cells; T1 is the Conjugate-Kill time for the first NK cell contact, and T2 the time for the second. Each symbol represents an individual TC. Data are shown as mean ± SD and are pooled from the three time-lapse experiments. (F) Two examples of ineffective conjugates formed with HL60 cells; Overlaid time-lapse images taken as described above at 40× magnification: (left) initial contact, (right) NK cell leaving the HL60 cell undamaged (no blue fluorescence). Images are representative of 25 unkilled HL60 cells. (G) Percentage of K562 or HL60 cells surviving after forming at least one conjugate was manually counted in images. Data are pooled from the three time-lapse experiments. * p = 0.04 (Fisher's exact test). c ( m 0 29 This article is protected by copyright. All rights reserved. 29 protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. 30 30 31 31 FIGURE 4. Lytic granule polarization in the NK-AML immunological synapse. Purified NK cells and AML cells were incubated together for 1 h at 37 °C, resuspended gently, seeded on slides, fixed and stained. Lytic granule (stained in green with antibody against CD107a) polarization was scored by counting under Axiovert 200M microscope (Carl Zeiss) in at least 50 NK/target cell conjugates in which F-actin had polymerized (left panel). Conjugates were considered polarized when CD107a+ lytic granules were located in the quarter of the NK cell nearest the target cell (schematized in the middle panel). The right panel shows merge images with DAPI stain for nuclei. (A) Examples of conjugates with K562, THP1 and HL60 cells. Images were acquired with an LSM 510 confocal microscope at 64×/4 magnification, and are representative of 6 independent experiments. (B) Percentage of conjugates showing polarized lytic granules with AML cell lines (upper graph) or with AML patient (Pt) blasts (lower graph) compared with K562, K562-Cw*0304 and K562-E the control cell lines, each symbol represents an experiment and means ± SD are indicated. **p <0.01 (Wilcoxon's signed rank test). FIGURE 4. Lytic granule polarization in the NK-AML immunological synapse. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved. References: Purified NK cells and AML cells were incubated together for 1 h at 37 °C, resuspended gently, seeded on slides, fixed and stained. Lytic granule (stained in green with antibody against CD107a) polarization was scored by counting under Axiovert 200M microscope (Carl Zeiss) in at least 50 NK/target cell conjugates in which F-actin had polymerized (left panel). Conjugates were considered polarized when CD107a+ lytic granules were located in the quarter of the NK cell nearest the target cell (schematized in the middle panel). The right panel shows merge images with DAPI stain for nuclei. (A) Examples of conjugates with K562, THP1 and HL60 cells. Images were acquired with an LSM 510 confocal microscope at 64×/4 magnification, and are representative of 6 independent experiments. (B) Percentage of conjugates showing polarized lytic granules with AML cell lines (upper graph) or with AML patient (Pt) blasts (lower graph) compared with K562, K562-Cw*0304 and K562-E the control cell lines, each symbol represents an experiment and means ± SD are indicated. **p <0.01 (Wilcoxon's signed rank test). 32 32 FIGURE 5. Recruitment of LFA-1 and CD3ζ in the NK-AML immunological synapse. LFA- 1 (CD11a) or CD3ζ recruitment to the contact area was estimated on slides prepared as indicated in Figure 4. Images were acquired with an Axiovert 200M microscope at 64×/1.6 magnification. The relative recruitment index (RRI) in NK cells conjugated with K562, THP1 or HL60 cells was calculated as described in Materials and methods, where polymerized F- actin was visible (red, left). (A-B) Examples of NKIS with THP1 or HL60 cells. Recruitment of CD11a (green, middle) is shown in (A), and recruitment of CD3ζ (green, middle) is shown in (B). The right panel shows merge images with DAPI stain for nuclei. Images are representative of three independent experiments. (C-D) RRI values for CD11a or CD3ζ. Data are shown as mean ± SD and are pooled from three different experiments, each point represents a cell. ***p <0.001 (unpaired t test). In the THP1 image in (B), an example of an NK cell (+) with no actin polarization is visible. This type of cell was not included in the counted conjugates. 33 This article is protected by copyright. All rights reserved. 33 FIGURE 6. Relation of lytic granule polarization and CD3ζ recruitment in the NK-AML synapse. This article is protected by copyright. All rights reserved. References: (A) Purified NK cells and AML cells were incubated together for 1 h at 37 °C, resuspended gently, seeded on slides, fixed and stained for F-actin (red), CD3ζ (green), and CD107a (blue). The right panel shows merge images. Images are representative of four independent experiments and were acquired with an LSM 510 confocal microscope at 64×/4 magnification. (B) Lytic granule polarization was evaluated in NK-AML conjugates with (RRI > 2) or without (RRI < 2) CD3ζ recruitment, as described in Figure 4. The number of NK cells with or without CD107a polarization is shown in conjugates with THP1 cells (upper graph) or with HL60 cells (lower graph). Data are pooled from three experiments. *** p < 0.001 (Fisher's exact test). FIGURE 6. Relation of lytic granule polarization and CD3ζ recruitment in the NK-AML synapse. (A) Purified NK cells and AML cells were incubated together for 1 h at 37 °C, resuspended gently, seeded on slides, fixed and stained for F-actin (red), CD3ζ (green), and CD107a (blue). The right panel shows merge images. Images are representative of four independent experiments and were acquired with an LSM 510 confocal microscope at 64×/4 magnification. (B) Lytic granule polarization was evaluated in NK-AML conjugates with (RRI > 2) or without (RRI < 2) CD3ζ recruitment, as described in Figure 4. The number of NK cells with or without CD107a polarization is shown in conjugates with THP1 cells (upper graph) or with HL60 cells (lower graph). Data are pooled from three experiments. *** p < 0.001 (Fisher's exact test). This article is protected by copyright. All rights reserved. 34 34 This article is protected by copyright. All rights reserved. 34 FIGURE 7. Effects of AML cell lines treatment with lenalidomide or NK-cell pre-activation with IL-2 or IL-15 on NK-cell degranulation. (A-C) NK cells were activated overnight with 10 ng/mL IL-15 or 1000 UI/mL IL-2 before (A) using flow cytometry to assess the number of conjugates formed as indicated in Figure 2, (B) determining the NK degranulation rate by flow cytometry as described in Figure 1, and (C) determining the percentage of conjugates showing granule polarization by microscopy as indicated in Figure 4. Data are shown as mean + SD of six independent experiments. (D) Lenalidomide was added to THP1 and HL60 cell lines cultures at 10 µM 18-24 hours before using microscopy to assess lytic granule polarization as described in Figure 4. References: Results are compared with control where only the solvent DMSO was added. Each symbol represents an individual experiment from six performed. *p < 0.05 (Wilcoxon's signed rank test). FIGURE 7. Effects of AML cell lines treatment with lenalidomide or NK-cell pre-activation with IL-2 or IL-15 on NK-cell degranulation. (A-C) NK cells were activated overnight with 10 ng/mL IL-15 or 1000 UI/mL IL-2 before (A) using flow cytometry to assess the number of conjugates formed as indicated in Figure 2, (B) determining the NK degranulation rate by flow cytometry as described in Figure 1, and (C) determining the percentage of conjugates showing granule polarization by microscopy as indicated in Figure 4. Data are shown as mean + SD of six independent experiments. (D) Lenalidomide was added to THP1 and HL60 cell lines cultures at 10 µM 18-24 hours before using microscopy to assess lytic granule polarization as described in Figure 4. Results are compared with control where only the solvent DMSO was added. Each symbol represents an individual experiment from six performed. *p < 0.05 (Wilcoxon's signed rank test). 35 35

https://openalex.org/W2594668300

https://www.intechopen.com/citation-pdf-url/53847

English

Clinical Lung Cancer Mutation Detection

InTech eBooks

cc-by

Chapter 5 Chapter 5 Chapter 5 Abstract As the promise of personalized medicine in the treatment of cancer begins to be realized, the diagnostic techniques needed to drive that revolution have continued to evolve. What started as optical imaging of banded chromosomes for karyotyping has progressed to DNA sequencing and now next‐generation sequencing capable of producing billions of reads. There are currently a large number of techniques that are used in the clinical labo‐ ratory for assessing the presence of mutations in lung tumors, all with their own strengths and weaknesses. Here, we survey the technology that is available and take a closer look at next‐generation sequencing. We discuss the instruments that are currently on the market and demonstrate the common workflow from patient to data. Additionally, the outside factors that influence the use of these technologies, from government regulation to insur‐ ance reimbursement, are presented. Keywords: detection, lung cancer, mutations, next‐generation sequencing © 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, © 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons © 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Clinical Lung Cancer Mutation Detection Stephan C. Jahn and Petr Starostik Stephan C. Jahn and Petr Starostik Stephan C. Jahn and Petr Starostik Stephan C. Jahn and Petr Starostik Stephan C. Jahn and Petr Starostik Stephan C. Jahn and Petr Starostik Additional information is available at the end of the chapter p Additional information is available at the end of the chapter Additional information is available at the end of the chapter http://dx.doi.org/10.5772/67168 http://dx.doi.org/10.5772/67168 1. Introduction Ultimately, cancer is a genetic disease of the DNA. Changes in chromosomal sequences result in altered gene expression, protein structure, and enzyme activity, leading to increased cell growth, motility, and the associated clinical symptoms. Every cancer is different, and under‐ standing the mutations present in each case is crucial in choosing the proper treatment. The field of molecular diagnostics, the detection of DNA abnormalities, has come a long way since 1902 when Theodor Boveri first noted through microscopic observations that cancer likely came from abnormal chromosomes [1]. From the discovery that DNA consists of nucle‐ otides, to the ability to sequence those nucleotides and the technology to do it more and more efficiently, our knowledge of cancerous mutations and their role in treating the disease © 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons © 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. tive Comm ve Comm A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 84 has grown. We have progressed from manually generating one data point at a time to using automation to create billions of them in a matter of hours. DNA abnormalities may be present in a number of forms. These range from the gain or loss of whole chromosomes to the substitution of one DNA nucleotide for another, leading to a change in the resulting protein. Between these extremes are chromosomal translocations, small or large deletions and insertions, amplifications, and inversions. There are multiple generations of techniques currently being utilized in the clinical labora‐ tory for the characterization of lung cancer. Each has its own inherent advantages and dis‐ advantages that must be considered prior to ordering tests. The results of these diagnostics have profound impacts on the treatment of individual patients. They are also being used for research into better understanding the disease and have been responsible for the generation of the International Cancer Genome Consortium and The Cancer Genome Atlas [2]. Their impact on medicine will certainly only increase in the future. 2. Scope of mutation testing in lung cancer The scope of mutation testing in lung cancer is determined by the mutation landscape described in these tumors. Considering which mutations need to be detected during diagnostic workup, two differentiations must be applied to mutations: driver mutations versus passenger muta‐ tions and therapeutically accessible versus inaccessible mutations. Driver mutations directly cause the development of cancer and allow it to grow and metastasize [3]. Other mutations that may either enhance the driver mutation or have no functional role in cancer progression are known as passenger mutations. As our knowledge of mutations is improving, the percent‐ age of known driver mutations detected in lung cancer is increasing. The percentage of lung adenocarcinomas with no detected driver mutation dropped from 40% in 2013 to 24% in 2016 [4], increasing the number of known driver genes from 10 to 15 in the same time span. Some of the diver mutations are “actionable” in that they are therapeutically accessible and may be targeted in treatment. It is such mutations that are primary candidates for mutation screening assays. A nice summary of gene mutations, treatment options, and clinical trials in lung can‐ cer is available from the mycancergenome.org website. Developing a lung cancer mutation assay with full coverage of the genes listed in the mycancergenome database would comprise a fairly comprehensive assay (Table 1). The absolute minimum of genes to be tested for muta‐ tions in lung cancer can be determined from the National Comprehensive Cancer Network (NCCN) Guidelines on Non‐Small Cell Lung Cancer [5], which recommends at least EGFR mutation and ALK translocation testing. The molecular characterization of lung cancer has focused primarily on those of the ade‐ nocarcinoma type for two main reasons. It is responsible for more deaths than any other form of cancer [6], and adenocarcinomas tend to be under the strong influence of identifiable driver mutations, making testing and treatment less nebulous than other cases. The three most common such mutations are of the epidermal growth factor receptor (EGFR), KRAS, Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 85 and anaplastic lymphoma kinase (ALK), with ERBB2 mutations also being quite common [7]. Less common mutations in lung cancer are reviewed in Ref. [8]. 2. Scope of mutation testing in lung cancer Gene Alteration Frequency in NSCLC (%) Available drugs AKT1 Mutation 1 Drugs in clinical development ALK Rearrangement 3–7 Drugs approved in NSCLC BRAF Mutation 1–3 Drugs approved in other cancer DDR2 Mutation ∼4 Drugs approved in other cancer EGFR Mutation 10–35 Drugs approved in NSCLC FGFR1 Amplification 20 Drugs in clinical development ERBB2 Mutation 2–4 Drugs approved in other cancer KRAS Mutation 15–25 Drugs in clinical development MEK1 Mutation 1 Drugs approved in other cancer MET Amplification 2–4 Drugs approved in NSCLC but for other molecular subtype NRAS Mutation 1 Drugs in clinical development PIK3CA Mutation 1–3 Drugs in clinical development PTEN Mutation 4–8 Drugs in clinical development RET Rearrangement 1 Drugs approved in other cancer ROS1 Rearrangement 1 Drugs approved in NSCLC Table 1. Frequency of mutations and availability of targeted therapies in NSCLC. Drug availability for the most frequent mutations detected in NSCLC is shown. Data are from mycancergenome.org. Table 1. Frequency of mutations and availability of targeted therapies in NSCLC. Drug availability for the most frequent mutations detected in NSCLC is shown. Data are from mycancergenome.org. Table 1. Frequency of mutations and availability of targeted therapies in NSCLC. Drug availability for the most frequent mutations detected in NSCLC is shown. Data are from mycancergenome.org. EGFR‐activating mutations represent a critical determinant for proper therapy selection in patients with lung cancer. There is a significant association between EGFR mutations, especially exon 19 deletions and exon 21 (L858R, L861), exon 18 (G719X, G719), and exon 20 (S768I) mutations, and sensitivity to EGFR inhibitors [9, 10]. A secondary mutation, T790M, is present in approximately 60% of tumors with acquired resistance to EGFR inhibitors [11] and confers resistance through steric interactions in the inhibitor binding site [12]. Primary resistance to EGFR‐targeted therapy is associated with KRAS mutations. That is why some laboratories choose to add KRAS mutation testing to their assays, as concurrent EGFR and KRAS mutations occur in <1% of patients with lung cancer, and KRAS mutations are associated with intrinsic EGFR inhibitor resistance. KRAS undergoes mutations resulting in single amino acid changes. In lung cancer, those mutations are most commonly present at the 12th amino acid position [13]. While KRAS muta‐ tions are the second most common mutation in lung cancer [7] and are especially prevalent in adenocarcinoma of the lung [13], their use as a prognostic marker or therapeutic guide in lung cancer has been limited. 2. Scope of mutation testing in lung cancer Studies have shown both a correlation between a KRAS mutation and a worse prognosis [14], but also no association with outcome [15]. ALK rearrangements A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 86 represent the fusion between ALK and various partner genes, including echinoderm micro‐ tubule‐associated protein‐like 4 (EML4) [16]. ALK fusions have been identified in a subset of patients with NSCLC and represent a unique subset of NSCLC patients for whom ALK inhibitors may represent a very effective therapeutic strategy. That is the reason why ALK translocation testing should be included in any basic lung cancer mutation panel. For the most part, ALK translocations and EGFR mutations are mutually exclusive. The parameters of tests employed in mutation analysis should fulfill the criteria suggested by professional organizations like the College of American Pathologists, International Association for the Study of Lung Cancer, and the Association for Molecular Pathology that have recently offered guidance on the use of molecular pathology in the screening of lung can‐ cer patients. Their recommendations focus largely on the importance of EGFR mutations and ALK rearrangements. While they encourage testing for these abnormalities in patients with early stage lung adenocarcinoma, leaving the decision to the physician and laboratory, they require testing in advanced lung adenocarcinoma. Additionally, these tests may be performed on lung tumors other than adenocarcinoma if there is reason to believe an oncogenic driver is likely to be found. 3. Current technologies to detect mutations There are multiple techniques that are currently used to detect mutations in clinical tumor samples. The specific technique used for each case is dependent on a number of factors including fundamental requirements such as the type of analysis to be performed (structural/ copy number changes vs sequence mutations), the length of sequence required, and the type of nucleic acid being sequenced (DNA or RNA), as well as practical considerations including amount and quality of the sample, turnaround time (TAT), and cost to carry out the analysis. While the ability exists to sequence a patient's entire genome, most clinical assays are more targeted, focusing on one specific nucleotide, a chosen gene, or a number of genes due to the cost and labor involved. 3.1.1. Cytogenetics Still widely used today, karyotyping was the first clinical mutation assay and was responsible for the identification of the Philadelphia chromosome in 1960, which is the result of a trans‐ location and is the defining aberration in chronic myeloid leukemia. When stained, generally with Giemsa stain, chromosomes develop a characteristic appearance of alternating light and dark “bands.” This allows for the documentation of gross changes in chromosome number or structure. Structural changes able to be identified include large insertions or deletions, sequence inversions, and translocations. The technique has been considerably improved over the years, including the ability to digitally photograph the chromosomes; however, the preparatory work‐up is still very labor intensive, and the analysis requires highly trained individuals. This limits the number of cells that can be analyzed for each sample, increas‐ ing the risk of false‐negative results in samples that are highly heterogeneous. Additionally, Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 87 87 as cells must be cultured to generate the necessary metaphase spreads, genotyping is mostly limited to rapidly dividing hematological malignancies and is only used sparingly in solid tumors such as lung cancer. 3.1.2. Fluorescence in situ hybridization Rather than using a generic dye, fluorescence in situ hybridization (FISH) uses single‐ stranded, fluorophore‐labeled DNA probes that hybridize to complementary regions in the genome and can be visualized as fluorescent spots on a metaphase spread or in an interphase nucleus. Due to the paired nature of human chromosomes, a FISH probe would be expected to identify two instances of the target sequence in each cell. Any numeric deviance from that indicates a loss or amplification of that locus (Figure 1A). Abnormal distances between probes targeting neighboring regions on the same chromosome can also indicate a translocation or an inversion has occurred. Figure 1. Techniques for measuring gross changes in chromosomes. (A) Fluorescent in situ hybridization (FISH) probes are applied to metaphase spreads or interphase cells. Complementary locations on the target chromosomes are identified as fluorescent spots due to the binding of the probe. Anything other than one locus on each of the sister chromosomes, represented by one fluorescent spot, is abnormal. This is shown at the chromosomal (top panel) and cellular (bottom panel) levels. (B) Array‐comparative genomic hybridization binds fluorescently dyed and fragmented genomic DNA to immobilized probes. As tumor and normal DNA are dyed different colors, when both are applied to the array, it will be visualized as yellow if there are equal quantities of DNA from each source. Overrepresentation of green or red indicates overrepresentation of the corresponding DNA source. The fluorescence is then plotted as a graph of copy number versus chromosome position. (C) The single‐strand conformation polymorphism utilizes the fact that wild‐type (WT) and mutant (M) DNA strands will adopt different conformations due to differing intramolecular forces. This will lead them to migrate at different rates during polyacrylamide gel electrophoresis. If a sample of tumor DNA contains a band that migrates at a different rate than a control sample, it is indicative of a mutation. A “normal” band is also expected from the tumor sample due to the likely heterogeneity of the sample. (D) As mutations will alter the forces holding complementary strands of DNA together, a melting curve analysis measures the fluorescence emitted by DNA with increasing temperature. Since the DNA dye only fluoresces when bound to double‐stranded DNA, as the DNA denatures the fluorescence decreases. Figure 1. Techniques for measuring gross changes in chromosomes. (A) Fluorescent in situ hybridization (FISH) probes are applied to metaphase spreads or interphase cells. 3.1.2. Fluorescence in situ hybridization Complementary locations on the target chromosomes are identified as fluorescent spots due to the binding of the probe. Anything other than one locus on each of the sister chromosomes, represented by one fluorescent spot, is abnormal. This is shown at the chromosomal (top panel) and cellular (bottom panel) levels. (B) Array‐comparative genomic hybridization binds fluorescently dyed and fragmented genomic DNA to immobilized probes. As tumor and normal DNA are dyed different colors, when both are applied to the array, it will be visualized as yellow if there are equal quantities of DNA from each source. Overrepresentation of green or red indicates overrepresentation of the corresponding DNA source. The fluorescence is then plotted as a graph of copy number versus chromosome position. (C) The single‐strand conformation polymorphism utilizes the fact that wild‐type (WT) and mutant (M) DNA strands will adopt different conformations due to differing intramolecular forces. This will lead them to migrate at different rates during polyacrylamide gel electrophoresis. If a sample of tumor DNA contains a band that migrates at a different rate than a control sample, it is indicative of a mutation. A “normal” band is also expected from the tumor sample due to the likely heterogeneity of the sample. (D) As mutations will alter the forces holding complementary strands of DNA together, a melting curve analysis measures the fluorescence emitted by DNA with increasing temperature. Since the DNA dye only fluoresces when bound to double‐stranded DNA, as the DNA denatures the fluorescence decreases. Figure 1. Techniques for measuring gross changes in chromosomes. (A) Fluorescent in situ hybridization (FISH) probes are applied to metaphase spreads or interphase cells. Complementary locations on the target chromosomes are identified as fluorescent spots due to the binding of the probe. Anything other than one locus on each of the sister chromosomes, represented by one fluorescent spot, is abnormal. This is shown at the chromosomal (top panel) and cellular (bottom panel) levels. (B) Array‐comparative genomic hybridization binds fluorescently dyed and fragmented genomic DNA to immobilized probes. As tumor and normal DNA are dyed different colors, when both are applied to the array, it will be visualized as yellow if there are equal quantities of DNA from each source. Overrepresentation of green or red indicates overrepresentation of the corresponding DNA source. The fluorescence is then plotted as a graph of copy number versus chromosome position. 3.1.2. Fluorescence in situ hybridization (C) The single‐strand conformation polymorphism utilizes the fact that wild‐type (WT) and mutant (M) DNA strands will adopt different conformations due to differing intramolecular forces. This will lead them to migrate at different rates during polyacrylamide gel electrophoresis. If a sample of tumor DNA contains a band that migrates at a different rate than a control sample, it is indicative of a mutation. A “normal” band is also expected from the tumor sample due to the likely heterogeneity of the sample. (D) As mutations will alter the forces holding complementary strands of DNA together, a melting curve analysis measures the fluorescence emitted by DNA with increasing temperature. Since the DNA dye only fluoresces when bound to double‐stranded DNA, as the DNA denatures the fluorescence decreases. A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 88 In lung cancer, FISH is the current FDA‐approved gold standard method for detecting trans‐ locations involving the ALK gene. ALK translocations were the second driver mutation described in NSCLC [16]. ALK partners in the translocation can vary, but the most frequently involved is the EML4 gene. The FISH test based on the Vysis ALK Break‐Apart FISH Probe by Abbott detects the translocation of ALK, but does not identify the fusion partner. This test consists of two probes that bind on either side of the common 2p23 break point. In the absence of a translocation, the probes will be seen as adjacent or overlapping. After a translation, the probes will be visualized with distinct separation. The test results are not always clear cut, and there is a large gray zone area of inconclusive results which can be influenced by subjec‐ tive judgment. The test is thus highly dependent on optimal sample quality and experienced medical technologist staff. With appropriate setup, FISH can be used for detection of other less frequent translocations or copy number changes in lung cancer. The technique has a fast TAT and is widely used. The largest drawback of FISH is throughput. Generally, one sample/set of probes is used per slide, making analysis of multiple samples/targets labor intensive. 3.1.3. Comparative genomic hybridization 3.1.3. Comparative genomic hybridization The sensitivity of karyotyping and FISH is limited by the magnification of the microscope since the results are visualized by the human eye. Comparative genomic hybridization (CGH) is an extension of karyotyping that utilizes fluorescent dyes and microscopy to improve the ability to detect smaller changes in chromosome structure. In this method, chromosomal material from the tumor is dyed with a fluorescent dye, such as red, while chromosomal material from a normal sample is labeled with a different fluorophore, such as green. The two samples are combined, denatured, and allowed to hybridize to a reference metaphase spread in classical CGH. Differential fluorescence (i.e., greater red than green) indicates more or less of that chromosomal segment in the tumor. This technique has now evolved into an array‐based assay. Array‐CGH utilizes the same prin‐ ciples, except the labeled input DNA is fragmented and hybridized to an array consisting of small probes rather than a metaphase spread. Where FISH hybridizes fluorescent probes to immobilized chromosomes, array‐CGH hybridizes fluorescent genomic DNA to immobilized probes. If the target locus is neither under‐ or over‐represented in the tumor sample, there will be equal numbers of normal and tumor fragments binding to the designated location on the array, and the equal mixture of red and green fluorescence will yield a yellow light. An imbalance of the target DNA in the tumor will lead to a shift in the observed fluorescence to red or green (Figure 1B). While a novel technique, array‐CGH has a number of limitations that have prevented it from widespread use in cancer diagnostics. The fundamental principles of the assay allow it to only detect quantitative differences in the amount of the target sequence. Chromosomal abnor‐ malities that involve only the rearrangement of DNA, such as translocations and inversions, cannot be detected. Additionally, the presence of non‐tumor cells in the tumor sample will dilute the mutation signal with normal background making the method less sensitive. For this reason, array‐CGH is commonly used in oncologic hematology for chronic lymphocytic Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 89 leukemia (CLL) [17], where a highly pure tumor sample can be obtained, but is rarely used in solid tumors such as lung cancer, which are generally very heterogeneous. Array‐CGH is slowly being phased out of the cancer field entirely due to the emergence of next‐genera‐ tion sequencing (NGS), which is discussed in‐depth below. 3.1.3. Comparative genomic hybridization The long TAT of NGS currently requires some samples to be analyzed with array‐CGH instead, but that will surely improve as the technology evolves. Outside of oncology, array‐CGH is routinely used for postnatal diagnostics where a homogeneous sample is readily available [18]. 3.2.1. Sequence mutation screening Prior to analyzing samples for specific mutations, it is possible to screen samples for the pres‐ ence of mutations using a qualitative assay yielding a yes or no answer. This is particularly useful when carrying out a large sequencing study as it allows the targeted use of more labor intensive and expensive techniques for only those samples most likely to contain the targeted mutation. These methods can be efficiently scaled to high throughput assays and are based on the chemical and physical principles of DNA structure. The conformation of single‐stranded DNA is a direct function of its nucleotide sequence; there‐ fore, mutant and wild‐type DNA adopt different conformations that result in differing migra‐ tion rates in polyacrylamide gel electrophoresis which can be detected using the single‐strand conformation polymorphism technique (Figure 1C). Similarly, denaturing high pressure liq‐ uid chromatography hybridizes potentially mutant DNA to wild‐type DNA. Subsequent ion pair, reverse phase chromatography, can then detect heterodimers as additional peaks due to differences in retention time [19]. Another common technique is melting curve analysis. Changes in sequence will alter the forces holding the double‐stranded DNA together, result‐ ing in differences in the temperature required to denature various loci [20]. By staining DNA with intercalating dyes that fluoresce only when bound to double‐stranded DNA, the loss of this signal can be monitored with increasing temperature, generating a graph of fluorescence versus temperature (Figure 1D). While unique fingerprints are sometimes seen for specific mutations, analysis generally consists of simply looking for the presence of any differences compared to the wild‐type DNA. If they are identified, other methods may be used to find the exact mutation present. 3.2.2. Allele‐specific PCR Techniques for analyzing single nucleotide mutations can be divided into those that give a direct yes/no readout as to the presence of a specific mutation and those that provide the exact DNA sequence, allowing the user to examine it for various mutations. Allele‐specific PCR is an example of the first type of assay. The original allele‐specific PCR utilized the fact that Taq polymerase is inefficient when there is a mismatch at the 3′ end of a primer hybridized to the target DNA. Therefore, if the primer is designed such that the final nucleotide will line up exactly with the genomic nucleotide in question, and its sequence is complimentary to the suspected mutation, a much larger quantity of PCR product will be yielded by a sample that contains the target mutation than by a sample that does not (Figure 2A). Output can be quan‐ A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 90 titated in a number of ways including electrophoresis or a follow‐up round of qPCR. Due to the exponential nature of PCR, this method is very sensitive, allowing for the detection of mutations in a sample diluted by a large number of normal cells. Figure 2. Allele‐specific PCR (A) In the original assay, a primer with a 3′ end that aligns perfectly with the nucleotide in question is designed such that it is complementary to the mutant allele and has a mismatch with the wild‐type allele. During PCR, elongation will only occur if the mutant allele is present. (B) The TaqMan approach to allele‐specific PCR utilizes a probe that will bind only to the mutant allele. The probe contains both fluorescent (F) and quencher (Q) epitopes and does not fluoresce when it is intact. A mutually complementary primer is used to initiate elongation upstream of the probe in both wild‐type and mutant alleles. If the elongation encounters the probe, it will release the fluorescent moiety, emitting a detectable signal. Figure 2. Allele‐specific PCR (A) In the original assay, a primer with a 3′ end that aligns perfectly with the nucleotide in question is designed such that it is complementary to the mutant allele and has a mismatch with the wild‐type allele. During PCR, elongation will only occur if the mutant allele is present. (B) The TaqMan approach to allele‐specific PCR utilizes a probe that will bind only to the mutant allele. 3.2.2. Allele‐specific PCR The probe contains both fluorescent (F) and quencher (Q) epitopes and does not fluoresce when it is intact. A mutually complementary primer is used to initiate elongation upstream of the probe in both wild‐type and mutant alleles. If the elongation encounters the probe, it will release the fluorescent moiety, emitting a detectable signal. Since its introduction, this method has been simplified, allowing the user to quantitate dur‐ ing the initial PCR step. This is typified by the TaqMan probe system from ThermoFisher Scientific. In this technique, a probe that is complimentary to a mutant, but not wild‐type, allele contains a fluorochrome reporter attached to its 5′ end and a quencher moiety on the 3′ end that prevents a fluorescent signal from the reporter being detected. An unlabeled primer upstream of the targeted sequence is used to initiate PCR. As the DNA polymerase elon‐ gates the strand starting from the primer, it degrades the oligonucleotide backbone of the TaqMan probe. The fluorochrome reporter emits fluorescence which can now be detected due to this spatial separation from the quencher [21] (Figure 2B). This method is highly efficient as it is one step, can easily be done in a 386‐well format with multiple probes and samples, and can be automated, resulting in a fast TAT. The assay itself requires only a qPCR instru‐ ment, and the reagents are relatively inexpensive. This ease and low cost have resulted in many labs creating their own laboratory‐developed tests using this technology, and there are a number of FDA‐approved kits available for purchase. As the TaqMan probe assay requires a large amount of genomic DNA per reaction, the number of different probes tested or other diagnostic procedures possible for the sample may be limited. Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 91 Having the lowest limit of detection and highest specificity, the Therascreen system from Qiagen combines both of the above methods. A mutation‐specific primer is used to amplify the mutated region, and then a Scorpion probe, containing fluorescent and quencher moi‐ eties, again binds to this same site. As elongation occurs, the epitopes are separated, and fluorescence occurs. FDA‐approved Therascreen kits are available for both EGFR and KRAS genes. However, these assays look only at a limited number of mutations. The EGFR assay can detect 5 point mutations, exon 19 deletions, and exon 20 insertions; the KRAS kit can detect 7 point mutations. 3.2.3. Sanger sequencing The first widely used sequencing method, Sanger sequencing employs PCR in the presence of unlabeled nucleotides and labeled (i.e., radio‐ or fluorescent labels) chain terminating nucleo‐ tides [24]. Since each form of nucleotide (A, T, C, G) is differentially labeled, incorporation of chain terminating nucleotides stops the elongation step and labels the product, indicating the identity of the final nucleotide added (Figure 3A). The sample is then subjected to electropho‐ resis, either gel or capillary, which separates the amplicons by size, allowing the sequence to be read from smallest to largest simply by the label present. 3.2.2. Allele‐specific PCR The largest flaw of these allele‐specific PCR assays is their pro‐ pensity for false‐negatives due to large mutations. The primers and probes are designed to detect single nucleotide mutations and anything greater than that will cause a mismatch and prevent hybridization. A 1% rate of false negatives has been reported for the FDA‐approved Therascreen KRAS assay [22]. These assays are promoted as companion assays for therapeu‐ tics, such as the BRAF assay from Roche, which is the companion diagnostic assay for therapy with BRAF inhibitor vemurafenib. The test only detects the c.1799T>A p.V600E mutation, though up to 20% of BRAF mutations are non‐V600E mutations [23]. 3.2.3. Sanger sequencing 3.2.4. Pyrosequencing Where Sanger sequencing uses chain termination followed by electrophoresis, pyrose‐ quencing uses a real‐time process of sequencing by synthesis [25]. Here, the four possible nucleotides are added and removed sequentially. When the proper nucleotide is added, elon‐ gation occurs, releasing pyrophosphate that is converted to light through bioluminescence (Figure 3B). The template DNA is immobilized through binding to a biotinylated primer bound to sepharose beads. The technique provides semiquantitative data in regard to preva‐ lence of the mutant allele in the sample [26]. As each template strand is elongated to comple‐ tion, smaller sample quantities are needed, generally only requiring 10 ng of DNA. For a number of reasons, read lengths obtained with pyrosequencing are not as long as those from Sanger sequencing, generally being well below one hundred bases [27]. This prevents its use for sequencing large‐scale mutations. Due to the fact that the technique requires a pre‐PCR step and manual analysis of the results, it is time intensive. However, the method is routinely used clinically, both as laboratory‐developed tests [28] and commercial kits [29] for analysis of a number of forms of cancer, including lung cancer. Its much lower limit of detection of 5–10% sets it apart from Sanger sequencing and makes it an optimal assay to confirm NGS variants occurring at low variant allele frequency. A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 2 92 Figure 3. Sequencing technologies. (A) Sanger sequencing consists of PCR of template DNA in the presence of standard and fluorescently labeled chain terminating nucleotides. If a chain terminating nucleotide is incorporated into the strand, it is permanently labeled with a tag that identifies the final nucleotide added. Following PCR, fragments are separated by size, and the sequence is read from smallest to largest strand. (B) Pyrosequencing utilizes a sequencing‐by‐synthesis approach during PCR. Nucleotides (A, T, C, G) are added sequentially and then removed. When the correct nucleotide is added, elongation occurs, and pyrophosphate is released, resulting in a luciferase reaction and the emission of light. Figure 3. Sequencing technologies. (A) Sanger sequencing consists of PCR of template DNA in the presence of standard and fluorescently labeled chain terminating nucleotides. If a chain terminating nucleotide is incorporated into the strand, it is permanently labeled with a tag that identifies the final nucleotide added. 3.2.4. Pyrosequencing Following PCR, fragments are separated by size, and the sequence is read from smallest to largest strand. (B) Pyrosequencing utilizes a sequencing‐by‐synthesis approach during PCR. Nucleotides (A, T, C, G) are added sequentially and then removed. When the correct nucleotide is added, elongation occurs, and pyrophosphate is released, resulting in a luciferase reaction and the emission of light. 3.2.5.1. Clinical NGS workflow Most NGS runs are conducted on either genomic DNA or mRNA. While formalin‐fixed and paraffin‐embedded tumor blocks contain suitable quality genomic DNA, the quality of mRNA obtained from these samples will be variable and yield unsatisfactory results in up to 15% of cases. Fresh tissue is required to ensure high‐quality mRNA for RNASeq experiments. Prior to beginning the workup, the tumor cell content of the sample must be determined. This requires an experienced and knowledgeable set of eyes. Between 5 and 20% tumor cell content is generally sufficient for NGS [31]. After isolation, DNA is sheared to produce uniform frag‐ ments which then undergo multiplex PCR or are hybridized to immobilized probes. The labor required for NGS is dependent on the level of automation available in the lab carrying out the sequencing. Illumina requires 1–2 days of workup, while Ion Torrent requires up to 3 days, although the Ion Torrent instrument has shorter run times. This means that individual labs need to identify their rate‐limiting step in order to choose the instrument that will provide the highest throughput and shortest TAT. 3.2.5. Next‐generation sequencing If one focuses on detection of only small mutations in a limited number of genes, single‐ana‐ lyte assays will deliver reliable and informative results. However, running mutation assays aiming to capture mutations in all 15 driver mutation‐prone genes requires a multianalyte assay, which is why the current direction of the sequencing field is to multiplex. While differ‐ ent technologies that would all be considered next‐generation sequencing (NGS) utilize very different sequencing methods, the similarity is that they are able to sequence incredibly large numbers of DNA strands at the same time. This massively parallel sequencing is capable of providing millions or billions of short reads. This ability lends NGS to a wide range of appli‐ cations including untargeted sequencing of entire genomes, exomes, or transcriptomes, as well as targeted sequencing of large numbers of cancer mutation locations at the same time. The NGS instrument market space is currently occupied by four competitors utilizing slightly different approaches. By far, the largest player is currently Illumina, who offers a wide range of instruments ranging, among others, from the miniSeq to the HiSeq X Ten, the latter of which is capable of producing up to six billion reads. A small range instrument, the MiSeqDx, is FDA approved for in vitro diagnostic use. These instruments use chemistry that is a million‐ wise multiplexed variant of Sanger sequencing with clonal amplification and sequencing by synthesis chemistry. The color of the signal detected when a nucleotide is added indicates the identity of the base added and thus the sequence. Currently, the only other company with a significant market share is ThermoFisher with their Ion Torrent instruments. Similar to traditional pyrosequencing, the Ion Torrent technology sequentially adds and removes Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 93 nucleotides looking for elongation. Instead of converting pyrophosphate to fluorescence, Ion Torrent measures small‐scale changes in pH caused by the release of H+ during the polym‐ erization of nucleotides [30]. Similarly, their Ion PGM™ Dx System is FDA approved. While both Pacific Biosciences and Qiagen have offerings for NGS, they are not currently widely used. The Pacific Biosciences instruments suffer from low throughput and high cost, while Qiagen only recently entered the market. 3.2.5.3. Whole exome sequencing While assays for sequencing the entire exome are available, they are not widely used in the clinical setting. As each run is capable of producing a set number of reads, sequencing a greater number of loci reduces the number of reads for each locus. Mutations in a highly het‐ erogeneous sample may not be detected. Exome sequencing may identify rare or previously unidentified mutations, though clinicians would likely not have the knowledge on how to use that information to better care for their patient, so it would have no impact on patient care. Therefore, large‐scale exome sequencing has found a better home in the research laboratory than the clinical lab setting. 3.2.5.2. Targeted sequencing While NGS has great potential, its cost and massive data generation currently limit its clinical use to targeted sequencing of mutation hotspots rather than genome, exome, or transcrip‐ tome sequencing. Disease‐specific gene panels are routinely used. The genes are chosen based on their immediate impact on patient care, either due to their use in determining prognosis or in selecting the proper treatment regimen. Sequencing of genes beyond the minimum is determined on a case‐by‐case basis according to the capability and throughput capacity of the laboratory as well as insurance coverage of the patient. Library preparation kits for NGS generally fall into two categories: those that use PCR amplification or hybridization probes. The first class is typified by Illumina's TruSeq Amplicon Cancer Panel, the Ion AmpliSeq panel from Ion Torrent, and HaloPlex kits from Agilent. The second group contains kits avail‐ able from sequencing instrument manufactures and also from Agilent, Roche, and Integrated DNA Technologies, along with other third parties. Direct hybridization gives more reliable quantitative data since the possibility of bias introduced by the pre‐PCR step is eliminated. These reagents can include probes for intronic regions rearranged in translocations which are detectable in this way [32], though larger rearrangements may lead to poor hybridization and false negatives [33] or misleading copy number data. The internal tandem duplications in the FLT3 gene in acute myeloid leukemia [34] are an excellent example of mutations that require a special reagent and bioinformatics approach. Generally, with adequate expertise, wet lab and bioinformatics support labs are capable of developing their own lab‐developed assays A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 94 as evidenced by cancer panels developed by Foundation Medicine (Foundation One Assay) [35], Memorial Sloan Kettering (Integrated Mutation Profiling of Actionable Cancer Targets) [36, 37], and many others. 3.2.5.4. Bioinformatics Perhaps the largest hurdle to the adoption of NGS is data analysis. When an instrument is providing billions of reads, it can be a challenge not only to store the data, but also to utilize them. Instrument manufacturers have greatly improved their software offerings, making it easier to view the data. However, the challenge still remains of how to use the data. With so much information at your fingertips, how do you best utilize it to better care for the patient? Without a meaningful approach, substantial time may be devoted to analyzing irrelevant data. The significance of identified mutations must be denoted, and their impact on treat‐ ment must be delineated. A number of companies have begun offering services that would annotate a mutation report with clinically relevant content, including N‐of‐One, Genome Oncology, and PireanDx. Once the report is generated, commonly outdated electronic medi‐ cal records systems often make the sharing of the results difficult. Due to file incompat‐ ibilities, highly complex documents are all too often printed, scanned, and uploaded as PDF documents. This has prompted data to be stored in third‐party systems that greatly enhance the user experience [37]. 3.2.5.5. Reimbursement and government regulation In addition to the difficult science, the advancement of NGS in the United States has been hampered by reduced reimbursement rates for molecular pathology testing. Reimbursement is now only available for the test itself, not the interpretation of the results. Additionally, the testing classification has been greatly simplified, failing to differentiate between similar tests that may have very disparate costs while not being redundant [38]. Diagnostic procedures that are conducted in clinical laboratories are regulated by the federal Clinical Laboratory Improvement Amendments (CLIA) that were established in the 1970s. These regulations are administered by federal agencies and by the states, which require laboratories to be properly licensed. Additionally, the College of American Pathologists [39], Association for Molecular Pathology [40], and the American College of Medical Genetics [41] all help to draft Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 95 technique‐specific regulations and offer certification programs of their own and require pro‐ ficiency testing. To date, most NGS tests have been developed by individual laboratories, over which the Food and Drug Administration (FDA) has no explicit regulatory oversight. The FDA has warned that NGS technologies, as with older technologies, will be subject to FDA regulation if they advance toward marketable in vitro diagnostic status, over which the FDA exerts considerable control. This is generally established on a case‐by‐case basis. Regulation by all of these bodies is complicated by the fact that NGS is continually evolving and improving. Guidelines must either be routinely revisited or generalized such that their interpretation may change to match the current technology. A fine line must be walked in order to protect the patients for whom the assays are being used, while at the same time not stifling the development of a technology that promises to usher in the era of personalized medicine. 4.1. Single‐analyte vs multianalyte assays While NGS will likely become the dominant technique for evaluating mutations in cancer moving forward, the other methods discussed here are hardly headed toward imminent obsolescence. Clinical labs must focus not only on the future, but also providing the best diag‐ nostics for current patients. Today's limitations of NGS, including substantial labor require‐ ments and a long TAT, mean that single‐analyte methods are still widely used. While NGS may be able to provide a greater quantity of data and increased sample throughput, it has not been found to be qualitatively better than other technologies including allele‐specific PCR and Sanger sequencing, among others, with all providing around 96% agreement [42, 43]. 4.3. Future sequencing technologies Even with NGS technologies still not up to full speed, even better techniques are already being developed. A number of these sequencing instruments are able to decipher the sequence of single, existing DNA strands directly, rather than being based on PCR syn‐ thesis [50]. Perhaps showing the most promise, Oxford nanopore technologies has devel‐ oped a new instrument, the MinION. By passing individual strands through nanopores containing ionic currents, the instrument is able to detect minute changes in current that differ depending on which nucleotide is passing by. The device is small enough to easily hold in the palm of the hand. It is capable of 60 kb read lengths, with 16,000 reads per run [51, 52]. While the portability and capabilities offer amazing promise, the accuracy is cur‐ rently not sufficient for clinical use. It is gaining significant traction in the research space, however [53]. 4.2. Liquid biopsy Direct testing of tumor biopsies is preferred; however, circulating tumor DNA may be used to test for mutations after a lung adenocarcinoma diagnosis through direct biopsy has been established. For solid tumors, mutation detection has traditionally required a direct biopsy. In recent years, research has been conducted on the presence of circulating tumor cells, excreted miRNAs, and even tumor genomic DNA floating freely in the bloodstream. The ease of obtaining a blood sample makes sequencing these samples an appealing future technique not only for tumor characterization, but also screening and evaluation of therapeutic efficacy during treatment. The ratio of mutant DNA to normal DNA or tumor to normal cells collected is often very low and seems to be partially dependent on the tumor type, being detectable in greater than 75% of patients with breast, colorectal, and hepatocellular cancers, among others, but in less than half of patients with cancer of the brain, kidney, or prostate [44, 45]. Circulating tumor cells were detected in 78% of small‐cell lung cancer patients [46], and both circulating DNA [47] and tumor cells [48] have been successfully used to screen for EGFR mutations in lung cancer. These technologies have not progressed to the point of widespread clinical use and are currently at the research stage. They are more thoroughly reviewed in Ref. [49]. A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 96 5. Conclusions Clinical mutation testing in lung cancer is driven by the available therapies. Currently, there is only a limited armoire of targeted therapy options available which can be employed. If there is not a drug targeting a specific mutated gene available, the fact that the gene shows a mutation is not very meaningful. EGFR‐targeted therapeutics may not be used without the presence of EGFR mutations. If the tumor continues to progress after treatment with an EGFR inhibitor, the presence of a resistance‐conferring EGFR T790M mutation must be determined prior to giving a third‐generation EGFR inhibitor to overcome the resistance. Similarly, ALK inhibitors are only to be given to patients showing ALK rearrangements; however, if treatment fails, there is no need to re‐evaluate mutational status. ROS1 inhibi‐ tors receive a similar recommendation, while BRAF, RET, ERBB2, KRAS, and MET screen‐ ings are only recommended if included as a larger panel or if EGFR, ALK, and ROS1 tests come back negative. Physicians and laboratories currently have a large number of techniques at their disposal for the interrogation of mutations present in lung cancer. The tests to be performed can be tailored to each patient based on a number of factors including type and number of data points needed, TAT, cost, and availability of quality biopsy material. The absolutely minimal mutation workup includes detection of SNVs and indels in EGFR by Sanger sequencing, pyrosequencing, allele‐ specific PCR or any other method, and detection of ALK translocations by FISH. However, NGS has the potential to replace many, if not all, of these techniques, but must first continue to improve. Its use in the clinic is currently limited by the fact that it is still quite expensive and has a relatively long TAT, and the results require special skills and reference databases in order to be fully utilized. In the future, it is the hope that every lung cancer patient may have their entire tumor genome sequenced cheaply and efficiently, allowing an in‐depth understanding of all mutations driving the tumor and providing information to tailor the best possible treatment for that patient. Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 97 Author details Stephan C. Jahn and Petr Starostik* *Address all correspondence to: [email protected] Department of Pathology, Immunology, and Laboratory Medicine, University of Florida College of Medicine, Gainesville, FL, USA Department of Pathology, Immunology, and Laboratory Medicine, University of Florida College of Medicine, Gainesville, FL, USA Department of Pathology, Immunology, and Laboratory Medicine, University of Florida College of Medicine, Gainesville, FL, USA References [1] Boveri T (2008) Concerning the origin of malignant tumours by Theodor Boveri. Translated and annotated by Henry Harris. J Cell Sci 121(Suppl 1): 1–84. [2] Cancer Genome Atlas Research Network (2013) Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med 368: 2059–2074. Stratton MR, Campbell PJ, Futreal PA (2009) The cancer genome. Nature 458: 719–724 [4] Johnson BE, Kris MG, Berry LD, Kwiatkowski DJ, Iafrate AJ, et al. (2013) A multicenter effort to identify driver mutations and employ targeted therapy in patients with lung adenocarcinomas: The Lung Cancer Mutation Consortium (LCMC). ASCO Annual Meeting Proceedings 31: 8019. [5] Network NCC NCCN Clinical Practice Guidelines in Oncology. Non‐Small Cell Lung Cancer (Version 4.2016). [6] Cancer Genome Atlas Research Network (2014) Comprehensive molecular profiling of lung adenocarcinoma. Nature 511: 543–550. [7] Tan D, Lynch HT (2013) Principles of Molecular Diagnostics and Personalized Cancer Medicine. Wolters Kluwer Health/Lippincott Williams & Wilkins. [8] Patel JN, Ersek JL, Kim ES (2015) Lung cancer biomarkers, targeted therapies and clinical assays. Transl Lung Cancer Res 4: 503–514. [9] Pao W, Miller V, Zakowski M, Doherty J, Politi K, et al. (2004) EGF receptor gene mutations are common in lung cancers from “never smokers” and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci USA 101: 13306–13311. [10] Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, et al. (2004) Activating mutations in the epidermal growth factor receptor underlying responsiveness of non‐ small‐cell lung cancer to gefitinib. N Engl J Med 350: 2129–2139. [11] Yu HA, Arcila ME, Rekhtman N, Sima CS, Zakowski MF, et al. (2013) Analysis of tumor specimens at the time of acquired resistance to EGFR‐TKI therapy in 155 patients with EGFR‐mutant lung cancers. Clin Cancer Res 19: 2240–2247. A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 98 [12] Wang J, Wang B, Chu H, Yao Y (2016) Intrinsic resistance to EGFR tyrosine kinase inhibi‐ tors in advanced non‐small‐cell lung cancer with activating EGFR mutations. Onco Targets Ther 9: 3711–3726. [13] Rodenhuis S, Slebos RJ (1990) The ras oncogenes in human lung cancer. Am Rev Respir Dis 142: S27–30. [14] Mascaux C, Iannino N, Martin B, Paesmans M, Berghmans T, et al. (2005) The role of RAS oncogene in survival of patients with lung cancer: a systematic review of the litera‐ ture with meta‐analysis. References Br J Cancer 92: 131–139. [15] D'Angelo SP, Janjigian YY, Ahye N, Riely GJ, Chaft JE, et al. (2012) Distinct clinical course of EGFR‐mutant resected lung cancers: results of testing of 1118 surgical specimens and effects of adjuvant gefitinib and erlotinib. J Thorac Oncol 7: 1815–1822. [16] Soda M, Choi YL, Enomoto M, Takada S, Yamashita Y, et al. (2007) Identification of the transforming EML4‐ALK fusion gene in non‐small‐cell lung cancer. Nature 448: 561–566. [17] Alsolami R, Knight SJ, Schuh A (2013) Clinical application of targeted and genome‐wide technologies: can we predict treatment responses in chronic lymphocytic leukemia? Per Med 10: 361–376. [18] Miller DT, Adam MP, Aradhya S, Biesecker LG, Brothman AR, et al. (2010) Consensus statement: chromosomal microarray is a first‐tier clinical diagnostic test for individ‐ uals with developmental disabilities or congenital anomalies. Am J Hum Genet 86: 749–764. [19] OD'Donovan MC, Oefner PJ, Roberts SC, Austin J, Hoogendoorn B, et al. (1998) Blind analysis of denaturing high‐performance liquid chromatography as a tool for mutation detection. Genomics 52: 44–49. [20] Ririe KM, Rasmussen RP, Wittwer CT (1997) Product differentiation by analysis of DNA melting curves during the polymerase chain reaction. Anal Biochem 245: 154–160. [21] De la Vega FM, Lazaruk KD, Rhodes MD, Wenz MH (2005) Assessment of two flexible and compatible SNP genotyping platforms: TaqMan SNP Genotyping Assays and the SNPlex Genotyping System. Mutat Res 573: 111–135. [22] Tol J, Dijkstra JR, Vink‐Borger ME, Nagtegaal ID, Punt CJ, et al. (2010) High sensitivity of both sequencing and real‐time PCR analysis of KRAS mutations in colorectal cancer tissue. J Cell Mol Med 14: 2122–2131. [23] Millington GW (2013) Mutations of the BRAF gene in human cancer, by Davies et al. (Nature 2002; 417: 949–54). Clin Exp Dermatol 38: 222–223. [24] Sanger F, Coulson AR (1975) A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J Mol Biol 94: 441–448. [25] Ronaghi M, Uhlen M, Nyren P (1998) A sequencing method based on real‐time pyro‐ phosphate. Science 281: 363, 365. Clinical Lung Cancer Mutation Detection http://dx.doi.org/10.5772/67168 99 [26] Kringen MK (2015) Analysis of Copy Number Variation by Pyrosequencing(R) Using Paralogous Sequences. Methods Mol Biol 1315: 115–121. [27] Mashayekhi F, Ronaghi M (2007) Analysis of read length limiting factors in pyrose‐ quencing chemistry. Anal Biochem 363: 275–287. [28] Poehlmann A, Kuester D, Meyer F, Lippert H, Roessner A, et al. References (2007) K‐ras mutation detec‐ tion in colorectal cancer using the pyrosequencing technique. Pathol Res Pract 203: 489–497. [29] Sundstrom M, Edlund K, Lindell M, Glimelius B, Birgisson H, et al. (2010) KRAS analy‐ sis in colorectal carcinoma: analytical aspects of pyrosequencing and allele‐specific PCR in clinical practice. BMC Cancer 10: 660. [30] Boland JF, Chung CC, Roberson D, Mitchell J, Zhang X, et al. (2013) The new sequencer on the block: comparison of Life Technology's Proton sequencer to an Illumina HiSeq for whole‐exome sequencing. Hum Genet 132: 1153–1163. [31] Tsongalis GJ, Peterson JD, de Abreu FB, Tunkey CD, Gallagher TL, et al. (2014) Routine use of the Ion Torrent AmpliSeq Cancer Hotspot Panel for identification of clinically actionable somatic mutations. Clin Chem Lab Med 52: 707–714. [32] Asan, Xu Y, Jiang H, Tyler‐Smith C, Xue Y, et al. (2011) Comprehensive comparison of three commercial human whole‐exome capture platforms. Genome Biol 12: R95. [33] Sboner A, Elemento O (2016) A primer on precision medicine informatics. Brief Bioinform 17: 145–153. [34] Rustagi N, Hampton OA, Li J, Xi L, Gibbs RA, et al. (2016) ITD assembler: an algo‐ rithm for internal tandem duplication discovery from short‐read sequencing data. BMC Bioinformatics 17: 188. [35] Frampton GM, Fichtenholtz A, Otto GA, Wang K, Downing SR, et al. (2013) Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nat Biotechnol 31: 1023–1031. [36] Cheng DT, Mitchell TN, Zehir A, Shah RH, Benayed R, et al. (2015) Memorial Sloan Kettering‐Integrated Mutation Profiling of Actionable Cancer Targets (MSK‐IMPACT): a hybridization capture‐based next‐generation sequencing clinical assay for solid tumor molecular oncology. J Mol Diagn 17: 251–264. [37] Hyman DM, Solit DB, Arcila ME, Cheng DT, Sabbatini P, et al. (2015) Precision medi‐ cine at Memorial Sloan Kettering Cancer Center: clinical next‐generation sequencing enabling next‐generation targeted therapy trials. Drug Discov Today 20: 1422–1428. [38] Ahlman JB, Connelly J, Crosslin R, Dimovski B, Espronceda M, et al. (2015) Current Procedural Terminology 2016. Chicago: American Medical Association. [39] Aziz N, Zhao Q, Bry L, Driscoll DK, Funke B, et al. (2015) College of American Pathologists’ laboratory standards for next‐generation sequencing clinical tests. Arch Pathol Lab Med 139: 481–493. A Global Scientific Vision - Prevention, Diagnosis, and Treatment of Lung Cancer 100 [40] Schrijver I, Aziz N, Farkas DH, Furtado M, Gonzalez AF, et al. References (2012) Opportunities and challenges associated with clinical diagnostic genome sequencing: a report of the Association for Molecular Pathology. J Mol Diagn 14: 525–540. [41] Rehm HL, Bale SJ, Bayrak‐Toydemir P, Berg JS, Brown KK, et al. (2013) ACMG clinical laboratory standards for next‐generation sequencing. Genet Med 15: 733–747. [42] Malapelle U, Sgariglia R, De Stefano A, Bellevicine C, Vigliar E, et al. (2015) KRAS mutant allele‐specific imbalance (MASI) assessment in routine samples of patients with metastatic colorectal cancer. J Clin Pathol 68: 265–269. [43] Tsiatis AC, Norris‐Kirby A, Rich RG, Hafez MJ, Gocke CD, et al. (2010) Comparison of Sanger sequencing, pyrosequencing, and melting curve analysis for the detection of KRAS mutations: diagnostic and clinical implications. J Mol Diagn 12: 425–432. [44] Bettegowda C, Sausen M, Leary RJ, Kinde I, Wang Y, et al. (2014) Detection of circulating tumor DNA in early‐ and late‐stage human malignancies. Sci Transl Med 6: 224ra224. [45] Leon SA, Shapiro B, Sklaroff DM, Yaros MJ (1977) Free DNA in the serum of cancer patients and the effect of therapy. Cancer Res 37: 646–650. [46] Punnoose EA, Atwal S, Liu W, Raja R, Fine BM, et al. (2012) Evaluation of circulating tumor cells and circulating tumor DNA in non‐small cell lung cancer: association with clinical endpoints in a phase II clinical trial of pertuzumab and erlotinib. Clin Cancer Res 18: 2391–2401. [47] Luo J, Shen L, Zheng D (2014) Diagnostic value of circulating free DNA for the detection of EGFR mutation status in NSCLC: a systematic review and meta‐analysis. Sci Rep 4: 6269. [48] Maheswaran S, Sequist LV, Nagrath S, Ulkus L, Brannigan B, et al. (2008) Detection of mutations in EGFR in circulating lung‐cancer cells. N Engl J Med 359: 366–377. [49] Speicher MR, Pantel K (2014) Tumor signatures in the blood. Nat Biotechnol 32: 441–443. [50] Thompson JF, Milos PM (2011) The properties and applications of single‐molecule DNA sequencing. Genome Biol 12: 217. [51] Jain M, Fiddes IT, Miga KH, Olsen HE, Paten B, et al. (2015) Improved data analysis for the MinION nanopore sequencer. Nat Methods 12: 351–356. [52] Ip CL, Loose M, Tyson JR, de Cesare M, Brown BL, et al. (2015) MinION Analysis and Reference Consortium: Phase 1 data release and analysis. F1000Res 4: 1075. [53] Laver T, Harrison J, OD'Neill PA, Moore K, Farbos A, et al. (2015) Assessing the perfor‐ mance of the Oxford Nanopore Technologies MinION. References Biomol Detect Quantif 3: 1–8.

https://openalex.org/W2790555614

https://revistas.ucm.es/index.php/CLAC/article/download/59065/4564456546500

es

El discurso público de la universidad: estudio socio pragmático contrastivo de Camerún, España y Francia

Círculo de lingüística aplicada a la comunicación

cc-by

ARTÍCULOS Círculo de Lingüística Aplicada a la Comunicación ISSN: 1576-4737 http://dx.doi.org/10.5209/CLAC.59065 El discurso público de la universidad: un estudio socio pragmático contrastivo de Camerún, España y Francia Carmen Pérez-Sabater1; Ginette Maguelouk Moffo2 Recibido: 20 de diciembre de 2016/ Aceptado: 22 de enero de 2018 Resumen. La informalización y conversalización del discurso público en la última parte del siglo XX es un fenómeno complejo y bien documentado en el mundo anglosajón. Este artículo parte de las propuestas de Fairclough (1995) para analizar si este fenómeno tiene lugar también en lenguas románicas con un desarrollo tecnológico similar en comparación con las prácticas discursivas de un país con una penetración de Internet en la vida de sus habitantes meramente simbólica. El estudio que presentamos compara el nivel de informalización del discurso público en Camerún, España y Francia estudiando las cartas de bienvenida a la universidad publicadas en las páginas web oficiales de estas instituciones académicas en español y francés. Los resultados muestran que el proceso de informalización es parejo en el caso de los dos países europeos, aunque las cartas de los rectores españoles tienen un grado de informalidad mayor. Asimismo, los resultados confirman que la baja penetración de Internet en Camerún parece ser una de las razones por las que los escritos en este país siguen estilos muy tradicionales y formales en lo que se refiere a las firmas y al uso de los pronombres de segunda persona. En general, la estructura y el estilo empleados en Camerún no se corresponden con las nuevas prácticas discursivas en Internet, a diferencia de lo que vemos en España y Francia. Finalmente nos gustaría decir que somos conscientes de las limitaciones de este trabajo que tienen que ver claramente con las restricciones que los cameruneses tienen en su acceso a Internet y ha circunscrito nuestro estudio a las páginas web oficiales de universidades. No obstante, esperamos que nuestro análisis del discurso electrónico sirva de lanzadera para futuros estudios del discurso electrónico en África y en otras sociedades no tan desarrolladas tecnológicamente pero que, sobre todo, enriquezca los estudios lingüísticos en lenguas románicas. Palabras clave: socio pragmática; comunicación electrónica; estudios contrastivos; informalización del discurso; discurso académico; “marketización” de la universidad. [en] The public discourse of the university: A contrastive sociopragmatic discourse analysis among Cameroon, France and Spain Abstract. The informalization and conversationalization of public discourse in the latter part of the twentieth century is a complex phenomenon that has been well documented in English speaking societies. This article parts from Fairclough’s (1995) premises to analyse how this phenomenon is taking place in Romance languages with a similar technological development in comparison to the discourse practices of a country where the penetration of the Internet in the everyday life of its _____________ 1 2 Universitat Politècnica de València E-mail: [email protected] Universitat Politècnica de València E-mail: [email protected] CLAC 73 2018: 177-196 177 178 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 inhabitants is merely symbolic. The study thus compares the level of informalization of the public discourse in Cameroon, France and Spain by studying the welcoming letters of the principals of universities to their students published on the official websites of these higher institutions. We see the formality and informality of greetings, sign-offs and signatures of letters as well as the use of the second person pronoun as clear parameters that can confirm the tendency towards informalization in present public discourse. The results have shown that the discourse practices of these European university authorities are similar in their degree of informality, although there seems to be a stronger preference for informal styles in Spain. The results also confirm that the low penetration of the Internet must have a decisive influence on the fact that the discourse practices in Cameroonian universities follow a very traditional style with very formal salutations and farewells and a very formal use of signatures and pronouns, since most welcoming letters have not been written for the web, unlike the letters published on the French and Spanish websites. We understand the limitations of the study, which are clearly provoked by the poor access of Cameroonians to the Internet and have restricted our study to the official webpages of universities. However, we hope that this novel analysis of Cameroonian Computer-mediated Communication discourse (CMC) can serve as a point of departure for similar studies on CMC in technologically under-developed societies, but, above all, we wish this piece of research to enrich the field of linguistic studies in Romance languages. Keywords: socio-pragmatics; Computer-mediated Communication (CMC); cross-cultural studies; informalization of discourse; academic discourse; marketization of universities Cómo citar: Pérez-Sabater, C. y G. Maguelouk Moffo (2018): El discurso público de la universidad: un estudio socio pragmático contrastivo de Camerún, España y Francia, en Círculo de lingüística aplicada a la comunicación 73, 177-196. Índice. 1. Introducción. 2. Contexto teórico. 2.1. La comunicación electrónica. 2.2. La comunicación electrónica en África y Camerún. 2.3. El multilingüismo en Camerún. 3. Hipótesis de investigación. 4. Metodología. 4.1. Justificación del corpus y su descripción. 4.2. Método de análisis llevado a cabo y parámetros empleados. 5. Resultados e interpretación. 5.1. Saludos iniciales. 5.2. Despedidas. 5.3. Firmas. 5.4. Pronombres personales de segunda persona. 6. Reflexiones finales. Bibliografía. Anejo 1. 1. Introducción Las nuevas formas de comunicación que nacieron y se desarrollaron en Europa y América del Norte pronto suscitaron el interés científico de lingüistas occidentales, interés especialmente notorio durante la última década del siglo XX y la primera del XXI. Las publicaciones de Herring (1996, 2004), Baron (1998, 2008) o Yus (2010) son claros ejemplos de la atención que el mundo académico presta a las nuevas formas de comunicación generadas a raíz del desarrollo de Internet y que implican nuevas prácticas discursivas de relevancia para la lingüística, la pragmática y la sociolingüística. A pesar del gran alcance académico de estas publicaciones, su área de influencia se limita, en gran medida, al mundo occidental, mientras que en otros continentes como el africano, todavía no han alcanzado su máximo desarrollo. En África, en efecto, el estudio de la repercusión de Internet en la vida diaria de los ciudadanos y en el ámbito académico está en un estadio incipiente, exceptuando algunos casos como el de Chiluwa (2010a, 2010b), aunque poco a poco va ganando espacio y despertando el interés de la comunidad universitaria. En Camerún, la ausencia de investigaciones sobre la comunicación electrónica es aún si cabe más patente por la escasa penetración de Internet que ha obligado a que nuestro estudio se centre en un género Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 179 específico de la comunicación electrónica: las cartas de bienvenida a las instituciones de educación superior. En estos momentos, otras herramientas de comunicación como el correo electrónico o las redes sociales no forman parte de la vida diaria de la universidad camerunesa. Para esta investigación partimos de los estudios de autores como Biber y Finnegan (1989) quienes han constatado la tendencia a utilizar estilos más orales e informales en distintos registros del inglés escrito. Al respecto, Leech (1966) habla de lo que denomina coloquialización, es decir, el uso de cada vez más formas coloquiales en el discurso público en inglés. Fairclough (1995) estudia la coloquialización del discurso público en inglés y la denomina informalización, proceso que conlleva la mezcla de géneros y que está provocado, en parte, por el desarrollo de las nuevas tecnologías. En mucha menor medida, este proceso de informalización del discurso público ha llamado la atención de académicos de lenguas románicas. Por ejemplo, Cervera (2001) destaca varios casos donde se aprecia la oralización del discurso escrito en español y para el francés, Pires (2003) analiza la informalización del discurso público de la prensa y los anuncios en Francia. Sin embargo, mientas que el fenómeno de la informalización del discurso, asociado, entre otras cosas, al desarrollo de las nuevas herramientas tecnológicas, se ha documentado extensamente en el mundo académico anglosajón, requiere más dedicación en las lenguas románicas. En el caso de África, la tendencia a la coloquialización o informalización del discurso público universitario no se ha estudiado apenas. Partiendo del escaso desarrollo tecnológico del continente y, en especial de Camerún, el objetivo de esta investigación será comprobar hasta qué punto influye el nivel de desarrollo tecnológico de un país en el grado de formalidad de los escritos públicos académicos. Para tal fin analizaremos en las webs de centros educativos de enseñanza superior de Camerún los saludos de los rectores, directores o decanos a modo de cartas de presentación de sus universidades, escuelas o facultades. Estos documentos electrónicos en francés se compararán con presentaciones de centros de enseñanza superior de España y Francia, que, al igual que la parte francófona de Camerún, son países de lenguas románicas pero con una penetración de Internet en la vida diaria de sus habitantes y del mundo académico muy superior, con niveles parecidos al mundo anglosajón. Según http://www.internetworldstats.com/ en, febrero de 2015 sólo el 6.4% de los cameruneses tiene acceso a Internet mientras que en España y Francia ronda el 80% de la población. En 2016, el 17,7% de los cameruneses accede a Internet. El estudio pragmalingüístico llevado a cabo comparará las estrategias discursivas en estos tres ámbitos y analizará el grado de formalidad/informalidad en la comunicación (cfr. Pérez-Sabater et al. 2008, Yus 2010 y Lorenzo-Dus y Bou-Franch 2013). El foco de atención en elementos opcionales de la comunicación en línea como son los saludos, despedidas y firmas es relevante para estudiar las relaciones jerárquicas en la comunicación electrónica, como también lo es su inclusión o no en este tipo de discurso (Bou-Franch, 2011). Nuestra investigación se enmarca en lo que llamamos “cuarta generación” de estudios electrónicos, aquellos que tratan lenguas diferentes del inglés y 180 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 comparan prácticas discursivas en varias lenguas. La necesidad académica de realizar análisis lingüísticos contrastivos de textos electrónicos se defiende, por ejemplo, en Lorenzo-Dus y Bou-Franch (2013), que tendría como objetivo recopilar y promover el desarrollo de un conjunto significativo de investigaciones sobre comunicación electrónica comparada con un marcado carácter socio cultural. Pasamos ahora a justificar y enmarcar esta investigación en su contexto teórico. 2. Contexto teórico 2.1. La comunicación electrónica Este trabajo se orienta en los estudios sobre la comunicación electrónica que se han llevado a cabo en las tres últimas décadas, investigaciones que han ido evolucionando de consideraciones más generales a cuestiones más específicas. Por ejemplo, si en 1998 Baron buscaba cómo caracterizar el discurso electrónico usado en el correo electrónico, en 2015 McKeown y Zhang limitan su análisis a los saludos y despedidas de una sola compañía británica. Además, aunque hasta hace apenas diez años el análisis del discurso electrónico se centró casi exclusivamente en la comunicación electrónica en lengua inglesa, es palpable la evolución de esta rama de la lingüística que alcanza una dimensión más global. El papel de la revista Journal of Computer-Mediated Communication en este cambio ha sido determinante al publicar un número especial sobre el Internet multilingüe. Este número editado por Danet y Herring en 2003 incluye investigaciones en las que se comparan el español y el catalán (Climent et al., 2003), así como análisis relativos al japonés, al griego, al tailandés y al árabe, entre otras lenguas. Otros trabajos que ejemplifican las investigaciones llevadas a cabo en esta generación son los de Montero-Fleta et al. (2009) donde se analiza el grado de formalidad de varias lenguas románicas y el inglés en foros de periódicos; o los trabajos de Lorenzo-Dus y Bou-Franch (2013), en los que se compara el papel de la formalidad/informalidad en el español peninsular y el inglés británico en correos académicos. Estas investigaciones son ejemplo del impacto que en los últimos años ha adquirido el estudio contrastivo como el que presentamos a continuación, relativo a la comunicación electrónica en África y en Camerún. Antes cabe hacer una pequeña introducción a la situación de multilingüismo en este país africano. 2.2. La comunicación electrónica en África y Camerún Mientras que en los países occidentales la repercusión de los estudios lingüísticos sobre la comunicación en Internet es manifiesta, en África podemos destacar la labor llevada a cabo por algunos autores nigerianos como Chiluwa (2010a, 2010b), quien ha publicado y editado varias obras con repercusión internacional que tratan de la creación de la identidad nigeriana en los correos electrónicos en inglés de Nigeria o sobre la pragmática de las llamadas cartas nigerianas. En Camerún, los estudios sobre la comunicación por Internet están en un estadio muy inicial. Sólo hemos encontrado tres publicaciones rigurosas: Feussi (2007), Tabe (2008) y Mbah (2010). En Feussi (2007) se analizan las prácticas discursivas del francés de Camerún en los mensajes de texto, correos Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 181 electrónicos y chats, y las elecciones lingüísticas conscientes de grafía y léxico. Tabe (2008) estudia el correo electrónico y propone varias prácticas en el aula para aprender lenguas. Por su parte, Mbah (2010) trata el impacto de las nuevas tecnologías en la enseñanza y el aprendizaje de la lengua, y pone de manifiesto el papel que las nuevas tecnologías tienen en la mejora del aprendizaje en la universidad, comparando su uso en hombres y mujeres. 2.3. El multilingüismo en Camerún En África, gran parte de la sociedad es multilingüe, a diferencia de Europa y los países occidentales donde el multilingüismo y la educación multilingüe no son una verdadera necesidad sino más bien una elección (Harrow y Mpoche, 2008). Como señalan estos autores, Camerún en particular es uno de los países más multilingües del planeta. Echu (2004) detalla que Camerún es un país en el que se hablan 247 lenguas indígenas que conviven con el inglés y el francés (las dos lenguas oficiales) y también el pidgin inglés (la principal lengua franca). El francés sigue siendo el idioma oficial dominante y el idioma más usado en la vida cotidiana del país, dado que se habla y se comprende por el 80% de la población (Bema-Nemedeu, 2010), mientras que el inglés es relegado a un segundo plano. Echu (2004) saca a la luz el desequilibrio en la práctica de las dos lenguas oficiales a pesar del bilingüismo oficial. Esto ha generado en noviembre de 2016 varias revueltas políticas con huelgas generales y manifestaciones en las zonas anglófonas del país. En este trabajo, nos centraremos en el francés estándar como principal lengua de comunicación en las universidades del país, especialmente a nivel escrito. 3. Hipótesis de la investigación Tradicionalmente se ha asumido que las prácticas discursivas en tipos específicos de comunicación electrónica son homogéneas, como indica BouFranch (2011). Veremos si las prácticas discursivas empleadas en las cartas de rectores y decanos de Camerún, España y Francia comparten el mismo estilo cuando presentan su institución a los internautas. A tal fin, como ya se ha comentado en la introducción, nuestro estudio empírico parte de Fairclough (1995) quien relaciona manifiestamente los avances tecnológicos con la informalidad del discurso público y veremos si el escaso desarrollo de Internet en Camerún influye de manera directa en el grado de formalidad de los escritos públicos académicos en Internet. La cuestión central que nos ocupa es si la comunicación electrónica en francés de Camerún, español de España y francés de Francia presenta un estilo homogéneo en el mismo género electrónico. Este análisis contrastivo entre comunidades que comparten una lengua, Camerún y Francia, y otra lengua románica como el español peninsular plantea una hipótesis que parte de estudios previos sobre la informalidad del discurso (por ejemplo, Fairclough, 1995): en Camerún encontraremos un mayor grado de formalidad en la presentación de las universidades y centros de enseñanza superior que en las presentaciones de centros educativos de enseñanza superior de España y Francia debido al poco desarrollo de Internet en ese país, a la poca penetración de Internet en la población y a la difícil integración de esta 182 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 tecnología por motivos sociales, económicos, políticos y culturales. Por el contrario, en España y Francia el proceso de informalización del discurso, originado en países anglosajones (Fairclough, 1995), está documentado y se manifiesta en los últimos años como señalan Kerbrat-Orecchioni (2007) y Briz (2014). 4. Metodología 4.1. Justificación del corpus y su descripción Como ya se detalló en el marco teórico, en Camerún se hablan varias lenguas locales pero hay dos lenguas oficiales: inglés y francés, lengua esta última que prevalece en el entorno social, en la política y en la economía (Echu 2004). Esto se debe a que de las diez regiones de este país existen sólo dos que tienen el inglés como lengua oficial (en las otras siete predomina el francés). En las regiones anglófonas, la información institucional también se debe trasmitir en francés. El papel predominante de la lengua francesa en la universidad camerunesa es la razón principal por la que hemos escogido mensajes originalmente escritos en francés con el fin de obtener la mayor cantidad posible de ejemplos para el corpus; en total, hemos encontrado en todas las universidades y centros de educación superior del país 20 cartas que contienen 16280 palabras (muchos centros educativos cameruneses no tienen páginas web o páginas web que funcionen regularmente). Con el fin de equilibrar el corpus en la medida de lo posible, al igual que en Camerún, en España y Francia hemos seleccionado 20 mensajes de universidades españolas que contienen 7215 palabras en total y 20 bienvenidas francesas con 7389 palabras (el listado de las universidades se encuentra en el Anejo 1). Estas instituciones están situadas en distintas zonas del país con el fin de formar un corpus más representativo del país que no se circunscribiera a una región concreta. Se trata de 20 cartas de rectores, decanos y directores de universidades, facultades y centros de educación superior en francés estándar de Camerún, en español peninsular de España y en francés de Francia. Sentimos que en este artículo no se puedan relacionar otras variedades como el francés de Canadá o el español de América por ser inabarcable en este momento. Esperamos y deseamos que su inclusión pueda ser factible en otro análisis más amplio y con más investigadores a su cargo en un futuro cercano. 4.2. Método de análisis llevado a cabo y parámetros empleados. Muchos de los análisis de la comunicación electrónica de los últimos años se basan en las propuestas de Herring (2004), quien recoge la variedad de niveles, temas, fenómenos y métodos de análisis de esta rama de la lingüística y los engloba en lo que denomina CMDA, siglas en inglés de Computer-mediated Discourse Analysis o análisis del discurso de la comunicación electrónica. En este trabajo partimos de lo que propone Herring (2004) para estructurar nuestro examen pragmalingüístico. Así pues, partimos de la observación de la estructura del discurso electrónico teniendo como objeto de estudio el continuum que va de la formalidad a la informalidad. Sabemos que hay muchos otros parámetros que pueden ayudar a identificar estilos informales en el discurso electrónico (véase, por ejemplo la categorización del texto escrito Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 183 oralizado en Yus, 2010) pero, por razones básicamente de espacio, para esta investigación el fenómeno examinado son las convenciones de uso para los saludos, despedidas y las firmas. Usamos el método descriptivo, analizamos el estilo empleado, el contexto lingüístico y las relaciones pragmáticas subyacentes. En concreto, los saludos y las despedidas, como parámetros indicadores de cortesía, permiten expresar familiaridad y distancia con el receptor del mensaje. Son estrategias necesarias para personalizar mensajes y el medio idóneo para reforzar el estatus del autor (Waldvogel, 2007). Además, como indica BouFranch (2011), debido a su opcionalidad en la comunicación electrónica, el estudio de las convenciones estilísticas empleadas en los saludos y las despedidas es de interés académico, como también lo es su inclusión o no en el discurso electrónico. En las siguientes tablas se explica cómo estos elementos estructurales de cortesía se han contabilizado partiendo de la escala del grado de formalidad e informalidad utilizada por Pérez-Sabater et al. (2008) basada en estudios anteriores sobre estos elementos de cortesía como los de Duthler (2006) o Waldvogel (2007). Siguiendo los valores de formalidad/informalidad que estos autores asignaban a correos electrónicos institucionales, hemos adaptado estos parámetros a nuestro corpus según detalla la Tabla I. Saludos y despedidas Valor Saludo muy formal y/o introducción al centro educativo muy formal 1 Despedida muy formal Saludo formal y/o introducción al centro educativo formal 0.7 Despedida formal Saludo informal y/o introducción informal Despedida informal 0.5 Sin saludos ni introducciones Sin despedida 0 Tabla I. Tabla de formalidad e informalidad adaptada de Pérez-Sabater et al. (2008). Por ejemplo, si el rector o rectora de la universidad empezara su bienvenida diciendo: “Estimados alumnos y alumnas, les deseo que su incorporación a nuestra universidad …” le asignaríamos el máximo valor de formalidad porque usa la introducción típica de las cartas en papel y se dirige a los futuros alumnos de usted. Mientras que si el rector se despidiera así: “Estos son los rasgos generales de nuestro centro educativo, ahora solo espero que la conozcas personalmente. Hasta pronto”, la puntuación asignada sería un poco menor debido al uso de la segunda persona informal o familiar. De igual forma, la Tabla II refleja los valores establecidos para las firmas de las cartas institucionales de nuestro corpus como indicadores de cortesía opcionales y a la vez claros marcadores de las relaciones jerárquicas en una institución (BouFranch, 2011). 184 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 Firma Nombre del rector/decano/a y cargo académico Sólo nombre o cargo académico Ni nombre ni cargo académico Valor 1 0.5 0 Tabla II. Valores asignados a la firma de la carta de presentación. A firmas como la siguiente: “Dra. Encarna Fernández Igual. Rectora de la Universidad de X” se le asignaría la máxima puntuación. En el caso de terminar la carta de bienvenida con sólo el nombre o el cargo, la puntuación sería de 0.5. Por otra parte, tendremos también en cuenta otro parámetro ligado a la formalidad/informalidad del discurso público apuntado por Fairclough (1995): el uso de los pronombres de segunda persona en el discurso público académico que contribuyen en gran manera a la apariencia de que se reducen marcadores claros de poder, lo que él denomina “synthetic personalization” o el engaño que consiste en hacer creer que el discurso público es una conversación privada cara a cara. Si Fairclough señala la importancia de los pronombres de segunda persona como mecanismos decisivos en el proceso de informalización en inglés, estos son más determinantes aún si cabe en español y francés, ya que el autor puede elegir la segunda persona formal o informal a la hora de dirigirse al lector, como subraya Pires (2003). En el artículo emblemático de Brown y Gilman (1960) sobre los pronombres se distinguía entre el tú de intimidad y el usted de formalidad y asociaban su uso a las relaciones de poder simétricas y no simétricas de los interlocutores, llamándole a esta distinción T/V (del francés tu/vous). Blas Arroyo (1994) señala que los significados corteses en español peninsular se suelen equiparar con usted mientras que tú se asocia con parámetros psicosociales como la cercanía, la solidaridad, la confianza y la familiaridad. Este autor comenta el auge progresivo del uso familiar en las relaciones, es decir, el tuteo. Sin embargo, hay muchos factores que entran en juego a la hora de elegir el pronombre formal o el informal en las prácticas discursivas actuales. Al respecto, Peeters (2004) concluye su artículo sobre el uso formal o informal de la segunda persona de hoy en día en francés diciendo que la elección del pronombre formal o informal es más complicada que nunca. En la comunicación electrónica, Baron (1998) destaca la frecuencia de segundas personas en el correo electrónico. Asimismo, otros autores como Shortis (2001) han destacado que el uso de la segunda persona, que hace referencia directa al lector, marca el carácter conversacional de estos escritos electrónicos y es clave en el proceso de informalización del discurso actual. Para el estudio de los pronombres personales de segunda persona utilizaremos una herramienta de análisis automático de corpus, en concreto WordSmith Tools, que nos permitirá medir cuantitativamente la frecuencia de la segunda persona formal e informal. 5. Resultados e interpretación Los resultados obtenidos se recogen en la Tabla III que engloba las cifras sobre formalidad en los tres países y las dos lenguas observadas en los saludos iniciales, las despedidas y la firma. Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 Parámetros Saludos iniciales Despedidas Firma Camerún 0.91 0.91 0.95 185 España 0.76 0.74 0.55 Francia 0.81 0.86 0.92 Tabla III: Grado de formalidad. Por otra parte, hemos examinado si estas cartas presentan algún tipo de saludo, despedida y firma en cada corpus, que son segmentos opcionales según BouFranch (2011) y, por consiguiente, elementos estructurales de interés. La Tabla IV resume el porcentaje de estos elementos estructurales de cortesía en las cartas institucionales. Frecuencia Saludos iniciales Despedidas Firma Camerún 100% 100% 100% España 100% 90% 55% Francia 100% 100% 95% Tabla IV. Frecuencia de elementos estructurales de cortesía. En general, el análisis realizado muestra que los escritos electrónicos que nos ocupan presentan un estilo muy formal en Camerún y Francia y un poco menos formal en España. Las diferencias entre los tres corpus son evidentes y no necesitan de ningún estudio estadístico adicional: los saludos iniciales de las bienvenidas españolas son 15% menos formales que en Camerún, las despedidas un 17% y las firmas un 40% menos formales. Comparativamente con las bienvenidas francesas, éstas son un 5% menos formales en la introducción, un 12% menos formales en la conclusión y un 37% en la firma final. Es importante señalar a su vez que, globalmente, la mayor parte de las cartas analizadas recuerdan a las cartas tradicionales en la estructura y estilo, pero no presentan saludos al uso de las cartas formales, ya que sólo hemos encontrado algunos casos de Estimados alumnos/alumnas. La segunda tabla nos indica que se puede concluir que los tres corpus incluyen normalmente algún tipo de introducción y saludo inicial dando la bienvenida a la web. En el parámetro de las firmas las cifras demuestran diferencias muy destacadas entre los tres corpus, de especial intereses es el hecho de que sólo la mitad de las cartas españolas las incluye. Por otro lado, el análisis de los pronombres de segunda persona mediante WordSmith refuerza los resultados de los elementos estructurales de cortesía: tanto en las cartas de Camerún como las de Francia encontramos sólo pronombres formales/plurales de segunda persona, ningún tous aparece en los dos corpus en francés. Por el contrario, en España se recogen varios usos de la segunda persona informal. Por consiguiente, nuestra hipótesis de partida, que postulaba más grado de formalidad en Camerún que en España y Francia debido, principalmente, al escaso desarrollo de las nuevas tecnologías en ese país, ha sido confirmada con los resultados obtenidos. Los datos han remarcado el carácter formal de la 186 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 comunicación escrita camerunesa en el entorno académico estudiado. Por otra parte, el análisis también ha dado lugar a algunos resultados menos esperados: la relativa informalidad de este género en las universidades españolas en comparación con el grado de formalidad tan elevado en Francia, ambos países con un desarrollo tecnológico similar y un proceso de informalización parecido que, en principio, tendrían que arrojar resultados muy similares. Investigaciones previas han señalado que el grado de formalidad en las lenguas románicas es más elevado que en otras lenguas como el inglés (véase, por ejemplo, Montero-Fleta et al., 2010). Esto es debido, principalmente, a que tradicionalmente en estas lenguas la división entre géneros orales y escritos es más marcada mientras que en prácticas discursivas anglosajonas, la frontera entre lo oral e informal o lo escrito formal no está definida en muchos casos, como demuestran Machin y van Leeuwen (2007) quienes hablan de fronteras bien definidas en las prácticas discursivas en español. Los resultados presentados ahora sugieren que el grado de oralización o informalización del discurso electrónico en español es patente en un género tradicionalmente muy formal, a diferencia de publicaciones anteriores sobre la comunicación electrónica en español en otros géneros en línea (véase, por ejemplo, Montero-Fleta et al., 2009). A continuación relacionaremos los resultados obtenidos en cada parámetro con trabajos similares que puedan servir para contextualizar nuestros datos en un entorno más amplio. Nos gustaría aclarar que, a consecuencia del escaso número de publicaciones sobre la comunicación electrónica en Camerún, la referencia a las prácticas discursivas utilizadas por estas comunidades es muy escasa. 5.1. Saludos iniciales Chiluwa (2010b) indica en su análisis del estilo de correos electrónicos de cartas nigerianas que las despedidas generalmente son actos de habla muy expresivos. En nuestro caso, los resultados de los tres corpus estudiados resaltan dos aspectos: su formalidad y el hecho de que la gran mayoría de las cartas de bienvenida contiene algún tipo de saludo inicial y/o introducción. La formalidad imperante en el corpus está en consonancia con estudios similares en los que los mensajes destinados a varios interlocutores obtienen índices de formalidad claros, como en el caso de los correos institucionales estudiados por Pérez-Sabater et al. (2008). Esto es así porque la persona que actúa como máxima figura de la institución académica es consciente de que en su escrito representa a la universidad o facultad y, de ahí, su alto grado de formalidad; es una forma de construir la identidad institucional, como señala Bou-Franch (2011). Los resultados obtenidos relacionan claramente la formalidad de los saludos iniciales de los nuevos géneros electrónicos con los ya existentes, en este caso, con las cartas comerciales (Yates, 2000). Al igual que en las cartas comerciales tradicionales, los saludos iniciales son claros indicadores estructurales de cortesía que buscan establecer o mantener una relación grupal, en cierta forma, comercial, como futuros clientes de dicha organización. Así, la totalidad de las cartas de bienvenida presenta algún tipo de saludo inicial o introducción, como sucede en el 93% de los correos electrónicos en español recibidos por profesores (cfr. BouFranch 2011): estas cifras están en clara oposición con otros trabajos realizados con corpus de organizaciones anglosajonas, como el caso de Waldvogel (2007), en el Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 187 que sólo el 59% de los correos electrónicos analizados contiene un marcador de cortesía inicial. En cuanto al estilo, las prácticas discursivas en las lenguas románicas suelen ser mucho más formales que en inglés. Estudios relacionados con el tema han subrayado que en español y francés los internautas recurren a indicadores de cortesía estilísticamente más formales que los escritos electrónicos en inglés por razones de índole social y cultural (Montero-Fleta et al., 2010). La formalidad es clara en la universidad francesa, aunque se observan ciertos cambios en la cortesía de los intercambios escritos en entornos académicos. Al respecto, KerbratOrecchioni (2007) comenta que los saludos iniciales de los correos electrónicos formales y semi-formales tienen un marcado carácter oral y se parecen más a los intercambios de cortesía cara a cara que a los saludos propios de las cartas tradicionales, ya que, en la mayoría de casos estudiados, se inicia la comunicación con un Bonjour. En nuestro caso, sin embargo, se observan claros marcadores de formalidad en este parámetro como el uso de Madame, Monsieur o Chères étudiantes, chers étudiants en la bienvenida al centro académico: (1) Chères étudiantes, chers étudiants, Je vous souhaite la bienvenue à l'Université Joseph Fourier sur le site de Valence e vous souhaite la bienvenue à l'Université Joseph Fourier sur le site de Valence. Ejemplo 1. Saludo inicial en la Universidad Joseph Fourier en Francia. En el corpus de Camerún, el grado de formalidad es, aún si cabe, mayor que en los otros dos países. Véase, por ejemplo, el saludo inicial de una universidad privada del país: (2) Message du recteur C’est pour nous un réel plaisir de vous souhaiter la bienvenue à l’Université Adventiste Cosendai (UAC), la deuxième université privée à avoir été agréée par le Gouvernement Camerounais. Nous avons l’assurance que l’information que vous allez lire sur ce site vous sera d’une grande utilité pour vous permettre de faire le bon choix pour vos études. Ejemplo 2. Saludo inicial de la Universidad Adventiste Cosendai (UAC). La necesidad de reafirmar la formalidad de la institución académica a la que se representa es mayor porque la distinción entre géneros formales e informales está muy patente en la sociedad camerunesa, donde la universidad requiere de los ritos necesarios como máxima institución académica del país. Además, sólo se usa en todos los casos francés estándar y no se ha encontrado ninguna palabra de la lengua franca o de cualquier otra lengua local, como es frecuente en otros géneros en línea del país como los correos electrónicos o los chats (Feussi, 2007). Por último, es de destacar que, a diferencia de trabajos similares sobre el español, el estudio de las cartas de rectores españoles arroja resultados imprevistos: el grado de formalidad de los saludos es sorprendentemente más bajo de lo esperado. Tradicionalmente, investigaciones sobre el español han demostrado que existen y se mantienen fronteras bien controladas entre la alta cultura y la popular, rehusandose, en la mayoría de casos, la mezcla de géneros formales e informales (Machin y van Leewen, 2007). En nuestro estudio empírico, por el contrario, sí se observan ciertos estilos informales en contextos muy formales, como sería el caso 188 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 de esta carta de bienvenida que comienza directamente con esta oración separada del resto del cuerpo de la carta: “Quiero darte la bienvenida a la Universidad de La Rioja”. 5.2. Despedidas En cuanto a las despedidas, el estilo elegido en cada grupo del corpus es similar a lo obtenido en los saludos. En Camerún y Francia tenemos despedidas mucho más formales que en España. Lo más característico de este apartado es la gran variedad de fórmulas empleadas para terminar la carta de bienvenida: desde las neutras, en las que se vuelve a recordar a los lectores que son bienvenidos y bienvenidas a la institución universitaria, como la del rector de la Universidad de Salamanca: “¡Os doy la más cordial bienvenida a la Universidad de Salamanca!”, a las que subrayan el papel de dicha institución en la sociedad que la rodea, como la despedida de una facultad de la Universidad de Douala en Camerún: “A la Faculté des Sciences à Douala, nous investissons surtout pour l’avenir de notre pays et de l’humanité toute entière”. Chiluwa (2010b) comenta que la formalidad de las despedidas en las cartas nigerianas analizadas tiene clara relación con la seriedad del mensaje, al igual que en las cartas comerciales. De igual modo, en el estudio de Kerbrat-Orecchioni (2007) sobre correos electrónicos formales o semi-formales entre académicos, en contraposición al marcado carácter oral de los saludos, las despedidas recuerdan mayoritariamente a las cartas convencionales francesas con el uso generalizado del adverbio cordialement como fórmula predominante de despedida. Sin embargo, en general, nuestros resultados presentan despedidas que tienen un estilo similar a los saludos, resultados que están en consonancia con Yongyan (2002) donde generalmente las despedidas van acorde con el estilo de los saludos iniciales. En cuanto a la estructura, hay que destacar que en la mayoría de casos (100% en Camerún y Francia, y 90% en España) observamos algún tipo de fórmula que sirve para concluir lo expuesto, fórmula de cortesía indispensable en las despedidas de los correos electrónicos académicos franceses en palabras de Kerbrat-Orecchioni (2007). De nuevo, como ocurría en los saludos iniciales, la incorporación de despedidas en la mayoría de las cartas es destacable si comparamos nuestros resultados con el bajo número de despedidas encontradas en los correos electrónicos académicos analizados por Waldvogel (2007), donde sólo el 34% presentaba algún tipo de despedida; los correos intercambiados en una fábrica obtenían aún índices más bajos, sólo el 17% incluía una despedida. Por el contrario, nuestros resultados son similares a otros estudios en contextos académicos, como el de Bou-Franch y Lorenzo-Dus (2008), donde casi el 100% de los mensajes de correo enviados en instituciones académicas británicas y españolas tenían una despedida. En el artículo posterior de Bou-Franch (2011) sobre mensajes en español, los resultados sobre el porcentaje de despedidas encontradas son algo inferiores, 78%. Por último, hay que dedicar mención especial a cómo los rectores y decanos de las instituciones camerunesas deciden terminar sus escritos. La identificación de la universidad con la nación camerunesa implica que estas despedidas terminen, en algún caso, como el discurso oral de un ministro o del presidente del gobierno. Así, leemos que en la Universidad de Douala, la carta de bienvenida se despide con arengas propias de un discurso oral ante una masa de gente: “Vive l'Université de Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 189 Douala! y Vive le Cameroun!”, despedida de marcado carácter oral en un contexto escrito muy formal. En otros casos, los logros de la universidad se identifican con los del país y también con los del continente africano. La universidad es, por tanto, el emblema de una cultura e identidad nacional que emerge de la formación de la nación-estado (Giménez, 1993). 5.3. Firmas De los tres parámetros estudiados como elementos opcionales en la comunicación oral según Bou-Franch (2011) son las firmas las que más pueden aportar a los estudios pragmalingüísticos de la comunicación electrónica, ya que en las cartas de rectores y decanos su inclusión sí es verdaderamente redundante. En la mayoría de casos, la página web donde está la carta de presentación incorpora la foto del rector y su nombre como encabezado de la página. Por lo tanto, la inclusión de nombre y cargo nuevamente como final de la presentación de la universidad refuerza aún más si cabe la formalidad del escrito y, por ende, de la institución a la que se representa. Asimismo, el uso de la firma para terminar este episodio social conlleva y refuerza las relaciones de poder, como señala Bou-Franch (2011). En general, los resultados muestran que las firmas son estilísticamente muy formales en Camerún y Francia, obteniendo en ambos casos un mayor nivel de formalidad que los saludos y las despedidas estudiadas. En España, por el contrario, la incorporación de firmas a estas cartas no es consistente ya que en muchos casos no hay firma y, cuando se incluye, no aporta toda la información de la máxima autoridad educativa, es decir, no muestra su titulación académica sino que simplemente consiste en el nombre y cargo de forma muy escueta. Esto demuestra que la comunicación electrónica en español peninsular presenta un grado menor de formalidad en estos casos. Finalmente, cabe señalar que lo más destacado es el estilo elegido para terminar este episodio social en Camerún. En España y Francia se suele firmar con el nombre y cargo como “Bertrand Monthubert, Président de l'Université Toulouse III - Paul Sabatier” o “Francisco J. Mora Mas, Rector de la UPV”. Sólo en el caso de la rectora de la universidad de Málaga hemos encontrado un Dª que precede al nombre, en ningún caso el tratamiento que les corresponde oficialmente en España, el de ilustrísimo señor o señora. Esto puede deberse a que tanto en España como en Francia el rector tiene que ser doctor y/o catedrático, con lo que añadir esta información no aportaría nada nuevo al lector. Sin embargo, en Camerún se indica en casi la totalidad de los casos el grado académico y la categoría profesional del firmante como en la bienvenida de la Universidad Cosendai: “Prof. Joseph I. Masinda, PhD”. En lo que respecta a las relaciones de poder, Bema-Nemedeu (2010) afirma que los africanos y los cameruneses no suelen ser distantes en el habla coloquial. En la universidad, sin embargo, la firma sirve, como bien señalaba Bou-Franch (2011), para marcar y reforzar el poder social del firmante y su distancia social con el futuro estudiante. En este corpus, la distancia social entre el autor y el lector es más marcada en la comunicación camerunesa en clara oposición a los textos en español donde, en muchos casos, no hay firma. Esto forma parte del proceso de informalización de las prácticas discursivas en español y de lo que Fairclough (1995) también denomina democratización discursiva que conlleva la reducción de marcadores de diferencias de poder entre personas de poder institucional distinto. 190 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 Tendencia que empezó en países anglófonos en las últimas décadas del siglo pasado y que, al menos en el género de comunicación estudiado, no ha comenzado aún en Camerún. 5.4. Pronombres personales de segunda persona En Camerún, la estadística de WordSmith nos señala que no hay ningún ejemplo de la segunda persona singular informal, lo que recalca la formalidad de la comunicación electrónica académica del país africano. En cuanto al plural, en francés no hay diferencia entre ustedes y vosotros por lo que no podemos aportar ningún dato. En Francia, al igual que en Camerún, no hemos encontrado ningún caso del pronombre de segunda persona informal. En España, se muestran otros niveles de formalidad, ya que es común, a diferencia de los demás casos estudiados, encontrar la segunda persona del singular tú, tus o te. Como recoge el siguiente ejemplo, el decano y la administración se dirigen a los alumnos con te, invitándoles a formar parte del proyecto común representado por nuestro, un posesivo claramente inclusivo (Labbé, 1998). Lo que añade un matiz de informalidad a una bienvenida, por otra parte, bastante formal. (3) La página web de la Facultad de Biología de la Universidad de Salamanca es la puerta de entrada a nuestro Centro, permíteme que en primer lugar te agradezca haber cruzado su umbral y que como Decano te transmita una cálida bienvenida y el deseo de que tu interés no sea defraudado… Quedo a tu disposición, hasta pronto. Ejemplo 3. Carta del decano de la Facultad de Biología de la Universidad de Salamanca. Con todo, lo más destacado del análisis ha sido ver cómo tanto en España como en Francia estas cartas recuerdan el estilo del lenguaje publicitario que adopta las características de la comunicación impersonal cuya finalidad es la de persuadir al grupo e influir en él a partir del uso de los términos que fortalecen las relaciones sociales como calidad, reconocimiento, etc. En español se observa el uso de los adjetivos posesivos tú, sus, nuestros que da una falsa imagen de familiaridad entre la universidad y los alumnos con el fin de reforzar las relaciones del grupo. Los verbos de modalidad en la primera persona del plural valoramos, apreciamos, invitamos enfatizan la nueva orientación de la universidad centrada en el valor al alumno de la universidad. En francés, el siguiente ejemplo muestra un evidente lenguaje publicitario, marcado, entre otras estrategias, por la inclusión del lector en la institución universitaria con el énfasis puesto en la institución. Además, el emisor trasmite al lector el entusiasmo, lo privilegiado que es y que su incorporación a la universidad cumplirá con todas sus expectativas. Por tanto, el uso de los pronombres personales vosotros y los pronombres posesivos vuestras llaman la atención de los alumnos sobre la importancia de su presencia en la universidad. Asimismo, nosotros y nuestros establecen las relaciones de familiaridad que se entablaran entre la administración y los estudiantes con un marcado carácter promocional. Igualmente, el uso de los verbos afectivos (plaisir, heureux, souhaiter, accueillir) sirve para venderles a los alumnos su elección por esa universidad. (4) À quelques jours de la rentrée universitaire étudiante, j’ai le plaisir de vous accueillir, au nom de l’établissement et de l’équipe présidentielle, et Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 191 de vous souhaiter la bienvenue à l’Université Lumière Lyon 2, qui devient aujourd’hui «votre» établissement. Parmi vous se trouvent peutêtre des habitués des couloirs de Lyon 2 qui, au gré des réussites aux concours (et je les en félicite chaleureusement), restent dans nos murs. Quoi qu’il en soit, que cette université vous soit familière, que vous connaissiez déjà l’enseignement supérieur ou non, que vous releviez des corps techniques, administratifs, enseignants, chercheurs ou personnels de bibliothèque, je suis heureux de vous compter parmi nous. Ejemplo 4. Pronombres de la Universidad Lumière Lyon 2 en Francia. La tendencia actual a considerar la universidad gobernada y determinada por el mercado (Bayne y Ross, 2007), tiene una clara relación con lo que llamábamos personalización sintética (Fairclough, 1995), es decir, la falsa creencia de que el rector o persona que gobierna una institución se dirige a nosotros personalmente creando una falsa idea de informalidad y cercanía, en español, además, claramente con el uso de la segunda persona informal. Esta tendencia ha sido analizada ampliamente en los medios de comunicación de masas anglosajones, pero poco en los medios de masas en lenguas románicas. Ya en 1993, Fairclough analizaba varios ejemplos de textos universitarios donde se destacaba la informidad de las entonces nuevas prácticas discursivas universitarias. En este trabajo también se remarcaba la nueva orientación comercial de estas instituciones británicas donde la función promocional de la universidad ya era omnipresente en comparación con los textos tradicionales. En nuestro corpus, como ejemplificaba Fairclough (1993) en el suyo, la personalización de la institución mediante nosotros marca la actividad promocional ligada al marketing, principalmente en las universidades españolas y francesas. Las universidades camerunesas todavía no han entrado de lleno en esta exigencia promocional por lo que se desprende del análisis de estas cartas de bienvenida que, en muchos casos, no han sido escritas para la web, sino que son discursos que el rector ha utilizado con anterioridad, por ejemplo, para inaugurar unos juegos universitarios, y alguien ha decidido que podrían servir de bienvenida o son simplemente resúmenes que recuerdan memorandos tradicionales como esta bienvenida de la Universidad de Ngaounderé de la que incluimos el final: (5) Fort de ces atouts, l’Université de Ngaoundéré s’ouvre sur de nombreux projets. Ceux en vue, sont le lancement du concours du meilleur chercheur, l’ouverture de l’Ecole de Chimie, la création de la filière sciences de l’éducation, la sécurisation complète du campus universitaire, l’achat des cars du transport pour enseignants et étudiants, l’acquisition des équipements neufs et dignes pour sportifs, l’intégration pleine et définitive de la jeune fille ou de l’étudiant handicapé; toutes choses qui concourent à offrir un cadre de vie agréable et adéquat pour les études. Professeur AMVAM ZOLLO Paul Henri Recteur de l'Université de Ngaoundéré Ejemplo 5. Bienvenida Universidad de Ngaoundéré. Por último, cabe mencionar brevemente la longitud del corpus recopilado. Si en Camerún 20 cartas de bienvenida incluyen 16280 palabras pero en España y Francia esas 20 cartas sólo abarcan la mitad de palabras, nos encontramos que los rectores de los escritos cameruneses necesitan el doble de palabras que el resto para la misma función promocional o no de su institución. Esto está íntimamente ligado 192 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 con la informalidad del discurso escrito y la evolución de las prácticas discursivas actuales que priman la brevedad y que han provocado que la longitud media de frase en el mundo anglosajón se haya reducido mucho en los últimos años (Baron, 2000). Escribir para la red exige que los textos que se van a leer tengan una organización que facilite el acceso a la información de forma rápida y lógica (Nielsen, 1999). Leer la bienvenida a la universidad en una pantalla y no varias como en el caso de Camerún, facilita la lectura; lo contrario indica que lo que se incluye en esas páginas web no se ha concebido expresamente para ese sitio en línea. 6. Reflexiones finales El análisis lingüístico de las cartas de bienvenida a la universidad nos ha revelado que el proceso de informalización del discurso, en marcha en la universidad española pero menos desarrollado en la universidad de Francia, es prácticamente inexistente en países como Camerún donde la penetración de Internet en la sociedad es meramente simbólica. Además, el énfasis en la función promocional del discurso público de la universidad, claramente relacionado con la informalización del discurso según Fairclough (1993), tampoco se encuentra al mismo nivel en los tres corpus estudiados en este momento. Estudios posteriores pueden dar lugar a otros resultados donde esta tendencia pueda variar. Sabemos que las limitaciones provocadas por la escasa penetración de Internet en la vida diaria y académica de Camerún, que ha restringido nuestro análisis de la comunicación electrónica en la universidad a las cartas de bienvenida, puede restar valor a nuestros resultados al no poder compararlas con otros textos en línea del mundo académico. Sin embargo, pensamos que, con todo, esta investigación tiene relevancia y puede servir de lanzadera para el estudio de la comunicación electrónica en ese país y en cualquier otro donde el desarrollo de Internet sea similar. Con el tiempo, el interés de los cameruneses por las comunicaciones en línea crecrá, promoverá un mayor uso de las nuevas tecnologías y posibilitará el estudio del discurso académico en Facebook o Twitter, por ejemplo. Siguiendo los estudios de Feussi (2007) sobre el correo electrónicos y los chats, veremos entonces si en nuestra investigación futura el plurilingüismo y multilingüismo de Camerún también tienen repercusión en escritos electrónicos publicados en herramientas que, en un futuro posiblemente cercano, se incluyan en las páginas web de las universidades del país como Facebook o Twitter, al estilo de las universidades occidentales de hoy en día. Concluimos con la esperanza y el deseo de que las investigaciones realizadas sean de interés para la comunidad académica de los países del estudio y, en general, para los estudios lingüísticos aplicados a las prácticas discursivas de la comunicación electrónica. Bibliografía Baron, N. S. (1998). “Letters by Phone or Speech by other Means: the Linguistics of Email.” Language and Communication, 18, 133-170. Baron, N. S. (2000). Alphabet to email: How written English evolved and where it's heading. Londres: Routledge. Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 193 Baron, N. S. (2008). Always on: Language in an online and mobile world. Nueva York: Oxford University Press. Bayne, S., y Ross, J. (2007). “The ‘digital native’ and ‘digital immigrant’: A dangerous opposition”. Paper presented at the Annual Conference of the Society for Research into Higher Education (SRHE), Brighton, Sussex. Bema-Nemedeu, C. E. (2010). “Pratiques linguistiques, statut des langues et interactions entre locuteurs sur les marchés de Douala (Cameroun)” (Tesis doctoral, Universidad de Besançon). Biber, D. y Finegan, E. (1989). “Drift and the evolution of English style: a history of three genres”. Language, 65/3, 487-517. Blas Arroyo, J. L. (1994). “Los pronombres de tratamiento y la cortesía”. Revista de Filología de la Universidad de la Laguna, 13, 7-36. Bou-Franch, P. (2011). “Openings and closings in Spanish email conversations”, Journal of Pragmatics, 43/6, 1772-1785. Bou-Franch, P. y Lorenzo-Dus, N. (2008). “Natural versus elicited data in crosscultural speech act realisation: The case of requests in Peninsular Spanish and British English”, Spanish in Context, 5/2, 246-277. Briz, A. (2014). “Hablar electrónicamente por escrito”, CHIMERA: Journal of Romance Corpora and Linguistic Studies, 1, 77-89. Brown, R. y Gilman, A. (1960). “The Pronouns of Power and Solidarity”, en T.A. Sebeok (ed) Style in Language. Cambridge, Ma: MIT Press, 253-275. Cervera Rodríguez, A. (2001). “La Irrupción del Coloquialismo en Internet y las Nuevas Tecnologías”, en http://cvc.cervantes.es/obref/congresos/valladolid/ponencias/nuevas_fronteras_del_ es. [fecha acceso: 20.1.2015]. Climent, S., Moré, J., Oliver, A., Salvatierra, M., Sànchez, I., Taulé, M. y Vallmanya, L. (2003). “Bilingual newsgroups in Catalonia: A challenge for machine translation”, Journal of Computer‐Mediated Communication, 9/1. Chiluwa, I. (2010a). “Nigerian English in informal email messages”, English WorldWide, 31/1, 40-61. Chiluwa, I. (2010b). “The pragmatics of hoax email business proposals”, Linguistik Online, 43/3, 4. Danet, B. y Herring, S. C. (2003). “Introduction: The multilingual internet”, Journal of Computer-Mediated Communication, 9/1. Duthler, K. W. (2006). “The Politeness of Requests made Via Email and Voicemail: Support for the Hyperpersonal Model.” Journal of Computer-Mediated Communication, 11/2. Echu, G. (2004). “The language question in Cameroon”, Linguistik Online 18/1, en https://bop.unibe.ch/linguistik-online/article/view/765/1309. [fecha de acceso: 17/02/2015]. Fairclough, N. (1993). “Critical discourse analysis and the marketization of public discourse: The universities”. Discourse & Society, 4/2, 133-168. Fairclough, N. (1995). Critical Discourse Analysis. Londres: Longman. Feussi, V. (2007). “A travers textos, courriels et tchats: des pratiques de français au Cameroun”, Via text messages, emails and chat [French language practices in Cameroon”] Glottopol, 10, 70-85. Giménez, G. (1993). “Apuntes para una teoría de la identidad nacional”, Sociológica, 8/21, 26-27. Harrow, K. W., y Mpoche, K. (2008). “Language, literature and education in multicultural societies: Collaborative research on Africa”, Cambridge Scholars Publishing, 7, 381, en https://www.h-net.org/reviews/showrev.php?id=32558 [fecha acceso: 10.4.2015] 194 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 Herring, S. C. (Ed.). (1996). Computer-mediated communication: Linguistic, social, and cross-cultural perspectives. Amsterdam: John Benjamins Publishing 39. Herring, S. C. (2004). “Computer-mediated discourse analysis: An approach to researching online behavior”, en S. A. Barab, R. Kling y J. H. Gray (eds.) Designing for virtual Communities in the Service of Learning. Nueva York: Cambridge University Press, 338–376. Kerbrat-Orecchioni, C. (2007). “La “cyberpolitesse”: Formes de l’adresse, ouverture et clôture dans les courriers électroniques”, Quaderns de Filologia-estudis Lingüístics, 12, 35-56. Labbé, D. (1998). “Le ‘nous’ du Général de Gaulle”, Quaderni di studi linguistici, 4, 331-354. Leech, G.N. (1966). English in advertising. Londres: Longmans. Lorenzo-Dus, N. y Bou-Franch, P. (2013). “A cross-cultural investigation of email communication in Peninsular Spanish and British English: The role of (in) formality and (in) directness”, Pragmatics and Society, 4/1, 1-25. Machin, D. y van Leeuwen, T. (2007). Global Media Discourse: a Critical Introduction. Londres y Nueva York: Routledge. Mbah, T. B. (2010). “The Impact of ICT on students study habits. Case study: University of Buea, Cameroon”, International Journal of Science and Technology Education Research, 1/5, 107-110. McKeown, J. y Zhang, Q. (2015). “Socio-pragmatic influence on opening salutation and closing valediction of British workplace email”. Journal of Pragmatics, 85, 92107. Montero-Fleta, B., Montesinos-López, A., Pérez-Sabater, C., y Turney, E. (2009). ‟Computer mediated communication and informalization of discourse: The influence of culture and subject matter”, Journal of Pragmatics, 41/4, 770-779. Montero-Fleta, B., Turney, E., Pérez-Sabater, C., y Peña-Martinez, G. (2010). ‟Fronteras borrosas entre géneros en la comunicación electrónica: un estudio multilingüe”, Revista Española de Lingüística, 40/2, 47-74 Nielsen, J. (1999). Designing web usability: The practice of simplicity. Thousand Oaks, Ca: New Riders Publishing. Peeters, B. (2004). “Tu ou vous?”, Zeitschrift für französische Sprache und Literatur, 114, 1-17. Pérez Sabater, C., Turney, E., y Montero Fleta, B. (2008). ‟Orality and literacy, formality and informality in email communication”, Ibérica, 15, 71-88. Pires, M. (2003). Usages et stratégies de tutoiement dans l’écrit public. Actas del Coloquio Pronombres de segunda persona y formas de tratamiento en las lenguas de Europa. Tabe, C. A. (2008). “Orality and Literacy in Cameroon E-Mail Discourse”. Language, Literature and Education in Multicultural Societies: collaborative Research on Africa. Newcastle: Cambridge Scholars Publishing, 194-209. Yates, S. J. (2000). “Computer-Mediated Communication, The Future of the Letter?” en D. Barton y N. Hall (eds.) Letter Writing as a Social Practice. Amsterdam y Filadelfia: John Benjamins. Yongyan, L. (2002). “Writing for international publication: The perception of Chinese doctoral researchers”, Asian Journal of English Language Teaching, 12, 179-193. Yus, F. (2010). Ciberpragmática 2.0: El uso del Lenguaje en Internet. Barcelona: Ariel Lingüística. Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 195 Anejo Listado de las universidades cuyas cartas de bienvenida han formado el corpus: I- En Camerún 1. Université de Douala 2. Faculté des Sciences (U. de Douala) 3. Faculté de Génie Industriel (U. de Douala) 4. Université de Ngaoundéré 5. Faculté des Sciences Juridiques et Politiques (U. Ngaoundéré) 6. Université Adventiste Cosendai (UAC) 7. Université de Yaoundé I 8. Institut Universitaire de la Côte 9. Institut National de la Jeunesse et des Sports (INJS) 10. Ecole de Faune de Garoua (EFG) 11. Ecole Supérieure des Sciences Economiques et Commerciales (ESSEC) 12. Institut Siantou Supérieur 13. Institut des Relations Internationales du Cameroun (IRIC) 14. Ecole Nationale d’Administration et de Magistrature (ENAM) 15. Institut Supérieure de Traducteurs et d’Interprétariat (ISTI) 16. Université Evangélique du Cameroun (UEC) 17. Institut Supérieur de Traduction, d’Interprétation et de Communication de Yaoundé (ISTIC) 18. Université des Montagnes (UdM) 19. Université de Yaoundé II 20. Institut Supérieur de Commerce et de Gestion Siantou (ISCGS) II. En España 1. Universitat Pompeu Fabra 2. Universidad Carlos III 3. Universidad de Salamanca 4. Facultad de Biología (U. de Salamanca) 5. Universidad de Vigo 6. Universidad de Zaragoza 7. La Universidad Politécnica de Madrid 8. Universitat Politècnica de Catalunya 9. Universitat Politècnica de València 10. Universitat Jaume I 11. Universitat de Barcelona 12. Universidad de Sevilla 13. Universidad de Cádiz 14. Universidad del País Vasco 15. Universidad de Santander 16. Universidad de Castilla-La Mancha 17. Universidad de León 196 Pérez-Sabater, C.; Maguelouk Moffo, G. CLAC 73 2018: 177-196 18. Universidad de La Rioja 19. Universidad de Extremadura 20. Universidad de Málaga III.En Francia 1. Université Toulouse - Jean Jaurès 2. Université Toulouse III- Paul Sabatier 3. Université Joseph Fourier Valence 4. Université d’Artois 5. Université Nice Sophia Antipolis 6. Université Savoie Mont Blanc 7. IUT de Bourges, Université d’Orléans 8. Université d'Orléans 9. Centre de recherches historiques de l’Université de Rennes 1 10. Université Lyon 3 11. Université Savoie Mont Blanc (Chambéry) 12. Université Montpellier I 13. Ecole Supérieure d’Ingénieurs en Agroalimentaire de Bretagne Atlantique 14. Université de Strasbourg 15. Université de Lille 16. Université de Bordeaux 17. Université Lumière Lyon 2 18. Université de Toulon 19. Université de Corse Pasquale Paoli 20. Université Paul-Valéry – Montpellier 3 

W1973249030.txt

https://bmcresnotes.biomedcentral.com/track/pdf/10.1186/1756-0500-3-66

en

Replicated association of the single nucleotide polymorphism in EDG1 with marbling in three general populations of Japanese Black beef cattle

BMC research notes

cc-by

Sukegawa et al. BMC Research Notes 2010, 3:66 http://www.biomedcentral.com/1756-0500/3/66 SHORT REPORT Open Access Replicated association of the single nucleotide polymorphism in EDG1 with marbling in three general populations of Japanese Black beef cattle Shin Sukegawa1, Takeshi Miyake2, Yoichi Takahagi1, Hiroshi Murakami1, Fumiki Morimatsu1, Takahisa Yamada2*, Yoshiyuki Sasaki3 Abstract Background: Marbling, defined by the amount and the distribution of intramuscular fat and measured as beef marbling score (BMS), is an economically important trait of beef cattle in Japan. We recently reported that a single nucleotide polymorphism (SNP), namely, c.-312A>G, in the endothelial differentiation, sphingolipid G-protein-coupled receptor, 1 (EDG1) gene was associated with the BMS level in the Japanese Black beef cattle population of Oita prefecture, with the G allele being associated with a high level of the BMS. Thus, the c.-312A>G SNP seems to be a candidate marker for marker-assisted selection. In this study, we investigated whether this association could be replicated in 3 other independent Japanese Black cattle populations and analyzed the effect of the SNP genotypes on the carcass traits other than the BMS. Findings: Statistically significant differences in the BMS level were detected among the genotypes of the c.-312A>G SNP in the Japanese black beef cattle populations of Miyazaki (P = 0.0377) and Nagasaki (P = 0.0012) prefectures, and marginal difference was detected in the Kagoshima prefecture population (P = 0.0786). The G allele in the SNP was associated with an increase in the BMS level. The G allele also seemed to have a favorable influence, if any, on the carcass weight, rib eye area and rib thickness of the cattle populations. Conclusions: These findings suggest that the association of the c.-312A>G SNP with the BMS level in the Japanese Black beef cattle population was replicated in other beef cattle populations, and revealed favorable effects of the G allele on the beef productivity in the general Japanese Black beef cattle population. Thus, we concluded that the c.-312A>G SNP is useful for effective marker-assisted selection to increase the BMS level in Japanese Black beef cattle. Background Intramuscular fat deposition (marbling) measured as beef marbling score (BMS) is one of the economically important traits of beef cattle [1]. A high level of the BMS enriches the taste and tenderness of beef, improving the palatability [2-4], therefore, the BMS affects the evaluation of beef quality [1]. The BMS is regarded as the most important trait especially in Japan. Thus, it would be desirable to construct a more effective marker-assisted breeding scheme for increasing the BMS level in Japanese Black beef cattle. * Correspondence: [email protected] 2 Laboratory of Animal Breeding and Genetics, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan We recently showed that a single nucleotide polymorphism (SNP), namely, c.-312A>G, in the endothelial differentiation, sphingolipid G-protein-coupled receptor, 1 (EDG1) gene was associated with the BMS level in a Japanese Black beef cattle population in Oita prefecture, with the G allele of the SNP being associated with a high level of the BMS [5]. Thus, it was necessary to investigate whether this association could be replicated in other independent populations of Japanese Black beef cattle and to analyze the effects of the SNP genotypes on the carcass traits other than the BMS, in order to confirm the application of the c.-312A>G SNP to effective marker-assisted selection. © 2010 Yamada et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Sukegawa et al. BMC Research Notes 2010, 3:66 http://www.biomedcentral.com/1756-0500/3/66 Samples and data We used 3 independent Japanese Black cattle populations, namely, the Kagoshima, Miyazaki and Nagasaki prefecture populations, and studied the association of the c.-312A>G SNP with the BMS, subcutaneous fat thickness (SFT), carcass weight (CWT), rib eye area (REA), and rib thickness (RT). For the Kagoshima, Miyazaki and Nagasaki prefecture populations, respectively, 489, 160 and 191 paternal half-sib progeny steers (1 to 110, 1 to 18, and 1 to 36 steers per sire) produced from 72, 45, and 40 sires were used. Adipose tissue specimens of the progeny steers were collected for genotyping the SNP. DNA samples were prepared from the materials using DNeasy Blood & Tissue Kit (QIAGEN, Hilden, Germany). The BMS, SFT, CWT, REA and RT were measured according to the Japanese meat grading system by certified graders from the Japan Meat Grading Association (Tokyo, Japan) [1]. The predicted breeding values of the BMS, SFT, CWT, REA and RT for the progeny steers were used as the phenotypic values in this study. The breeding values were predicted separately for each of the 3 populations using carcass records of Japanese Black steers and heifers fattened in the Kagoshima, Miyazaki and Nagasaki prefectures. Data were analyzed by the REML method using the MTDFREML program [6], and the genetic and environmental variances were estimated. The BLUP option in the program using the estimated variance components was chosen to predict the breeding values for animals with a single trait model. The sex, market-year, and farm of the animals were considered as the fixed effects. The fattening period and slaughter age were also considered as up to quadratic covariates. The fattening period denotes the period from the start of the fattening to shipping to the market for each animal. Random effects included additive genetic effects of the individuals; that is, the animal model was adopted to predict the breeding values. This study conformed to the guidelines for animal experimentation of the Graduate School of Agriculture, Kyoto University (Kyoto, Japan). SNP genotyping The c.-312A>G SNP was genotyped by the PCR-restriction fragment length polymorphism method as described previously [5]. Using this method, 378-bp PCR fragments containing the SNP site were amplified and MscI-digested into 163- and 215-bp fragments at the A allele, but not the G allele: the GG homozygotes, the AA homozygotes and the AG heterozygotes yielded 1 band (378 bp), 2 bands (163 and 215 bp) and 3 bands (163, 215, and 378 bp), respectively. Page 2 of 3 Table 1 Frequencies of the c.-312A>G SNP genotypes in the 3 populations Population Genotype No. of animals Frequency Kagoshima (n = 489) GG 166 0.339 AG 242 0.495 AA GG 81 44 0.166 0.275 AG 91 0.569 AA 25 0.156 GG 33 0.173 AG 94 0.492 AA 64 0.335 Miyazaki (n = 160) Nagasaki (n = 191) The observed frequencies of the SNP genotypes in the 3 populations are shown in Table 1. The observed and expected heterozygosity values at the SNP conformed to the Hardy-Weinberg equilibrium in all the populations. The frequencies of the G allele of c.-312A>G SNP in the Kagoshima and Miyazaki prefecture populations were consistent with the frequency of this allele in the Oita prefecture population determined in our previous study [7], whereas the allele frequency in the Nagasaki prefecture population was lower [7]. Association analysis Statistically significant differences in the BMS level were detected among the genotypes of the SNP in the Miyazaki and Nagasaki prefecture populations, by analysis using a model that included the SNP genotype as the fixed effect and the sire as the random effect (Table 2). The BMS level in the GG homozygotes was significantly higher than that in the AA homozygotes, and the values in the heterozygotes were intermediate between those in the 2 homozygotes (Table 2). Theses results were consistent with the data obtained in our previous study in the Oita prefecture population [5]. In the Kagoshima prefecture population, the effects of the SNP genotypes reached marginal significance for the BMS level, with the GG homozygotes showing a tendency to exhibit the highest values among the 3 genotypes. These results suggest that the association of the c.-312A>G SNP with the BMS level was replicated in the general Japanese Black beef cattle populations. The effect of the SNP genotype on the BMS level was not statistically significant in the Kagoshima prefecture population (P > 0.05). The predominant breeding objective is the BMS in the Miyazaki and Nagasaki prefectures, while it is the CWT in the Kagoshima prefecture (Fujita T, personal communication). Thus, many quantitative trait loci (QTL) for the BMS are thought to be fixed in the Miyazaki and Nagasaki prefecture populations, but not in the Kagoshima prefecture population. Sukegawa et al. BMC Research Notes 2010, 3:66 http://www.biomedcentral.com/1756-0500/3/66 Page 3 of 3 Table 2 Effects of the c.-312A>G SNP genotypes on BMS, BFT, CWT, REA and RT in the 3 populations Genotype§ Trait* Population P-value BMS SFT GG AG AA Kagoshima 0.0786 0.42 ± 0.05 0.28 ± 0.04 0.37 ± 0.07 Miyazaki Nagasaki 0.0377 0.0012 0.43 ± 0.13a 0.36 ± 0.09a -0.07 ± 0.17b 0.52 ± 0.18a 0.01 ± 0.11b -0.32 ± 0.13b Kagoshima 0.1256 -0.59 ± 0.22 -0.89 ± 0.18 -1.36 ± 0.31 Miyazaki 0.0619 0.88 ± 0.26 0.45 ± 0.18 -0.14 ± 0.34 -0.07 ± 0.32 Nagasaki 0.7401 0.05 ± 0.45 0.25 ± 0.27 Kagoshima 0.1927 8.83 ± 1.60 6.62 ± 1.32 3.83 ± 2.28 Miyazaki 0.1811 -0.19 ± 0.50 0.25 ± 0.35 -1.12 ± 0.66 Nagasaki 0.0014 16.30 ± 3.69a 5.78 ± 2.19a -0.49 ± 2.65b REA Kagoshima Miyazaki 0.0029 0.2642 1.26 ± 0.23a 0.45 ± 0.19b -0.02 ± 0.34b 0.71 ± 0.26 0.48 ± 0.18 0.00 ± 0.35 Nagasaki 0.0960 0.13 ± 0.30 -0.48 ± 0.18 RT Kagoshima 0.2868 1.14 ± 0.20 0.73 ± 0.16 0.92 ± 0.28 Miyazaki 0.4833 -0.12 ± 0.42 0.45 ± 0.29 -0.12 ± 0.55 Nagasaki 0.0002 3.30 ± 0.96a 1.50 ± 0.57b -1.32 ± 0.69b CWT -0.67 ± 0.22 *BMS, Beef marbling score; SFT, Subcutaneous fat thickness (mm); CWT, Carcass weight (kg); REA, Rib eye area (cm2); RT, Rib thickness (mm). § The breeding values are shown as estimates ± SE. a, b Estimates at different genotypes without a common letter in their superscripts significantly differ (P < 0.05). Our present study might not have sufficient power to detect the association of the c.-312A>G SNP with the BMS level in the Kagoshima prefecture population, because of the larger number of segregating QTL for the BMS in the Kagoshima prefecture population as compared with that in the other two populations. Further study using a larger number of samples is needed for the Kagoshima prefecture population. The effect of the SNP genotype was not statistically significant for the SFT (Table 2), consistent with the results obtained in the Oita prefecture population [5]. In addition, the SNP genotype had no significant effect on the CWT in the Kagoshima and Miyazaki prefecture populations, on the REA in the Miyazaki and Nagasaki prefecture populations, and on the RT in Kagoshima and Miyazaki prefecture populations (Table 2). However, the effect of the SNP genotype reached statistical significance for the CWT in the Nagasaki prefecture population, for the REA in the Kagoshima prefecture population, and for the RT in the Nagasaki prefecture population. In these cases, the CWT, REA and RT values were significantly higher in the GG homozygotes than in the AA homozygotes, and the values in the heterozygotes were intermediate between those in the 2 homozygotes. Thus, just like its effect on the BMS, the effects of the G allele on carcass traits other than the BMS, if any, seemed to be favorable rather than deleterious. In this study, we demonstrated replication of the association of the c.-312A>G SNP with the BMS level, and revealed a favorable effect of the G allele on the beef productivity in general Japanese Black beef cattle populations. Together with the results of our previous study [5], these findings suggest that the c.-312A>G SNP in EDG1 is useful for effective marker-assisted selection to increase the level of marbling in Japanese Black beef cattle. Acknowledgements We are grateful to Nipponfeed Inc. for kindly providing us with the adipose tissue samples and phenotypic data. Author details 1 Research and Development Center, Nippon Meat Packers, Inc, Tsukuba, Ibaraki 300-2646, Japan. 2Laboratory of Animal Breeding and Genetics, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan. 3Kyoto Cattle Genetics Section, Beef Information & Genetics Institute, Inc, Otsu, Shiga 520-0865, Japan. Authors’ contributions SS carried out the genotyping and statistical analyses and drafted the manuscript. TM carried out further statistical analyses. YT, HM and FM participated in the sample and data collection. TY and YS participated in the design and coordination of the study and helped draft the manuscript. All authors read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Received: 28 December 2009 Accepted: 11 March 2010 Published: 11 March 2010 References 1. Japan Meat Grading Association: New Beef Carcass Grading Standards Tokyo, Japan: Japan Meat Grading Association 1988. 2. Boylston TD, Morgan SA, Johnson KA, Busboom JR, Wright RW Jr, Reeves JJ: Lipid content and composition of Wagyu and domestic breeds of beef. J Agric Food Chem 1995, 43:1202-1207. 3. Busboom JR, Jeremiah LE, Gibson LL, Johnson KA, Gaskins CT, Reeves JJ, Wright RW Jr: Effects of biological source on cooking and palatability attributes of beef produced for the Japanese market. Meat Sci 1993, 35:241-258. 4. Matsuishi M, Fujimori M, Okitani A: Wagyu beef aroma in Wagyu (Japanese Black cattle) beef preferred by the Japanese over imported beef. Anim Sci J 2001, 72:498-504. 5. Yamada T, Itoh M, Nishimura S, Taniguchi Y, Miyake T, Sasaki S, Yoshioka S, Fujita T, Shiga K, Morita M, Sasaki Y: Association of single nucleotide polymorphisms in the endothelial differentiation, sphingolipid G-proteincoupled receptor, 1 gene with marbling in Japanese Black beef cattle. Anim Genet 2009, 40:209-216. 6. Boldman KG, Kriese LA, Van Vleck LD, Van Tassell CP, Kachman SD: A Manual for Use of MTDFREML, A Set of Programs to Obtain Estimates of Variances and Covariances Washinton, DC: Agricultural Research Servise, United States Department of Agriculture 1995. 7. Watanabe N, Yoshioka S, Itoh M, Satoh Y, Furuta M, Komatsu S, Sumio Y, Fujita T, Yamada T, Sasaki Y: The G allele at the c.-312A>G in the EDG1 gene associated with high marbling in Japanese Black cattle is at a low frequency in breeds not selected for marbling. Anim Genet 2009, 40:579. doi:10.1186/1756-0500-3-66 Cite this article as: Sukegawa et al.: Replicated association of the single nucleotide polymorphism in EDG1 with marbling in three general populations of Japanese Black beef cattle. BMC Research Notes 2010 3:66. 

https://openalex.org/W2980812793

https://scholarworks.utrgv.edu/cgi/viewcontent.cgi?article=1033&context=eems_fac

English

Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama

Biogeosciences

cc-by

Benchmarking and parameter sensitivity of physiological and Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Benchmarking and parameter sensitivity of physiological and Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Ecosystem Simulator (FATES) at Barro Colorado Island, Panama See next page for additional authors Follow this and additional works at: https://scholarworks.utrgv.edu/eems_fac University of Texas Rio Grande Valley University of Texas Rio Grande Valley ScholarWorks @ UTRGV ScholarWorks @ UTRGV School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations College of Sciences 6-15-2020 Benchmarking and parameter sensitivity of physiological and Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Charles D. Koven Ryan G. Knox Rosie A. Fisher Jeffrey Q. Chambers Bradley O. Christoffersen The University of Texas Rio Grande Valley, [email protected] See next page for additional authors Follow this and additional works at: https://scholarworks.utrgv.edu/eems_fac Part of the Earth Sciences Commons, Environmental Sciences Commons, and the Marine Biology Commons Recommended Citation Recommended Citation Koven, Charles D., Ryan G. Knox, Rosie A. Fisher, Jeffrey Q. Chambers, Bradley O. Christoffersen, Stuart J. Davies, Matteo Detto, et al. 2020. “Benchmarking and Parameter Sensitivity of Physiological and Vegetation Dynamics Using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama.” Biogeosciences 17 (11): 3017–44. https://doi.org/10.5194/bg-17-3017-2020 This Article is brought to you for free and open access by the College of Sciences at ScholarWorks @ UTRGV It has https://doi.org/10.5194/bg-17-3017-2020 This Article is brought to you for free and open access by the College of Sciences at ScholarWorks @ UTRGV. It has been accepted for inclusion in School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations by an authorized administrator of ScholarWorks @ UTRGV. For more information, please contact [email protected], [email protected]. This Article is brought to you for free and open access by the College of Sciences at ScholarWorks @ UTRGV. It has been accepted for inclusion in School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations by an authorized administrator of ScholarWorks @ UTRGV. For more information, please contact [email protected], [email protected]. University of Texas Rio Grande Valley University of Texas Rio Grande Valley ScholarWorks @ UTRGV ScholarWorks @ UTRGV School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations College of Sciences 6-15-2020 Benchmarking and parameter sensitivity of physiological and Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Charles D. Koven Ryan G. Knox Rosie A. Fisher Jeffrey Q. Chambers Bradley O. Christoffersen The University of Texas Rio Grande Valley, [email protected] See next page for additional authors Follow this and additional works at: https://scholarworks.utrgv.edu/eems_fac Part of the Earth Sciences Commons, Environmental Sciences Commons, and the Marine Biology Commons Recommended Citation Recommended Citation Koven, Charles D., Ryan G. Knox, Rosie A. Fisher, Jeffrey Q. Chambers, Bradley O. Christoffersen, Stuart J. Davies, Matteo Detto, et al. 2020. “Benchmarking and Parameter Sensitivity of Physiological and Vegetation Dynamics Using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama.” Biogeosciences 17 (11): 3017–44. https://doi.org/10.5194/bg-17-3017-2020 This Article is brought to you for free and open access by the College of Sciences at ScholarWorks @ UTRGV It has University of Texas Rio Grande Valley University of Texas Rio Grande Valley ScholarWorks @ UTRGV ScholarWorks @ UTRGV School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations College of Sciences 6-15-2020 Benchmarking and parameter sensitivity of physiological and Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Charles D. Koven Ryan G. Knox Rosie A. Fisher Jeffrey Q. Chambers Bradley O. Christoffersen The University of Texas Rio Grande Valley, [email protected] See next page for additional authors Follow this and additional works at: https://scholarworks.utrgv.edu/eems_fac Part of the Earth Sciences Commons, Environmental Sciences Commons, and the Marine Biology Commons Recommended Citation Recommended Citation Koven, Charles D., Ryan G. Knox, Rosie A. Fisher, Jeffrey Q. Chambers, Bradley O. Christoffersen, Stuart J. Davies, Matteo Detto, et al. 2020. “Benchmarking and Parameter Sensitivity of Physiological and Vegetation Dynamics Using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama.” Biogeosciences 17 (11): 3017–44. https://doi.org/10.5194/bg-17-3017-2020 This Article is brought to you for free and open access by the College of Sciences at ScholarWorks @ UTRGV It has University of Texas Rio Grande Valley University of Texas Rio Grande Valley ScholarWorks @ UTRGV ScholarWorks @ UTRGV School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations College of Sciences Sciences Faculty Publications and Presentations College of Sciences 6-15-2020 Benchmarking and parameter sensitivity of physiological and Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Charles D. Koven Ryan G. Knox Rosie A. Fisher Jeffrey Q. Chambers Bradley O. Christoffersen The University of Texas Rio Grande Valley, [email protected] See next page for additional authors Follow this and additional works at: https://scholarworks.utrgv.edu/eems_fac Part of the Earth Sciences Commons, Environmental Sciences Commons, and the Marine Biology Commons Recommended Citation Recommended Citation Koven, Charles D., Ryan G. Knox, Rosie A. Fisher, Jeffrey Q. Chambers, Bradley O. Christoffersen, Stuart J. Davies, Matteo Detto, et al. 2020. “Benchmarking and Parameter Sensitivity of Physiological and Vegetation Dynamics Using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama.” Biogeosciences 17 (11): 3017–44. Authors Authors Charles D. Koven, Ryan G. Knox, Rosie A. Fisher, Jeffrey Q. Chambers, Bradley O. Christoffersen, Stuart J. Davies, and Matteo Detto Recommended Citation Recommended Citation Koven, Charles D., Ryan G. Knox, Rosie A. Fisher, Jeffrey Q. Chambers, Bradley O. Christoffersen, Stuart J. Davies, Matteo Detto, et al. 2020. “Benchmarking and Parameter Sensitivity of Physiological and Vegetation Dynamics Using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama.” Biogeosciences 17 (11): 3017–44. https://doi.org/10.5194/bg-17-3017-2020 This Article is brought to you for free and open access by the College of Sciences at ScholarWorks @ UTRGV. It has been accepted for inclusion in School of Earth, Environmental, and Marine Sciences Faculty Publications and Presentations by an authorized administrator of ScholarWorks @ UTRGV. For more information, please contact [email protected], [email protected]. Correspondence: Charles D. Koven ([email protected]) Correspondence: Charles D. Koven ([email protected]) Correspondence: Charles D. Koven ([email protected]) Received: 9 October 2019 – Discussion started: 14 October 2019 Revised: 15 February 2020 – Accepted: 20 February 2020 – Published: 15 June 2020 Received: 9 October 2019 – Discussion started: 14 October 2019 Revised: 15 February 2020 – Accepted: 20 February 2020 – Published: 15 June 2020 distribution of plant trait variation, derived primarily from observations in Panama, and define plant functional types (PFTs) as random draws from this distribution. We com- pare several model ensembles, where individual ensemble members vary only in the plant traits that define PFTs, and separate ensembles differ from each other based on either model structural assumptions or non-trait, ecosystem-level parameters, which include (a) the number of competing PFTs present in any simulation and (b) parameters that govern dis- Abstract. Plant functional traits determine vegetation re- sponses to environmental variation, but variation in trait val- ues is large, even within a single site. Likewise, uncertainty in how these traits map to Earth system feedbacks is large. We use a vegetation demographic model (VDM), the Function- ally Assembled Terrestrial Ecosystem Simulator (FATES), to explore parameter sensitivity of model predictions, and com- parison to observations, at a tropical forest site: Barro Col- orado Island in Panama. We define a single 12-dimensional Authors Authors This article is available at ScholarWorks @ UTRGV: https://scholarworks.utrgv.edu/eems_fac/34 Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 © Author(s) 2020. This work is distributed under the Creative Commons Attribution 4.0 License. Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama Charles D. Koven1, Ryan G. Knox1, Rosie A. Fisher2,3, Jeffrey Q. Chambers1,4, Bradley O. Christoffersen5, Stuart J. Davies6, Matteo Detto7,8, Michael C. Dietze9, Boris Faybishenko1, Jennifer Holm1, Maoyi Huang10, Marlies Kovenock11, Lara M. Kueppers1,12, Gregory Lemieux1, Elias Massoud13, Nathan G. McDowell10, Helene C. Muller-Landau6,7, Jessica F. Needham1, Richard J. Norby14, Thomas Powell1, Alistair Rogers15, Shawn P. Serbin15, Jacquelyn K. Shuman2, Abigail L. S. Swann11,16, Charuleka Varadharajan1, Anthony P. Walker14, S. Joseph Wright7, and Chonggang Xu17 1Climate and Ecosystem Sciences Division, Lawrence Berkeley National Lab, Berkeley, CA, USA 2Climate and Global Dynamics Division, National Center for Atmospheric Research, Boulder, CO, USA 3Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique, Toulouse, France p 4Department of Geography, University of California, Berkeley, CA, USA 4Department of Geography, University of California, Berkeley, CA, USA 5Department of Biology, University of Texas, Rio Grande Valley, Edinburg, TX, USA 6 6Forest Global Earth Observatory, Smithsonian Tropical Research Institute, Washington, DC, US 7Smithsonian Tropical Research Institute, Apartado 0843–03092 Balboa, Republic of Panama 8Department of Ecology and Evolutionary Biology, Princeton University, Princeto 9Department of Earth and Environment, Boston University, Boston, MA, USA 10Atmospheric Sciences and Global Change Division, Pacific Northwest National Laboratory, Richland, WA, USA 11Department of Biology, University of Washington, Seattle, WA, USA 10Atmospheric Sciences and Global Change Division, Pacific Northwest National Laboratory 11 Sciences and Global Change Division, Pacific Northwest National Laboratory, Richland, WA, USA p g y 11Department of Biology, University of Washington, Seattle, WA, USA 11Department of Biology, University of Washington, Seattle, WA, USA 12Energy and Resources Group, University of California, Berkeley, USA 3Jet Propulsion Laboratory, Pasadena, CA, USA 14Climate Change Science Institute, Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA Environmental and Climate Sciences Department, Brookhaven National Laboratory, Upton, NY, US D t t f At h i S i U i it f W hi t S ttl WA USA 15Environmental and Climate Sciences Department, Brookhaven National Laboratory, Upton, NY, USA 16Department of Atmospheric Sciences, University of Washington, Seattle, WA, USA 17Earth and Environmental Sciences Division, Los Alamos National Laboratory, Los Alamos, NM, USA 15Environmental and Climate Sciences Department, Brookhaven National Laboratory, Upton, NY, USA 16Department of Atmospheric Sciences, University of Washington, Seattle, WA, USA 17 p y p 16Department of Atmospheric Sciences, University of Washington, Seattle, WA, USA 17 17Earth and Environmental Sciences Division, Los Alamos National Laboratory, Los Alamos, NM C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3018 McDowell et al., 2018; Powell et al., 2013; Walker et al., 2015), which may be already underway in tropical forests (Brienen et al., 2015). turbance and height-based light competition. While single- PFT simulations are roughly consistent with observations of productivity at Barro Colorado Island, increasing the num- ber of competing PFTs strongly shifts model predictions towards higher productivity and biomass forests. Different ecosystem variables show greater sensitivity than others to the number of competing PFTs, with the predictions that are most dominated by large trees, such as biomass, being the most sensitive. Changing disturbance and height-sorting pa- rameters, i.e., the rules of competitive trait filtering, shifts regimes of dominance or coexistence between early- and late-successional PFTs in the model. Increases to the extent or severity of disturbance, or to the degree of determinism in height-based light competition, all act to shift the community towards early-successional PFTs. In turn, these shifts in com- petitive outcomes alter predictions of ecosystem states and fluxes, with more early-successional-dominated forests hav- ing lower biomass. It is thus crucial to differentiate between plant traits, which are under competitive pressure in VDMs, from those model parameters that are not and to better un- derstand the relationships between these two types of model parameters to quantify sources of uncertainty in VDMs. McDowell et al., 2018; Powell et al., 2013; Walker et al., 2015), which may be already underway in tropical forests (Brienen et al., 2015). ( ) In addition to the above structural problems in existing ESM vegetation representations, there are enormous uncer- tainties due to the representation of parameters in such mod- els (Booth et al., 2012). Typically, ESMs are run with a single set of parameters that are chosen through processes that range from formal (but limited-scope) optimization ap- proaches to ad hoc selection of values that give acceptable results. These parameters may or may not be measurable, and if they are measurable, the values used in a given model may need to be scaled up and may or may not agree with ob- served ranges (Bonan et al., 2012; Rogers, 2014). C. D. Koven et al.: Benchmarking and sensitivity of FATES model It is crucial to benchmark ecosystem models against a wide range of ob- servations (Collier et al., 2018; Luo et al., 2012), and at the same time to understand how sensitive model predictions are to uncertainty in the model parameters (Dietze et al., 2014; Raczka et al., 2018), so that we may better assess how much to trust a given model prediction. Land surface models (LSMs), by virtue of their enormous scope – which typically includes aspects of boundary layer turbulence, radiative transfer, soil hydrology, soil biogeo- chemistry, plant physiology, land management, and commu- nity ecology – have many parameters, all of which are un- certain. In this paper, which focuses on vegetation processes, we broadly separate these model parameters as belonging to two sets: the parameters that comprise a plant functional type (PFT), which we refer to as plant traits, and the parameters that govern the environment in which PFTs exist, which we refer to as ecosystem-level parameters. The importance of this distinction is that, in a dynamic vegetation model with more than one competing PFT, while we can specify the val- ues of the traits of each PFT, the overall trait distributions are controlled by both the trait values of the PFTs and the rel- ative abundance of each PFT. Because the PFT abundances are themselves emergent outcomes that result from the trait values (Fisher et al., 2015), complex feedbacks exist that am- plify or attenuate the influence of any given trait value on model predictions as well as trade-offs or other interactions between traits. These feedbacks greatly complicate the as- sessment of parameter sensitivity in the models. It is thus im- portant to distinguish between the parameter uncertainty as- sociated with plant traits and that associated with ecosystem- level parameters to better understand how they relate to each other and contribute in different ways to model dynamics. C. D. Koven et al.: Benchmarking and sensitivity of FATES model croft et al., 2001) describes an approach to represent a spa- tially heterogeneous forest canopy comprised of individual trees existing on a complex disturbance history by approxi- mating the forest as a set of partial differential equations in a two-dimensional space comprised of plant size and the age of a given location since its last disturbance event. These con- tinuous equations are then solved numerically by discretiz- ing the ecosystem along each of these two dimensions: plant growth and mortality are discretized by tracking cohorts of individual trees that have a similar size, and disturbance his- tory is tracked as a set of patches with shared disturbance his- tories, such that each patch may have several cohorts growing on it. The number of patches and cohorts varies in time. New cohorts are generated by recruitment, existing cohort number densities are reduced by mortality, cohorts are merged if they grow to be sufficiently similar, and cohorts are split by any process – such as light competition – that leads to divergence in outcomes across plants at a similar stage. New patches are generated during disturbance events by reducing the area of existing patches, and patches may be merged if their distur- bance history or composition is sufficiently similar. rameter uncertainty. Because this parameter uncertainty can show up in a number of different ways in a VDM like FATES, we are interested in trying to separate three distinct types of parametric uncertainty: (1) the direct effects of traits on phys- iological predictions by the model, (2) the indirect effects of trait control on competitive outcomes, which further affect ecosystem-level processes, and (3) how non-trait parameters interact with each of these trait uncertainties to further affect model dynamics. rameter uncertainty. Because this parameter uncertainty can show up in a number of different ways in a VDM like FATES, we are interested in trying to separate three distinct types of parametric uncertainty: (1) the direct effects of traits on phys- iological predictions by the model, (2) the indirect effects of trait control on competitive outcomes, which further affect ecosystem-level processes, and (3) how non-trait parameters interact with each of these trait uncertainties to further affect model dynamics. To do this, we first describe the model and the data that comprise the test bed used to drive the model. 2.1 Description of the FATES model Since the original version of CLM ED described in Fisher et al. (2015), there have been numerous develop- ments in the FATES model, which we briefly summarize here. These relate to five main areas: (1) the overall struc- ture of the model and its modularization from the CLM, (2) changes to canopy biophysics, (3) changes to alloca- tion and allometry, (4) changes to the representation of dis- turbance, and (5) changes to the canopy sorting approach. For a complete reference of the FATES model, see https: //fates-docs.readthedocs.io/en/latest/index.html (last access: 1 October 2019), and for a schematic of key processes and their linkages in FATES, see Fig. 1. FATES is a size- and age-structured vegetation model whose foundations are based on a representation of ecosystem bio- physics from CLM4.5 (Oleson et al., 2013), a discretization of individual plant and forest disturbance dynamics based on the ecosystem demography (ED) approach (Moorcroft et al., 2001), and an approach to scale from individual plants to a forest canopy based on the perfect plasticity approximation (PPA; Purves et al., 2008), all of which were first brought to- gether in the CLM ED model (Fisher et al., 2015). Following the development of CLM4.5, FATES was created by sepa- rating the demographic components of the CLM ED model from CLM itself to facilitate a more modular structure, to combat the “shanty-town syndrome” prevalent in land sur- face models (Clark et al., 2017), whereby new model features are added without a clear infrastructure for supporting the ad- ditional complexity that they bring, and to enable FATES to be used within multiple ESMs, initially both the CLM and ELM. A key distinction between CLM ED and FATES is the modularization of the code into a separate repository, with clearly identified boundary conditions between the demog- raphy code and the rest of the LSM into which FATES is embedded. Information is passed between FATES and the LSM at two different frequencies: a biophysics frequency, with a default time step of 30 min, and a vegetation dy- namics frequency, with a default time step of 1 d. Within each biophysics time step, the LSM provides FATES with information about the current state of the soil moisture, atmospheric radiation inputs, atmospheric thermodynamic state, and some time-averaged functions of the environment. C. D. Koven et al.: Benchmarking and sensitivity of FATES model This test bed includes distributions of plant traits, most of which are based directly on observations across research sites in Panama. We then describe a series of numerical experiments aimed at exploring the structural and parametric uncertainty in the model. These include (1) assessing direct control of trait un- certainty on model predictions using an ensemble of model runs with only one PFT per ensemble member, (2) separat- ing ensembles where we embed FATES within two related but divergent land surface models, the Energy Exascale Earth System Model (E3SM) Land Model (ELM) and the Commu- nity Land Model (CLM), (3) ensembles where we add greater numbers of competing PFTs (from 1 to 2 to 10) into each en- semble member, and (4) a set of ensembles where we com- pete two PFTs against each other in each ensemble member while also varying a set of ecosystem-level parameters that govern competition and disturbance in the model. The PPA (Purves et al., 2008) describes an approach of or- ganizing trees (or, equivalently, cohorts) into discrete canopy strata by rank-ordering the trees from tallest to shortest and defining canopy trees as those whose cumulative crown area equals that of the ground (or, when combined with ED, patch area) that they occupy. Fisher et al. (2010) added a modi- fied form of the PPA, whereby the cohorts, rather than being strictly rank-ordered in their separation between canopy and understory, were probabilistically sorted into the canopy and understory based on a function of their height. 1 Introduction Climate-change-related feedbacks from the terrestrial bio- sphere are an important and highly uncertain component of global change (Friedlingstein et al., 2013; Gregory et al., 2009). Tropical forests may contribute substantially to these feedbacks, as vegetation dynamics within these ecosystems may lead to biome shifts and resulting changes to carbon stocks (Cox et al., 2000; Huntingford et al., 2013; Malhi et al., 2009). The majority of Earth system models (ESMs) rep- resent vegetation through conceptual structures that are likely to inhibit realistic or accurate ecosystem responses to global change. In particular, most ESMs use prescribed vegetation distributions, and/or do not represent the functional diversity that exists within tropical forests, and/or impose static veg- etation turnover times. Each of these assumptions may sub- stantially bias model results. Prescribed biogeography does not allow models to project either the abrupt changes (Cox et al., 2000) or the long-term committed ecosystem changes (Jones et al., 2009) that may result from vegetation shifts. Conversely, assuming all tropical forests are comprised of a single set of plant traits may lead to overly abrupt changes in response to an imposed forcing, as compared to approaches that allow community-wide shifts in the trait composition of forests (Levine et al., 2016; Powell et al., 2018; Sakschewski et al., 2016). Lastly, assuming fixed turnover times for vege- tation may bias the responses to both elevated CO2 and cli- mate change, as doing so does not permit changes to mor- tality rates that may result from changes to climate and re- source competition (Friend et al., 2014; Koven et al., 2015; This paper has three goals. The first is to describe a vege- tation demographic model (VDM; Fisher et al., 2018) for use in ESMs, which we call the Functionally Assembled Terres- trial Ecosystem Simulator (FATES). A VDM is a size- and age-structured representation of vegetation dynamics within an LSM and may also be coupled within an ESM. The second is to describe FATES behavior at a test bed site at Barro Col- orado Island (BCI), Panama. The third goal is to explore the sensitivity of mean-state model predictions by FATES to pa- Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3019 C. D. Koven et al.: Benchmarking and sensitivity of FATES model g y Figure 1. Overall process schematic of FATES. Boxes represent major processes, and arrows represent linkages in the form of material flows or other dependencies. Fast time step processes are resolved every 30 min, and slow time step processes are resolved daily. Figure 1. Overall process schematic of FATES. Boxes represent major processes, and arrows represent linkages in the form of material flows or other dependencies. Fast time step processes are resolved every 30 min, and slow time step processes are resolved daily. biomass pools. All of these are prescribed as functions of a cohort’s stem diameter, which thus serves as the basic index for all allometry. FATES currently has six separate allometric target biomass pools: leaf, stem, coarse root, fine root, seed, and storage. Of these, FATES also assumes that the target values of fine-root and storage pools are both linearly pro- portional to the target leaf biomass pool and that the target coarse-root pool is linearly proportional to the target stem biomass pool. Thus only three index target pools exist: leaf, stem, and seed. As a further simplification, FATES currently assumes that sapwood cross-sectional area at breast height is a constant fraction of a cohort’s target leaf area, and thus the sapwood allometry follows the leaf area allometry. FATES solves the photosynthesis equations for sunlit and shaded leaves, separately for each PFT, along vertical gra- dients both within each cohort’s canopy and between cohorts residing in different canopy layers, to calculate water and car- bon fluxes at the level of individual leaves. FATES then pro- vides the LSM with transpiration, integrated canopy conduc- tance, and albedo terms, which the LSM then uses to calcu- late the energy fluxes at the whole-canopy level. FATES also calculates autotrophic respiration at the biophysics time steps and integrates the resulting net primary productivity (NPP) over the day to end up with an increment of carbon per cohort at the end of each day. Heterotrophic respiration is handled outside of FATES by the LSM it is embedded within. At the daily time step, FATES sequentially allocates the daily carbon increment per cohort. If this carbon increment is negative, the amount is subtracted from the cohort’s storage pool. If the increment is positive, then the cohort allocates it, first to replenish storage and then to compensate for tissue turnover. 2.1 Description of the FATES model The two key structural components that FATES adds to a traditional land surface model, the ED and PPA approaches, are described elsewhere in greater detail (e.g., Fisher et al., 2018), so we only briefly summarize them here. ED (Moor- Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 3020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model have been met. This represents a biomass flux from the indi- vidual cohorts to the site-level seed pool, which then serves as a basis for recruitment flux from the seed pool to new co- horts. This fraction is a plant trait that varies among PFTs. This approach represents an extremely simplified view of re- production, which we plan to develop further, but does at least allow us to test baseline sensitivity of the current con- figuration. C =  f Dg D < Dmax f Dg max D ≥Dmax , (2) (2) We treat the crown area allometry coefficient (f ) and expo- nent (g) in the above equation as plant traits that we vary based on species-level values, as described below, and we use a single maximum size for crown allometry (Dmax), for all model runs, of 200 cm. Plants can continue to grow past Dmax, but they do so at a progressively slower rate because the gross primary productivity (GPP) per individual becomes capped by the crown allometry, while the carbon cost of growth continues to increase with increasing stem size. In early versions of FATES, the presence of understory trees that persist for long periods of time but only grow very slowly – as is observed in real forests – was difficult to achieve because of the lack of any stabilizing term on an individual cohort’s carbon dynamics. If a given cohort’s NPP was even slightly negative for sufficiently long, then its storage pool would eventually be reduced to below zero, at which point the entire cohort would die. In order to prevent this and allow the model to produce multiple canopy strata, we added a stabilizing term to the carbon budgets of trees, whereby when their storage pools become depleted, we si- multaneously increase the rate of carbon starvation mortal- ity and decrease the rate of maintenance respiration. This re- duction of maintenance respiration during carbon starvation is consistent with observations of trees under acute carbon stress (Sevanto et al., 2014). Because the physiologic basis and form of this process is poorly constrained, we use heuris- tic functions here to define these processes. First, we define a target carbon storage pool (St): For the target stem biomass allometric model, FATES in- cludes several options, including a power law of diameter, as well as the functional forms of Saldarriaga et al. (1988) and Chave et al. C. D. Koven et al.: Benchmarking and sensitivity of FATES model (2015), as well as the relative ability of a plant to actually achieve its target leaf biomass, can lead to large differences in crown depth between the canopy and understory strata, and thus dif- ferences in crown depth can occur along growth trajectories. where the leaf allometric coefficient m is a plant trait that we allow to vary, as described below, and the target leaf al- lometric exponent g has the same value as the crown area allometric exponent above. As with the crown area, we set a maximum stem diameter above which target leaf biomass remains constant and use the same maximum diameter for both allometries. Setting the exponent on leaf biomass to be the same as that of crown area is equivalent to asserting that a tree’s (target) crown depth and leaf area index (LAI) within the footprint of its crown does not vary over the course of its growth trajectory. This holds true – within a given canopy strata – even though FATES does allow specific leaf area (SLA) to vary vertically through the canopy. However, the canopy trimming logic described in Fisher et al. (2015), as well as the relative ability of a plant to actually achieve its target leaf biomass, can lead to large differences in crown depth between the canopy and understory strata, and thus dif- ferences in crown depth can occur along growth trajectories. Mcs =  Mcs,max(1 −S/L) S < L 0 S ≥L , (6) R =  1 −qS/L
/(1 −q) S < L 1 S ≥L , (7) (6) (7) where Mcs,max is a trait that defines the maximum rate of car- bon starvation mortality, and q is a parameter that governs the curvature of the respiration reduction function. Thus we implicitly assume that there is a critical storage pool Sc = L that sets the total-plant storage level where mortality begins; the implied parameter Sc/L = 1 could be made explicit, but we left this as an implicit parameter here due to the generally weak data constraints on it at present. For the experiments described here, we use a single value, 0.01, of the q param- eter and allow the maximum rate of carbon starvation mor- tality Mcs,max to be a PFT trait. C. D. Koven et al.: Benchmarking and sensitivity of FATES model (2014), which both relate target aboveground biomass to both the stem diameter and height. For all of the experiments described below, we use the Chave et al. (2014) aboveground biomass allometry, expressed in units of kilo- grams of carbon per individual tree: AGB = cj(ρwD2h)p, (3) (3) AGB = cj(ρwD2h)p, where the parameters j and p have values of 0.0673 and 0.976, respectively, ρw is the plant trait wood density, and c is the carbon-to-biomass ratio in wood, which we set as 0.5 for all cases. We did not yet fully explore the sensitiv- ity of model dynamics to the alternate aboveground biomass allometries; this remains future work. St = nL, (5) (5) St = nL, where n is a parameter that linearly relates the target storage pool to the target leaf biomass L. If a given plant is unable to achieve its target carbon storage because of having a negative NPP at any given time, then its actual storage pool S will drop below the target storage pool St. Then we set both the carbon starvation mortality rate (Mcs) and the fractional rate of maintenance respiration (R) on the ratio of S to L: For target leaf biomass, we use a power law allometric model: L = mDg, (4) (4) L = mDg, where the leaf allometric coefficient m is a plant trait that we allow to vary, as described below, and the target leaf al- lometric exponent g has the same value as the crown area allometric exponent above. As with the crown area, we set a maximum stem diameter above which target leaf biomass remains constant and use the same maximum diameter for both allometries. Setting the exponent on leaf biomass to be the same as that of crown area is equivalent to asserting that a tree’s (target) crown depth and leaf area index (LAI) within the footprint of its crown does not vary over the course of its growth trajectory. This holds true – within a given canopy strata – even though FATES does allow specific leaf area (SLA) to vary vertically through the canopy. However, the canopy trimming logic described in Fisher et al. C. D. Koven et al.: Benchmarking and sensitivity of FATES model If the remaining carbon increment is still positive, the cohort will then allocate to any organ pools that are be- low their allometric targets, which are intrinsic functions for a given PFT that are defined relative to the cohort’s stem di- ameter. If, after this, carbon still remains to be allocated, the cohort will grow its stem diameter, allocating to each pool proportionally to that pool’s derivative with respect to stem diameter. FATES currently allows several allometric models for de- termining tree height. These include a generic power law relationship as well as the models described in O’Brien et al. (1995), Poorter et al. (2006), Chave et al. (2014), and Martínez Cano et al. (2019). For the simulations described here, we use the Martínez Cano et al. (2019) allometry for all cases, which uses a Michaelis–Menten form to calculate height (H) from stem diameter (D): H = aDb k + Db . (1) H = aDb k + Db . (1) We use a single mean set of height allometry parameters for all PFTs in this model, with the mean based on the results from Martínez Cano et al. (2019): a = 57.6, b = 0.74, and k = 21.6. A key development since Fisher et al. (2015) has been modularizing all allometry functions so that PFTs of dif- ferent allometric functional forms and parameters can exist and compete against each other. FATES requires four dis- tinct types of allometric models to be defined for each PFT: height, crown area, sapwood cross-sectional area, and target Crown allometry (C) in FATES is set as a two-parameter power law of diameter, subject to a maximum stem diameter for crown allometry: Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 3021 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model Though these traits are highly coor- dinated in plants (Wright et al., 2004), we allow this coor- dination to occur in FATES at the point of defining a PFT that has a specific set of trait values rather than by imposing the trait coordination within the model itself. Exceptions to this rule include that we do define the maximum rate of elec- tron transport at reference temperature (Jmax,25) as a direct function of Vc,max,25. Also, FATES scales leaf traits verti- cally through the canopy so that Vc,max,25, leaf N per unit area, and LMA decrease exponentially with overlying leaf area, following Lloyd et al. (2010) and Kovenock (2019). This allows shaded leaves, which are deeper in the canopy, to be thinner and have lower maximum photosynthetic rates (Vc,max,25, Jmax) than sun-exposed, top-of-canopy leaves and maintains a fixed leaf C : N throughout the canopy, following observations (Lloyd et al., 2010). We generalized some aspects of canopy sorting and dis- turbance in FATES, as compared to their CLM ED repre- sentations, where some strong assumptions were implicit in the model structure. For example, gap-phase disturbance in FATES occurs when canopy trees die. When a given canopy tree dies, or more precisely, when the rate of mortality in a canopy cohort, mc (as measured by the total crown area of trees that died; in m2 ha−1 yr−1), is greater than zero, the patch that previously contained the canopy trees may or may not split off newly disturbed patch area. A pair of ecosystem- level parameters, the fraction of newly dead crown area that becomes a new patch (a new parameter, fd, a unitless ratio), and the fractional understory mortality during a transition to a new patch due to disturbance, mu,d, control the outcomes of disturbance, as described below and in Fig. 2a. The rate of new patch area formation, rd (m2 ha−1 yr−1), equals A last set of modifications since Fisher et al. (2015) are in regards to the canopy sorting via the PPA. As described above, the original PPA (Purves et al., 2008) used a deter- ministic ranking of trees based on their heights and sepa- rated them in each time step based on whether their height was above or below the height, z∗, equal to the tree whose cumulative crown area equaled the area of the ground that trees occupied. Fisher et al. C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3022 storage buffer that determines how much cumulative negative NPP a plant can experience before it begins to suffer from carbon starvation. equals 1, then the existing patch area shrinks in tandem with the reduction in tree crown area within the patch’s canopy. What this means is that it is effectively not possible for trees in the understory to be “promoted” to the canopy while re- maining in a patch – their only route to the canopy is to survive that disturbance event, whereupon they are promoted into the canopy of the new patch. We refer here to this end- member as a “pure-ED” representation of disturbance (on ac- count of its similarity to the original ecosystem demography approach). At the other extreme, if fd equals 0, then no new patch area is created and there is no horizontal heterogeneity in the system (i.e., there is only ever one patch). In this case, when canopy trees die, the entire void in the canopy created by the loss of their crown area is filled through promotion of trees from the understory within the patch. We refer to this endmember as the “pure-PPA” endmember of disturbance. Intermediate cases exist between these endmembers, where a fraction of understory trees may be promoted from within a patch while a fraction of new patch area is generated. A spe- cial intermediate case considered here is a “bare-ground in- termediate”, where mu,d equals 1 – i.e., all cohorts in the un- derstory that are transferred to a new patch are killed during the disturbance event, and thus the new patch area starts from bare ground. This bare-ground intermediate, with mu,d = 1 and fd = 0.5, is equivalent to the equations and PPA-type model described in Farrior et al. (2016). We will consider each of these three special cases – the two endmembers and the bare-ground intermediate – below. In FATES, we separate as distinct traits the top-of-canopy values of maximum carboxylation at reference temperature (Vc,max,25,top), leaf carbon-to-nitrogen ratios (C : N), and leaf mass per area (LMA). (8) rd = mc · fd. When new patch area is created from an existing (“donor”) patch, the new patch is initialized with a fraction of the under- story plants and litter from the donor patch. The pools from the donor patch are thus split in proportion to the fraction of the old patch area transferred to the new patch. Thus when new patch area is created, all understory cohorts in the exist- ing patch are split, with resulting number densities in the cor- responding cohorts in the new and old patches proportional to the fraction of patch area disturbed. Formerly understory trees in this newly disturbed patch may, however, be killed in the disturbance event itself; thus the mu,d term is applied during the disturbance event. C. D. Koven et al.: Benchmarking and sensitivity of FATES model Because both the increase in mortality and the decrease in respiration begin when S drops below L, the parameter (n−1) thus sets the size of the carbon where Mcs,max is a trait that defines the maximum rate of car- bon starvation mortality, and q is a parameter that governs the curvature of the respiration reduction function. Thus we implicitly assume that there is a critical storage pool Sc = L that sets the total-plant storage level where mortality begins; the implied parameter Sc/L = 1 could be made explicit, but we left this as an implicit parameter here due to the generally weak data constraints on it at present. For the experiments described here, we use a single value, 0.01, of the q param- eter and allow the maximum rate of carbon starvation mor- tality Mcs,max to be a PFT trait. Because both the increase in mortality and the decrease in respiration begin when S drops below L, the parameter (n−1) thus sets the size of the carbon For seed production, FATES uses as its target a constant fraction of NPP once tissue turnover and storage demands https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model (2010) modified this to create a probabilistic PPA whereby the relative probability of trees in a cohort (or, equivalently, the fractional number density of trees of a given cohort) being assigned to the canopy was pro- portional to their size raised to a parameter called the com- petitive exclusion parameter cexcl. In FATES, we generalized the height sorting so that it can use either the deterministic- or probabilistic-sorting approach to the PPA and discuss both versions below. rd = mc · fd. (8) (8) 2.2 Site description and driving data Trees can be promoted from the understory to the canopy either in the old patch or if they are transferred to the new patch area as survivors of the disturbance event. Endmembers of this case are the “pure-ED” case, in which all crown area becomes new patch area, and the “pure-PPA” case, in which no newly disturbed patch area is created. We also consider an intermediate case, in which half of the dead tree canopy becomes disturbed, but with no survivorship of trees in the newly disturbed patch. (b) Representation of height sorting. When canopy tree crown area exceeds the patch area that the trees are on due to crown growth, canopy trees are “demoted” to the understory. In the deterministic case, trees are rank-ordered by height and the shortest cohort is split at the point where total tree crown area equals patch area, and the remaining cohort is demoted. In the probabilistic case, all canopy trees are demoted, with the fraction of each cohort demoted based on the cohorts’ relative heights. Figure 2. Schematic of how disturbance and height sorting are represented in the FATES model. (a) Representation of disturbance. When canopy trees die, some fraction of the crown area of the dead trees is transferred to a newly disturbed patch, while the remainder remains in the old patch. Trees can be promoted from the understory to the canopy either in the old patch or if they are transferred to the new patch area as survivors of the disturbance event. Endmembers of this case are the “pure-ED” case, in which all crown area becomes new patch area, and the “pure-PPA” case, in which no newly disturbed patch area is created. We also consider an intermediate case, in which half of the dead tree canopy becomes disturbed, but with no survivorship of trees in the newly disturbed patch. (b) Representation of height sorting. When canopy tree crown area exceeds the patch area that the trees are on due to crown growth, canopy trees are “demoted” to the understory. In the deterministic case, trees are rank-ordered by height and the shortest cohort is split at the point where total tree crown area equals patch area, and the remaining cohort is demoted. In the probabilistic case, all canopy trees are demoted, with the fraction of each cohort demoted based on the cohorts’ relative heights. 2.2 Site description and driving data All model experiments here are conducted at Barro Colorado Island (BCI), Panama (9.151◦N, 79.855◦W). The environ- ment at BCI has a mean precipitation of 2600±480 mm yr−1, with a 4-month dry season during which precipitation drops below 100 mm yr−1. The ecosystem at BCI is a primary forest, with a disturbance regime characterized by primar- ily small-scale disturbance and subject to elevated mortality rates during ENSO-driven droughts. The site includes a 50 ha The fd parameter thus allows FATES to scale continuously between two endmembers in how the simulated ecosystem responds to gap-phase disturbance dynamics (Fig. 2a). If fd Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 3023 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3023 Figure 2. Schematic of how disturbance and height sorting are represented in the FATES model. (a) Representation of disturbance. When canopy trees die, some fraction of the crown area of the dead trees is transferred to a newly disturbed patch, while the remainder remains in the old patch. Trees can be promoted from the understory to the canopy either in the old patch or if they are transferred to the new patch area as survivors of the disturbance event. Endmembers of this case are the “pure-ED” case, in which all crown area becomes new patch area, and the “pure-PPA” case, in which no newly disturbed patch area is created. We also consider an intermediate case, in which half of the dead tree canopy becomes disturbed, but with no survivorship of trees in the newly disturbed patch. (b) Representation of height sorting. When canopy tree crown area exceeds the patch area that the trees are on due to crown growth, canopy trees are “demoted” to the understory. In the deterministic case, trees are rank-ordered by height and the shortest cohort is split at the point where total tree crown area equals patch area, and the remaining cohort is demoted. In the probabilistic case, all canopy trees are demoted, with the fraction of each cohort demoted based on the cohorts’ relative heights. Figure 2. Schematic of how disturbance and height sorting are represented in the FATES model. (a) Representation of disturbance. When canopy trees die, some fraction of the crown area of the dead trees is transferred to a newly disturbed patch, while the remainder remains in the old patch. 2.3 Plant trait data and application to FATES PFT definition A key conceptual point in this study is that we define a PFT only as a vector of plant traits; we do not make any further a priori assumptions about the ecological role that a given PFT plays. In some of these experiments, we do diagnose properties of a PFT that allow us to – in certain cases – make post hoc distinctions such as “early-successional” or “late-successional” PFTs, and in this paper all PFTs may be thought of as belonging to tropical forest tree communities, but we essentially take a probabilistic view of PFTs here as being random draws from some continuous trait covariance matrix. To define this matrix, we assemble several datasets and cross-reference them based on mean values per plant species, with Latin binomials used as the reference index. In addition to the observed traits that allow us to generate prior distributions on values based on data, we also want to include parameter variation in a small set of traits that are poorly observed but that we expect to be important in model dynamics. We thus add four more unobserved trait values: the allometric coefficient of relationship of leaf biomass to stem diameter (kg cm−1), the allometric ratio of fine-root biomass to leaf biomass (unitless), the fractional allocation to reproduction (unitless), and the maximum rate of carbon starvation mortality (yr−1). For each of these, we assume no correlations with other observed traits, and we assume the first three of these to be normally distributed and that the last (maximum rate of carbon starvation mortality) is lognor- mally distributed as we do for the background mortality trait. The choice of these additional traits are to extend the possible range of dynamics to include crown thickness, plant carbon use efficiency, understory mortality rates and thus shade tol- erance, and reproductive fecundity as possible determinants in the competitiveness of a given PFT. Table 1 lists each of the parameters varied, and the most closely associated pro- cess box is shown in Fig. 1. We start with two datasets describing plant traits at BCI, and at two other sites across a precipitation gradient in Panama, Parque Nacional Metropolitano (PNM) and Fort San Lorenzo (SLZ), which are originally described in Os- nas et al. (2018) and Wright et al. (2010). 2.3 Plant trait data and application to FATES PFT definition Data from these sets used here include leaf lifespan, leaf mass per unit area (LMA), wood density, mortality of 10 cm and larger trees, and leaf N content. For these datasets, we only use values for trees in the canopy stratum. Where a given species occurs in more than one site, we use mean values across the sites. Because these are the only datasets that include leaf lifespan estimates, where other datasets also include an estimate of LMA for a given species, we only use the estimates in these datasets, as they will correspond to the specific individuals with which leaf lifespan is also measured. We add two further datasets on leaf traits, both based on canopy crane measurements at PNM and SLZ sites: Gu et al. (2016) and Rogers et al. (2017) and Wu et al. (2019). Each of these contain estimates of Vc,max, LMA, wood density, and leaf N content. We use FATES temperature scaling func- tions to calculate Vc,max at the reference temperature (25 ◦C) based on the temperature at which specific Vc,max observa- tions were made. Together these sets of traits describe plant variation along the leaf and wood economic spectra, two crit- ical axes of functional diversity (Baraloto et al., 2010; Wright et al., 2004). We thus define a single 12 × 12 trait covariance matrix as the basis of all experiments described below, representing the data-constrained hypervolume from which we sample plant functional types. In all experiments, the vector of trait values that defines a PFT is sampled as a single random draw from this 12 × 12 trait covariance matrix. An example of resam- pled trait matrix from a single model ensemble is shown in Fig. 4. The traits considered in this study are not meant to be comprehensive but are meant to cover a range of processes in the model, including (a) physiology and the leaf economic spectrum; (b) allocation of biomass within a whole plant to leaves, roots, and reproduction; (c) patterns of acquisition of the primary resource, light, through crown area allometry; and (d) mortality rates in both the canopy and understory. Lastly, we add a dataset on crown area allometry from trees at BCI (Martínez Cano et al., 2019). C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3024 cessing and cleaning raw datasets; formatting timestamps; detecting and removing duplicates, bad data, and outliers; gap filling of missing data; and flagging quality-controlled data. For each simulation, we recycled meteorology over the 1986–2017 period. of plants, and how it changes over plant size, and thus are important determinants of both baseline growth rates (for co- efficient g) and the derivative of growth rates with respect to plant size (for exponent d). In total, we thus use eight traits from the observa- tional datasets: Vc,max,25,top (µmol CO2 m−2 s−1), wood den- sity (g cm−3), LMA (m2 g−1), leaf N per unit area (Leaf N/area; g m−2), leaf lifespan (year), background tree mor- tality (yr−1), crown area coefficient (m2 cm−1), and crown area intercept (unitless). We assume lognormal distributions for Vc,max,25,top, LMA, Leaf N/area, leaf lifespan, and back- ground tree mortality and normal distributions for wood den- sity, crown area coefficient, and crown area intercept, with correlations between these traits as determined from the data. The full matrix of observed traits is shown in Fig. 3, where each dot represents a pair of mean trait values for a given species, and the histograms across the diagonal show the full distribution of species-mean values for each trait. 2.2 Site description and driving data using the R (https://www.r-project.org/, RStudio v1.1, last access: 9 October 2017) software, with the application of libraries “zoo” (Zeileis et al., 2019), “xts” (Ryan et al., 2018), “tsoutliers” (López-de-Lacalle, 2019), and “Rssa” (Korobeynikov et al., 2017). The procedure includes the fol- lowing major steps – the identification of problems in the datasets (QA) and then data cleaning, flagging, and gap fill- ing of missing data (QC). Step 1 (QA) includes an initial visual inspection and cataloging data, determining the tem- poral frequency of sampling to assess data availability and preliminary assessing data quality. Step 2 (QC) includes pro- census plot, in which every stem ≥1 cm diameter has been measured every 5 years since 1982, with 321 species iden- tified (Condit et al., 2017), as well as eddy covariance and other observations. We force the model with drivers measured at the BCI me- teorological station for the period 1986–2017; these data are available at https://biogeodb.stri.si.edu/physical_monitoring/ research/barrocolorado (last access: November 2017). All site-level data were scanned for quality assurance and quality control (QA/QC) as described by Faybishenko et al. (2018). The QA/QC procedure of time-series data was performed https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 2.3 Plant trait data and application to FATES PFT definition The crown area allometry in FATES is defined with crown area, C, set as a power law relationship with diameter, D, as described above, so for each species we use the crown area coefficient g and exponent d as reported in Martínez Cano et al. (2019). These crown area traits control the overall light interception ability https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 3025 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 2 Figure 3. Matrix of plant trait data used to inform FATES ensembles. Measured traits are as follows: leaf Vc,max,25,top (µmol CO2 m−2 s−1), wood density (g cm−3), leaf mass per unit area (m2 g−1), leaf N per unit area (g m−2), leaf lifespan (year), plant mortality rate (yr−1), coefficient of relationship of crown area to stem diameter (m2 cm−1), and exponent of relationship of crown area to stem diameter (unitless). Each dot represents one pair of species-level trait values where both traits are measured for a given species; histograms show the distributions of all species-level values for a given trait. Figure 3. Matrix of plant trait data used to inform FATES ensembles. Measured traits are as follows: leaf Vc,max,25,top (µmol CO2 m−2 s−1), wood density (g cm−3), leaf mass per unit area (m2 g−1), leaf N per unit area (g m−2), leaf lifespan (year), plant mortality rate (yr−1), coefficient of relationship of crown area to stem diameter (m2 cm−1), and exponent of relationship of crown area to stem diameter (unitless). Each dot represents one pair of species-level trait values where both traits are measured for a given species; histograms show the distributions of all species-level values for a given trait. 2.4 Model testing data initial imprint of these initial size distributions on the model output, we integrate FATES for 200 or 300 years (depend- ing on the experiment); after this spin-up time the model dy- namics have diverged from the initialization (e.g., Fig. 5b). (2) We compare model predictions of size distributions to the census data of the forest as a whole. (3) We compare model predictions of aboveground biomass against observa- tions, which are also derived from the BCI census data, that are reported in Meakem et al. (2018), which are approxi- mately 13.6 kg C m−2; we assign ±10 % uncertainty to these We make use of the long-term forest dynamics plot cen- sus data at BCI (Hubbell et al., 1999). We use a total of five censuses here, beginning with the 1985 census. We use the census data in three ways in this paper. (1) To more rapidly equilibrate the model, we initialize the forest with observed size distributions (from the 2005 census); in sim- ulations with more than one PFT present, we use the same initial size distribution for each PFT. In order to remove the https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3026 g y Figure 4. Resampled trait matrix, including eight observed and four unobserved traits, as used to define PFTs in FATES simulations. Eig observed traits are as in Fig. 3 (with background mortality set to be the observed plant mortality). Four additional unobserved traits a as follows: allometric coefficient of relationship of leaf biomass to stem diameter (kg cm−1), allometric ratio of fine-root biomass to le biomass (unitless), fractional NPP allocated to reproduction (unitless), and maximum rate of carbon starvation mortality (yr−1). Bleaf is le biomass. Figure 4. Resampled trait matrix, including eight observed and four unobserved traits, as used to define PFTs in FATES simulations. Eight observed traits are as in Fig. 3 (with background mortality set to be the observed plant mortality). Four additional unobserved traits are as follows: allometric coefficient of relationship of leaf biomass to stem diameter (kg cm−1), allometric ratio of fine-root biomass to leaf biomass (unitless), fractional NPP allocated to reproduction (unitless), and maximum rate of carbon starvation mortality (yr−1). Bleaf is leaf biomass. Figure 4. C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model Table 1. Traits varied within each ensemble in order to define plant functional types, their units, and the process or processes (as diagrammed in Fig. 1) most closely associated with each trait. Table 1. Traits varied within each ensemble in order to define plant functional types, their units, and the process or processes (as diagrammed in Fig. 1) most closely associated with each trait. Trait Units Associated process(es) Vcmax,25 top leaf layer (µmol CO2 m−2 s−1) Photosynthesis Wood density (g cm−3) Stem growth Leaf mass per unit area (m2 g−1) Leaf growth and turnover Leaf N per unit area (g m−2) Plant respiration Leaf lifespan (year) Leaf growth and turnover Background plant mortality rate (yr−1) Mortality Coefficient of relationship of crown area to stem diameter (m2 cm−1) Crown area allometry Exponent of relationship of crown area to stem diameter (unitless) Crown area allometry Allometric coefficient of relationship of leaf biomass to stem diameter (kg cm−1) Allometry and allocation; leaf growth and turnover Allometric ratio of fine-root biomass to leaf biomass (unitless) Allometry and allocation; root growth and turnover Fractional NPP allocated to reproduction (unitless) Allometry and allocation; seed production Maximum rate of carbon starvation mortality (yr−1) Mortality cluded). Lastly, we compare LAI as predicted by FATES to observations of LAI reported in Detto et al. (2018). Figure 5. Joint distribution of modeled GPP and LAI (a) and modeled time series of biomass trajectories (b) for a 576-member ensemble of site-scale FATES simulations, where each ensemble member, represented here as an individual dot, has a single PFT de- fined as a random draw from the 12-trait covariance matrix shown in Fig. 3. Site-level observations of LAI and GPP (mean ± 1 SD) are shown in (a) as a grey ellipse, and observed mean biomass is shown in (b) as black line. 2.4 Model testing data Resampled trait matrix, including eight observed and four unobserved traits, as used to define PFTs in FATES simulations. Eight observed traits are as in Fig. 3 (with background mortality set to be the observed plant mortality). Four additional unobserved traits are as follows: allometric coefficient of relationship of leaf biomass to stem diameter (kg cm−1), allometric ratio of fine-root biomass to leaf biomass (unitless), fractional NPP allocated to reproduction (unitless), and maximum rate of carbon starvation mortality (yr−1). Bleaf is leaf biomass. biomass observations to account for allometric uncertainty (Chave et al., 2003). by a data logger (CR1000, Campbell Scientific) and stored on a local PC. Data were processed with a custom program using a standard routine described in Detto et al. (2010). GPP was derived from daytime values of net ecosystem exchange (NEE) by adding the corresponding mean daily ecosystem respiration obtained as the intercept of the light response curve (Lasslop et al., 2010). The light curve was fitted on a 15 d moving window using a rectangular hyperbolic func- tion (runs with friction velocity less than 0.4 m s−1 were ex- We compare fluxes of GPP as well as the sensible and latent heat fluxes to observations made with an eddy- covariance system. The tower used for these measurements is 41 m a.g.l. (meters above ground level) on a plateau on BCI. The eddy-covariance system includes a sonic anemometer (CSAT3, Campbell Scientific, Logan, UT) and an open-path infrared CO2–H2O gas analyzer (LI-7500, LI-COR, Lincoln, NE). High-frequency (10 Hz) measurements were acquired https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3027 Table 1. Traits varied within each ensemble in order to define plant functional types, their units, and the process or processes (as diagrammed in Fig. 1) most closely associated with each trait. 2.4 Model testing data Trait Units Associated process(es) Vcmax,25 top leaf layer (µmol CO2 m−2 s−1) Photosynthesis Wood density (g cm−3) Stem growth Leaf mass per unit area (m2 g−1) Leaf growth and turnover Leaf N per unit area (g m−2) Plant respiration Leaf lifespan (year) Leaf growth and turnover Background plant mortality rate (yr−1) Mortality Coefficient of relationship of crown area to stem diameter (m2 cm−1) Crown area allometry Exponent of relationship of crown area to stem diameter (unitless) Crown area allometry Allometric coefficient of relationship of leaf biomass to stem diameter (kg cm−1) Allometry and allocation; leaf growth and turnover Allometric ratio of fine-root biomass to leaf biomass (unitless) Allometry and allocation; root growth and turnover Fractional NPP allocated to reproduction (unitless) Allometry and allocation; seed production Maximum rate of carbon starvation mortality (yr−1) Mortality 3027 2.5 Model experiment descriptions Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3028 Table 2. Experimental matrix used in this study. Each ensemble above consists of 576 ensemble members, with one or more PFTs per ensemble member chosen as a random draw from the 12 × 12 trait covariance matrix. Table 2. Experimental matrix used in this study. Each ensemble above consists of 576 ensemble members, with one or more PFTs per ensemble member chosen as a random draw from the 12 × 12 trait covariance matrix. Number of PFTs competing per ensemble member Height sorting Recruitment Disturbance LSM Purpose 1 Deterministic Normal Bare-ground intermediate: fd = 0.5, mu,d = 1 CLM Control 1 Deterministic Normal Bare-ground intermediate ELM Understand sensitivity to driving model 3 Deterministic Mixing Bare-ground intermediate CLM Understand sensitivity to number of PFTs 10 Deterministic Mixing Bare-ground intermediate CLM Understand sensitivity to number of PFTs 10 Deterministic Normal Bare-ground intermediate CLM Understand sensitivity to intergenerational trait filtering 2 Deterministic Normal Bare-ground intermediate CLM Reference case for looking at regimes of coex- istence 2 Probabilistic, cexcl = 3 Normal Bare-ground intermediate CLM Understand sensitivity of coexistence to repre- sentation of height sorting 2 Probabilistic, cexcl = 1 Normal Bare-ground intermediate CLM Understand sensitivity of coexistence to repre- sentation of height sorting 2 Deterministic Normal Pure ED: fd = 1, mu,d = 0.5 CLM Understand sensitivity of coexistence to repre- sentation of disturbance 2 Deterministic Normal Pure ED: fd = 1, mu,d = 1.0 CLM Understand sensitivity of coexistence to repre- sentation of disturbance 2 Deterministic Normal Pure PPA: fd = 0 CLM Understand sensitivity of coexistence to repre- sentation of disturbance LSM Purpose much of the remaining variance is explained by the trait of interest (Xu, 2013; Xu and Gertner, 2008). runs with three PFTs and 576 runs with 10 PFTs, in each case drawing all PFTs at random from the multivariate trait distributions. We refer to these as the three-PFT ensemble and the 10-PFT ensemble, respectively. As a next experiment, we add increasing numbers of com- peting PFTs to the model. 2.5 Model experiment descriptions We define a series of model experiments here to explore para- metric and structural uncertainty in the model and how trait uncertainty can combine with vegetation dynamics to feed back on model predictions (Table 2). We first begin with single-PFT experiments, randomly drawing a set of PFTs and running each of them as a separate FATES simulation. We refer to the set of such simulations, which differ only in their PFT specification, as an ensemble of simulations and each separate FATES run as an ensemble member. The size of each of the ensembles here is 576 members, chosen as a somewhat arbitrary number but one which balances compu- tational costs against statistical sampling depth while allow- ing one simulation per CPU on the 36-core computing nodes used for most simulations. We compare outputs from these ensembles against a set of observations (of biomass, LAI, and eddy covariance) at BCI to assess patterns of variabil- ity in the model and comparisons to observations. We per- form these single-PFT ensembles, with an identical set of ensemble members each for two different model configura- tions, CLM-FATES and ELM-FATES, in order to further test the structural uncertainty of embedding FATES within two closely related, yet divergent, land models. Figure 5. Joint distribution of modeled GPP and LAI (a) and modeled time series of biomass trajectories (b) for a 576-member ensemble of site-scale FATES simulations, where each ensemble member, represented here as an individual dot, has a single PFT de- fined as a random draw from the 12-trait covariance matrix shown in Fig. 3. Site-level observations of LAI and GPP (mean ± 1 SD) are shown in (a) as a grey ellipse, and observed mean biomass is shown in (b) as black line. We assess parameter sensitivity via direct trait control of model predictions in the one-PFT simulations by fitting splines of each of the model predictions that we analyze as a function of each of the traits that we vary across the ensem- ble. We calculate the maximum potential variance explained as the fraction of variance in the predictions across the en- semble that is predicted by the fitted spline. Because some of the traits are correlated, we also assess the minimum vari- ance explained, which we calculate by first subtracting the variance explained by all other traits and then assessing how https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 Biogeosciences, 17, 3017–3044, 2020 C. D. 3.1 Single-PFT simulations and comparison to observations A first question is how the distributions of ecosystem-level properties – such as biomass; size distributions; LAI; and carbon, water, and energy fluxes – from a set of single-PFT simulations compare with observations at the site. To answer this, we conduct an ensemble of single-PFT simulations to generate a set of possible forests, each of which is comprised of trees sharing a single set of traits. Results from this single- PFT ensemble are shown in Figs. 5–7. There is a broad range of model predictions, ranging from some ensemble members that fail to establish to others which grow to highly produc- tive forests. Figure 6. Comparison of FATES simulations of mean annual cy- cles in gross primary productivity, latent heat (LH), and sensible heat (SH) with eddy-covariance observations from Barro Colorado Island flux tower. Green lines correspond to the mean annual cycle from each FATES ensemble member. Blue lines show individual years of eddy-covariance data. The joint distribution of GPP and LAI (Fig. 5a) shows that the overall ensemble spread is roughly centered around the observed values (shown as ellipse in Fig. 5a), though with wide spread and a tail that extends to low-productivity, low- LAI simulations. Likewise, trajectories of biomass in these simulations (Fig. 5b), where each simulation is initialized with observed size distributions and is then integrated for 200 years to come into a quasi-steady state that is determined by the ensemble parameters, converge towards a distribution in biomass that spans the observed estimates (black line in Fig. 5b). While the ensemble distributions in LAI and GPP are roughly symmetric, albeit with a tail extending to the low- GPP, low-LAI zone in Fig. 5a, the distribution of biomass shows a tail in the other direction towards extremely high biomass forests, with some ensemble members converging towards values that are several times those observed. show both the wide spread of ensemble members, as dis- cussed above, and two systematic model–data mismatches. The first of these is in the shape and amplitude of the seasonal cycles: FATES simulations systematically predict a decrease in GPP during the dry season (February–April) as compared to the eddy-covariance data that do not show a systematic de- crease in productivity during the dry season. The second bias is that the FATES simulations here systematically predict a lower latent heat flux and a higher sensible heat flux than the observations. C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3029 Figure 6. Comparison of FATES simulations of mean annual cy- cles in gross primary productivity, latent heat (LH), and sensible heat (SH) with eddy-covariance observations from Barro Colorado Island flux tower. Green lines correspond to the mean annual cycle from each FATES ensemble member. Blue lines show individual years of eddy-covariance data. where each ensemble member is comprised of PFTs that are randomly drawn from the same trait covariance matrix. In this case, we also explore different values of the ecosystem structural parameters that govern light competition and gap- phase disturbance dynamics, as described above and shown in Fig. 2. The control for this set of ensembles uses the “deterministic-PPA” mode for height sorting and a bare- ground intermediate representation of disturbance (which we also use in all preceding experiments). Two additional en- sembles vary light competition parameters to use probabilis- tic PPA height sorting with cexcl = 3 and probabilistic PPA height sorting with cexcl = 1. In three further ensembles, we vary the disturbance parameters fd and mu,d to explore the two extreme representations of disturbance to the pure-ED and pure-PPA endmembers, and in the pure-ED case, we ex- plore the sensitivity of the model to mu,d, or how many un- derstory plants are killed during a disturbance event. This pa- rameter mu,d has no effect in the pure-PPA case, since there is no disturbance when fd = 0. 2.5 Model experiment descriptions The premise of this is that a model can represent plant trait diversity either through mul- tiple realizations of the model where plants with each set of traits only interact with plants of the same type or alter- nately through allowing plants with different traits to inter- act with each other through competition for resources. In a PFT-based model such as FATES, these options exist on a continuum: as we add further PFTs to a given simulation, we increase the diversity that is resolved within each sim- ulation and thus, in principle, should reduce the variability across simulations. The goal here is to ask how increasing the diversity that is resolved within any specific simulation changes the distribution of model predictions, as compared to an ensemble approach, where we only account for diver- sity by non-interacting PFTs. Again, we construct each ex- periment as a perturbed-parameter ensemble – where we use the random draws of parameter values to construct new pa- rameter vectors for each model run – but instead of including a single PFT in each ensemble member, we do 576 model We conduct the last (10-PFT) experiment twice. In the first instance, we force the model to maintain functional diversity by evenly recruiting from a mixed-PFT seed pool into each PFT, thus preventing competitive exclusion and intergenera- tional trait filtering. This approach still allows trait filtering to occur within the lifespan of an individual plant but prevents any PFT from completely excluding the others, thus acting as a discrete-PFT analog to the continuous generation of the trait diversity approach used in the model of Sakschewski et al. (2016). In the second instance, we allow the normal inter- generational trait filtering to occur; i.e., each PFT reproduces only recruits of its own PFT with no supplement so that PFTs may go extinct. Lastly we perform a series of two-PFT ensembles aimed at asking whether we can identify regimes where trade-offs in general, and in particular early–late-successional trade- offs, lead to a degree of coexistence, after 300 years, in the model. To do this, we again conduct 576-member ensembles, Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3030 Figure 7. Stem size distributions (in number – n – of trees per hectare per centimeter width of size class bin), of single-PFT en- semble members, as compared to census data from the BCI forest dynamics plot. Parametric control by plant traits on several ecosystem- level model predictions is shown via variance decomposition in Fig. 8. While the analysis here is not meant to be as com- prehensive as that of Massoud et al. (2019), Fig. 8 nonethe- less shows that each of these model predictions shows sen- sitivity to a different set of traits, thus highlighting the com- plex mapping of trait variation onto model predictions. Fur- ther, many of these model predictions show a high degree of sensitivity across several variables. To some degree, this arises because of the correlations between trait values, such as through the leaf economic spectrum, which can be seen by the large spread between the maximum potential vari- ance explained by a given trait (closed circles) and the min- imum variance explained by that trait (open circles). How- ever, in other instances, such as for tree growth rates in the canopy, sets of relatively uncorrelated parameters, such as wood density and the set of leaf-economic-spectrum traits, jointly control the rates. And in other cases, individual traits directly affect the rates predicted; an example of this is the canopy mortality rates, which in this mean-state configura- tion effectively show only the background mortality rates. Understory mortality rates are slightly more complex, with joint control of both the background mortality rates and the maximum rate of carbon starvation as well as small contribu- tions from the leaf and stem physiological traits. Trait con- trol over LAI shows that, because of the combined effects of within-cohort leaf optimization, and the potential for multi- ple canopy strata to exist, there is a relatively weak direct control on ecosystem-level LAI by the direct leaf-to-stem allometric-coefficient trait; LAI is equally constrained by the leaf-economic-spectrum traits that control the marginal costs and benefits of additional leaves at the bottom of the canopy, as well as a small contribution from the reproductive alloca- tion trait, which sets how the recruitment rate and thus many small plants are contributing to the understory LAI. Figure 7. C. D. Koven et al.: Benchmarking and sensitivity of FATES model Stem size distributions (in number – n – of trees per hectare per centimeter width of size class bin), of single-PFT en- semble members, as compared to census data from the BCI forest dynamics plot. as replacement of old leaves with leaves with greater photo- synthetic capacity (Wu et al., 2017); and/or (d) biases in the soil hydrologic modules in which hillslope hydrologic pro- cesses are largely ignored (Fan et al., 2019). A fuller analysis of plant hydrologic traits, as well as the structural changes to represent plant hydrodynamics and photosynthetic seasonal- ity, is underway in FATES but beyond the scope of this paper. Observed tree size distributions are an emergent outcome resulting from the growth rates, death rates, and light compe- tition parameters in a forest. In principle, the accurate predic- tion of diameter distributions, which follows a Weibull dis- tribution (approximately power function at small diameters, dropping off at larger sizes), is possible in a vegetation de- mographic model using the combined hypotheses of ED and PPA (Farrior et al., 2016) or through the combined ED and plant hydrodynamic hypotheses (Powell et al., 2018). The en- semble of FATES simulations shown here roughly capture the shape of the curve (Fig. 7) though, again, with consider- able spread and some systematic biases. The wide spread in simulations shows that some trait combinations lead to out- comes with either too many or two few trees at the larger end of the tree size distribution. The more systematic bias is that most of the ensemble members show too many very large trees, and too few small trees, as compared to obser- vations, suggesting an overall bias in the rates of establish- ment, growth, and death. The degree of this ensemble-level bias – close to an order of magnitude – shows some sen- sitivity to ecosystem-level parameters, as discussed further below, which suggests modest control by representation of gap-phase disturbance dynamics and light competition pa- rameters. 3.1 Single-PFT simulations and comparison to observations Similar biases are also documented in Huang et al. (2019). In this paper, we do not try to correct these biases, which likely arise from a combination of (a) not including a broader set of plant traits that govern ecohydrological pro- cesses, such as those traits that govern stomatal conductance, canopy turbulence, or rooting depth distributions; (b) not us- ing a full plant hydraulics model (Christoffersen et al., 2016; Xu et al., 2016); (c) not including processes known to in- crease GPP during the dry seasons of tropical forests, such Seasonal cycles of ecosystem fluxes, as compared to ob- servations from the eddy-covariance tower at BCI (Fig. 6), https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 3.3 Sensitivity of results to the number of competing PFTs Koven (2020). Such an approach can allow, for example, one to ask how the representation of soil physics of bio- geochemistry feeds back onto an identical representation of plant physiology in order to better separate the contributions of each to total model prediction uncertainty. Here we begin to test this approach by testing the exact same representation of FATES within the alternate LSMs it can be run in. The above experiments each contained a single PFT in each ensemble member, and so the ensemble spread of the pre- dictions demonstrates the global trait sensitivity of monocul- tural forests, in the absence of competition effects. In real tropical forest ecosystems, the enormous trait diversity exists as a mosaic of plants of different species, each competing for resources and contributing to ecosystem-level dynamics. A key goal of models such as FATES is to explore how this heterogeneity in traits at the scale of individual cohorts of plants interacts with atmospheric and soil processes to gov- ern ecosystem fluxes and structure. Thus we want to move away from the monocultural representation to ask how trait diversity affects model predictions in the presence of compet- itive interactions. To do this, we conduct experiments to add greater amounts of trait diversity into each ensemble by in- creasing the number of PFTs in each run. We first hold distur- bance and light competition parameters constant; in Sect. 3.4 below, we vary these parameters to explore their role in gov- erning competitive outcomes. We repeated the ensemble described above using FATES embedded within ELMv1 and compare the ensemble predic- tions between the two models in Fig. 9. The ensembles used identical plant traits, forcing data, and other FATES parame- ters; however, many aspects of the LSMs differ, particularly including soil depth and the number of soil layers. Thus the two ensembles can be considered an experiment to the sen- sitivity of the structural representation of the physical soil environment that the vegetation is growing in. Distributions between mean GPP (Fig. 9a), LAI (Fig. 9b), and biomass (Fig. 9c) are all similar, as are the final size distributions of the plant community (Fig. 9d). This demonstrates that the diversity of plant traits used here, at least in this generally well-watered site, has a stronger control on model predic- tions than whatever structural divergences have accumulated in the representation of the soil environment between these models. C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3031 Figure 8. Variance decomposition of trait control on ecosystem states and vital rates. Shown are the fractions of variance explained by each of the 12 traits for seven ecosystem variables. Filled circles and associated lines show the maximum potential fraction of variance explained by each trait, without considering trait-trait correlations. Open circles show the minimum fraction of variance explained by each trait after first subtracting out the variance explained by all other traits. Figure 8. Variance decomposition of trait control on ecosystem states and vital rates. Shown are the fractions of variance explained by each of the 12 traits for seven ecosystem variables. Filled circles and associated lines show the maximum potential fraction of variance explained by each trait, without considering trait-trait correlations. Open circles show the minimum fraction of variance explained by each trait after first subtracting out the variance explained by all other traits. 3.2 Sensitivity of results to land surface model FATES is designed to work as a modular representation of plant biophysical and community assembly processes within a host land surface model rather than being a land surface model on its own. It has been developed out of the CLM ED framework described by Fisher et al. (2015) and cur- rently works within two related but distinct LSMs: CLM5 (Lawrence et al., 2019; Wieder et al., 2019) and ELMv1 (Go- laz et al., 2019). This modularity of FATES and its ability to work within alternate LSMs represents an important capabil- ity. As LSMs have grown ever more complex, the infrastruc- ture for managing model complexity and attributing model behavioral differences to structural and parametric assump- tions has not grown equivalently; a potential strategy for ad- dressing this complexity problem is to separate the represen- tation of processes in such a way that they can be explored as conditional on alternate boundary conditions, following the “modular complexity” approach described in Fisher and Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3032 C. D. Koven et al.: Benchmarking and sensitivity of FATES model Figure 9. Comparison of FATES simulations as embedded within two land surface models: ELM-FATES and CLM-FATES. (a) GPP, (b) LAI, and (c) aboveground biomass. Observational range shown as grey band in (a)–(c). Figure 9. Comparison of FATES simulations as embedded within two land surface models: ELM-FATES and CLM-FATES. (a) GPP, (b) LAI, and (c) aboveground biomass. Observational range shown as grey band in (a)–(c). We can quantify these competitive effects on ensemble spread by looking at how the standard deviation of the en- semble shrinks as we add more PFTs (Fig. 10e). We can formulate a null model: if competition did not matter for a given trait, then we would expect that the narrowing of the distribution upon adding further PFTs would follow a sta- tistical sampling relationship for independent variables and therefore decrease as proportional to n−1/2, where n is the number of PFTs. This null model thus represents the “selec- tion effect” of Tilman et al. (1997). In practice, what is ob- served here is a rate of narrowing with additional PFTs that is much smaller than this null model – i.e., the null model narrows much faster than the realized model outcomes. This shows that competition is an important component of the higher PFT cases, both in maintaining variability within an ensemble and in increasing the ensemble mean productivity by weighting the overall ecosystem function towards the part of the trait distribution that is more productive. plants, we first “force” some degree of coexistence between functional types by recruiting equally into the smallest-size cohorts of all PFTs, as described above. Figure 10a–d show a key set of model predictions for each of these simulations. For all outputs (GPP, LAI, aboveground biomass, size distri- butions), adding additional PFTs to each ensemble member both narrows the ensemble distribution and induces a shift to- wards values indicative of a higher-productivity forest com- prised of larger trees. This narrowing and shifting of the ensemble distributions are separate but related outcomes of resolving trait diversity and competitive interactions. In the single-PFT case, func- tional diversity is only resolved across ensemble members, which are each comprised of monoculture forests. 3.3 Sensitivity of results to the number of competing PFTs We calculate further ensembles, drawing plant traits from the same distribution as before, but with either three or 10 PFTs per ensemble member. To separate competition during the recruitment process from competition by larger-statured https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 3032 C. D. Koven et al.: Benchmarking and sensitivity of FATES model As we add PFTs, each ensemble member better samples the observed functional diversity, so we expect that the differences be- tween ensemble members should decrease as a result. But at the same time, competitive dynamics mean that some traits will be more competitive and therefore more strongly repre- sented in each ensemble member. Thus the single-PFT en- semble will most evenly sample throughout the possible trait distribution, while ensembles comprised of greater numbers of interacting PFTs will unevenly sample the parts of the dis- tribution that are more competitive. Different variables are more strongly affected by compet- itive dynamics than others: of the three we show here, and comparing the one-PFT and 10-PFT cases, the competitive effects on LAI are smaller than those for GPP, which are in turn smaller than for biomass, where an increasing PFT num- ber has very little effect on the ensemble spread. An explana- Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 3033 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model Figure 10. Variation between ensemble members as a function of the number of competing PFTs in each ensemble member. (a–c) Histog of mean GPP (a), LAI (b), and aboveground biomass (c). Observational range shown as grey band in (a)–(c). (d) Size distributio ensembles. (e) Standard deviation across ensemble members as a function of the number of competing PFTs, as compared to a null m which considers the distribution changes purely to be a sampling problem, for expected reduction in variation between ensemble mem in the absence of competition effects. Figure 10. Variation between ensemble members as a function of the number of competing PFTs in each ensemble member. (a–c) Histograms of mean GPP (a), LAI (b), and aboveground biomass (c). Observational range shown as grey band in (a)–(c). (d) Size distributions of ensembles. (e) Standard deviation across ensemble members as a function of the number of competing PFTs, as compared to a null model, which considers the distribution changes purely to be a sampling problem, for expected reduction in variation between ensemble members in the absence of competition effects. cause crown area scales with diameter to the power of ∼1.3 (Fig. 3 and Martínez Cano et al., 2019), and the relative pro- portion of trees in the canopy to the understory will further be dominated by larger trees, GPP will be more dominated by larger trees than their relative contribution to LAI. Biomass is even further dominated by large trees: combining allometry Eqs. (1) and (3) above implies that a given plant’s contribu- tion will scale with its diameter to the power of roughly 2.1, tion for why the competitive effects have stronger effects on some variables than others may be the relative control of a given prediction by very competitive – and thus very large – trees. Leaf area is provided by trees of both canopy strata and so is represented most evenly across the spectrum of the com- petitiveness. The relative contribution by a given PFT to GPP at the ecosystem level is roughly proportional to the frac- tion of the canopy that the given PFT crown occupies. Be- 3.4 Regimes of coexistence and their sensitivity to disturbance and light competition parameters We can represent such a trade-off as a line connect- ing two points that represent two sets of PFT vital rates in a growth–mortality space (Fig. 12); while we expect that only the combinations that define a trade-off – i.e., a positive slope – between growth and mortality can stably coexist, we do not know what the range of possible stable lines might be. To in- vestigate these questions, we conducted a series of six sets of paired-PFT ensembles (last six rows of Table 2), each us- ing the same 576 pairs of two-PFT, but with different values of ecosystem-level parameters that govern light competition and disturbance. yp p We further investigate the degree of competitive filtering within and between generations by re-running the 10-PFT ensemble, but in this instance, we allow species to go extinct by re-coupling the rate of recruitment of a given PFT to the seed production by that PFT. Comparisons of the resulting predictions (Fig. 11a–d) show only subtle differences in the ecosystem-level rates investigated here: biomass and GPP are barely shifted, while the distribution of LAI is slightly ex- panded towards higher values, and the number of small trees is slightly decreased when we allow intergenerational com- petition to play out. Thus the effects of trait filtering during recruitment are much more muted in the model than the trait filtering that happens after recruitment has occurred. This can further be illustrated if we compare ranked abundance curves for the two ensembles of trees greater than 1 cm vs. trees greater than 10 cm (Fig. 11e–f): at 1 cm, the presence or absence of recruitment filtering leads to a marked change in the slope of ranked abundance curves, whereas at 10 cm the slopes of the two cases are roughly similar. Even when we force the model to allow neutral filtering during recruit- ment, by the time trees grow to 10 cm, the resolved filtering is strongly evident. There are many different ways that a plant can grow quickly or slowly (Fig. 8). This creates a problem in trying to map sets of plant traits directly onto the potential for a given pair of trait combinations to coexist with each other. 3.4 Regimes of coexistence and their sensitivity to disturbance and light competition parameters The convergence of the model with increasing numbers of PFTs towards higher-productivity forests than are observed demonstrates that, even with the strong assertion of neutral filtering between generations that we use in these ensem- bles, either the competitive filtering within each generation is still too strong or other biases in the model which are com- pensated for in single-PFT simulations become evident in the more diverse simulations. This is most apparent in the tree size distributions (Fig. 10d), where the 10-PFT ensem- bles generate many more large-statured trees than either the lower-PFT-number simulations or the sizes that are observed. Possible causes for this bias include the following: (a) that the marginal competitiveness associated with a given trait ad- vantage in the model is too strong, as compared to more neu- tral dynamics that may occur in real forests (Hubbell, 2011); (b) that additional, unmeasured trade-offs associated with the set of possible strategies – which might constrain the set of possible trait combinations to remove superspecies or loser species – are insufficiently represented (Clark et al., 2018); (c) that processes which govern tree vital rates at the large end of the size distribution are poorly represented, such as senescence strategies that are observed in forest demography (Johnson et al., 2018; Needham et al., 2020); or (d) that other density-dependent effects such as herbivory or pathogen load act to reduce the competitive success of any given species in real forests (Connell, 1971; Janzen, 1970), though such ef- fects should be weaker for functional types than species. In order to represent shifts in plant trait distributions at a given location under global change pressures, a model like FATES must be capable of maintaining some degree of trait heterogeneity in the first instance. The maintenance of func- tional diversity in ecosystems is a complex topic (Chave, 2004; Chesson, 2000), and its analysis in the context of Earth-system-type models such as FATES is in its infancy (Fisher et al., 2018). Here we seek to first understand which combinations of traits within FATES allow stable coexis- tence of PFTs in the mean state and whether there is other ecosystem-level parametric control on these regimes of co- existence. In particular, we expect that a model that resolves heterogeneity in the light environment can accommodate at least two niches, for fast-growing early-successional plants and shade-tolerant, slow-growing plants (Moorcroft et al., 2001). C. D. Koven et al.: Benchmarking and sensitivity of FATES model which would imply that trees that are extremely large should more seriously impact biomass than either GPP or LAI. https://doi.org/10.5194/bg-17-3017-2020 https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3034 Biogeosciences, 17, 3017–3044, 2020 3.4 Regimes of coexistence and their sensitivity to disturbance and light competition parameters To overcome this, we first want to reduce the problem from the high-dimensional set of plant traits that we use to define a PFT to a lower-dimensional set of PFT vital rates that may act to determine the coexistence dynamics. The simplest set of rates to propose are growth and mortality rates of canopy trees. For each set of traits that comprise a PFT, we eval- uate the mean growth and mortality rates for a tree of that PFT, conditional on the tree being approximately 20 cm in size and located within the canopy strata of the forest. This permits a mapping between the 12-dimensional trait space and a two-dimensional growth vs. mortality space (Fig. 8). Within this reduced space, we can evaluate the slope of lines connecting pairs of competing PFTs, as in Fig. 12, to iden- tify the range of slopes that permit coexistence between PFT pairs. An example of this is shown in Fig. 13. In Fig. 13a we show the lines connecting paired PFTs for a subset of ensemble members in the reference (determin- istic sorting, intermediate bare ground) case. We first iden- tify the canopy growth and mortality rates (of 20 cm diam- eter trees) and examine only combinations with a positive slope in a growth–mortality space, i.e., ones where we can classify an early- and late-successional PFT where the early- Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3035 C. D. Koven et al.: Benchmarking and sensitivity of FATES model g y redictions from 10-PFT ensembles with and without recruitment filtering. (a–c) Histograms of mean GPP biomass (c). Observational range shown as grey band in (a)–(c). (d) Size distributions of ensembles. (e–f) FTs. (e) PFT abundances for all trees greater than 1 cm diameter (DBH – diameter at breast height). (f) PF ter than 10 cm diameter. For all cases in (e)–(f), solid line is the median ensemble member, and shaded rang cross 576 ensembles. Figure 11. Predictions from 10-PFT ensembles with and without recruitment filtering. (a–c) Histograms of mean GPP (a), LAI (b), and aboveground biomass (c). Observational range shown as grey band in (a)–(c). (d) Size distributions of ensembles. (e–f) Rank abundance curves for PFTs. (e) PFT abundances for all trees greater than 1 cm diameter (DBH – diameter at breast height). (f) PFT abundances for all trees greater than 10 cm diameter. 3.4 Regimes of coexistence and their sensitivity to disturbance and light competition parameters For all cases in (e)–(f), solid line is the median ensemble member, and shaded range is the 5th–95th percentiles across 576 ensembles. early-successional PFTs, moderate slopes having some coex- istence, and low slopes dominated by late-successional PFTs. To begin to quantitatively compare the effects of the ecosystem-level parameters on these competitive outcomes, we can first plot the fraction of biomass in each ensem- ble member existing in an early-successional PFT against the log of the slope of the line connecting the two PFTs successional PFT has both higher growth and higher mortal- ity rates than the late-successional PFT. We color the lines based on whether, after 300 years, there is a degree of co- existence (which we define as having less than 95 % of the biomass in either of the PFTs) and, if not, which PFT is dominant. The slope of the lines shows evident control on the competitive outcome, with high slope lines dominated by early-successional PFTs, moderate slopes having some coex- istence, and low slopes dominated by late-successional PFTs. To begin to quantitatively compare the effects of the ecosystem-level parameters on these competitive outcomes, we can first plot the fraction of biomass in each ensem- ble member existing in an early-successional PFT against the log of the slope of the line connecting the two PFTs C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3036 Figure 12. Growth–mortality trade-off and possible regimes of co- existence in a model like FATES along a successional axis. In a growth–mortality space, if a line connecting two PFTs comprising the system is negative as in (a), one PFT should be competitively dominant and exclude the other. If the slope of the line is positive as in (b), coexistence may be possible; however the range of slopes that may permit coexistence in tropical forests is not known a priori. ploit. But if we reduce the intensity of disturbance by al- lowing a fraction of trees in the understory to survive dis- turbance events (by setting mu,d to 0.5), doing so effectively counteracts the increased niche area for the fast-growing, fast-dying trees by giving slow-growing understory trees a chance to end up in newly created patches and dominate them. Thus the disturbance cases that are bare-ground inter- mediate (fd = 0.5, mu,d = 1) and pure ED with fractional un- derstory mortality (fd = 1, mu,d = 0.5) are relatively similar in their relative tendency to promote success between early- and late-successional PFTs. Figure 12. Growth–mortality trade-off and possible regimes of co- existence in a model like FATES along a successional axis. In a growth–mortality space, if a line connecting two PFTs comprising the system is negative as in (a), one PFT should be competitively dominant and exclude the other. If the slope of the line is positive as in (b), coexistence may be possible; however the range of slopes that may permit coexistence in tropical forests is not known a priori. These ecosystem-level parametric differences in the bal- ance between competitive outcomes are large: over an order of magnitude in growth–mortality trade-off slopes separates the midpoint of the logistic regressions between the various cases in Fig. 13c. Because parameters such as fd and mu,d are poorly constrained at present, they represent a signifi- cant source of uncertainty in model predictions; constraining these parameters with census data thus represents an oppor- tunity for reducing this uncertainty. Furthermore, looking at the sensitivity of the relative success of species with different growth and mortality rates across gradients of disturbance intensity or frequency may provide further benchmarks of models of this type. in this growth–mortality trade-off space (Fig. 13b). C. D. Koven et al.: Benchmarking and sensitivity of FATES model The points follow a roughly sigmoidal shape, again showing that low slopes (i.e., small difference in growth, large difference in mortality) lead to a competitive exclusion by the late- successional PFT, large slopes (large difference in growth, small difference in mortality) lead to competitive exclusion by the early-successional PFT, and intermediate slopes can either lead to coexistence or exclusion by either of the PFTs. Following this pattern, we then fit a logistic function to the ensemble of growth–mortality trade-off slopes and coexis- tence states. Canopy growth and mortality rates are only one possible set of plant vital rates that may determine coexistence. If, instead of using canopy growth and mortality rates as the de- pendent variables to explain competitive outcomes produced by FATES, we substitute canopy growth rates and understory mortality rates, as may be expected given the importance of shade tolerance in defining successional strategies, we do not see a clear sigmoidal pattern as in Fig. 13. Thus, within the FATES predictions explored here, canopy mortality rates are more determinative of success than understory mortality rates. We can then compare the effects of the different ecosys- tem structural parameters by comparing the resulting fitted logistic curves for each ensemble (Fig. 13c). The differences between these curves indicate the tendency for a given set of ecosystem parameters to favor PFTs with traits and the resulting set of vital rates that make them either early or late successional: curves with a midpoint that is shifted to the left in Fig. 13c favor early-successional PFTs, and those with a midpoint shifted to the right favor late-successional PFTs. For height-sorting parameters, the more probabilistic the height sorting, the more it favors late-successional PFTs. This makes sense: at the margin, if growing tall more quickly than its neighbors is less likely to make a tree end up in the canopy, then that means that outliving its neighbors becomes relatively more important. The converse is also true in that the rapid growth of early-successional trees becomes more important if even a tiny difference in growth pays off with a position in the canopy. Overall distributions of ecosystem-level model predictions (Fig. 14) are relatively similar to the earlier experiments, though some differences can be seen. GPP distributions are similar between the cases. https://doi.org/10.5194/bg-17-3017-2020 https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3037 Figure 13. Competitive outcomes between two PFTs as a function of PFT growth and mortality rates along an early–late-successional continuum. Only ensemble members where a successional trade-off, i.e., one of the PFTs both grows and dies faster than the other PFT, are shown. (a) The lines connecting paired PFTs in a growth–mortality trade-off space, for a random subset of ensemble members in the reference case, are colored by the competitive outcome in each member to show the importance of the slope of the trade-off line in determining the outcome. (b–c) The relative fraction of total ecosystem biomass in the faster, early-successional PFT is plotted against the log of the ratios of the slope of the growth–mortality trade-off in each PFT pairing. Curves in (b)–(c) show a continuous logistic regression as applied to the PFT biomass fractions in each experiment. Panel (b) shows the individual ensemble members as well as the logistic regression for the reference case. Panel (c) shows only the logistic regression fit for each of the cases, demonstrating that the parameter uncertainty related to disturbance and height sorting that differentiates each ensemble leads to divergent outcomes in the relative success between early- and late-successional PFTs. See Fig. 1 for qualitative schematics of the different structural cases. Figure 13. Competitive outcomes between two PFTs as a function of PFT growth and mortality rates along an early–late-successional continuum. Only ensemble members where a successional trade-off, i.e., one of the PFTs both grows and dies faster than the other PFT, are shown. (a) The lines connecting paired PFTs in a growth–mortality trade-off space, for a random subset of ensemble members in the reference case, are colored by the competitive outcome in each member to show the importance of the slope of the trade-off line in determining the outcome. (b–c) The relative fraction of total ecosystem biomass in the faster, early-successional PFT is plotted against the log of the ratios of the slope of the growth–mortality trade-off in each PFT pairing. Curves in (b)–(c) show a continuous logistic regression as applied to the PFT biomass fractions in each experiment. Panel (b) shows the individual ensemble members as well as the logistic regression for the reference case. C. D. Koven et al.: Benchmarking and sensitivity of FATES model LAI distributions are slightly shifted towards higher values in the probabilistic-height- sorting relative to the deterministic-height-sorting cases and are lower in the pure-PPA disturbance case, likely because of overall suppression of the understory in the absence of disturbance. Biomass distributions are shifted towards lower values in the probabilistic-height-sorting cases, as well as in the pure-ED case with mu,d = 1, and towards higher values in the pure-PPA disturbance case. The height sorting appears to have little effect on size distributions, while the disturbance parameters have a strong effect: the pure-PPA disturbance case has a greater deficit of small trees, while the pure-ED disturbance case has greater number of trees in the smaller size classes (but still not as many as observed). These ef- fects on size distributions make sense from the perspective of small trees in each of these cases. In the pure-PPA dis- turbance case, no new patches are created, so there are no For disturbance parameters, the story is slightly more com- plicated: in the case of no gap-generating disturbance (the pure-PPA disturbance case), early-successional strategies are highly unlikely to pay off, as there is no environmental niche for those PFTs to occupy. The converse is also true for the high-disturbance pure-ED case, which is the most conducive to early-successional PFTs as long as the disturbance gen- erates bare ground (i.e, mu,d = 1) for new recruits to ex- Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model Experiments are as in Figs. 13–14 and ordered from most promoting of early-successional to late-successional PFTs: (a) deterministic sorting, pure- ED, bare-ground (fd = 1, mu,d = 1) disturbance; (b) deterministic sorting, intermediate bare-ground disturbance; (c) deterministic sorting, pure-ED medium-intensity (fd = 1, mu,d = 0.5) disturbance; (d) probabilistic sorting (cexcl = 3), intermediate bare-ground disturbance; (e) probabilistic sorting (cexcl = 1), intermediate bare-ground disturbance; and (f) deterministic sorting, pure-PPA (fd = 0) disturbance. The same ensemble member was used for each panel so that plant traits are identical across experiments. Each panel depicts individual cohorts, arranged from tallest to shortest within a patch from right to left. The horizontal axis of each panel shows area: both cohort crown area and patch area. Older patches are to the right, with thin vertical lines separating patches. Cohort widths in the figure are proportional to the crown area of each cohort. Within the canopy, different PFTs are given different colors, with an early-successional PFT in light green and a late-successional PFT in blue–green. Understory cohorts are shaded darker than canopy cohorts. Figure 15. Forest structure and composition at year 600 of one ensemble member for each structural sensitivity experiment. Experiments are as in Figs. 13–14 and ordered from most promoting of early-successional to late-successional PFTs: (a) deterministic sorting, pure- ED, bare-ground (fd = 1, mu,d = 1) disturbance; (b) deterministic sorting, intermediate bare-ground disturbance; (c) deterministic sorting, pure-ED medium-intensity (fd = 1, mu,d = 0.5) disturbance; (d) probabilistic sorting (cexcl = 3), intermediate bare-ground disturbance; (e) probabilistic sorting (cexcl = 1), intermediate bare-ground disturbance; and (f) deterministic sorting, pure-PPA (fd = 0) disturbance. The same ensemble member was used for each panel so that plant traits are identical across experiments. Each panel depicts individual cohorts, arranged from tallest to shortest within a patch from right to left. The horizontal axis of each panel shows area: both cohort crown area and patch area. Older patches are to the right, with thin vertical lines separating patches. Cohort widths in the figure are proportional to the crown area of each cohort. Within the canopy, different PFTs are given different colors, with an early-successional PFT in light green and a late-successional PFT in blue–green. Understory cohorts are shaded darker than canopy cohorts. each of the cases in Fig. 13, which maintained some degree of early–late PFT coexistence in each of the different cases. C. D. Koven et al.: Benchmarking and sensitivity of FATES model Panel (c) shows only the logistic regression fit for each of the cases, demonstrating that the parameter uncertainty related to disturbance and height sorting that differentiates each ensemble leads to divergent outcomes in the relative success between early- and late-successional PFTs. See Fig. 1 for qualitative schematics of the different structural cases. Figure 13. Competitive outcomes between two PFTs as a function of PFT growth and mortality rates along an early–late-successional continuum. Only ensemble members where a successional trade-off, i.e., one of the PFTs both grows and dies faster than the other PFT, are shown. (a) The lines connecting paired PFTs in a growth–mortality trade-off space, for a random subset of ensemble members in the reference case, are colored by the competitive outcome in each member to show the importance of the slope of the trade-off line in determining the outcome. (b–c) The relative fraction of total ecosystem biomass in the faster, early-successional PFT is plotted against the log of the ratios of the slope of the growth–mortality trade-off in each PFT pairing. Curves in (b)–(c) show a continuous logistic regression as applied to the PFT biomass fractions in each experiment. Panel (b) shows the individual ensemble members as well as the logistic regression for the reference case. Panel (c) shows only the logistic regression fit for each of the cases, demonstrating that the parameter uncertainty related to disturbance and height sorting that differentiates each ensemble leads to divergent outcomes in the relative success between early- and late-successional PFTs. See Fig. 1 for qualitative schematics of the different structural cases. Figure 14. Ecosystem-level model results of paired PFT competition experiments. Ensemble distributions of (a) GPP, (b) LAI, (c) biomass, and (d) size distributions for each of the paired-PFT cases. Figure 14. Ecosystem-level model results of paired PFT competition experiments. Ensemble distributions of (a) GPP, (b) LAI, (c) biomass, Figure 14. Ecosystem-level model results of paired PFT competition experiments. Ensemble distributions of (a) GPP, (b) LAI, (c) biomass, and (d) size distributions for each of the paired-PFT cases. Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3038 3038 C. D. Koven et al.: Benchmarking and sensitivity of FATES model Figure 15. Forest structure and composition at year 600 of one ensemble member for each structural sensitivity experiment. 4 Conclusions Land surface and ecosystem carbon models are highly depen- dent on parameters that are both imperfectly known and that may have highly diverse values within any given ecosystem. We attempt to separate some of these different controls on model dynamics by distinguishing plant trait variation from other ecosystem parameters, to explore how representing di- versity in plant traits affects predictions made by a VDM, and to determine how ecosystem-level parameters govern com- petitive outcomes and other predictions by the model. In a single-PFT configuration, where competitive pres- sures on trait values are not present, the model exhibits both some agreement and some biases as compared to a set of ob- servations that span from physiological processes to ecosys- tem structure. The degree of agreement with observations is not sensitive to the choice of two related land models in which we run FATES, which both behave similarly. Code availability. The FATES model is available at https://github.com/NGEET/fates (last access: 15 May 2020; https://doi.org/10.5281/zenodo.3825474, FATES Development Team, 2020). Experiments here are based on git commit 0bc7a5d on the fork: https://github.com/ckoven/fates (last ac- cess: 4 June 2020; https://doi.org/10.5281/zenodo.3875687, FATES Development Team, 2019). FATES is run here within two host land surface models, CLM5 and ELMv1, available at https://github.com/ESCOMP/ctsm (git commit b9c92b7, last access: 15 May 2020; https://doi.org/10.5281/zenodo.3739617, CTSM Development Team, 2020) and https://github.com/ E3SM-Project/E3SM (git commit 544db3b, last access: 15 May 2020; https://doi.org/10.11578/E3SM/dc.20180418.36, E3SM Project, 2018), respectively. Scripts to initialize parame- ter files and analyze model output shown here are available at https://github.com/NGEET/testbeds (last access: 15 May 2020; https://doi.org/10.5281/zenodo.3785705, Koven, 2020a), and scripts to run the all model experiments here are available at https://github.com/ckoven/runscripts (last access: 15 May 2020; https://doi.org/10.5281/zenodo.3785703, Koven, 2020b). As we add the effects of competitive pressures on param- eter uncertainty, by increasing the number of PFTs compet- ing within any given simulation, these shift the distributions of model predictions in a systematic way. Productivity and biomass increase as we add further PFTs to a simulation, in ways that push the model, which agrees roughly with ob- servations of biomass and productivity in a single-PFT con- figuration, further from the observations as we add more di- versity, even though such increased diversity in the model should better represent processes that exist in species-rich tropical forests. C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3039 lation numbers to be passed on from one generation to the next. ture of the forests. The fractional coverage of PFTs roughly follows the pattern in Fig. 13. The relative heterogeneity of patch area follows the fd parameter, with most hetero- geneous patches when fd = 1 and no heterogeneity when fd = 0. Reducing the disturbance intensity parameter mu,d from 1 to 0.5 causes a small number of large trees, which had been in the understory prior to disturbance, to remain even in newly disturbed patches, thus making the character of patches more similar across ages. Shifting the height sorting to a more probabilistic treatment shifts the relative size dis- tributions of canopy and understory trees within any patch. Animations of annual snapshots of one of these ensemble members are in supplementary Video SV1 (Koven, 2019b), which shows the emergence of heterogeneity in structure and composition over time. Figures 13–15 demonstrate the wide range of outcomes, both in terms of PFT composition and in the size and age structure of the forest, that result from these ecosystem-level height-sorting and disturbance parameters. We further explore the effects of non-trait parameter vari- ability on competitive outcomes in a set of paired-PFT ex- periments to show how the competitively successful strategy between early- and late-successional traits shifts as a result of ecosystem-level parameters. In particular, the parameters that govern both disturbance and competitiveness for light have strong effects on the balance between early- and late- successional PFTs: increases to either the extent or sever- ity of disturbance, or to the efficiency of height-based light competition, all act to shift the community towards early- successional PFTs. These differences in the PFT composi- tion of the modeled forests feed back onto ecosystem-level predictions of states and fluxes by the model. In order to understand how global change pressures will affect ecosystems, and in turn how ecosystem response will further feed back on global change, we must consider the role of shifts in community structure. VDMs are a promising tool to resolve these processes; however VDMs bring a high degree of complexity that adds greater uncertainty to model predictions than more simple model frameworks that may be more easily tuned to match observations. C. D. Koven et al.: Benchmarking and sensitivity of FATES model The results shown here underscore the need to better understand the roles that uncertainty in model parameters plays – both the direct role as well as the indirect roles that govern how parameter uncer- tainty changes competitive pressures on trait distributions at the ecosystem level. It is thus crucial to understand and inte- grate these types of uncertainty into projections of the Earth system. https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model In each panel in Fig. 14a–f, the FATES patches and cohorts are drawn out and rank-ordered by height, with the tallest to the right within each patch, with cohort width proportional to the crown area occupied by each cohort, and with patches simi- larly arrayed with oldest to the right and the patch width pro- portional to the patch area. Thus the width of all canopy co- horts in a closed-canopy patch equals the width of the patch that they occupy. Cohorts are colored by PFT (color) and canopy position (shading), with yellow–green representing an early-successional PFT, blue–green representing a late- successional PFT, and darker shading of each for the under- story cohorts. Shown is the final year of a 600-year set of simulations, started from bare-ground initial conditions. Dif- ferences between the cases are evident in the resulting struc- gaps in which small trees can grow. In the pure-ED distur- bance case, when the mu,d – the parameter that controls the fraction of small understory trees that both survive the death of a canopy tree above them and find themselves in a newly opened patch – is 0.5 (thus representing a medium intensity to disturbance), it provides an additional pathway for plants that are recruited into older patches to make it to the canopy. In the higher-intensity bare-ground (mu,d = 1) and pure-PPA disturbance cases, the only such pathway for plants recruited into older patches is for them to persist in the understory and grow through to the canopy, which fewer of them are able to do. The difference in forest structure that results from these ecosystem-level parameters can be further seen in Fig. 15, which shows, in a quantitative way, the ecosystem structure as sketched out in Fig. 2 for a single ensemble member of Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model Competing interests. The authors declare that they have no conflict of interest. Competing interests. The authors declare that they have no conflict of interest. Acknowledgements. This research was supported as part of the Next-Generation Ecosystem Experiments – Tropics, funded by the US Department of Energy, Office of Science, Office of Biologi- cal and Environmental Research. Charles D. Koven also acknowl- edges support from the DOE Early Career Research Program. LBNL is managed and operated by the Regents of the University of California under prime contract no. DE-AC02-05CH11231. Al- istair Rogers and Shawn P. Serbin were also supported through the United States Department of Energy contract no. DE-SC0012704 to Brookhaven National Laboratory. Michael C. Dietze was supported by NSF 1458021. Oak Ridge National Laboratory is operated by UT–Battelle, LLC, under contract no. DE-AC05-00OR22725 with the US Department of Energy. Pacific Northwest National Labo- ratory is operated by the DOE by the Battelle Memorial Institute under contract DE-AC05-76RL01830. Matteo Detto was supported by the Carbon Mitigation Initiative at Princeton University. N., Keeling, H., Killeen, T. J., Laurance, W. F., Laurance, S., Li- cona, J., Magnussen, W. E., Marimon, B. S., Marimon-Junior, B. H., Mendoza, C., Neill, D. A., Nogueira, E. M., Nunez, P., Pal- lqui Camacho, N. C., Parada, A., Pardo-Molina, G., Peacock, J., Pena-Claros, M., Pickavance, G. C., Pitman, N. C. A., Poorter, Prieto, A., Quesada, C. A., Ramirez, F., Ramirez-Angul Restrepo, Z., Roopsind, A., Rudas, A., Salomao, R. P., Schwarz, M., Silva, N., Silva-Espejo, J. E., Silveira, M., Stropp, J., Tal- bot, J., ter Steege, H., Teran-Aguilar, J., Terborgh, J., Thomas- Caesar, R., Toledo, M., Torello-Raventos, M., Umetsu, R. K., van der Heijden, G. M. F., van der Hout, P., Guimaraes Vieira, I. C., Vieira, S. A., Vilanova, E., Vos, V. A., and Zagt, R. J.: Long-term decline of the Amazon carbon sink, Nature, 519, 344–348, 2015. Chave, J.: Neutral theory and community ecology, Ecol. Lett., 7, 241–253, 2004. Chave, J., Condit, R., Lao, S., Caspersen, J. P., Foster, R. B., and Hubbell, S. P.: Spatial and temporal variation of biomass in a tropical forest: results from a large census plot in Panama, J. Ecol., 91, 240–252, 2003. Financial support. This research has been supported by the US De- partment of Energy, Office of Biological and Environmental Re- search (Next-Generation Ecosystem Experiments – Tropics), the DOE Early Career Research Program, the United States Depart- ment of Energy (Brookhaven National Laboratory (contract no. DE- SC0012704)), and the NSF (grant no. 1458021). References Data availability. FATES output files from all simulations de- scribed here are archived at https://doi.org/10.15486/ngt/1569647 (Koven, 2019a). Data availability. FATES output files from all simulations de- scribed here are archived at https://doi.org/10.15486/ngt/1569647 (Koven, 2019a). Baraloto, C., Timothy Paine, C. E., Poorter, L., Beauchene, J., Bonal, D., Domenach, A.-M., Hérault, B., Patiño, S., Roggy, J.- C., and Chave, J.: Decoupled leaf and stem economics in rain forest trees, Ecol. Lett., 13, 1338–1347, 2010. Video supplement. SV1 is an animated version of Fig. 15d, show- ing 600 years of forest development from bare ground for one en- semble member. Bonan, G. B., Oleson, K. W., Fisher, R. A., Lasslop, G., and Reich- stein, M.: Reconciling leaf physiological traits and canopy flux data: Use of the TRY and FLUXNET databases in the Com- munity Land Model version 4, J. Geophys. Res., 117, G02026, https://doi.org/10.1029/2011JG001913, 2012. The video is available at: https://doi.org/10.5446/43627 (Koven, 2019b). Booth, B. B. B., Jones, C. D., Collins, M., Totterdell, I. J., Cox, P. M., Sitch, S., Huntingford, C., Betts, R. A., Harris, G. R., and Lloyd, J.: High sensitivity of future global warming to land carbon cycle processes, Environ. Res. Lett., 7, 024002, https://doi.org/10.1088/1748-9326/7/2/024002, 2012. Author contributions. CDK, RGK, RAF, JQC, BOC, MCD, JH, MH, MK, LMK, GL, EM, JFN, TP, SPS, JKS, ALSS, APW, and CX contributed to the development of the FATES model. SJD, MD, BF, NGM, HCML, RJN, AR, SPS, CV, APW, and SJW contributed to observational data used for input to or testing of the FATES model. CDK designed and performed model experiments and an- alyzed model output, with input from RGK and RAF. CDK wrote the paper, with input from all coauthors. Author contributions. CDK, RGK, RAF, JQC, BOC, MCD, JH, MH, MK, LMK, GL, EM, JFN, TP, SPS, JKS, ALSS, APW, and CX contributed to the development of the FATES model. SJD, MD, BF, NGM, HCML, RJN, AR, SPS, CV, APW, and SJW contributed to observational data used for input to or testing of the FATES model. CDK designed and performed model experiments and an- alyzed model output, with input from RGK and RAF. CDK wrote the paper, with input from all coauthors. Brienen, R. J. W., Phillips, O. L., Feldpausch, T. R., Gloor, E., Baker, T. R., Lloyd, J., Lopez-Gonzalez, G., Monteagudo- Mendoza, A., Malhi, Y., Lewis, S. 4 Conclusions This emphasizes the need to better repre- sent trade-offs that equalize competitive performance among species, so as to limit the competitive ability of any given functional type to outcompete other types. These effects of competition are only partially dependent on filtering that may occur from one generation to the next, as they are also strongly present even when we prevent advantages in popu- Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3040 References L., Vasquez Martinez, R., Alexiades, M., Alvarez Davila, E., Alvarez-Loayza, P., Andrade, A., Aragao, L. E. O. C., Araujo-Murakami, A., Arets, E. J. M. M., Arroyo, L., Aymard C., G. A., Banki, O. S., Baraloto, C., Bar- roso, J., Bonal, D., Boot, R. G. A., Camargo, J. L. C., Castilho, C. V., Chama, V., Chao, K. J., Chave, J., Comiskey, J. A., Cornejo Valverde, F., da Costa, L., de Oliveira, E. A., Di Fiore, A., Er- win, T. L., Fauset, S., Forsthofer, M., Galbraith, D. R., Grahame, E. S., Groot, N., Herault, B., Higuchi, N., Honorio Coronado, E. N., Keeling, H., Killeen, T. J., Laurance, W. F., Laurance, S., Li- cona, J., Magnussen, W. E., Marimon, B. S., Marimon-Junior, B. H., Mendoza, C., Neill, D. A., Nogueira, E. M., Nunez, P., Pal- lqui Camacho, N. C., Parada, A., Pardo-Molina, G., Peacock, J., Pena-Claros, M., Pickavance, G. C., Pitman, N. C. A., Poorter, L., Prieto, A., Quesada, C. A., Ramirez, F., Ramirez-Angulo, H., Restrepo, Z., Roopsind, A., Rudas, A., Salomao, R. P., Schwarz, M., Silva, N., Silva-Espejo, J. E., Silveira, M., Stropp, J., Tal- bot, J., ter Steege, H., Teran-Aguilar, J., Terborgh, J., Thomas- Caesar, R., Toledo, M., Torello-Raventos, M., Umetsu, R. K., van der Heijden, G. M. F., van der Hout, P., Guimaraes Vieira, I. C., Vieira, S. A., Vilanova, E., Vos, V. A., and Zagt, R. J.: Long-term decline of the Amazon carbon sink, Nature, 519, 344–348, 2015. C. D. Koven et al.: Benchmarking and sensitivity of FATES model I.: As- sessing uncertainties in a second-generation dynamic vegetation model caused by ecological scale limitations, New Phytol., 187, 666–681, 2010. Connell, J. H.: On the role of natural enemies in preventing compet- itive exclusion in some marine animals and in rain forest trees, Dynamics of populations, 298, 298–312, 1971. Fisher, R. A., Muszala, S., Verteinstein, M., Lawrence, P., Xu, C., McDowell, N. G., Knox, R. G., Koven, C., Holm, J., Rogers, B. M., Spessa, A., Lawrence, D., and Bonan, G.: Taking off the training wheels: the properties of a dynamic vegetation model without climate envelopes, CLM4.5(ED), Geosci. Model Dev., 8, 3593–3619, https://doi.org/10.5194/gmd-8-3593-2015, 2015. Cox, P. M., Betts, R. A., Jones, C. D., Spall, S. A., and Totterdell, I. J.: Acceleration of global warming due to carbon-cycle feed- backs in a coupled climate model, Nature, 408, 184–187, 2000. CTSM Development Team: ESCOMP/CTSM: Update documen- tation for release-clm5.0 branch, and fix issues with no-anthro surface dataset creation (Version release-clm5.0.34), Zenodo, https://doi.org/10.5281/zenodo.3739617, 2020. Fisher, R. A., Koven, C. D., Anderegg, W. R. L., Christoffersen, B. O., Dietze, M. C., Farrior, C. E., Holm, J. A., Hurtt, G. C., Knox, R. G., Lawrence, P. J., Lichstein, J. W., Longo, M., Matheny, A. M., Medvigy, D., Muller-Landau, H. C., Powell, T. L., Serbin, S. P., Sato, H., Shuman, J. K., Smith, B., Trugman, A. T., Viskari, T., Verbeeck, H., Weng, E., Xu, C., Xu, X., Zhang, T., and Moor- croft, P. R.: Vegetation demographics in Earth System Models: A review of progress and priorities, Glob. Change Biol., 24, 35–54, 2018. Detto, M., Baldocchi, D., and Katul, G. G.: Scaling Properties of Biologically Active Scalar Concentration Fluctuations in the At- mospheric Surface Layer over a Managed Peatland, Bound.-Lay. Meteorol., 136, 407–430, 2010. M., Medvigy, D., Muller-Landau, H. C., Powell, T. L., Serbin, S. Detto, M., Wright, S. J., Calderón, O., and Muller-Landau, H. C.: Resource acquisition and reproductive strategies of tropi- cal forest in response to the El Niño–Southern Oscillation, Nat. Commun., 9, 913, https://doi.org/10.1038/s41467-018-03306-9, 2018. Friedlingstein, P., Meinshausen, M., Arora, V. K., Jones, C. D., Anav, A., Liddicoat, S. K., and Knutti, R.: Uncertainties in CMIP5 climate projections due to carbon cycle feedbacks, J. Cli- mate, 27, 511–526, https://doi.org/10.1175/JCLI-D-12-00579.1, 2013. Dietze, M. C., Serbin, S. P., Davidson, C., Desai, A. C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3041 Hillslope Hydrology in Global Change Research and Earth Sys- tem Modeling, Water Resour. Res., 55, 1737–1772, 2019. cuccini, M., McDowell, N. G., and Meir, P.: Linking hydraulic traits to tropical forest function in a size-structured and trait- driven model (TFS v.1-Hydro), Geosci. Model Dev., 9, 4227– 4255, https://doi.org/10.5194/gmd-9-4227-2016, 2016. Hillslope Hydrology in Global Change Research and Earth Sys- tem Modeling, Water Resour. Res., 55, 1737–1772, 2019. cuccini, M., McDowell, N. G., and Meir, P.: Linking hydraulic traits to tropical forest function in a size-structured and trait- driven model (TFS v.1-Hydro), Geosci. Model Dev., 9, 4227– 4255, https://doi.org/10.5194/gmd-9-4227-2016, 2016. Farrior, C. E., Bohlman, S. A., Hubbell, S., and Pacala, S. W.: Dom- inance of the suppressed: Power-law size structure in tropical forests, Science, 351, 155–157, 2016. Clark, A. T., Lehman, C., and Tilman, D.: Identifying mechanisms that structure ecological communities by snapping model param- eters to empirically observed tradeoffs, Ecol. Lett., 21, 494–505, 2018. FATES Development Team: Functionally-Assembled Ter- restrial Ecosystem Simulator (commit 0bc7a5d), Zenodo, https://doi.org/10.5281/zenodo.3875687, 2019. FATES Development Team: The Functionally Assembled Terrestrial Ecosystem Simulator (FATES) (Version sci.1.35.5_api.11.0.0), Zenodo, https://doi.org/10.5281/zenodo.3825474, 2020. Clark, M. P., Bierkens, M. F. P., Samaniego, L., Woods, R. A., Ui- jlenhoet, R., Bennett, K. E., Pauwels, V. R. N., Cai, X., Wood, A. W., and Peters-Lidard, C. D.: The evolution of process-based hydrologic models: historical challenges and the collective quest for physical realism, Hydrol. Earth Syst. Sci., 21, 3427–3440, https://doi.org/10.5194/hess-21-3427-2017, 2017. Faybishenko, B., Paton, S., Powell, T., Knox, R., Pastorello, G., Varadharajan, C., Christianson, D., and Agarwal, D.: QA/QC-ed BCI meteorological drivers, NGEE – Tropics, https://doi.org/10.15486/NGT/1423307, 2018. Collier, N., Hoffman, F. M., Lawrence, D. M., Keppel-Aleks, G., Koven, C. D., Riley, W. J., Mu, M., and Randerson, J. T.: The International Land Model Benchmarking (ILAMB) System: De- sign, Theory, and Implementation, J. Adv. Model. Earth Sy., 10, 2731–2754, 2018. Fisher, R. A. and Koven, C. D.: Perspectives on the future of Land Surface Models and the challenges of represent- ing complex terrestrial systems, J. Adv. Model. Earth Sy., https://doi.org/10.1029/2018MS001453, online first, 2020. Condit, R., Pérez, R., Lao, S., Aguilar, S., and Hubbell, S. P.: Demographic trends and climate over 35 years in the Barro Colorado 50 ha plot, Forest Ecosystems, 4, 17, https://doi.org/10.1186/s40663-017-0103-1, 2017. Fisher, R. A., McDowell, N., Purves, D., Moorcroft, P., Sitch, S., Cox, P., Huntingford, C., Meir, P., and Woodward, F. Competing interests. The authors declare that they have no conflict of interest. Chave, J., Réjou-Méchain, M., Búrquez, A., Chidumayo, E., Col- gan, M. S., Delitti, W. B. C., Duque, A., Eid, T., Fearnside, P. M., Goodman, R. C., Henry, M., Martínez-Yrízar, A., Mugasha, W. A., Muller-Landau, H. C., Mencuccini, M., Nelson, B. W., Ngomanda, A., Nogueira, E. M., Ortiz-Malavassi, E., Pélissier, R., Ploton, P., Ryan, C. M., Saldarriaga, J. G., and Vieilledent, G.: Improved allometric models to estimate the aboveground biomass of tropical trees, Glob. Change Biol., 20, 3177–3190, 2014. Review statement. This paper was edited by Sönke Zaehle and re- viewed by Emilie Joetzjer and one anonymous referee. Chesson, P.: Mechanisms of Maintenance of Species Diversity, Annu. Rev. Ecol. Syst., 31, 343–366, 2000. Christoffersen, B. O., Gloor, M., Fauset, S., Fyllas, N. M., Gal- braith, D. R., Baker, T. R., Kruijt, B., Rowland, L., Fisher, R. A., Binks, O. J., Sevanto, S., Xu, C., Jansen, S., Choat, B., Men- https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model H., Xie, Koven, C.: 600 Years of Forest Dynamics using FATES model at Barro Colorado Island, Panama, TIB, https://doi.org/10.5446/43627, 2019b. S., Yang, Y., Yoon, J., Zelinka, M. D., Zender, C. S., Zeng, X., Zhang, C., Zhang, K., Zhang, Y., Zheng, X., Zhou, T., and Zhu, Q.: The DOE E3SM Coupled Model Version 1: Overview and Evaluation at Standard Resolution, J. Adv. Model. Earth Sy., 108, 2089–2129, https://doi.org/10.1029/2018MS001603, 2019. Koven, C.: NGEET/testbeds: version 1.0 of testbeds scripts, as used in FATES BCI analysis (Version v1.0), Zenodo, https://doi.org/10.5281/zenodo.3785705, 2020a. Gregory, J. M., Jones, C. D., Cadule, P., and Friedlingstein, P.: Quantifying Carbon Cycle Feedbacks, J. Climate, 22, 5232– 5250, 2009. Koven, C.: ckoven/runscripts: version 1.0 of ckoven/runscripts (Ver- sion v1.0), Zenodo, https://doi.org/10.5281/zenodo.3785703, 2020b. Gu, L., Norby, R., Haworth, I., Jensen, A., Turner, B., Walker, A., Warren, J., Weston, D., and Winter, K.: Photosynthetic parame- ters and nutrient content of trees at the Panama crane sites, NGEE – Tropics, https://doi.org/10.15486/NGT/1255260, 2016. Koven, C. D., Chambers, J. Q., Georgiou, K., Knox, R., Negron- Juarez, R., Riley, W. J., Arora, V. K., Brovkin, V., Friedlingstein, P., and Jones, C. D.: Controls on terrestrial carbon feedbacks by productivity versus turnover in the CMIP5 Earth System Models, Biogeosciences, 12, 5211–5228, https://doi.org/10.5194/bg-12- 5211-2015, 2015. Huang, M., Xu, Y., Longo, M., Keller, M., Knox, R., Koven, C., and Fisher, R.: Assessing impacts of selective logging on water, energy, and carbon budgets and ecosystem dynamics in Amazon forests using the Functionally Assembled Terrestrial Ecosystem Simulator, Biogeosciences Discuss., https://doi.org/10.5194/bg- 2019-129, in review, 2019. Kovenock, M.: Ecosystem and large-scale climate impacts of plant leaf dynamics, PhD, University of Washington, avail- able at: https://digital.lib.washington.edu/researchworks/handle/ 1773/44061 (last access: 15 May 2020), 2019. Hubbell, S. P.: The Unified Neutral Theory of Biodiversity and Biogeography (MPB-32), in: Monographs in Population Biol- ogy, Vol. 32, Princeton University Press, New Jersey, USA, https://doi.org/10.1515/9781400837526, 2011. Lasslop, G., Reichstein, M., Papale, D., Richardson, A. D., Ar- neth, A., Barr, A., Stoy, P., and Wohlfahrt, G.: Separation of net ecosystem exchange into assimilation and respiration using a light response curve approach: critical issues and global evalu- ation, Glob. Change Biol., 16, 187–208, 2010. Hubbell, S. P., Foster, R. B., O’Brien, S. T., Harms, K. E., Con- dit, R., Wechsler, B., Wright, S. J., and de Lao, S. L.: Light-Gap Disturbances, Recruitment Limitation, and Tree Diversity in a Neotropical Forest, Science, 283, 554–557, 1999. Lawrence, D. M., Fisher, R. A., Koven, C. C. D. Koven et al.: Benchmarking and sensitivity of FATES model D., Oleson, K. W., Swen- son, S. C., Bonan, G., Collier, N., Ghimire, B., van Kampenhout, L., Kennedy, D., Kluzek, E., Lawrence, P. J., Li, F., Li, H., Lom- bardozzi, D., Riley, W. J., Sacks, W. J., Shi, M., Vertenstein, M., Wieder, W. R., Xu, C., Ali, A. A., Badger, A. M., Bisht, G., van den Broeke, M., Brunke, M. A., Burns, S. P., Buzan, J., Clark, M., Craig, A., Dahlin, K., Drewniak, B., Fisher, J. B., Flanner, M., Fox, A. M., Gentine, P., Hoffman, F., Keppel-Aleks, G., Knox, R., Kumar, S., Lenaerts, J., Leung, L. R., Lipscomb, W. H., Lu, Y., Pandey, A., Pelletier, J. D., Perket, J., Randerson, J. T., Ricci- uto, D. M., Sanderson, B. M., Slater, A., Subin, Z. M., Tang, J., Thomas, R. Q., Val Martin, M., and Zeng, X.: The Community Land Model version 5: Description of new features, benchmark- ing, and impact of forcing uncertainty, J. Adv. Model. Earth Sy., 11, 4245–4287, https://doi.org/10.1029/2018MS001583, 2019. Huntingford, C., Zelazowski, P., Galbraith, D., Mercado, L. M., Sitch, S., Fisher, R., Lomas, M., Walker, A. P., Jones, C. D., Booth, B. B. B., Malhi, Y., Hemming, D., Kay, G., Good, P., Lewis, S. L., Phillips, O. L., Atkin, O. K., Lloyd, J., Gloor, E., Zaragoza-Castells, J., Meir, P., Betts, R., Harris, P. P., Nobre, C., Marengo, J., and Cox, P. M.: Simulated resilience of tropi- cal rainforests to CO2-induced climate change, Nat. Geosci., 6, 268–273, 2013. Craig, A., Dahlin, K., Drewniak, B., Fisher, J. B., Flanner, M., Fox, A. M., Gentine, P., Hoffman, F., Keppel-Aleks, G., Knox, R., Kumar, S., Lenaerts, J., Leung, L. R., Lipscomb, W. H., Lu, Y., Pandey, A., Pelletier, J. D., Perket, J., Randerson, J. T., Ricci- Janzen, D. H.: Herbivores and the Number of Tree Species in Trop- ical Forests, Am. Nat., 104, 501–528, 1970. to, D. M., Sanderson, B. M., Slater, A., Subin, Z. M., Tang, J., Thomas, R. Q., Val Martin, M., and Zeng, X.: The Community Land Model version 5: Description of new features, benchmark- ing, and impact of forcing uncertainty, J. Adv. Model. Earth Sy., 11, 4245–4287, https://doi.org/10.1029/2018MS001583, 2019. Johnson, D. J., Needham, J., Xu, C., Massoud, E. C., Davies, S. J., Anderson-Teixeira, K. J., Bunyavejchewin, S., Chambers, J. Q., Chang-Yang, C.-H., Chiang, J.-M., Chuyong, G. B., Condit, R., Cordell, S., Fletcher, C., Giardina, C. P., Giambelluca, T. C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3042 G., Griffin, B. M., Hannay, C., Harrop, B. E., Hoffman, M. J., Hunke, E. C., Jacob, R. L., Jacobsen, D. W., Jeffery, N., Jones, G., Griffin, B. M., Hannay, C., Harrop, B. E., Hoffman, M. J., Hunke, E. C., Jacob, R. L., Jacobsen, D. W., Jeffery, N., Jones, P. W., Keen, N. D., Klein, S. A., Larson, V. E., Leung, L. R., Li, H., Lin, W., Lipscomb, W. H., Ma, P., Mahajan, S., Maltrud, M. E., Mametjanov, A., McClean, J. L., McCoy, R. B., Neale, R. B., Price, S. F., Qian, Y., Rasch, P. J., Reeves Eyre, J. E. J., Ri- ley, W. J., Ringler, T. D., Roberts, A. F., Roesler, E. L., Salinger, A. G., Shaheen, Z., Shi, X., Singh, B., Tang, J., Taylor, M. A., Thornton, P. E., Turner, A. K., Veneziani, M., Wan, H., Wang, H., Wang, S., Williams, D. N., Wolfram, P. J., Worley, P. H., Xie, S., Yang, Y., Yoon, J., Zelinka, M. D., Zender, C. S., Zeng, X., Zhang, C., Zhang, K., Zhang, Y., Zheng, X., Zhou, T., and Zhu, Q.: The DOE E3SM Coupled Model Version 1: Overview and Evaluation at Standard Resolution, J. Adv. Model. Earth Sy., 108, 2089–2129, https://doi.org/10.1029/2018MS001603, 2019. Jones, C., Lowe, J., Liddicoat, S., and Betts, R.: Committed terres- trial ecosystem changes due to climate change, Nat. Geosci., 2, 484–487, 2009. P. W., Keen, N. D., Klein, S. A., Larson, V. E., Leung, L. R., Li, H., Lin, W., Lipscomb, W. H., Ma, P., Mahajan, S., Maltrud, Korobeynikov, A., Shlemov, A., Usevich, K., and Golyandina, N.: A Collection of Methods for Singular Spectrum Analy- sis, Package “Rssa.”, available at: https://cran.r-project.org/web/ packages/Rssa/index.html (last access: 15 May 2020), 2017. M. E., Mametjanov, A., McClean, J. L., McCoy, R. B., Neale, R. B., Price, S. F., Qian, Y., Rasch, P. J., Reeves Eyre, J. E. J., Ri- ley, W. J., Ringler, T. D., Roberts, A. F., Roesler, E. L., Salinger, A. G., Shaheen, Z., Shi, X., Singh, B., Tang, J., Taylor, M. A., Koven, C.: FATES Parameters and Output for Parameter Sensitivity at the Panama Barro Colorado Island Testbed, NGEE – Tropics, https://doi.org/10.15486/ngt/1569647, 2019a. Thornton, P. E., Turner, A. K., Veneziani, M., Wan, H., Wang, H., Wang, S., Williams, D. N., Wolfram, P. J., Worley, P. H., Xie, H., Wang, S., Williams, D. N., Wolfram, P. J., Worley, P. C. D. Koven et al.: Benchmarking and sensitivity of FATES model R., Feng, X., Kelly, R., Kooper, R., LeBauer, D., Mantooth, J., McHenry, K., and Wang, D.: A quantitative assessment of a terrestrial biosphere model’s data needs across North American biomes, J. Geophys. Res.-Biogeo., 119, 286–300, https://doi.org/10.1002/2013JG002392, 2014. Friend, A. D., Lucht, W., Rademacher, T. T., Keribin, R., Betts, R., Cadule, P., Ciais, P., Clark, D. B., Dankers, R., Falloon, P. D., Ito, A., Kahana, R., Kleidon, A., Lomas, M. R., Nishina, K., Ostberg, S., Pavlick, R., Peylin, P., Schaphoff, S., Vuichard, N., Warsza- wski, L., Wiltshire, A., and Woodward, F. I.: Carbon residence time dominates uncertainty in terrestrial vegetation responses to future climate and atmospheric CO2, P. Natl. Acad. Sci. USA, 111, 3280–3285, 2014. E3SM Project: Energy Exascale Earth System Model, Com- puter Software, https://doi.org/10.11578/E3SM/dc.20180418.36, 2018. Fan, Y., Clark, M., Lawrence, D. M., Swenson, S., Band, L. E., Brantley, S. L., Brooks, P. D., Dietrich, W. E., Flores, A., Grant, G., Kirchner, J. W., Mackay, D. S., McDonnell, J. J., Milly, P. C. D., Sullivan, P. L., Tague, C., Ajami, H., Chaney, N., Hart- mann, A., Hazenberg, P., McNamara, J., Pelletier, J., Perket, J., Rouholahnejad-Freund, E., Wagener, T., Zeng, X., Beighley, E., Buzan, J., Huang, M., Livneh, B., Mohanty, B. P., Nijssen, B., Safeeq, M., Shen, C., Verseveld, W., Volk, J., and Yamazaki, D.: Golaz, J., Caldwell, P. M., Van Roekel, L. P., Petersen, M. R., Tang, Q., Wolfe, J. D., Abeshu, G., Anantharaj, V., Asay-Davis, X. S., Bader, D. C., Baldwin, S. A., Bisht, G., Bogenschutz, P. A., Branstetter, M., Brunke, M. A., Brus, S. R., Burrows, S. M., Cameron-Smith, P. J., Donahue, A. S., Deakin, M., Easter, R. C., Evans, K. J., Feng, Y., Flanner, M., Foucar, J. G., Fyke, J. n, J., Huang, M., Livneh, B., Mohanty, B. P., Nijssen, https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model L. M.: Optimisation of photosynthetic carbon gain and within- canopy gradients of associated foliar traits for Amazon forest trees, Biogeosciences, 7, 1833–1859, https://doi.org/10.5194/bg- 7-1833-2010, 2010. J. A., Makana, J., Malhi, Y., Marthews, T. R., McShea, W., Mo- hamad, M. Bt., Morecroft, M. D., Norden, N., Parker, G., Shringi, A., Sukumar, R., Sun, I., Suresh, H., Tan, S., Thomas, D. W., Thompson, J., Uriarte, M., Valencia, R., Yao, T., Yap, S. L., Yuan, Z., Yuehua, H., Zimmerman, J. K., Zuleta, D., and McMahon, S. M.: Demographic functional types determine carbon dynamics of global forests in preparation 2020 A., Sukumar, R., Sun, I., Suresh, H., Tan, S., Thomas, D. W., Thompson, J., Uriarte, M., Valencia, R., Yao, T., Yap, S. L., Yuan, p p Z., Yuehua, H., Zimmerman, J. K., Zuleta, D., and McMahon, S. López-de-Lacalle, J.: Detection of Outliers in Time Series, Package “tsoutliers”, available at: https://cran.r-project.org/web/ packages/tsoutliers/index.html (last access: 15 May 2020), 2019. M.: Demographic functional types determine carbon dynamics of global forests, in preparation, 2020. O’Brien, S. T., Hubbell, S. P., Spiro, P., Condit, R., and Foster, Luo, Y. Q., Randerson, J. T., Abramowitz, G., Bacour, C., Blyth, E., Carvalhais, N., Ciais, P., Dalmonech, D., Fisher, J. B., Fisher, R., Friedlingstein, P., Hibbard, K., Hoffman, F., Huntzinger, D., Jones, C. D., Koven, C., Lawrence, D., Li, D. J., Ma- hecha, M., Niu, S. L., Norby, R., Piao, S. L., Qi, X., Peylin, P., Prentice, I. C., Riley, W., Reichstein, M., Schwalm, C., Wang, Y. P., Xia, J. Y., Zaehle, S., and Zhou, X. H.: A framework for benchmarking land models, Biogeosciences, 9, 3857–3874, https://doi.org/10.5194/bg-9-3857-2012, 2012. R. B.: Diameter, Height, Crown, and Age Relationship in Eight Neotropical Tree Species, Ecology, 76, 1926–1939, 1995. Oleson, K. W., Lawrence, D. M., Bonan, G. B., Drewniak, B., Huang, M., Koven, C. D., Levis, S., Li, F., Riley, W. J., Subin, Z. M., Swenson, S. C., Thornton, P. E., Bozbiyik, A., Fisher, R., Heald, C. L., Kluzek, E., Lamarque, J.-F., Lawrence, P. J., Le- ung, L. R., Lipscomb, W., Muszala, S., Ricciuto, D. M., Sacks, W., Tang, J., and Yang, Z.-L.: Technical Description of version 4.5 of the Community Land Model (CLM), National Center for Atmospheric Research, Boulder, Colorado, 2013. Malhi, Y., Aragão, L. E. O. C. D. Koven et al.: Benchmarking and sensitivity of FATES model C., Galbraith, D., Huntingford, C., Fisher, R., Zelazowski, P., Sitch, S., McSweeney, C., and Meir, P.: Exploring the likelihood and mechanism of a climate-change- induced dieback of the Amazon rainforest, P. Natl. Acad. Sci. USA, 106, 20610–20615, 2009. Osnas, J. L. D., Katabuchi, M., Kitajima, K., Wright, S. J., Reich, P. B., Van Bael, S. A., Kraft, N. J. B., Samaniego, M. J., Pacala, S. W., and Lichstein, J. W.: Divergent drivers of leaf trait variation within species, among species, and among functional groups, P. Natl. Acad. Sci. USA, 115, 5480–5485, 2018. Martínez Cano, I., Muller-Landau, H. C., Wright, S. J., Bohlman, S. A., and Pacala, S. W.: Tropical tree height and crown allome- tries for the Barro Colorado Nature Monument, Panama: a com- parison of alternative hierarchical models incorporating interspe- cific variation in relation to life history traits, Biogeosciences, 16, 847–862, https://doi.org/10.5194/bg-16-847-2019, 2019. Poorter, L., Bongers, L., and Bongers, F.: Architecture of 54 moist- forest tree species: traits, trade-offs, and functional groups, Ecol- ogy, 87, 1289–1301, 2006. Powell, T. L., Galbraith, D. R., Christoffersen, B. O., Harper, A., Imbuzeiro, H. M. A., Rowland, L., Almeida, S., Brando, P. M., da Costa, A. C. L., Costa, M. H., Levine, N. M., Malhi, Y., Saleska, S. R., Sotta, E., Williams, M., Meir, P., and Moorcroft, P. R.: Con- fronting model predictions of carbon fluxes with measurements of Amazon forests subjected to experimental drought, New Phy- tol., 200, 350–365, 2013. Massoud, E. C., Xu, C., Fisher, R. A., Knox, R. G., Walker, A. P., Serbin, S. P., Christoffersen, B. O., Holm, J. A., Kueppers, L. M., Ricciuto, D. M., Wei, L., Johnson, D. J., Chambers, J. Q., Koven, C. D., McDowell, N. G., and Vrugt, J. A.: Identification of key parameters controlling demographically structured vegeta- tion dynamics in a land surface model: CLM4.5(FATES), Geosci. Model Dev., 12, 4133–4164, https://doi.org/10.5194/gmd-12- 4133-2019, 2019. Powell, T. L., Koven, C. D., Johnson, D. J., Faybishenko, B., Fisher, R. A., Knox, R. G., McDowell, N. G., Condit, R., Hubbell, S. P., Wright, S. J., Chambers, J. Q., and Kueppers, L. M.: Variation in hydroclimate sustains tropical forest biomass and promotes functional diversity, New Phytol., 219, 932–946, 2018. McDowell, N., Allen, C. D., Anderson-Teixeira, K., Brando, P., Brienen, R., Chambers, J., Christoffersen, B., Davies, S., Doughty, C., Duque, A., Espirito-Santo, F., Fisher, R., Fontes, C. G., Galbraith, D., Goodsman, D., Grossiord, C., Hartmann, H., Holm, J., Johnson, D. C. D. Koven et al.: Benchmarking and sensitivity of FATES model W., Gu- natilleke, N., Gunatilleke, S., Hsieh, C.-F., Hubbell, S., Inman- Narahari, F., Kassim, A. R., Katabuchi, M., Kenfack, D., Litton, C. M., Lum, S., Mohamad, M., Nasardin, M., Ong, P. S., Os- tertag, R., Sack, L., Swenson, N. G., Sun, I. F., Tan, S., Thomas, D. W., Thompson, J., Umaña, M. N., Uriarte, M., Valencia, R., Yap, S., Zimmerman, J., McDowell, N. G., and McMahon, S. M.: Climate sensitive size-dependent survival in tropical trees, Nat. Ecol. Evol., 2, 1436–1442, 2018. Levine, N. M., Zhang, K., Longo, M., Baccini, A., Phillips, O. L., Lewis, S. L., Alvarez-Dávila, E., de Andrade, A. C. S., Brienen, R. J. W., Erwin, T. L., Feldpausch, T. R., Lorenzo Monteagudo Mendoza, A., Nuñez Vargas, P., Prieto, A., Silva-Espejo, J. E., Malhi, Y., and Moorcroft, P. R.: Ecosystem heterogeneity deter- mines the ecological resilience of the Amazon to climate change, P. Natl. Acad. Sci. USA, 113, 793–797, 2016. Lloyd, J., Patiño, S., Paiva, R. Q., Nardoto, G. B., Quesada, C. A., Santos, A. J. B., Baker, T. R., Brand, W. A., Hilke, I., Gielmann, H., Raessler, M., Luizão, F. J., Martinelli, L. A., and Mercado, Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 3043 C. D. Koven et al.: Benchmarking and sensitivity of FATES model 3044 forests emerges from plant trait diversity, Nat. Clim. Change, 6, 1032, https://doi.org/10.1038/nclimate3109, 2016. forests emerges from plant trait diversity, Nat. Clim. Change, 6, 1032, https://doi.org/10.1038/nclimate3109, 2016. Wright, S. J., Kitajima, K., Kraft, N. J. B., Reich, P. B., Wright, I. J., Bunker, D. E., Condit, R., Dalling, J. W., Davies, S. J., Díaz, Wright, S. J., Kitajima, K., Kraft, N. J. B., Reich, P. B., Wright, I. J., Bunker, D. E., Condit, R., Dalling, J. W., Davies, S. J., Díaz, S., Engelbrecht, B. M. J., Harms, K. E., Hubbell, S. P., Marks, C. O., Ruiz-Jaen, M. C., Salvador, C. M., and Zanne, A. E.: Func- tional traits and the growth–mortality trade-off in tropical trees, Ecology, 91, 3664–3674, 2010. Saldarriaga, J. G., West, D. C., Tharp, M. L., and Uhl, C.: Long- Term Chronosequence of Forest Succession in the Upper Rio Negro of Colombia and Venezuela, J. Ecol., 76, 938–958, 1988. S., Engelbrecht, B. M. J., Harms, K. E., Hubbell, S. P., Marks, C. O., Ruiz-Jaen, M. C., Salvador, C. M., and Zanne, A. E.: Func- tional traits and the growth–mortality trade-off in tropical trees, Ecology, 91, 3664–3674, 2010. Sevanto, S., McDowell, N. G., Dickman, L. T., Pangle, R., and Pockman, W. T.: How do trees die? A test of the hydraulic failure and carbon starvation hypotheses, Plant Cell Environ., 37, 153– 161, 2014. Wu, J., Serbin, S. P., Xu, X., Albert, L. P., Chen, M., Meng, R., Saleska, S. R., and Rogers, A.: The phenology of leaf quality and its within-canopy variation is essential for accurate modeling of photosynthesis in tropical evergreen forests, Glob. Change Biol., 23, 4814–4827, 2017. Tilman, D., Lehman, C. L., and Thomson, K. T.: Plant diversity and ecosystem productivity: theoretical considerations, P. Natl. Acad. Sci. USA, 94, 1857–1861, 1997. Wu, J., Rogers, A., Albert, L. P., Ely, K., Prohaska, N., Wolfe, B. T., Oliveira Jr., R. C., Saleska, S. R., and Serbin, S. P.: Leaf re- flectance spectroscopy captures variation in carboxylation capac- ity across species, canopy environment and leaf age in lowland moist tropical forests, New Phytol., 224, 663–674, 2019. Walker, A. P., Zaehle, S., Medlyn, B. E., De Kauwe, M. G., Asao, S., Hickler, T., Parton, W., Ricciuto, D., Wang, Y.-P., Wårlind, D., and Norby, R. J.: Predicting long-term carbon sequestra- tion in response to CO2 enrichment: How and why do current ecosystem models differ?, Global Biogeochem. Cy., 29, 476– 495, https://doi.org/10.1002/2014GB004995, 2015. C. D. Koven et al.: Benchmarking and sensitivity of FATES model J., Kassim, A. R., Keller, M., Koven, C., Kueppers, L., Kumagai, T. ’omi, Malhi, Y., McMahon, S. M., Mencuccini, M., Meir, P., Moorcroft, P., Muller-Landau, H. C., Phillips, O. L., Powell, T., Sierra, C. A., Sperry, J., Warren, J., Xu, C., and Xu, X.: Drivers and mechanisms of tree mortality in moist tropical forests, New Phytol., 219, 851–869, 2018. Purves, D. W., Lichstein, J. W., Strigul, N., and Pacala, S. W.: Predicting and understanding forest dynamics using a simple tractable model, P. Natl. Acad. Sci. USA, 105, 17018–17022, 2008. Raczka, B., Dietze, M. C., Serbin, S. P., and Davis, K. J.: What Lim- its Predictive Certainty of Long-Term Carbon Uptake?, J. Geo- phys. Res.-Biogeo., 123, 3570–3588, 2018. Rogers, A.: The use and misuse of Vc,max in Earth System Models, Photosynth. Res., 119, 15–29, 2014. Meakem, V., Tepley, A. J., Gonzalez-Akre, E. B., Herrmann, V., Muller-Landau, H. C., Wright, S. J., Hubbell, S. P., Condit, R., and Anderson-Teixeira, K. J.: Role of tree size in moist tropical forest carbon cycling and water deficit responses, New Phytol., 219, 947–958, 2018. Rogers, A., Serbin, S., Ely, K., Wu, J., Wolfe, B., Dick- man, T., Collins, A., Detto, M., Grossiord, C., McDow- ell, N., and Michaletz, S.: CO2 response (ACi) gas ex- change, calculated Vcmax & Jmax parameters, Feb2016– May2016, PA-SLZ, PA-PNM: Panama, NGEE – Tropics, https://doi.org/10.15486/NGT/1411867, 2017. Moorcroft, P. R., Hurtt, G. C., and Pacala, S. W.: A method for scaling vegetation dynamics: The ecosystem demography model (ED), Ecol. Monogr., 71, 557–585, 2001. Ryan, J. A., Ulrich, J. M., Bennett, R., and Joy, C.: eXtensible Time Series, Package “xts.”, available at: https://cran.r-project. org/web/packages/xts/ (last access: 15 May 2020), 2018. Needham, J. F., Johnson, D. J., Chang-Yang, C., Anderson-Teixeira, K. J., Bourg, N. A., Bunyavejchewin, S., Butt, N., Cao, M., Car- denas, D., Chen, Y., Chuyong, G., Dattaraja, H., Davies, S. J., Duque, A., Ewango, C. E. N., Fernando, E. S., Fisher, R. A., Fletcher, C. D., Foster, R., Hao, Z., Hart, T., Hsieh, C., Hubbell, S. P., Itoh, A., Kenfack, D., Koven, C. D., Larson, A. J., Lutz, Sakschewski, B., von Bloh, W., Boit, A., Poorter, L., Peña-Claros, M., Heinke, J., Joshi, J., and Thonicke, K.: Resilience of Amazon https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020 Biogeosciences, 17, 3017–3044, 2020 C. D. Koven et al.: Benchmarking and sensitivity of FATES model Xu, C.: Decoupling correlated and uncorrelated parametric uncer- tainty contributions for nonlinear models, Appl. Math. Model., 37, 9950–9969, 2013. Wieder, W. R., Lawrence, D. M., Fisher, R. A., Bonan, G. B., Cheng, S. J., Goodale, C. L., Grandy, A. S., Koven, C. D., Lom- bardozzi, D. L., Oleson, K. W., and Thomas, R. Q.: Beyond static benchmarking: Using experimental manipulations to evalu- ate land model assumptions, Global Biogeochem. Cy., 33, 1289– 1309, https://doi.org/10.1029/2018GB006141, 2019. Xu, C. and Gertner, G. Z.: Uncertainty and sensitivity analysis for models with correlated parameters, Reliab. Eng. Syst. Safe, 93, 1563–1573, 2008. Xu, X., Medvigy, D., Powers, J. S., Becknell, J. M., and Guan, K.: Diversity in plant hydraulic traits explains seasonal and inter- annual variations of vegetation dynamics in seasonally dry trop- ical forests, New Phytol., 212, 80–95, 2016. Wright, I. J., Reich, P. B., Westoby, M., Ackerly, D. D., Baruch, Z., Bongers, F., Cavender-Bares, J., Chapin, T., Cornelissen, J. H. C., Diemer, M., Flexas, J., Garnier, E., Groom, P. K., Gu- lias, J., Hikosaka, K., Lamont, B. B., Lee, T., Lee, W., Lusk, C., Midgley, J. J., Navas, M.-L., Niinemets, Ü., Oleksyn, J., Osada, N., Poorter, H., Poot, P., Prior, L., Pyankov, V. I., Roumet, C., Thomas, S. C., Tjoelker, M. G., Veneklaas, E. J., and Villar, R.: The worldwide leaf economics spectrum, Nature, 428, 821–827, 2004. Zeileis, A., Grothendieck, G., Ryan, J. A., Ulrich, J. M., and Andrews, F.: zoo: S3 Infrastructure for Regular and Irreg- ular Time Series (Z’s Ordered Observations), available at: https://cran.r-project.org/web/packages/zoo/index.html (last ac- cess: 15 May 2020), 2019. Biogeosciences, 17, 3017–3044, 2020 Biogeosciences, 17, 3017–3044, 2020 https://doi.org/10.5194/bg-17-3017-2020

https://openalex.org/W2167363548

https://bioone.org/journals/acta-palaeontologica-polonica/volume-59/issue-3/app.2011.0158/Miocene-Coniferous-Woods-of-the-Polish-Carpathian-Foredeep/10.4202/app.2011.0158.pdf

English

Miocene coniferous woods of the Polish Carpathian Foredeep

Acta Palaeontologica Polonica

cc-by

Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Introduction This article presents the results of studies of Miocene wood collected from the area of the Carpathian Foredeep. The re- search undertaken involved their anatomical analysis and taxonomic classification. Taxonomic identification of fossil woods contributes to the knowledge of assemblage com- position of fossil plants. It provides information about bio- geography, biodiversity, and floral dynamics in geological time. Fossil wood classification can also be employed in the study of morphological and anatomical changes of trees, their phylogeny and evolution. Determination of fossil wood taxa is often used to reconstruct palaeoenvironmental condi- tions that prevailed while the ancient trees were alive. More- over, woods provide also opportunity for the examination of their growth rings (Chaloner and Creber 1973; Creber and Chaloner 1984). The presence of growth rings, ring width, the ratio of early and latewood zones and the occurrence of any disturbances within the wood structure can be useful in identifying the environmental factors affecting tree growth (Francis and Poole 2002; Falcon-Lang 2005; Vaganov et al. 2011; Olano et al. 2012). In the present paper the analysed woods, particularly the observations made on their growth rings were useful to conclude about the palaeoenvironmental Institutional abbreviations.—MZ, Museum of the Earth, Warsaw, Poland. Institutional abbreviations.—MZ, Museum of the Earth, Warsaw, Poland. Institutional abbreviations.—MZ, Museum of the Earth, Warsaw, Poland. Other abbreviations.—RLS, radial longitudinal plane; TLS, tangential longitudinal plane; TS, transverse plane. MARZENA KŁUSEK Kłusek, M. 2014. Miocene coniferous woods of the Polish Carpathian Foredeep. Acta Palaeontologica Polonica 59 (3): 697–708. This paper presents the results of analysis of the Miocene woods collected from three sampling sites in the Polish part of the Carpathian Foredeep. The location in the stratigraphic profile and the state of fossil preservation indicate that the woods were transported by waters flowing from highland or from mountainous regions and then deposited and fossilised in delta areas. Fragments of 16 lignified or silicified woods were collected for anatomical research. The fossil woods were classified as Piceoxylon piceae, Juniperoxylon pachyderma, Cupressinoxylon canadense and Cupressinoxylon poloni- cum. Disturbance zones observed in Cupressinoxylon polonicum are interpreted as frost rings. Frost rings suggest that despite the documented existence of warm and humid climatic conditions during the Miocene period, the temperatures in mountainous areas could temporarily drop below freezing point. Key words: Coniferophyta, fossil woods, palaeoecology, wood anatomy, Miocene, Poland. Marzena Kłusek [[email protected]], BOKU University of Natural Resources and Applied Life Sciences Vien- na, University Research Center Tulln, Konrad Lorenz Strasse 24, 3430 Tulln an der Donau, Austria. Marzena Kłusek [[email protected]], BOKU University of Natural Resources and Applied Life Sciences Vien- na, University Research Center Tulln, Konrad Lorenz Strasse 24, 3430 Tulln an der Donau, Austria. eceived 12 October 2011, accepted 29 October 2012, available online 7 November 2012. Copyright © 2014 M. Kłusek. This is an open-access article distributed under the terms of the Creative Commons At- tribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright © 2014 M. Kłusek. This is an open-access article distributed under the terms of the Creative Commons At- tribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. conditions prevailing during the Miocene in the area of the Carpathian Foredeep. Miocene Coniferous Woods of the Polish Carpathian Foredeep Author: Kłusek, Marzena Source: Acta Palaeontologica Polonica, 59(3) : 697-708 Published By: Institute of Paleobiology, Polish Academy of Sciences URL: https://doi.org/10.4202/app.2011.0158 BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titles in the biological, ecological, and environmental sciences published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use. Usage of BioOne Complete content is strictly limited to personal, educational, and non - commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder. BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Acta Palaeontol. Pol. 59 (3): 697–708, 2014 Piceoxylon piceae Roessler, 1937 1913 Picea sp.; Kraeusel 1913: 29. 1920 Piceoxylon sp.; Kraeusel 1920: 437, pl. 28: 4–6; pl. 36: 1–4. Fig. 2. Lithological log of Hebdów exposure. 27 25 20 15 10 5 0 loess (Quaternary) boulder clay (Quaternary) silts and mudstone (Middle Miocene) silts and fine-grained sand (Middle Miocene) calcareous sand (Middle Miocene) diamicton (Middle Miocene) sand with gravel and malacofauna (Middle Miocene) sand and silt with detritus (Middle Miocene) m ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 698 Fig. 1. Early Badenian palinspastic paleogeography map of the Polish Car- pathian Foredeep (after Oszczypko 2006, simplified). A. Hebdów locality. B. Tarnów-Gumniska locality. C. Zgłobice locality. peneplained land littoral zone neritic zone bathial zone fan-delta 0 Kraków Lwów Przemyśl 50 km A B C highlands and mountains Warszawa For the purposes of anatomical analysis, standard petro- graphic thin sections were used. Sections were prepared for both coalified and silicified specimens along three perpen- dicular planes: transverse (TS), radial longitudinal (RLS) and tangential longitudinal (TLS). Microscopic observation and photographic documentation of wood was carried out using an Olympus BX51 transmission light microscope. Samples were deposited in the Museum of the Earth (Polish Academy of Science) in Warsaw. Warszawa B At the request of the Editors, both fossil and extant plant taxa and higher units of the botanic system are cited with ful- ly spelt authors’ names and the year of publication, according to the zoological usage. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Geological setting The fossil woods which are the subject of this study came from three sampling sites situated in the Polish Carpathian Foredeep (Fig. 1), which is a part of a large sedimentary basin that stretches from the Danube in Vienna to the Southern Car- pathians in Romania. The Early Miocene to Middle Miocene Carpathian Foredeep developed as a peripheral foreland basin and its genesis is related to the advancing Carpathian Front. The progressive subsidence, accompanied by the Early Bad- enian transgression, flooded the foredeep and the marginal part of the Carpathian Mountains (Oszczypko 1998). As a result during the Middle Miocene the area of the Carpathian Foredeep was the northern part of the Central Paratethys, which was a warm sea of epicontinental characteristics. This relatively narrow brackish or freshwater basin was filled with Acta Palaeontol. Pol. 59 (3): 697–708, 2014 http://dx.doi.org/10.4202/app.2011.0158 Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Systematic paleontology Order Pinales Dumortier, 1829 Family Pinaceae Lindley, 1836 Morphogenus Piceoxylon Gothan, 1905 Order Pinales Dumortier, 1829 Family Pinaceae Lindley, 1836 Morphogenus Piceoxylon Gothan, 1905 Fig. 1. Early Badenian palinspastic paleogeography map of the Polish Car- pathian Foredeep (after Oszczypko 2006, simplified). A. Hebdów locality. B. Tarnów-Gumniska locality. C. Zgłobice locality. coarse clastic sediments derived from the emergent front of the Carpathians and from the platform area (Rögl 1999; Rögl et al. 2008). The analysed woods were deposited in layers of clay, silt, and sand of deltaic origin. The beds from which the fossil wood was taken formed proximal mouth bars, del- ta-fed marine fans and submarine landslides lithofacies. The mineral and petrographic composition of sediments indicates that they had been transported by the waters flowing from the Carpathian area and then deposited in a sedimentary basin at the foothill of the mountains. Especially the petrographic composition of gravel particles shows predominance of Car- pathian-derived sandstone and granitoid rocks (Brud 2002). The wood-containing strata belong to Skawia and Grabowiec Beds and also to Machów Formation. They were dated using microfossils as Middle Miocene (16.4–11.5 Ma) (Otfinowski 1973; Porębski 1999). coarse clastic sediments derived from the emergent front of the Carpathians and from the platform area (Rögl 1999; Rögl et al. 2008). The analysed woods were deposited in layers of clay, silt, and sand of deltaic origin. The beds from which the fossil wood was taken formed proximal mouth bars, del- ta-fed marine fans and submarine landslides lithofacies. The mineral and petrographic composition of sediments indicates that they had been transported by the waters flowing from the Carpathian area and then deposited in a sedimentary basin at the foothill of the mountains. Especially the petrographic composition of gravel particles shows predominance of Car- pathian-derived sandstone and granitoid rocks (Brud 2002). The wood-containing strata belong to Skawia and Grabowiec Beds and also to Machów Formation. They were dated using microfossils as Middle Miocene (16.4–11.5 Ma) (Otfinowski 1973; Porębski 1999). Type species: Piceoxylon pseudotsugae Gothan, 1906; Tertiary; Sierra Nevada, California, USA. Material and methods In total, 16 specimens of wood were collected from the Tarnów-Gumniska, Zgłobice, and Hebdów localities (Fig. 1). The size of the analysed specimens varied from several to twenty centimetres in length. The small size of specimens, the absence of bark and the abrasion of the wood surface confirm the assumption that woods were moved by rivers from highland or mountainous areas. Simultaneously, the low degree of roundness of spec- imen’s surfaces indicates that transport of the material took place before they fossilisation process. sand and silt with detritus (Middle Miocene) The material was preserved in the form of coalified or silicified fragments of trunks and branches. Their origin from specific morphological organs of plants was established on the basis of wood size and anatomy, ring curvature and the divergence of rays. Fig. 2. Lithological log of Hebdów exposure. KŁUSEK—MIOCENE CONIFEROUS WOODS 699 A B C D E F Fig. 3. Microphotographs of fossil wood Piceoxylon piceae Roessler, 1937 from Tarnów-Gumniska, Middle Miocene; specimen MZ LXVa. A. Vertical resin canals within reaction wood (TS). B. Crossfield pitting (RLS). C. Pitting on radial walls of tracheids (RLS). D. Bordered pits on tangential walls of tracheids (TLS). E. Latewood tracheids with spiral thickening (TLS). F. Uniseriate rays and rays with resin canals (TLS). Scale bars 50 μm. B F D D E Fig. 3. Microphotographs of fossil wood Piceoxylon piceae Roessler, 1937 from Tarnów-Gumniska, Middle Miocene; specimen MZ LXVa. A. Vertical resin canals within reaction wood (TS). B. Crossfield pitting (RLS). C. Pitting on radial walls of tracheids (RLS). D. Bordered pits on tangential walls of tracheids (TLS). E. Latewood tracheids with spiral thickening (TLS). F. Uniseriate rays and rays with resin canals (TLS). Scale bars 50 μm. 1932 Picea sp.; Slijper 1932: 21. and they are obliquely oriented. Distinct spiral thickenings are confined to latewood tracheids (Fig. 3E). 1937 Piceoxylon piceae Roessler, 1937; Roessler 1937: 82, pl. 3: 10; pl. 4: 11–12. 1937 Piceoxylon piceae Roessler, 1937; Roessler 1937: 82, pl. 3: 10; pl. 4: 11–12. Rays are uniseriate or partially biseriate, they have 3–20 cells in height. Some rays have resin canals (Fig. 3F). Hori- zontal resin ducts are usually located in the central part of the ray. Transverse and tangential walls of rays are thick and well pitted. Ray parenchyma cells show 2–6 piceoid pits (7–9 μm diameter) in each cross-field. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Material and methods 22: 5–6. 1920 Juniperoxylon sp.; Kraeusel 1920: 420, pl. 29: 4; pl. 36: 8. 1931 Juniperoxylon sp.; Jurasky 1931: 161. 1937 Juniperoxylon sp.; Roessler 1937: 77. g ( g ) Because of the presence of spiral thickenings confined to latewood cells, classification to the genus Pseudotsuga can be ruled out. However, identification of the analysed spec- imens as larch or spruce is quite a difficult task. The most reliable anatomical feature which allows these genera to be distinguished is the relative proportion of pit types in the ray tracheids. Ray tracheids of spruce usually have thick pit bor- ders with angular extensions into the cell lumen and small pit apertures, while smooth pit borders and large pit apertures of ray tracheids are the characteristic of larch (Bartholin 1979; Anagnost et al. 1994). The state of preservation of the anal- ysed wood, however, did not allow the types of pit in the ray tracheids to be established. Therefore, to determine the taxo- nomic affiliation of the specimens, other observed anatomical features were used: gradual transition between the earlywood and latewood zone, uniseriate pits on radial walls of tracheids, and transverse resin ducts usually located in the central part of the rays. These are typical features of spruce wood (Jagels et al. 2001; Blokhina and Afonin 2009). Comparison with the fossil wood taxa allows the analysed specimens to be clas- sified as the species Piceoxylon piceae Roessler, 1937. The diagnosis of Piceoxylon piceae species was formed on the basis of Pliocene wood from Gleichenberg, Austria (Roessler 1937). Its nearest living relative is Norway spruce (Picea ab- ies [Linnaeus, 1753] Karsten, 1881) (Roessler 1937). 1944 Juniperoxylon neosibiricum (Schmalhausen, 1890) Selling, 1944; Selling 1944: 14. 1949 Juniperoxylon pachyderma (Goeppert, 1850) Kraeusel, 1949; Kraeusel 1949: 177. Material.—Eight small specimens (2–10 cm long) of coali- fied and to a small extent silicified wood (MZ LXXa–h) from Zgłobice site. Description is based on specimen MZ LXXc (Fig. 4A–F). Description.—The wood constitutes fragments of trunks or large branches. Growth rings are narrow. Transition between earlywood and latewood zones is gradual. Axial parenchyma is abundant and arranged in tangential rows (Fig. 4A). Trans- verse walls of parenchyma cells have beadlike thickenings (Fig. 4F). Description.—The wood constitutes fragments of trunks or large branches. Growth rings are narrow. Transition between earlywood and latewood zones is gradual. Axial parenchyma is abundant and arranged in tangential rows (Fig. 4A). Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Material and methods Trans- verse walls of parenchyma cells have beadlike thickenings (Fig. 4F). Tracheids are thick-walled and rounded. On radial and tangential walls of tracheids occur uniseriate bordered pits (Fig. 4C). Pits on radial walls of tracheids posses circular borders (9–11 μm in diameter) and oval apertures. On tan- gential walls of tracheids pits are scattered and circular in outline (diameter 6–7 μm). These pits have oval or slit-like, obliquely oriented apertures (Fig. 4D). Rays are uniseriate, 1–15 cells tall, usually between 4 and 9 cells high (Fig. 4E). They consist of very thick and usually smooth transverse walls. Tangential walls contain beadlike or dentate thickenings. Indentures are absent. Crossfield regions possess 1–2 pits, arranged in one row (Fig. 4B). Pits have circular or oval outlines (7–8 μm in diameter) and oblique apertures. Rays often include resinous filling. Stratigraphic and geographic range.— Tertiary; Europe. Stratigraphic and geographic range.— Tertiary; Europe. Family Cupressaceae Richard ex Bartling, 1830 Morphogenus Juniperoxylon (Houlbert, 1910) Kraeusel, 1919 Family Cupressaceae Richard ex Bartling, 1830 Morphogenus Juniperoxylon (Houlbert, 1910) Remarks.—The lack of resin ducts, abundant axial parenchy- ma, type of tracheids pitting, cupressoid pits in the cross field, lack of ray tracheids and presence of thickenings in tangential walls of ray cells makes it possible to establish that the anal- ysed wood belongs to the species Juniperoxylon pachyderma (Goeppert, 1850) Kraeusel, 1949. The basionym of this taxon is Cupressinoxylon pachyderma species which was created by Goeppert in 1850 on the basis of Miocene wood from Łaża- ny. Juniperoxylon pachyderma was established by Kraeusel (1949), who transferred to this new species numerous previ- ously described woods (see list of synonyms). However he has not determined its holotype. Juniperoxylon pachyderma does not have also an unambiguous nearest living relative. Most probably, it is closely related to the present-day Juni- perus virginiana Linnaeus, 1753 wood (Müller-Stoll 1951). However, it is quite anatomically diversified and encompass- es trees belonging to different contemporary genera (Prill and Kraeusel 1919). The specimens analysed have features Kraeusel, 1919 Type species: Juniperoxylon turonense Houlbert, 1910; Tertiary, Mio- cene; Les Faluns de Touraine, France. Type species: Juniperoxylon turonense Houlbert, 1910; Tertiary, Mio- cene; Les Faluns de Touraine, France. Material and methods These pits are circular and pos- sess slit-like or elliptic apertures. Cross-field pits are usually grouped in two rows (Fig. 3B). Marginal ray tracheids have no spiral thickenings. Material.—Six specimens from Tarnów-Gumniska, marked as MZ LXVa–d, MZ LVIII, MZ LXXI. Wood is silicified, white or black, 15–25 cm long. Description is based on spec- imen MZ LXVa (Fig. 3A–F). Description.—The wood corresponds to fragments of trunks of at least 15 cm in diameter. Growth rings are distinct, the transition from early- to latewood is gradual. The vertical res- in canals have thick-walled epithelial cells. The resin ducts are commonly paired (Fig. 3A). The axial parenchyma is absent. One of the specimens from the Tarnów-Gumniska lo- cality (MZ LXVd) has the characteristic features of reac- tion wood. Reaction wood in conifers reveals thick-walled, rounded tracheids, an increased angle of microfibrils and of spiral thickenings, distorted tips and short length of cells (Scurfield 1973; Schweingruber 1996). All these anatomical features were observed in the analysed specimen. The pitting on radial walls of tracheids is uniseriate and pits are loosely spaced (Fig. 3C). Bordered pits are circular and they measure 13–17 μm in diameter. Pit apertures are circular or elliptical and obliquely oriented. Tangential walls of tracheids have circular bordered pits (6.5–8.5 μm in diam- eter). These pits are arranged in a single irregular, scattered row (Fig. 3D). Pit apertures have slit-like or elliptical shape Remarks.—The wood is characterised by the presence of axial and radial resin canals with thick-walled epithelial cells, the 700 ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 existence of well marked pits on the transverse wall of ray cells and the appearance of loosely spaced pitting on radial and tangential walls of tracheids. Moreover wood possesses spiral thickenings of longitudinal tracheids and ray tracheids with- out thickenings. All these features indicate that the analysed woods belong to the genus Piceoxylon Gothan, 1905 (Philippe and Bamford 2008). This taxon consists of the contemporary wood of Larix Miller, 1754, Picea D. Don ex Loudon, 1838 and Pseudotsuga Carriere, 1867 (Greguss 1967). 1913 Cupressinoxylon juniperoides Kraeusel, 1913; Kraeusel 1913: 35. 1913 Juniperus silesiaca Prill, 1913; Prill 1913: 30. 1919 Juniperoxylon silesiacum (Prill, 1913) Kraeusel, 1919; Prill and Kraeusel 1919: 297, pl. 22: 10. p 1913 Juniperus pauciporosa Prill, 1913; Prill 1913: 35. 1919 Juniperoxylon pauciporosum (Prill, 1913) Kraeusel, 1919; Prill and Kraeusel 1919: 302, pl. 22: 5–6. and Kraeusel 1919: 302, pl. Juniperoxylon pachyderma (Goeppert, 1850) Krae- usel, 1949 Juniperoxylon pachyderma (Goeppert, 1850) Krae- usel, 1949 Fig. 4A–F. 1850 Cupressinoxylon pachyderma Goeppert, 1850; Goeppert 1850: 199, pl. 25: 1, 2. 1872 Cupressoxylon pachyderma (Goeppert, 1850) Kraus, 1872; Kraus 1872: 374. 1890 Cupressinoxylon (Glyptostrobus) neosibiricum Schmalhausen, 1890; Schmalhausen 1890: 21, pl. 2: 44–49. 1850 Cupressinoxylon fissum Goeppert, 1850; Goeppert 1850: 200, pl. 25: 3–5. 1872 Cupressoxylon fissum (Goeppert, 1850) Kraus, 1872; Kraus 1872: 375. KŁUSEK—MIOCENE CONIFEROUS WOODS 701 D C A B E F Fig. 4. Microphotographs of fossil wood Juniperoxylon pachyderma (Goeppert, 1850) Kraeusel, 1949 from Zgłobice, Middle Miocene; specimen MZ LXXc. A. Growth rings with abundant axial parenchyma (TS). B. Crossfield pitting (RLS). C. Uniseriate pits on tracheid walls (RLS). D. Bordered pits (arrow) on tangential walls of tracheids (TLS). E. Uniseriate rays (TLS). F. Beadlike thickenings (arrow) on axial parenchyma horizontal walls (TLS). Scale bars 50 μm. B C A D E Fig. 4. Microphotographs of fossil wood Juniperoxylon pachyderma (Goeppert, 1850) Kraeusel, 1949 from Zgłobice, Middle Miocene; specimen MZ LXXc. A. Growth rings with abundant axial parenchyma (TS). B. Crossfield pitting (RLS). C. Uniseriate pits on tracheid walls (RLS). D. Bordered pits (arrow) on tangential walls of tracheids (TLS). E. Uniseriate rays (TLS). F. Beadlike thickenings (arrow) on axial parenchyma horizontal walls (TLS). Scale bars 50 μm. tracheids can also signify the presence of compression wood (Schweingruber 1996). The analysed specimens have intercel- lular spaces, but do not otherwise show features typical of re- action wood. Thus this detail of anatomical structure supports its classification to Juniperus. characteristic of the group embracing Cupressus Linnaeus, 1753, Diselma Hooker, 1857, Fitzroya Hooker ex Lindley, 1851, Juniperus Linnaeus, 1753, and Libocedrus Endlicher, 1847 wood. In consideration of the significant similarity that exists within this group, the slight differences in their wood anatomy are used to identify the particular genera. Stratigraphic and geographic range.— Tertiary; Europe and East Asia. y y p g The species belonging to Diselma and Fitzroya are distin- guished from the analysed specimen because of the smaller amount of axial parenchyma, pitted transverse walls of ray cells, more pits in the cross field and the presence of inden- tures. Additionally, Diselma is characterised by the presence of ray tracheids, while Fitzroya often possesses vestured pits. Cupressus and Libocedrus, in turn, are usually devoid of thick- enings in tangential walls of ray cells; they often have pitted transverse walls of ray cells and possess indentures (Greguss 1955). Juniperoxylon pachyderma (Goeppert, 1850) Krae- usel, 1949 The structure that shows greatest similarity to the an- alysed specimen is that of species of the genus Juniperus. Juniper is characterised by the presence of pronounced bead- like thickenings in the tangential walls of ray cells and it also possesses intercellular spaces visible in the cross-section of wood (Phillips 1941). Large intercellular spaces and rounded Morphogenus Cupressinoxylon Goeppert, 1850 (nom. cons.) Type species: Cupressinoxylon gothanii Kraeusel, 1920; Tertiary, Mio- cene; Węgliniec, Silesia, Poland (typ. cons.). Type species: Cupressinoxylon gothanii Kraeusel, 1920; Tertiary, Mio- cene; Węgliniec, Silesia, Poland (typ. cons.). Cupressinoxylon canadense (Schroeter, 1880) Krae- usel, 1949 Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use 1919 Cupressinoxylon wellingtonioides (Prill, 1913) Kraeusel, 1919; Prill and Kraeusel 1919: 293, pl. 22: 7, 8. Cupressinoxylon canadense (Schroeter, 1880) Krae- usel, 1949 Fig. 5A–F. 1880 Sequoia canadensis Schroeter, 1880; Schroeter 1880: 17, pl. 2: 11–21; pl. 3: 22–26. p 1913 Sequoia wellingtonioides Prill, 1913; Prill 1913: 48. 1919 Cupressinoxylon wellingtonioides (Prill, 1913) Kraeusel, 1919; Prill and Kraeusel 1919: 293, pl. 22: 7, 8. ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 702 D C A B E F Fig. 5. Microphotographs of fossil wood Cupressinoxylon canadense (Schroeter, 1880) Kraeusel, 1949 from Hebdów, Middle Miocene; specimen MZ LXXII. A. Narrow growth rings (TS). B. Crossfield pitting (RLS). C. Bordered pits on radial walls of tracheids (RLS). D. Tangential walls of tracheids with bordered pits (arrow) (TLS). E. Tall, uniseriate rays (TLS). F. Horizontal walls of parenchyma cells with dentate thickenings (arrow) (TLS). Scale bars 50 μm. B A C B D Fig. 5. Microphotographs of fossil wood Cupressinoxylon canadense (Schroeter, 1880) Kraeusel, 1949 from Hebdów, Middle Miocene; specimen MZ LXXII. A. Narrow growth rings (TS). B. Crossfield pitting (RLS). C. Bordered pits on radial walls of tracheids (RLS). D. Tangential walls of tracheids with bordered pits (arrow) (TLS). E. Tall, uniseriate rays (TLS). F. Horizontal walls of parenchyma cells with dentate thickenings (arrow) (TLS). Scale bars 50 μm. 1949 Cupressinoxylon canadense (Schroeter, 1880) Kraeusel, 1949; Kraeusel 1949: 175. Rays are uniseriate or sometimes partially biseriate, 2–45 cells high (Fig. 5E). They are composed of oval or rectan- gular cells. Rays have thin and smooth, sporadically pitted transverse walls and smooth tangential walls with no inden- tures. Cross-field pits, 8–9 μm in diameter, in median cells form a single row, then 1–3 pits appear per field. In marginal cells, occur usually 2–6 pits organised in two rows (Fig. 5B). Pits have circular or oval outlines. Pit apertures are obliquely oriented in earlywood and latewood. Ray parenchyma often includes resinous contents. Ray tracheids occur rarely. Material.—One coalified fragment of wood, about 10 cm long, sourced from Hebdów locality and marked as MZ LXXII (Fig. 5A–F). Description.—The specimen comprises the part of the branch. Growth rings are narrow, with thin latewood lay- er (Fig. 5A). Axial parenchyma appears sparse. Transverse walls of parenchyma cells are smooth or sometimes with dentate thickenings (Fig. 5F). Description.—The specimen comprises the part of the branch. Growth rings are narrow, with thin latewood lay- er (Fig. 5A). Axial parenchyma appears sparse. Transverse walls of parenchyma cells are smooth or sometimes with dentate thickenings (Fig. 5F). Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Cupressinoxylon canadense (Schroeter, 1880) Krae- usel, 1949 Remarks.—The described features of wood anatomical structure, especially the absence of resin canals, character of cross-field pitting and bordered pits on the walls of trache- ids, indicate the affiliation of the examined specimen to Cu- pressinoxylon canadense (Schroeter, 1880) Kraeusel, 1949. Cupressinoxylon canadense was established by Kraeusel (1949) on the basis of previously defined taxa. This name referred to Sequoia canadensis originating from Miocene of North Canada (Schroeter 1880). Kraeusel (1949) interpret- ed as Cupressinoxylon canadense also Miocene wood from Tracheids have angular shape and thin walls. On radi- al walls of tracheids occur uniseriate or biseriate bordered pits, 8.5–13 μm in diameter, arranged in opposite, continu- ous rows. Pits are circular or sometimes slightly compressed along the sides of contact. Pit apertures are also circular in earlywood or slit-like, obliquely oriented in latewood (Fig. 5C). Bars of Sanio occur frequently. Tangential walls of tra- cheids have bordered pits arranged in a single irregular, scat- tered row (Fig. 5D). Pits possess circular shape (8.5–10.5 μm in diameter) and slit-like, oblique apertures. KŁUSEK—MIOCENE CONIFEROUS WOODS 703 D C A B E F Fig. 6. Microphotographs of fossil wood Cupressinoxylon polonicum (Kostyniuk, 1938) Kraeusel, 1949 from Hebdów, Middle Miocene; specimen MZ LIX. A. Slightly disturbed growth rings (TS). B. Frost ring with deformed tracheids and rays (TS). C. Pitting on radial walls of tracheids masked by spiral checking of tracheid walls (RLS). D. Crossfield pitting (RLS). E. Short, uniseriate rays (TLS). F. Axial parenchyma (TLS). Scale bars 50 μm. B A B A C A E D Fig. 6. Microphotographs of fossil wood Cupressinoxylon polonicum (Kostyniuk, 1938) Kraeusel, 1949 from Hebdów, Middle Miocene; specimen MZ LIX. A. Slightly disturbed growth rings (TS). B. Frost ring with deformed tracheids and rays (TS). C. Pitting on radial walls of tracheids masked by spiral checking of tracheid walls (RLS). D. Crossfield pitting (RLS). E. Short, uniseriate rays (TLS). F. Axial parenchyma (TLS). Scale bars 50 μm. Opole (Poland). The taxonomical position of this wood was revised many times. It had been initially attributed to Sequoia wellingtoniides Prill, 1913 later re-classified to Cupressinox- ylon wellingtonioides (Prill, 1913) Kraeusel, 1919 (Prill and Kraeusel 1919) and finally determined as Cupressinoxylon canadense (Kraeusel 1949). Taking into consideration living trees, analysed wood is most similar to the wood known from Glyptostrobus Endlicher, 1847, Metasequoia Hu and Cheng, 1948, Taxodium Richard, 1810, and Sequoia End- licher, 1847. Cupressinoxylon canadense (Schroeter, 1880) Krae- usel, 1949 Among the mentioned genera, Glyptostrobus wood differs from the examined specimen in the presence of glyptostroboid cross-field pitting, nodular transverse walls of wood parenchyma cells and the absence of ray tracheids. The rays are homogeneous also in Metasequoia and Taxodium wood. The wood of Metasequoia, Taxodium, and Sequoia, in distinction from the analysed specimen, possess taxodi- oid pits in the cross field. Moreover, the genus Taxodium is marked by the presence of conspicuous thickenings of trans- verse walls of axial parenchyma cells and occasionally up to triseriate bordered pits on radial walls of tracheids (Henry and Mc Intyre 1926; Visscher and Jagels 2003). The anatom- Opole (Poland). The taxonomical position of this wood was revised many times. It had been initially attributed to Sequoia wellingtoniides Prill, 1913 later re-classified to Cupressinox- ylon wellingtonioides (Prill, 1913) Kraeusel, 1919 (Prill and Kraeusel 1919) and finally determined as Cupressinoxylon canadense (Kraeusel 1949). Taking into consideration living trees, analysed wood is most similar to the wood known from Glyptostrobus Endlicher, 1847, Metasequoia Hu and Cheng, 1948, Taxodium Richard, 1810, and Sequoia End- licher, 1847. Among the mentioned genera, Glyptostrobus wood differs from the examined specimen in the presence of glyptostroboid cross-field pitting, nodular transverse walls of wood parenchyma cells and the absence of ray tracheids. The rays are homogeneous also in Metasequoia and Taxodium wood. The wood of Metasequoia, Taxodium, and Sequoia, in distinction from the analysed specimen, possess taxodi- oid pits in the cross field. Moreover, the genus Taxodium is marked by the presence of conspicuous thickenings of trans- verse walls of axial parenchyma cells and occasionally up to triseriate bordered pits on radial walls of tracheids (Henry and Mc Intyre 1926; Visscher and Jagels 2003). The anatom- ical structure of the analysed specimen is, in turn, consistent with giant sequoia wood (Sequoiadendron giganteum [Lind- ley, 1853] Buchholz, 1939). This species is characterised by the presence of cupressoid pits in the cross field. In fossil wood taxonomy, giant sequoia corresponds to Cupressinox- ylon canadense species (Prill and Kraeusel 1919; Kraeusel 1949). This fact confirms above established classification. Stratigraphic and geographic range.— Tertiary; Europe, North America. Stratigraphic and geographic range.— Tertiary; Europe, North America. Stratigraphic and geographic range.— Tertiary; Europe, North America. Cupressinoxylon polonicum (Kostyniuk, 1938) Krae- usel, 1949 Fig. 6A–F. 1938 Cupressinoxylon sp.; Kostyniuk 1938: 44, pl. 3: 3, 4. 1949 Cupressinoxylon polonicum (Kostyniuk, 1938) Kraeusel, 1949; Kraeusel 1949: 175. Fig. 6A–F. Discussion The fossil wood species described in this paper have not been known in the area of Polish Carpathian Foredeep so far. How- ever, their occurrence is registered in the close neighbour- hood of an analysed region (see SOM: Supplementary Online Material available at http://app.pan.pl/SOM/app59-Klusek_ SOM.pdf). The first described species, Piceoxylon piceae Roessler, 1937, is documented from the Miocene of Lower Silesia, from Węgliniec and Jaworów localities (Kraeusel 1913, 1920, 1949). The wood from Węgliniec had been ini- tially classified as Piceoxylon sp. (Kraeusel 1920), and later defined accurately, by the same author as Piceoxylon piceae (Kraeusel 1949). A similar situation occurred in the case of the wood from the Jaworów site. First it was described as Picea sp. (Kraeusel 1913), and then specified as Piceoxylon piceae (Kraeusel 1949). Uniseriate rays, typically 2–9 cells high (Fig. 6E), are characterised by smooth transverse and tangential walls without indentures. Cross fields contain cupressoid pits, oval or circular in outline, 6.5–8.5 μm in diameter. Pits usually are arranged in one row, 1–2 pits occur in cross field (Fig. 6D). Two separate damages, most likely frost rings, appear with- in this specimen (Fig. 6B). Frost rings in coniferous species are composed of underlignified, deformed tracheids, collapsed cells, bent rays and traumatic parenchyma cells (Glerum and Farrar 1966; LaMarche and Hirschboeck 1984). These distur- bances have been observed in the analysed wood (Fig. 6B). Apart from this taxon, there are plenty of other Tertiary Piceoxylon woods described from neighbouring areas. One of them is Piceoxylon arcticum (Schmalhausen, 1890) Krae- usel, 1949 that is the equivalent to the undetermined species of living larch (Kraeusel 1919). Piceoxylon arcticum is noted from the Miocene of Gliwice and Żarów. The woods coming from these sites had been initially ascribed to Larix silesiaca Kraeusel, 1913 (Kraeusel 1913). Then the specimen from Gliwice was reassigned to Piceoxylon laricinum Kraeusel, 1919 and the specimen from Żarów to Piceoxylon sp. (la- ricinum?) Kraeusel, 1919 (Prill and Kraeusel 1919). Finally, both of them were identified as Piceoxylon arcticum (Krae- usel 1949). Similarly, the Miocene wood Piceoxylon laric- inum originating from Posądza near Kraków (Rubczyńska and Zabłocki 1924) was also reclassified as Piceoxylon arcti- cum (Kraeusel 1949). Remarks.—The type of tracheary and cross-field pitting, the character of ray cells and the absence of resin canals refer to the Cupressinoxylon genus. Among fossil taxa, the most similar species is Cupressinoxylon polonicum (Kostyniuk, 1938) Kraeusel, 1949. Discussion It is characterised by the occurrence of low, uniseriate rays and by the absence of ray tracheids. Cupressinoxylon polonicum was described for the first time from Bielany near Grójec (Kostyniuk 1938). This wood, col- lected from Pliocene formations, had been initially named as Cupressinoxylon sp., later, however, it was classified as a separate species (Kraeusel 1949). Cupressinoxylon polon- icum does not have a defined counterpart among contempo- rary trees. This species is most similar to the group which encompasses Platycladus Spach, 1841, Fokienia Henry and Thomas, 1911, Thuja Linnaeus, 1753, Thujopsis Siebold and Zuccarini ex Endlicher, 1842, Chamaecyparis Spach, 1841, and Microbiota Komarov, 1923 woods. The first four gen- era differ from the analysed specimen because of the higher number of pits in the cross field and the presence of inden- tures. In the cases of Platycladus, Thuja, and Thujopsis dis- agreement results also from the existence of pitted transverse walls of ray cells. Moreover, Platycladus, Fokienia, and Thu- ja wood possesses uniseriate or biseriate bordered pits on ra- dial walls of tracheids, while Fokienia and Thuja sometimes have biseriate or triseriate pits on tracheid tangential walls. Meanwhile, Microbiota wood is marked by the presence of smooth transverse walls of axial parenchyma cells as well as the occurrence of pitted transverse walls of ray cells (Schwe- ingruber 1990; García Esteban et al. 2004). All these features allow those genera to be eliminated. The analysed specimen displays the highest similarity to Chamaecyparis wood. This fact is established on the basis of cupressoid type of cross- field pitting, the presence of uniseriate pits on radial and The next Piceoxylon species described from other areas of Poland are Piceoxylon pseudotsugae Gothan, 1906 from the Miocene of Konin (Grabowska 1956) and Piceoxylon macrocarpum (Prill, 1913) Kraeusel, 1919 from the Mio- cene of Gliwice (Reyman 1956). Piceoxylon macrocarpum is also detailed from Miocene locality in Opole (Prill and Kraeusel 1919), but in this case, it had been initially assigned as Pseudotsuga macrocarpa var. miocenica Prill, 1913 (Prill 1913). Both of these mentioned Piceoxylon species differ significantly from the spruce wood and both of them are closely related to the extant Pseudotsuga Carriere, 1867. Ac- cording to Kraeusel 1919, the first is most similar to Pseudot- suga douglasii Carriere, 1867 and the second corresponds to Pseudotsuga macrocarpa Mayr, 1900. From the close neighbourhood of the Carpathian Fore- deep are also known Piceoxylon woods with unspecified affinity. Cupressinoxylon canadense (Schroeter, 1880) Krae- usel, 1949 1938 Cupressinoxylon sp.; Kostyniuk 1938: 44, pl. 3: 3, 4. 1949 Cupressinoxylon polonicum (Kostyniuk, 1938) Kraeusel, 1949; Kraeusel 1949: 175. Material.—One fragment of coalified wood from the Heb- dów locality, 10 cm long and marked as MZ LIX (Fig. 6A–F). Material.—One fragment of coalified wood from the Heb- dów locality, 10 cm long and marked as MZ LIX (Fig. 6A–F). Material.—One fragment of coalified wood from the Heb- dów locality, 10 cm long and marked as MZ LIX (Fig. 6A–F). Description.—The wood is derived from branch and con- tains the pith. Growth rings have narrow latewood zone (Fig. 6A). Resin canals are absent. Axial parenchyma appears ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 704 abundantly (Fig. 6F) and forms tangential rows. Transverse walls of parenchyma cells usually are smooth or sometimes with beadlike thickenings. tangential walls of tracheids and the occurrence of uniseriate rays up to 9 cells high. Stratigraphic and geographic range.— Tertiary; Europe. Stratigraphic and geographic range.— Tertiary; Europe. Tracheids have thick walls. On the radial walls of tra- cheids appears uniseriate pitting, loosely spaced in latewood (Fig. 6C). Pits are bordered, circular or oval, 11–13.5 μm in diameter. Pit apertures have circular shape in earlywood and slit-like, oblique outline in latewood. Tangential walls of tracheids possess uniseriate, sparse pits. Pits are circular, 6.5–7.5 μm in diameter. They demonstrate narrow, obliquely oriented apertures. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use KŁUSEK—MIOCENE CONIFEROUS WOODS Similarly, the woods from Gliwice and Radomierzyce at the beginning were assigned to Juniperoxylon pauciporosum (Prill and Kraeusel 1919) and the woods from Zgorzelec had been ascribed to Juniperoxylon silesiacum (Kraeusel 1920). In turn, the wood from Nowa Wieś Królewska, had been identified as Juniperoxylon sp. (Kraeusel 1920). Finally, all these specimens were reassigned to Juniperoxylon pachyder- ma (Kraeusel 1949). ( ) g ( ) The woods of Cupressinoxylon type are exceptionally nu- merous in the fossil record (Bamford et al. 2002). Therefore, there are also many other Cupressinoxylon species originat- ed from close vicinity of the Carpathian Foredeep anatom- ically similar to the wood described in this paper. Miocene woods that belong to this morphotype are Cupressinoxylon gothani Kraeusel, 1919 from Węgliniec, Cupressinoxylon cupressoides Kraeusel, 1919 from Zaręba (Kraeusel 1920) and Cupressinoxylon hausricanum Hofmann, 1927 from Turów lignite mine (Zalewska 1953). The first of these is equivalent to the present-day genera of Callitris Ventenat, 1808 and Widdringtonia Endlicher, 1842, the second one cor- responds to Cupressus Linnaeus, 1753 and Chamaecyparis Spach, 1841, and the third one—to Cryptomeria japonica D. Don, 1839 wood (Kraeusel 1919, 1949). The next species of Cupressinoxylon is Cupressinoxylon luckense (Kostyniuk, 1938) Kraeusel, 1949 that was noted from Oligocene or Mio- cene sediments in Niebożka near Łuck (Kostyniuk 1938). The wood from Niebożka had initially been named as Cupressi- noxylon sp. Then, on this basis, the diagnosis of a new species was established (Kraeusel 1949). Cupressinoxylon luckense wood was afterwards recognised also among Miocene lig- nites of Konin (Grabowska 1956). As for Juniperoxylon pachyderma, numerous Miocene woods from Turów open pit lignite mine were also assigned to this taxon (Zalewska 1953; Kostyniuk 1967). Moreover, among the specimens from Turów new species belonging to the Juniperoxylon were defined. The first of these is Juni- peroxylon lusaticum Kostyniuk, 1967 having the anatomi- cal structure similar to the present-day Juniperus Linnaeus, 1753, but differing from it by taxodioid pits in cross-fields (Kostyniuk 1967). The next described species from Turów is Juniperoxylon glyptostroboides Kostyniuk, 1967, similar to living Juniperus and Heyderia Koch, 1873. Its specif- ic characteristics are simple pits in cross-fields (Kostyniuk 1967). Another species classified from Turów as Juniperox- ylon turoviense Kostyniuk, 1967, possesses the features of extant Juniperus and Athrotaxis D. Don, 1838 genera but has also taxodioid pits on cross-fields (Kostyniuk 1967). Later, however, two of these new species were transferred to Tax- odioxylon Hartig, 1848. KŁUSEK—MIOCENE CONIFEROUS WOODS 705 KŁUSEK—MIOCENE CONIFEROUS WOODS Piceoxylon sp. from the Pliocene of Krościenko (Kostyniuk 1950), as well as the wood of Piceoxylon sp. from the Mio- cene of Otwock (Kostyniuk 1938). Kraeusel (1919) defined as Piceoxylon also some Tertiary specimens from Ostrołęka and Łażany, assigned previously to Taxites ayckii Goeppert, 1850, as well as Tertiary woods originating from Lubliniec and earlier described as Pinites silesiacus Goeppert, 1850 (Goeppert 1850). Taxonomic classification of all these woods remains questionable and both larch and spruce may be their nearest living relatives (Kraeusel 1919). lon, but, similarly to the species of Turów, it has taxodioid pits in its cross-fields (Kownas 1951). Some Miocene woods from Konin open pit lignite mine were also assigned to this new spe- cies (Grabowska 1956). The name Taxodioxylon juniperoides, however, was deemed to be invalid and the woods were reclas- sified as Juniperoxylon juniperoides (Kownas, 1951) Huard, 1966 (Huard 1966; Suess and Velitzelos 1997). Miocene wood ascribed to Cupressinoxylon canadense (Schroeter, 1880) Kraeusel, 1949 have been found, in the vicinity of Opole and Węgliniec. The wood from Opole had been initially described as Sequoia wellingtoniides Prill, 1913 (Prill 1913), later reclassified to Cupressinoxylon wellingtoni- oides (Prill, 1913) Kraeusel, 1919 (Prill and Kraeusel 1919) and finally to Cupressinoxylon canadense (Kraeusel 1949). Also the specimen from Węgliniec had been originally as- signed to Cupressinoxylon wellingtonioides (Kraeusel 1920). Furthermore, Cupressinoxylon canadense is known from the Miocene of Dobrzyń on the Vistula River (Kownas 1951) (as Cupressinoxylon wellingtonioides) and from the area of the Turów open pit lignite mine (Zalewska 1953). In turn, Cupressinoxylon polonicum (Kostyniuk, 1938) Kraeusel, 1949 was described from Pliocene of Bielany near Grójec as Cupressinoxylon sp. (Kostyniuk 1938). Cupressinoxylon po- lonicum was also recorded from Miocene deposits of Turów (Zalewska 1953) and Konin lignite mines (Grabowska 1956). Miocene woods belonging to Juniperoxylon pachyderma (Goeppert, 1850) Kraeusel, 1949 are especially numerous from the Silesia region. Among them occur specimens orig- inating from Gliwice, Radomierzyce, Zielona Góra, Nowa Wieś Królewska, and Zgorzelec. The specimens from Zielona Góra had been initially described as Juniperus silesiaca Prill, 1913 and Juniperus pauciporosa Prill, 1913 (Prill 1913), and later reclassified as Juniperoxylon silesiacum (Prill, 1913) Kraeusel, 1919 and Juniperoxylon pauciporosum (Prill, 1913) Kraeusel, 1919 respectively (Prill and Kraeusel 1919). Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use Discussion This group includes the woods of Picea sp. and KŁUSEK—MIOCENE CONIFEROUS WOODS The other Miocene species, namely: Cupressinoxylon fissum Goeppert, 1850 from Zielona Góra, Cupressinoxylon pachy- derma Goeppert, 1850 from Łażany (Goeppert 1850) and Cupressinoxylon juniperoides Kraeusel, 1913 from Gliwice and Radomierzyce in Silesia (Kraeusel 1913), were assigned by Kraeusel (1919, 1949) as Juniperoxylon pachyderma. oecological conditions from taxonomical results obtained is the rather small number of wood species determined. In spite of this, some observed features of the analysed wood may be helpful in reconstructing the Miocene environ- ment. The first is the presence of growth rings in all analysed samples from the Carpathian Foredeep. This fact points to seasonal climate fluctuations (Creber and Chaloner 1984). However, the development of growth rings is not determined by weather effect alone. Some tree species form rings even in areas characterised by uniform external conditions. In these cases, the presence of growth interruptions depends on ge- netic legacy and is not influenced by external factors (Creber 1977; Brison et al. 2001). This kind of ring usually shows great variability in width and has a small portion of latewood. These increments distinguish themselves also by a symmetri- cal distribution of cell sizes on both sides of ring boundaries and by a lack of continuity around the circumference of the stem (Ash and Creber 1992; Falcon-Lang 2003). Neverthe- less, such features do not correspond to the way the growth rings developed in the analysed wood. Moreover, only the absence of true growth rings in the prevailing number of trees from the examined territory indicate that they grew under non-seasonal conditions. Therefore, the occurrence of regular growth rings, distinctly marked throughout the whole circumference of the stem, in all analysed woods from the Carpathian Foredeep, as well as in the other Miocene woods from neighbouring localities, proves the seasonal change- ability of the palaeoenvironment. y ( , ) p y p y The main problem with Cupressinoxylon classification is the huge number of species that are frequently very sim- ilar in anatomic features (Kraeusel 1949). It seems that a renewed, critical review of these taxa is necessary. In addi- tion, the diagnosis of Cupressinoxylon genus itself requires specification and updating (Bamford et al. 2002; Dolezych and Schneider 2006). Because of the existing terminology problems, many woods of Cupressinoxylon remain without precise species diagnosis. KŁUSEK—MIOCENE CONIFEROUS WOODS For example, Miocene woods originating from Skorogoszcz (Prill 1913; Prill and Kraeusel 1919), Dobrzyń on the Vistula River (Lilpop 1917; Kownas 1951), Turów (Zalewska 1953; Kostyniuk 1967), and Konin lignite mines (Grabowska 1956) were named only as Cu- pressinoxylon sp. The case of one of them, namely Cupressi- noxylon sp. from the Turów open pit mine (Kostyniuk 1967), shows that the outdated, inaccurate classification might be updated. The wood from Turów was believed to be identical to the specimens from the Welzow mine in Germany. On this basis, a diagnosis of a new species Cupressinoxylon kostyniukii Dolezych, 2006 (Dolezych and Schneider 2006) was defined. Seasonality of temperature and rainfall in the Miocene Epoch is also evidenced by the results of other palaeonto- logical research (Boucot and Gray 2001; Mosbrugger et al. 2005). This research demonstrates that a warm temperate and humid climate prevailed in the area of the Central Paratethys (Böhme 2003; Böhme et al. 2007; Syabryaj et al. 2007). On the other hand, the results of ectothermic vertebrate re- search proved also the occurrence of frost periods (Böhme 2003). The existence of presumable frost rings within Cu- pressinoxylon polonicum wood may support this assumption. Such rings are formed when, during cambium activity, the air temperature drops below freezing point (LaMarche and Hirschboeck 1984; Raitio 1992). Similar disturbance zones within growth rings could develop as a result of insect ac- tivity, bacterial infection, local damages caused by fire and mechanical injury or as an effect of diagenetical changes of wood (Glerum and Farrar 1966; Schweingruber 1996). However, the frost origin of the aforementioned damage is highly probable if one assumes that the wood had been trans- ported from the Carpathian area. This region was character- ised by a lower temperature of the cold months and more seasonal changeability in comparison with the lower-lying neighbourhood stands (Bruch et al. 2007; Utescher et al. 2007). Moreover, analysed rings occur within the fragment of branch or young stem. These parts of tree are covered with thin layer of bark and are predisposed to frost rings develop- ment (Chapman 1994). This additionally supports presented presumption. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use KŁUSEK—MIOCENE CONIFEROUS WOODS These were Taxodioxylon lusaticum (Kostyniuk, 1967) Suess and Rathner, 1998 and Taxodioxy- lon turoviense (Kostyniuk, 1967) Suess and Rathner, 1998 (Suess and Rathner 1998). On the other hand, the significant amount of Tertiary woods of the Foredeep vicinity, assigned initially to vari- ous species of Cupressinoxylon, was later re-classified to other taxa. Kraeusel (1919, 1949) assigned Miocene woods Cupressinoxylon aequele Goeppert, 1850 from Łażany and Żarów (Goeppert 1850) and Cupressinoxylon aequele (Goeppert, 1850) Conwentz, 1881 from Nowogrodziec in Silesia (Conwentz 1881) to Taxodioxylon gypsaceum (Goep- The following site, where wood of Juniperoxylon type was recorded, is Dobrzyń on the Vistula River. The Miocene woods from this site were classified as Juniperoxylon silesiacum and Juniperoxylon sp. (Kownas 1951). Furthermore, part of the specimens from Dobrzyń became the basis for the creation of a new species, Taxodioxylon juniperoides Kownas, 1951. This wood is closely related in its anatomic structure to Juniperoxy- 706 ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 pert, 1842) Kraeusel, 1949 or, conditionally, Taxodioxylon taxodii Gothan, 1906. Similar transformation involved also the following Miocene woods: Cupressinoxylon lep- totichum Goeppert, 1850 from Łażany and Zielona Góra, as well as Cupressinoxylon multiradiatum Goeppert, 1850 and Cupressinoxylon opacum Goeppert, 1850 from Łaża- ny (Goeppert 1850). More unambiguous classification, as Taxodioxylon gypsaceum, was possible for Cupressinoxy- lon fissum (Goeppert, 1850) Conwentz, 1882 from Miocene of Skorogoszcz (Conwentz 1882), and for Cupressinoxylon nodosum Goeppert, 1850 and Cupressinoxylon subaequele Goeppert, 1850 from Miocene of Łażany (Goeppert 1850). The other Miocene species, namely: Cupressinoxylon fissum Goeppert, 1850 from Zielona Góra, Cupressinoxylon pachy- derma Goeppert, 1850 from Łażany (Goeppert 1850) and Cupressinoxylon juniperoides Kraeusel, 1913 from Gliwice and Radomierzyce in Silesia (Kraeusel 1913), were assigned by Kraeusel (1919, 1949) as Juniperoxylon pachyderma. pert, 1842) Kraeusel, 1949 or, conditionally, Taxodioxylon taxodii Gothan, 1906. Similar transformation involved also the following Miocene woods: Cupressinoxylon lep- totichum Goeppert, 1850 from Łażany and Zielona Góra, as well as Cupressinoxylon multiradiatum Goeppert, 1850 and Cupressinoxylon opacum Goeppert, 1850 from Łaża- ny (Goeppert 1850). More unambiguous classification, as Taxodioxylon gypsaceum, was possible for Cupressinoxy- lon fissum (Goeppert, 1850) Conwentz, 1882 from Miocene of Skorogoszcz (Conwentz 1882), and for Cupressinoxylon nodosum Goeppert, 1850 and Cupressinoxylon subaequele Goeppert, 1850 from Miocene of Łażany (Goeppert 1850). References Definitive identification of Larix (Pinaceae) wood based on anatomy from the Middle Eocene, Axel Heiberg Island, Canadian High Arctic. IAWA Journal 22: 73–83. Ü Jagels, R., LePage, B.A. and Jiang M. 2001. Definitive identification of Larix (Pinaceae) wood based on anatomy from the Middle Eocene, Brison, A.L., Philippe, M., and Thevenard, F. 2001. Are Mesozoic wood growth rings climate induced? Paleobiology 27: 531–538. Bruch, A.A., Uhl, D., and Mosbrugger, V. 2007. Miocene climate in Eu- rope—patterns and evolution. A first synthesis of NECLIME. Palaeo- geography, Palaeoclimatology, Palaeoecology 253: 1–7. Jurasky, K.A. 1931. Über rezentes und fossiles Harz. Zeitschrift Brennst- off-Chemie 12: 161–163. Kostyniuk, M. 1938. Trzeciorzędowe drewna i pyłki z Mazowsza i Wołynia. Kosmos 63: 1–33. Brud, S. 2002. Ewolucja paleogeograficzna zachodniej części Kotliny Sandomierskiej w późnym neogenie i wczesnym czwartorzędzie. 97 pp. Unpublished PhD thesis, Instytut Nauk Geologicznych Uniwersytetu Jagiellońskiego, Kraków. Kostyniuk, M. 1950. Szczątki drewna szpilkowych flory plioceńskiej z Krościenka. Prace Wrocławskiego Towarzystwa Naukowego, seria B 22: 5–57. Boucot, A.J. and Gray, J. 2001. A critique of Phanerozoic climatic models involving changes in the CO content of the atmosphere. Earth-Science Reviews 56: 1–159. Kostyniuk, M. 1967. Pnie drzew iglastych z górnego pokładu węgla brunatnego w Turowie. Prace Muzeum Ziemi 10: 3–96. Chaloner, W.G. and Creber, G.T. 1973. Growth rings in fossil woods as ev- idence of past climates. In: D.H. Tarling and S.K. Runcorn S. K. (eds.), Implications of Continental Drift to the Earth Sciences, 425–437. Ac- ademic Press, New York. Kownas, S. 1951. Trzeciorzędowe drewna z Dobrzynia nad Wisłą. Studia Societatis Scientiarum Toruniensis 1: 67–121. Kraeusel, R. 1913. Beiträge zur Kenntnis der Hölzer aus der schlesischen Braunkohle I. 56 pp. Inaugural Dissertation, W.G. Korn, Breslau. Braunkohle I. 56 pp. Inaugural Dissertation, W.G. Korn, Br Chapman, J.L. 1994. Distinguishing internal developmental characteristics from external palaeoenvironmental effects in fossil woods. Review of Palaeobotany and Palynology 81: 19–32. Kraeusel, R. 1919. Die fossilen Koniferenhölzer (unter Ausschluβ von Ar- aucarioxylon Kraus). Palaeontographica 62: 185–275. aucarioxylon Kraus). Palaeontographica 62: 185–275. Kraeusel, R. 1920. Nachträge zur Tertiärflora Schlesiens II. Braunkohlen- hölzer. Jahrbuch der Preußischen Geologischen Landesanstalt 39: 418–460. Conwentz, H. 1881. Über ein in Markasit verwandeltes Braunkohlenholz. Abhandlungen der Naturforschenden Gesellschaft zu Görlitz 17: 138– 197. Kraeusel, R. 1949. Die fossilen Koniferenhölzer, II Teil. Palaeontograph- ica B 89: 83–203. Conwentz, H. 1882. Fossile Hölzer aus der Sammlung der Königlichen geol- ogischen Landesanstalt zu Berlin. References García Esteban, L., de Palacios de Palacios, P., Guindeo Casasús, A., and García Fernández, F. 2004. Characterisation of the xylem of 352 co- nifers. Investigación Agraria: Sistemas y Recursos Forestales 13: 452–478. Anagnost, S.E., Meyer, R.W., and DeZeeuw, C. 1994. Confirmation and significance of Bartholin’s method for the identification of the wood of Picea and Larix. IAWA Journal 15: 171–184. Glerum, C. and Farrar, J.L. 1966. Frost ring formation in the stems of some Coniferous species. Canadian Journal of Botany 44: 879–886. Ash, S. and Creber, G. 1992. Palaeoclimatic interpretation of the wood structures of the trees in the Chinle Formation (Upper Triassic) in the area of Petrified Forest National Park, Arizona, U.S.A. Palaeogeogra- phy, Palaeoclimatology, Palaeoecology 96: 299–317. Goeppert, H.R. 1850. Monographie der Fossilen Coniferen. 286 pp. Arnz and Company, Leiden. Bamford, M., Zijlstra, G., and Philippe, M. 2002. Proposal to conserve the name Cupressinoxylon against Retinodendron Zenker (fossil Gymnospermae, Coniferales), with a conserved type. Taxon 51: 205–206. Grabowska, I. 1956. Przewodnie lignity węgla brunatnego z obszaru Kon- ina. Prace Instytutu Geologicznego 15: 201–258. Greguss, P. 1955. Identification of Living Gymnosperms on the Basis of Xylotomy. 263 pp. Akademiai Kiado, Budapest. Bartholin, T. 1979. The Picea–Larix problem. IAWA Bulletin n. s. 1: 68–70. Greguss, P. 1967. Fossil Gymnosperm Woods in Hungary. 136 pp. Aka- demiai Kiado, Budapest. Blokhina, N.I. and Afonin, M.A. 2009. New Species of Piceoxylon Gothan (Pinaceae) from the Cretaceous and Paleogene of the Northwestern Kamchatka Peninsula. Paleontological Journal 43: 1190–1201. Harvey, H.T., Shellhammer, H.S., and Stecker, R.E. 1980. Giant sequoia ecology: fire and reproduction. US Department of the Interior, Nation- al Park Service, Scientific Monograph Series 12: 1–182. Böhme, M. 2003. The Miocene Climatic Optimum: evidence from ecto- thermic vertebrates of Central Europe. Palaeogeography, Palaeocli- matology, Palaeoecology 195: 389–401. Henry, A. and McIntyre, M. 1926. The swamp cypresses, Glyptostrobus of China and Taxodium of America, with notes on allied genera. Proceed- ings of the Royal Irish Academy 37 B: 90–116. Böhme, M., Bruch, A.A., and Selmeier, A. 2007. The reconstruction of Early and Middle Miocene climate and vegetation in Southern Germa- ny as determined from the fossil wood flora. Palaeogeography, Palae- oclimatology, Palaeoecology 253: 107–130. Huard, J. 1966. Étude anatomique des bois de conifères des couches à lignite néogènes des Landes. Mémoires de la Société géologique de France 105: 1–85. Jagels, R., LePage, B.A. and Jiang M. 2001. Concluding remarks The Miocene fossils studied in this paper were identified as Pi- ceoxylon piceae, Cupressinoxylon canadense, Juniperoxylon pachyderma, and Cupressinoxylon polonicum. The nearest living relatives of these species are, in sequence: the Norway spruce, the giant sequoia and undefined species of juniper and cypress. Apart from the giant sequoia, these taxa have wide present-day geographical and habitat ranges (Schweingruber 1993). Only the giant sequoia occurs in the region limited to the western slopes of the Sierra Nevada Mountains (Harvey et al. 1980). This is, however, a relict distribution of this tree, which does not permit an authoritative opinion about the re- quirements of Miocene trees. The other factor which limits the possibility of drawing conclusions concerning the palae- KŁUSEK—MIOCENE CONIFEROUS WOODS 707 Acknowledgements Dolezych, M. and Schneider, W. 2006. Inkohlte Hölzer und Cuticulae dispersae aus dem 2. Miozänen Flözhorizont im Tagebau Welzow (Lausitz) – Taxonomie und vergleichende feinstratigraphisch-fazielle Zuordnung. Zeitschrift für Geologische Wissenschaften 34: 165–259. I would like to thank to the late Stanisław Brud (Jagiellonian Univer- sity, Kraków, Poland) for donating the specimens of fossil woods and for describing their occurrence in the area of the Carpathian Foredeep. I thank the reviewers, Marc Philippe (Université Claude Bernard Lyon 1, Villeurbanne, France), Mihai E. Popa (University of Bucharest, Ro- mania) and an anonymous reviewer, for the detailed review and con- structive comments. Francis, J. and Poole, I. 2002. Cretaceous and early Tertiary climates from Antarctica: evidence from fossil wood. Palaeogeography, Palaeocli- matology, Palaeoecology 182: 47–64. Falcon-Lang, H.J. 2003. Growth interruptions in silicified conifer woods from the Upper Cretaceous Two Medicine Formation, Montana, USA: implications for paleoclimate and dinosaur palaeoecology. Palaeoge- ography, Palaeoclimatology, Palaeoecology 199: 299–314. Falcon-Lang, H.J. 2005. Global climate analysis of growth rings in woods, and its implications for deep-time paleoclimate studies. Paleobiology 31: 434–444. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use References Jahrbuch der Königlich Preußischen geologischen Landesanstalt und Bergakademie 14: 144–171. Kraus, G. 1872. Bois fossiles de conifères. In: W.P. Schimper (ed.), Traité de Paléontologie végétale II, 363–385. Bailliére J.B. et fils, Paris. LaMarche, V.C. and Hirschboeck, K.K. 1984. Frost rings in trees and re- cords of major volcanic eruptions. Nature 307: 121–126. Creber, G.T. 1977. Tree rings: a natural data-storage system. Biological Reviews 52: 349–383. Creber, G.T. and Chaloner, W.G. 1984. Influence of Environmental Fac- tors on the Wood Structure of Living and Fossil Trees. The Botanical Review 50: 357–448. Lilpop, J. 1917. Mikroskopowo-anatomiczne badania węgli kopalnych. Bulletin International de l’Academie des Sciences de Cracowie, B 1/3: 6–24. 708 ACTA PALAEONTOLOGICA POLONICA 59 (3), 2014 Mosbrugger, V., Utescher, T., and Dilcher, D.L. 2005. Cenozoic continental climatic evolution of Central Europe. Proceedings of the National Acad- emy of Sciences of the United States of America 102: 14964–14969. Mosbrugger, V., Utescher, T., and Dilcher, D.L. 2005. Cenozoic continental ichenberg in Steiermark. Mitteilungen des Naturwissenschaftlichen Vereines für Steiermark 74: 64–97. ichenberg in Steiermark. Mitteilungen des Naturwissenschaftlichen Vereines für Steiermark 74: 64–97. climatic evolution of Central Europe. Proceedings of the National Acad- emy of Sciences of the United States of America 102: 14964–14969. Rubczyńska, M. and Zabłocki, M. 1924. Über zwei fossile Koniferenhöl- zer von Posądza. Bulletin de l’Academie Polonaise des Sciences (B) 1924: 433–436. y f f f Müller-Stoll, W.R. 1951. Mikroskopie des zersetzten und fossilisierten Holzes. In: H. Freund (ed.), Handbuch der Mikroskopie in der Technik 5 (2), 725–816. Umschau Verlag, Frankfurt am Main. Schmalhausen, J. 1890. Tertiäre Pflanzen der Insel Neusibirien. Mémoires de l’Académie Impériale des Sciences de St. Pétersbourg, Series 7 37: 1–22. ( ) g Olano, J.M., Eugenio, M., Garcia-Cervigon, A.I., Folch, M., and Rozas, V. 2012. Quantitative tracheid anatomy reveals a complex environmental control of wood structure in continental Mediterranean climate. Inter- national Journal of Plant Sciences 173: 137–149. Schroeter, C. 1880. Untersuchung über Fossile Hölzer aus der arktischen Zone. Flora Fossilis Arctica 6: 1–38. Schweingruber, F.H. 1990. Anatomy of European Woods. 800 pp. Paul Haupt Publishers, Berne and Stuttgart. Oszczypko, N. 1998. The Western Carpathian Foredeep—development of the foreland basin in front of the accretionary wedge and its burial history (Poland). Geologica Carpathica 49: 415–431. Schweingruber, F.H. 1993. Trees and Wood in Dendrochronology. 402 pp. Springer Verlag, Berlin. y ( ) g p Oszczypko, N. 2006. References Development of the Polish sector of the Carpathian Foredeep [in Polish, with English abstract]. Przegląd Geologiczny 54: 396–403. Schweingruber, F.H. 1996. Tree Rings and Environment. Dendroecology. 609 pp. Paul Haupt Verlag, Berne. Otfinowski, M. 1973. A Sarmatian sea slide at Hebdów near Nowe Brzesko. Bulletin of the Polish Academy of Sciences, Earth Sciences 21: 125–137. Scurfield, G. 1973. Reaction wood: its structure and function. Science 179: 647–655. Selling, O.H. 1944. On cupressoid root remains of Mesozoic age from the Arctic. Arkiv for Botanik 31A: 1–20. Philippe, M. and Bamford, M. 2008. A key to morphogenera used for Me- sozoic conifer-like woods. Review of Paleeobotany and Palynology 148: 184–207. f Slijper E.J. 1932. Über pliozäne Hölzer aus dem Ton von Reuver (Lim- burg, Holland). Recueil des travaux botaniques néerlandais 29: 18–35. Phillips, E.W.J. 1941. The identification of coniferous woods by their microscopic structure. Journal of the Linnean Society of London 52: 259–320. Suess, H. and Rathner, U. 1998. Ein neues fossiles Holz, Juniperoxylon wagneri sp. nova, aus der miozänen Braunkohle von Wetro. Feddes Repertorium 109: 15–24. Porębski, S.J. 1999. Depositional setting of a supra-evaporite succession (Upper Badenian) in the Kraków-Brzesko area, Carpathian Foredeep basin [in Polish, with English abstract]. In: T.M. Peryt (ed.), Analiza basenu trzeciorzędowego Przedkarpacia. Prace Państwowego Instytu- tu Geologicznego 168: 97–118. Suess, H. and Velitzelos, E. 1997. Fossile Hölzer der Familie Taxodiaceae aus tertiären Schichten des Versteinerten Waldes von Lesbos, Griech- enland. Feddes Repertorium 108: 1–30. Syabryaj, S., Utescher, T., Molchanoff, S., and Bruch, A.A. 2007. Vegeta- tion and palaeoclimate in the Miocene of Ukraine. Palaeogeography, Palaeoclimatology, Palaeoecology 253: 153–168. Prill, W. 1913. Beiträge zur Kenntnis schlesischer Braunkohlenhölzer II. 68 pp. Inaugural Dissertation, W.G. Korn, Breslau. Utescher, T., Erdei, B., François, L., and Mosbrugger, V. 2007. Tree di- versity in the Miocene forests of Western Eurasia. Palaeogeography, Palaeoclimatology, Palaeoecology 253: 226–250. Prill, W. and Kraeusel, R. 1919. Die Pflanzen des Schlesischen Tertiärs. Jahrbuch der Preußischen Geologischen Landesanstalt 38: 191–338. Raitio, H. 1992. Anatomical symptoms in the wood of Scots pine damaged by frost and pine bark bugs. Flora 186: 187–193. Vaganov, E.A., Anchukaitis, K.J., and Evans, M.N. 2011. How well un- derstood are the processes that create dendroclimatic records? a mechanistic model of the climatic control on conifer tree-ring growth dynamics. In: M.K. Hughes, T.W. Swetnamm, and H.F. Diaz (eds.), Dendroclimatology: Progress and Prospects. Vol. 2, 37–75. Springer, New York. Reyman, M. 1956. Downloaded From: https://bioone.org/journals/Acta-Palaeontologica-Polonica on 23 Oct 2024 Terms of Use: https://bioone.org/terms-of-use References O drewnach kopalnych ze śląskiego miocenu. Acta So- cietatis Botanicorum Poloniae 25: 517–527. Rögl, F. 1999. Mediterranean and Paratethys. Facts and hypotheses of an Oligocene to Miocene paleogeography (short overview). Geologica Carpathica 50: 339–349. Rögl, F., Ćorić, S., Harzhauser, M., Jimenez-Moreno, G., Kroh, A., Schul- tz, O., Wessely, G., and Zorn, I. 2008. The Middle Miocene Badenian stratotype at Baden-Sooss (Lower Austria). Geologica Carpathica 59: 367–374. Visscher, G.E. and Jagels, R. 2003. Separation of Metasequoia and Glyp- tostrobus (Cupressaceae) based on wood anatomy. IAWA Journal 24: 439–450. Zalewska, Z. 1953. Trzeciorzędowe szczątki drewna z Turowa nad Nysą Łużycką. Część I. Acta Geologica Polonica 3: 481–543. Roessler, W. 1937. Pliozäne Koniferenholzer der Umgebung von Gle- Roessler, W. 1937. Pliozäne Koniferenholzer der Umgebung von Gle-

https://openalex.org/W2147310208

https://escholarship.org/content/qt3w74n5ws/qt3w74n5ws.pdf?t=rcrato

English

<i>NR2F1</i> deletion in a patient with a de novo paracentric inversion, inv(5)(q15q33.2), and syndromic deafness

American journal of medical genetics. Part A

cc-by

UC Irvine Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed UC Irvine UC Irvine Previously Published Works Title NR2F1 deletion in a patient with a de novo paracentric inversion, inv(5)(q15q33.2), and syndromic deafness Permalink https://escholarship.org/uc/item/3w74n5ws Journal American Journal of Medical Genetics Part A, 149A(5) ISSN 1552-4825 Authors Brown, Kerry K Alkuraya, Fowzan S Matos, Michael et al. Publication Date 2009-05-01 DOI 10.1002/ajmg.a.32764 Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed UC Irvine UC Irvine Previously Published Works Title NR2F1 deletion in a patient with a de novo paracentric inversion, inv(5)(q15q33.2), and syndromic deafness Permalink https://escholarship.org/uc/item/3w74n5ws Journal American Journal of Medical Genetics Part A, 149A(5) ISSN 1552-4825 Authors Brown, Kerry K Alkuraya, Fowzan S Matos, Michael et al. Publication Date 2009-05-01 DOI 10.1002/ajmg.a.32764 Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed UC Irvine UC Irvine Previously Published Works Title NR2F1 deletion in a patient with a de novo paracentric inversion, inv(5)(q15q33.2), and syndromic deafness Permalink https://escholarship.org/uc/item/3w74n5ws Journal American Journal of Medical Genetics Part A, 149A(5) ISSN 1552-4825 Authors Brown, Kerry K Alkuraya, Fowzan S Matos, Michael et al. Publication Date 2009-05-01 DOI 10.1002/ajmg.a.32764 Copyright Information This work is made available under the terms of a Creative Commons Attribution License availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed NIH Public Access Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Published in final edited form as: Am J Med Genet A. 2009 May ; 149A(5): 931–938. doi:10.1002/ajmg.a.32764. Am J Med Genet A. 2009 May ; 149A(5): 931–938. doi:10.1002/ajmg.a.32764. NR2F1 Deletion in a Patient with a de novo Paracentric Inversion, inv(5)(q15q33.2), and Syndromic Deafness Kerry K. Brown1, Fowzan S. Alkuraya2,3,4, Michael Matos5, Richard L. Robertson6, Virginia E. Kimonis4,7, and Cynthia C. Morton8 1Department of Genetics, Harvard Medical School, Boston, MA ript NIH-PA Author Manuscript 2Developmental Genetics Unit, Department of Genetics, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia 3Department of Pediatrics, King Khalid University Hospital and College of Medicine, King Saud University, Riyadh, Saudi Arabia NIH-PA Author Manuscript 4Division of Genetics, Children's Hospital and Harvard Medical School, Boston, MA Division of Genetics, Children s Hospital and Harvard Medical School, Boston, MA 5Department of Pediatrics, Huggins Hospital, Wolfeboro, NH 5Department of Pediatrics, Huggins Hospital, Wolfeboro, NH 5Department of Pediatrics, Huggins Hospital, Wolfeboro, NH 6Division of Neuroradiology, Department of Radiology, Children's Hospital and Harvard Medical School, Boston, MA 6Division of Neuroradiology, Department of Radiology, Children's Hospital and Harvard Medical School, Boston, MA 7Division of Genetics and Metabolism, Department of Pediatrics, University of California Irvine Medical Center, Orange, CA 7Division of Genetics and Metabolism, Department of Pediatrics, University of California Irvine Medical Center, Orange, CA 8Departments of Obstetrics, Gynecology, and Reproductive Biology and Pathology, Brigham and Women's Hospital and Harvard Medical School, Boston, MA 8Departments of Obstetrics, Gynecology, and Reproductive Biology and Pathology, Brigham and Women's Hospital and Harvard Medical School, Boston, MA Abstract In an effort to discover genes important for human development, we have ascertained patients with congenital anomalies and cytogenetically balanced chromosomal rearrangements. Herein, we report a four year-old girl with profound deafness, a history of feeding difficulties, dysmorphism, strabismus, developmental delay, and an apparently balanced de novo paracentric chromosome 5 inversion, inv(5)(q15q33.2). Molecular cytogenetic analysis of the inversion revealed the presence of microdeletions of approximately 400-500 kb at or near both breakpoints. The 5q15 microdeletion completely removes the nuclear receptor NR2F1 (COUP-TFI) from the inverted chromosome 5. We propose haploinsufficiency of NR2F1 to be the cause of the patient's deafness and many of the other associated anomalies based on striking similarity with the Nr2f1 null mouse. Additionally, this study further highlights the need for high resolution analysis of clinical samples with chromosomal rearrangements as associated deletions may be primarily responsible for the clinical features of these patients. NIH-PA Author Manuscript NIH-PA Author Manuscript Send correspondence to: Cynthia C. Morton, Ph.D. Departments of Obstetrics, Gynecology, and Reproductive Biology and Pathology Brigham and Women's Hospital and Harvard Medical School 77 Avenue Louis Pasteur, NRB 160 Boston, MA 02115 Phone: 617-525-4535 Fax: 617-525-4533 Email: [email protected]. Powered by the California Digital Library University of California eScholarship.org eScholarship.org Keywords NR2F1; deafness; chromosomal inversion; microdeletion; FISH; 5q Send correspondence to: Cynthia C. Morton, Ph.D. Departments of Obstetrics, Gynecology, and Reproductive Biology and Pathology Brigham and Women's Hospital and Harvard Medical School 77 Avenue Louis Pasteur, NRB 160 Boston, MA 02115 Phone: 617-525-4535 Fax: 617-525-4533 Email: [email protected]. Page 2 Page 2 Brown et al. NIH-PA Author Manuscript INTRODUCTION NIH-PA Author Manuscript Approximately 1 in 2000 newborns has a de novo balanced chromosomal rearrangement [Warburton, 1991]. The risk for a congenital anomaly in this population is estimated at 6.1% for translocations and 9.4% for inversions [Warburton, 1991]. Overall, the risk of a congenital anomaly in individuals with a de novo balanced rearrangement is two to three times higher than that observed in an unselected population of newborns, suggesting that the phenotype may be due to deletion, disruption, or dysregulation of gene(s) in the breakpoint regions. Through the Developmental Genome Anatomy Project (DGAP, NIH-PA Author Manuscript NIH-PA Au g g p y j http://dgap.harvard.edu) we have been studying patients with de novo balanced chromosomal rearrangements and congenital anomalies. The goal of DGAP is to identify genes at the chromosomal breakpoints that are critical for human development. Recent studies of patients with phenotypic abnormalities and chromosomal rearrangements using high resolution molecular cytogenetic techniques, such as array comparative genomic hybridization (aCGH), have revealed that many presumably balanced rearrangements include cryptic microdeletions at or near the breakpoints [Astbury et al., 2004; Gribble et al., 2005; De Gregori et al., 2007; Higgins et al., 2008; Fantes et al., 2008; Baptista et al., 2008]. While the majority of patients we have studied have balanced rearrangements, detailed molecular mapping has also revealed some rearrangements with cryptic microdeletions [Higgins et al., 2008]. High resolution characterization of these microdeletions is essential for identifying candidate genes responsible for the patient's phenotype and for distinguishing haploinsufficiencies with clinical consequences from nonpathogenic copy number polymorphisms. NIH-PA Author Manuscript Herein, we describe a 4-year-old girl (designated DGAP169) with profound sensorineural deafness, feeding difficulties, dysmorphism, strabismus, developmental delay, and an apparently balanced de novo paracentric inversion of chromosome 5 identified by GTG- banded analysis of metaphase chromosomes. Detailed FISH mapping of the inversion breakpoints revealed two microdeletions of ~400-500 kb, one at the 5q33 breakpoint and the other a short distance from the 5q15 breakpoint. While several genes are completely or partially deleted due to the rearrangement, NR2F1 is a compelling positional candidate gene for the patient's phenotype. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Clinical details The canal for the cochlear nerve was small suggesting that the cochlear nerve is either small or absent and the course of the facial nerve was bilaterally anomalous in the tympanic and mastoid segments (Figs 2c and 2d). A follow-up MRI to visualize the facial and vestibulocochlear nerve was performed. The facial nerve was visible but very small; however, the cochlear division of the vestibulocochlear nerve could not be seen due to suboptimal quality of the MRI. DGAP169 received a cochlear implant at 28 months with the hope that she would gain some functional hearing ability. Two years after implantation she still has no significant hearing. Although she shows some response to sound during conditioned play audiometry testing, it appears to be a vibrotactile response rather than a true auditory response. Continued implant use and continued adjustment of the programming of the implant will reveal if it is providing any meaningful auditory input. Temporal bone CT revealed malformations of the middle and inner ear, and suggested an abnormality of the cochlear division of the vestibulocochlear nerve. The cochlea was malformed bilaterally with only one to one and a quarter turns. The vestibular aqueduct was enlarged and the position of the lateral semicircular canal was abnormal. All three ossicles, particularly the stapes, were malformed and the incus and malleus appeared to be fused (Figs 2a and 2b). Additionally, the oval window was stenotic. The canal for the cochlear nerve was small suggesting that the cochlear nerve is either small or absent and the course of the facial nerve was bilaterally anomalous in the tympanic and mastoid segments (Figs 2c and 2d). A follow-up MRI to visualize the facial and vestibulocochlear nerve was performed. The facial nerve was visible but very small; however, the cochlear division of the vestibulocochlear nerve could not be seen due to suboptimal quality of the MRI. DGAP169 i d hl i l t t 28 th ith th h th t h ld i Recent follow-up of DGAP169 at four years of age has uncovered several additional abnormal features. Her feeding has improved significantly with age and her caloric intake has been consistently above her estimated needs for growth catch-up; however, she still shows poor growth, weighing only 28 lbs (3rd-5th centile). As an infant, she stopped breathing whenever her arms were elevated. Clinical details DGAP169 was ascertained as part of the Developmental Genome Anatomy Project. The human study protocol for this project has been reviewed and approved by the Partners HealthCare System Human Research Committee. NIH-PA Author Manuscript DGAP169 was born at term to healthy, unrelated parents following an uneventful pregnancy via planned cesarean due to prior cesarean delivery. Her G3P2SAB1 mother was 26 years old at the time of delivery and has one healthy son. At birth, DGAP169 was immediately noticed to have malformed ears. Major sucking and swallowing difficulties further complicated her neonatal course. GTG-banded chromosomal analysis of peripheral blood lymphocytes revealed the presence of an apparently balanced chromosome 5 inversion. Her karyotype was reported as 46,XX,inv(5)(q14q34). Both parents were subsequently determined to have normal karyotypes. DGAP169 was evaluated at Children's Hospital Boston at 11 weeks of age. Her weight of 3.22 kg had barely increased over her birth weight because of poor sucking and suspected uncoordinated sucking and swallowing. Swallow study confirmed the uncoordinated sucking and swallowing. Her scalp hair pattern was unusual in that the direction of hair Page 3 Brown et al. NIH-PA Author Manuscript growth was toward the midline resembling a “comb” that extended backward along the line of the sagittal suture, with very sparse temporal hair (Fig 1). Her face was somewhat asymmetric with her eyes positioned at slightly different levels. She had upslanted palpebral fissures and anteverted nares with a broad tip of the nose. Remarkable micrognathia was noted with no associated cleft palate. Her ear pinnae were low-set, small, and significantly malformed (Fig 1) but her external auditory canals were patent. Her neurological examination was notable for significant axial hypotonia with decreased deep tendon reflexes. Auditory brainstem response (ABR) performed to test her hearing showed no air conduction response for 1-4 kHz even at 100 dB and no bone conduction response for 0.5-1.5 kHz at 65 dB indicating bilateral profound sensorineural deafness. NIH-PA Author Manuscript NIH- Temporal bone CT revealed malformations of the middle and inner ear, and suggested an abnormality of the cochlear division of the vestibulocochlear nerve. The cochlea was malformed bilaterally with only one to one and a quarter turns. The vestibular aqueduct was enlarged and the position of the lateral semicircular canal was abnormal. All three ossicles, particularly the stapes, were malformed and the incus and malleus appeared to be fused (Figs 2a and 2b). Additionally, the oval window was stenotic. Clinical details The cause of this condition is unknown but seems to persist as she is still unwilling to lift her arms upward. Otherwise, she moves and uses her arms normally. Strabismus was noted at 18 months and persists at four years. She continues to have sparse temporal hair at four years of age, although this feature is less obvious since the rest of her hair has been grown long. Evaluation of developmental milestones at 28 months showed significant delay in gross motor skills (18-21 month level), cognitive skills (15.5-18 month level), and communication (5.5-7 month level), although she had recently started walking. By 35 months, she engaged in limited signing, was able to follow simple commands, could stack blocks, and enjoyed toys. At four years, she climbs stairs with alternating feet and has started toilet training. NIH-PA Author Manuscript Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Reverse transcriptase-polymerase chain reaction (RT-PCR) of NR2F1 RNA was isolated from the DGAP169 lymphoblastoid cell line using Trizol (Invitrogen). NR2F1 was reverse transcribed and amplified from 1 ug of RNA using SuperScript One- Step RT-PCR with Platinum Taq (Invitrogen). Sequencing of NR2F1 DNA was isolated from the DGAP169 lymphoblastoid cell line using the PureGene system (Gentra, Minneapolis, MN). M13-tagged primer pairs were used to amplify the three exons of NR2F1. PCR reactions contained 50 ng DNA, 0.4 uM each primer, 2 mM MgSO4, 0.4 mM dNTP, 1X High Fidelity PCR Buffer, 2.5% DMSO, and 2.5 units Platinum High Fidelity Taq DNA polymerase (Invitrogen). PCR was performed with an initial denaturation of 97°C for 5 minutes, 35 cycles of 97°C for 30 seconds, 55°C for 30 seconds, 72°C for 45 seconds, and a final extension of 72°C for 7 minutes. PCR products were purified using the QIAquick PCR Purification kit (Qiagen, Valencia, CA) and sequenced at the Dana-Farber/ Harvard Cancer Center DNA Resource Core. DNA was isolated from the DGAP169 lymphoblastoid cell line using the PureGene system (Gentra, Minneapolis, MN). M13-tagged primer pairs were used to amplify the three exons of NR2F1. PCR reactions contained 50 ng DNA, 0.4 uM each primer, 2 mM MgSO4, 0.4 mM dNTP, 1X High Fidelity PCR Buffer, 2.5% DMSO, and 2.5 units Platinum High Fidelity Taq DNA polymerase (Invitrogen). PCR was performed with an initial denaturation of 97°C for 5 minutes, 35 cycles of 97°C for 30 seconds, 55°C for 30 seconds, 72°C for 45 seconds, and a final extension of 72°C for 7 minutes. PCR products were purified using the QIAquick PCR Purification kit (Qiagen, Valencia, CA) and sequenced at the Dana-Farber/ Harvard Cancer Center DNA Resource Core. NIH-PA Author Manuscript Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Fluorescence in situ hybridization (FISH) analysis After informed consent was obtained, a peripheral blood sample was collected from the child. A lymphoblastoid cell line was generated from the blood sample at the Massachusetts General Hospital Cell Transformation Core using standard protocols. Based on the reported karyotype, BAC clones spanning the breakpoint regions were selected for FISH mapping using the University of California Santa Cruz Genome Browser (http://genome.ucsc.edu). BACs from the RP11 library were acquired from Children's Hospital Oakland Research Institute (Oakland, CA) and BACs from the CTD library were acquired from Invitrogen (Carlsbad, CA). Metaphase chromosome spreads were prepared from the lymphoblastoid Brown et al. Page 4 cells using standard cytogenetic protocols and FISH performed as previously described [Ney et al., 1993]. BACs were labeled directly with either Spectrum Orange or Spectrum Green conjugated dUTP using a nick translation kit (Vysis, Downers Grove, IL) and differentially labeled pairs were hybridized overnight to metaphase chromosome preparations. After washing, chromosomes were counterstained with DAPI and analyzed with a Zeiss Axioskop microscope (Thornwood, NY) and Applied Imaging CytoVision software (Santa Clara, CA). At least 10 metaphases were scored per BAC probe. NIH-PA Author Manuscript To interpret accurately the results of the child's FISH experiments, peripheral blood samples were obtained from the biological parents. The blood samples were cultured, lymphocytes harvested, and metaphase spreads prepared according to standard cytogenetic protocols. FISH was then performed as described above. RESULTS To map the inversion breakpoints, we performed FISH on metaphase chromosomes from patient lymphoblastoid cells. Based on the reported karyotype, inv(5)(q14q34), BACs localized to 5q13-q15 and 5q33-q35 on the UCSC website were selected for FISH. Sequential FISH experiments were performed to narrow the breakpoint regions. Results for a subset of BACs used for each breakpoint are listed in Table I. BAC clones RP11-108e6 and RP11-176p18 from 5q15 hybridized to 5q15 on the normal chromosome 5 and to both 5q15 and 5q33 on the inverted chromosome 5, narrowing the breakpoint region to about 135 kb (Fig 3a). The ends of a BAC centromeric to the breakpoint, RP11-213a8, and a BAC telomeric to the breakpoint, RP11-297g19, slightly overlap but neither shows a split hybridization signal, indicating that the breakpoint likely occurs within the 20 kb of overlapping sequence. Based on this localization, the 5q15 breakpoint disrupts a large gene, C5ORF21, which spans about 500 kb of genomic sequence (Fig 4a). NIH-PA Author Manuscript NIH-PA Author Manuscript While mapping the 5q15 breakpoint, a cryptic deletion of 360-480 kb was discovered approximately 220-330 kb centromeric to the breakpoint region. Three BAC clones, CTD-3236h3, RP11-829m2, and RP11-608g16, produced hybridization signals only on the normal chromosome 5, while the flanking BAC clones, RP11-92c19 and RP11-65f13, produced hybridization signals on both the normal 5 and inv(5) (Fig 3b). As a result, NR2F1 and AK124699 are completely deleted and C5ORF21 may be partially deleted from the inv(5) (Fig 4a). Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Page 5 Page 5 Brown et al. NIH-PA Author Manuscript A cryptic deletion of 450-560 kb was also detected at the 5q33.2 breakpoint. All BACs centromeric to the 5q33.2 deletion showed a hybridization signal at 5q33.2 on the normal chromosome 5 and at 5q15 on the inv(5). All BACs telomeric to the 5q33.2 deletion showed a hybridization signal at 5q33.2 on both the normal 5 and the inv(5). These results indicate that the deletion marks the inversion breakpoint. Six BAC clones, CTD-2338o23, CTD-2336d9, RP11-635k22, CTD-2314a16, RP11-314p8, and RP11-108c12, produced hybridization signals only on the normal chromosome 5, while the flanking BAC clones, CTD-2270b22 and RP11-790o1, produced hybridization signals on the normal 5 and the inv(5) (Figure 3c). As a result of this deletion, five additional genes, C5ORF3, MFAP3, GALNT10, SAP30L, and HAND1, are completely or partially deleted from the inv(5) (Fig 4b). RESULTS NIH-PA Author Manuscript NIH To determine if these deletions occurred de novo in the child, FISH was performed on chromosomes from both parents using two BAC probes within the 5q15 deleted region and two BAC probes within the 5q33.2 deleted region. Two chromosome 5 signals were detected for both parents with all BACs tested, indicating that the microdeletions are only present in the child. To determine if any of the deleted genes could be candidate deafness genes, we queried our human fetal cochlear cDNA library [Skvorak et al., 1999] for transcripts from all deleted genes. Four of the eight possibly deleted genes, NR2F1, C5ORF21, AK124699, and MFAP3, are expressed in fetal cochlea with NR2F1 among the top 50 most highly expressed cochlear transcripts. Because NR2F1 is highly expressed in the cochlea and Nr2f1 knockout mice have profound sensorineural deafness by three weeks of age [Wenzel et al., 2004], we investigated this gene further as a possible candidate for deafness in DGAP169. We sequenced the remaining NR2F1 allele in DGAP169 and found no additional mutations suggesting that any phenotype attributed to the gene would likely be due to haploinsufficiency. We also attempted to assess expression levels of NR2F1 in the DGAP169 lymphoblastoid cell line to determine whether the deletion leads to reduced NR2F1 expression. Using reverse transcriptase-polymerase chain reaction (RT-PCR) we were unable to detect NR2F1 expression in either control or patient lymphoblastoid cell lines indicating that the gene is not expressed in this cell type. The RT-PCR results are consistent with the NR2F1 profile in the UniGene database (http://www.ncbi.nlm.nih.gov) in which no expression was detected in blood. NIH-PA Author Manuscript Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. DISCUSSION Specifically, the paths of some of the cranial nerves are abnormal, both the saccule and organ of Corti of the inner ear are poorly innervated, and misguided axonal projections are present in the cochlear duct [Qiu et al., 1997; Pereira et al., 2001]. DGAP169 seems to have similar axonal guidance defects. Imaging revealed an abnormal course of the facial nerve bilaterally, and the limited response to cochlear implantation may result from poor innervation of the cochlea or other defects in the auditory nerve. Additionally, the patient's strabismus could be due to defects in cranial nerves III, IV, or VI. Nr2f1 null mice that survived to three weeks are described as ataxic, suggesting there could also be problems with connections of other motor neurons [Qiu et al., 1997]. Likewise, the hypotonia and delay in motor skills seen in DGAP169 suggest deficits in motor neurons. While no abnormalities in hair patterning were reported for Nr2f1 null mice, Nr2f1 can regulate Notch signaling [Tang et al., 2006], which is important for proper hair patterning [Uyttendaele et al., 2004; Vauclair et al., 2005]. Because of the strikingly similar phenotypic features observed in DGAP169 and Nr2f1 null mice, we propose that haploinsufficiency for NR2F1 is the primary cause of the DGAP169 phenotype. NIH-PA Author Manuscript NIH- While mice haploinsufficient for Nr2f1 appear to be normal [Qiu et al., 1997], this does not exclude our hypothesis that NR2F1 is primarily responsible for the DGAP169 phenotype, as humans seem to be more sensitive than mice to dosage differences in certain genes. For example, holoprosencephaly type 3 is caused by haploinsufficiency for Sonic Hedgehog (SHH). Shh null mice generally display an embryonic lethal phenotype with CNS abnormalities and midline defects, while Shh haploinsufficient mice appear completely normal [Chiang et al., 1996]. Additionally, mice haploinsufficient for Tbx1 show only cardiac abnormalities, while Tbx1 null mice exhibit most of the common features seen in haploinsufficient DiGeorge/Velocardiofacial syndrome patients, including cardiac abnormalities, hypoplasia of the thymus, abnormal facial structures, abnormal vertebrae, and cleft palate [Jerome and Papaioannou, 2001]. Consistent with our hypothesis, the complete absence of Nr2f1 in the null mouse causes a more severe phenotype than seen in DGAP169 who presumably has some residual NR2F1 function. DISCUSSION NIH-PA Author Manuscript Molecular cytogenetic analysis of an inv(5)(q15q33.2)dn in a female with syndromic deafness revealed microdeletions of about 400-500 kb in the vicinity of both inversion breakpoints. The two microdeletions result in partial or complete deletion of approximately eight genes, and haploinsufficiency for one or more of these genes is the likely etiology of the child's phenotype. Copy number gains within the centromeric portion of the 5q15 deletion overlapping AK124699 have been reported in normal individuals, but no additional copy number polymorphisms overlapping the other deleted genes have been reported (Database of Genomic Variants, http://projects.tcag.ca/variation/; USCS Genome Browser, http://genome.ucsc.edu; Ensembl Genome Browser, http://www.ensembl.org). Pathogenic interstitial deletions including bands 5q15 and 5q33.2 have been reported in the literature; however, all reported patients have much larger deletions than detected in DGAP169 and only a few of these deletions have been mapped molecularly making it difficult to compare this patient to those reports. Hearing defects have been noted in two patients with interstitial deletions involving 5q15 [Harprecht-Beato et al., 1983; Harreus and Issing, 2002], but it is unclear whether these deletions overlap with that of DGAP169. Only conductive hearing loss has been noted in patients with deletions involving 5q33.2 [Giltay et al., 1997]. Page 6 Brown et al. NIH-PA Author Manuscript We propose haploinsufficiency for NR2F1 as the primary cause of the patient's syndrome. NR2F1 is an orphan receptor of the steroid/thyroid hormone receptor superfamily, which consists of ligand inducible transcription factors. It plays an important role in neurogenesis and neural crest cell differentiation [Qiu et al., 1997]. Studies of mouse embryos show that Nr2f1 is expressed in the otic vesicle, forebrain, midbrain, hindbrain, branchial arches, and neural tube at E9.0 [Qiu et al., 1997; Jonk et al., 1994]. At later time points, Nr2f1 is expressed in the cochlea, middle ear, restricted brain regions, spinal cord, tongue, and other tissues [Jonk et al., 1994; Pereira et al., 1995]. Most Nr2f1 knockout mice die between 8 and 36 hours of birth due to inability to feed [Qiu et al., 1997]. Similar to DGAP169, the mice have difficulty swallowing. Nr2f1 null mice that survive to P20 are profoundly deaf displaying no response when tested for auditory brainstem response (ABR) or distortion product otoacoustic emissions (DPOAE) [Wenzel et al., 2004]. Null mice have a shortened cochlear duct and malformed vestibular chambers [Tang et al., 2006]. The null mice also have axonal guidance and projection defects. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. DISCUSSION MFAP3, C5ORF21, and AK124699 appear to be weakly expressed in fetal cochlea, although NR2F1 is the more compelling candidate human deafness gene. NIH-PA Author Manuscript NIH-PA Several recent reports have noted that individuals with congenital anomalies due to apparently balanced chromosomal rearrangements commonly have deletions near the rearrangement breakpoints [Astbury et al., 2004; Gribble et al., 2005; De Gregori et al, 2007; Higgins et al., 2008; Baptista et al., 2008; Fantes et al., 2008]. These additional deletions will presumably affect more genes than the rearrangement alone increasing the chances that the individual will have an abnormal phenotype. As of yet, none of these studies have assessed if there are differences in the rate of cryptic deletions associated with inversions versus translocations. Three of these reports [Astbury et al., 2004; Higgins et al., 2008; Fantes et al., 2008] contain data on patients whose congenital anomalies are associated with inversions. Combining the data on patients with congenital anomalies studied in these reports reveals that 5 of 11 (45%) inversions and 18 of 78 (23%) simple reciprocal translocations have associated deletions. These data suggest that patients with inversions may be more likely to have cryptic deletions. So far, the number of reported inversion patients whose rearrangements have been studied at high resolution is very small and, therefore, the difference between the frequency of deletions associated with inversions versus translocations does not reach statistical significance. Consequently, additional studies with larger numbers of patients may be warranted to investigate further this preliminary observation. Interestingly, the increased number of deletions seen in these inversion patients correlates with Warburton's study [Warburton, 1991] which estimates the risk of congenital anomaly in individuals with apparently balanced inversions to be 1.5 times higher than in individuals with apparently balanced translocations. Perhaps, a higher incidence of associated deletions contributes to the increased risk seen for inversion patients. While we cannot definitively determine the causal gene from a single patient, we suggest that haploinsufficiency for NR2F1 could be etiologic for all or at least most of the anomalies observed in DGAP169, including deafness, feeding difficulties, abnormal course of the facial nerve, strabismus, hypotonia, and developmental delay. Genetic analysis of other individuals with a similar phenotype will be necessary to confirm this prediction. DISCUSSION Additionally, we recommend high resolution screening for imbalances in patients with While we cannot definitively determine the causal gene from a single patient, we suggest that haploinsufficiency for NR2F1 could be etiologic for all or at least most of the anomalies observed in DGAP169, including deafness, feeding difficulties, abnormal course of the facial nerve, strabismus, hypotonia, and developmental delay. Genetic analysis of other individuals with a similar phenotype will be necessary to confirm this prediction. Additionally, we recommend high resolution screening for imbalances in patients with congenital anomalies and chromosomal rearrangements as associated deletions seem to be a common occurrence. NIH-PA Author Manuscript Additionally, we recommend high resolution screening for imbalances in patients with congenital anomalies and chromosomal rearrangements as associated deletions seem to be a common occurrence. DISCUSSION Furthermore, while some differences in phenotypic manifestations between species are expected, the tissues and systems affected in the Nr2f1 null mice and in DGAP169 are quite similar and, perhaps, a mouse model homozygous for an Nr2f1 hypomorphic mutation would mimic even more closely the human phenotype. NIH-PA Author Manuscript NIH-PA Author Manuscript While NR2F1 is the most attractive positional candidate gene, we cannot conclusively exclude the other deleted genes from playing a role in the patient's phenotype. Most of the deleted genes are widely expressed; however, HAND1 has a more restricted expression pattern. HAND1 is the only deleted gene other than NR2F1 that has been well studied, and it is a transcription factor essential for proper development of the placenta and the heart [Firulli et al., 1998; Riley et al., 1998]. MFAP3 localizes to microfibrils and could play a Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Page 7 Brown et al. NIH-PA Author Manuscript role in microfibril diseases [Abrams et al., 1995]. Genes encoding other microfibril components cause Marfan syndrome, congenital contractural arachnodactyly, familial mitral valve prolapse syndrome, annuloaortic ectasia, and familial ectopia lentis. SAP30L interacts with several components of the Sin3A corepressor complex and can induce transcriptional repression probably through recruitment of histone deacetylases [Viiri et al., 2006]. GALNT10 is an enzyme that catalyzes the first step of O-linked protein glycosylation [Nelson et al., 2002]. Very little is known about C5ORF3, C5ORF21, and AK124699, only that C5ORF3 contains an ATP/GTP binding site [Boultwood et al., 2000] and C5ORF21 has an endoplasmic reticulum targeting sequence [Simpson et al., 2000]. MFAP3, C5ORF21, and AK124699 appear to be weakly expressed in fetal cochlea, although NR2F1 is the more compelling candidate human deafness gene. role in microfibril diseases [Abrams et al., 1995]. Genes encoding other microfibril components cause Marfan syndrome, congenital contractural arachnodactyly, familial mitral valve prolapse syndrome, annuloaortic ectasia, and familial ectopia lentis. SAP30L interacts with several components of the Sin3A corepressor complex and can induce transcriptional repression probably through recruitment of histone deacetylases [Viiri et al., 2006]. GALNT10 is an enzyme that catalyzes the first step of O-linked protein glycosylation [Nelson et al., 2002]. Very little is known about C5ORF3, C5ORF21, and AK124699, only that C5ORF3 contains an ATP/GTP binding site [Boultwood et al., 2000] and C5ORF21 has an endoplasmic reticulum targeting sequence [Simpson et al., 2000]. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. REFERENCES [PubMed: 17766364] De Gregori M, Ciccone R, Magini P, Pramparo T, Gimelli S, Messa J, Novara F, Vetro A, Rossi E, Maraschio P, Bonaglia MC, Anichini C, Ferrero GB, Silengo M, Fazzi E, Zatterale A, Fischetto R, Previdere C, Belli S, Turci A, Calabrese G, Bernardi F, Meneghelli E, Riegel M, Rocchi M, Previdere C, Belli S, Turci A, Calabrese G, Bernardi F, Meneghelli E, Riegel M, Rocchi M, Guerneri S, Lalatta F, Zelante L, Romano C, Fichera M, Mattina T, Arrigo G, Zollino M, Giglio S, Lonardo F, Bonfante A, Ferlini A, Cifuentes F, Van Esch H, Backx L, Schinzel A, Vermeesch JR, Zuffardi O. Cryptic deletions are a common finding in “balanced” reciprocal and complex chromosome rearrangements: a study of 59 patients. J Med Genet. 2007; 44:750–762. [PubMed: 17766364] Guerneri S, Lalatta F, Zelante L, Romano C, Fichera M, Mattina T, Arrigo G, Zollino M, Giglio S, Lonardo F, Bonfante A, Ferlini A, Cifuentes F, Van Esch H, Backx L, Schinzel A, Vermeesch JR, Zuffardi O. Cryptic deletions are a common finding in “balanced” reciprocal and complex chromosome rearrangements: a study of 59 patients. J Med Genet. 2007; 44:750–762. [PubMed: 17766364] Fantes JA, Boland E, Ramsay J, Donnai D, Splitt M, Goodship JA, Stewart H, Whiteford M, Gautier P, Harewood L, Holloway S, Sharkey F, Maher E, van Heyningen V, Clayton-Smith J, Fitzpatrick DR, Black GC. FISH mapping of de novo apparently balanced chromosome rearrangements identifies characteristics associated with phenotypic abnormality. Am J Hum Genet. 2008; 82:916– 926. [PubMed: 18374296] Firulli AB, McFadden DG, Lin Q, Srivastava D, Olson EN. Heart and extra-embryonic mesodermal defects in mouse embryos lacking the bHLH transcription factor Hand1. Nat Genet. 1998; 18:266– 270. [PubMed: 9500550] Giltay JC, Gerssen-Schoorl KB, Luitse GH, Dauwerse HG. A case of de novo interstitial deletion of chromosome 5(q33q34). Clin Genet. 1997; 52:173–176. [PubMed: 9377807] NIH-PA Author Manuscript Gribble SM, Prigmore E, Burford DC, Porter KM, Ng BL, Douglas EJ, Fiegler H, Carr P, Kalaitzopoulos D, Clegg S, Sandstrom R, Temple IK, Youings SA, Thomas NS, Dennis NR, Jacobs PA, Crolla JA, Carter NP. The complex nature of constitutional de novo apparently balanced translocations in patients presenting with abnormal phenotypes. J Med Genet. 2005; 42:8–16. [PubMed: 15635069] NIH-PA Author Manuscript Harprecht-Beato W, Kaiser P, Steuber E, Reinhard W. Interstitial deletion in the long arm of chromosome No. 5. Clin Genet. 1983; 23:167–171. [PubMed: 6839530] Harreus UA, Issing WJ. Acknowledgments We would like to express our sincere gratitude to the patient and her family for making this study possible. We would also like to thank Heather Ferguson for assistance with patient enrollment, the Massachusetts General Hospital Cell Transformation Core for assistance with EBV transformation, the Brigham and Women's Hospital Clinical Cytogenetics Laboratory for assistance with processing the parental blood samples, and the Dana-Farber/ Harvard Cancer Center (DF/HCC) DNA resource core (P30CA06516) for assistance with DNA sequencing. This work was supported by grants from the National Institutes of Health, P01 GM061354 (to C.C.M.) and F31 DC007540 (to K.K.B). Page 8 Page 8 Brown et al. NIH-PA Author Manuscript REFERENCES Abrams WR, Ma RI, Kucich U, Bashir MM, Decker S, Tsipouras P, McPherson JD, Wasmuth JJ, Rosenbloom J. Molecular cloning of the microfibrillar protein MFAP3 and assignment of the gene to human chromosome 5q32-q33.2. Genomics. 1995; 26:47–54. [PubMed: 7782085] NIH-PA Author Manuscript Astbury C, Christ LA, Aughton DJ, Cassidy SB, Kumar A, Eichler EE, Schwartz S. Detection of deletions in de novo “balanced” chromosome rearrangements: further evidence for their role in phenotypic abnormalities. Genet Med. 2004; 6:81–89. [PubMed: 15017330] Baptista J, Mercer C, Prigmore E, Gribble SM, Carter NP, Maloney V, Thomas NS, Jacobs PA, Crolla JA. Breakpoint mapping and array CGH in translocations: comparison of a phenotypically normal and an abnormal cohort. Am J Hum Genet. 2008; 82:927–936. [PubMed: 18371933] Boultwood J, Fidler C, Strickson AJ, Watkins F, Kostrzewa M, Jaju RJ, Muller U, Wainscoat JS. Transcription mapping of the 5q- syndrome critical region: cloning of two novel genes and sequencing, expression, and mapping of a further six novel cDNAs. Genomics. 2000; 66:26–34. [PubMed: 10843801] Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H, Beachy PA. Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature. 1996; 383:407– 413. [PubMed: 8837770] De Gregori M, Ciccone R, Magini P, Pramparo T, Gimelli S, Messa J, Novara F, Vetro A, Rossi E, Maraschio P, Bonaglia MC, Anichini C, Ferrero GB, Silengo M, Fazzi E, Zatterale A, Fischetto R, Previdere C, Belli S, Turci A, Calabrese G, Bernardi F, Meneghelli E, Riegel M, Rocchi M, Guerneri S, Lalatta F, Zelante L, Romano C, Fichera M, Mattina T, Arrigo G, Zollino M, Giglio S, Lonardo F, Bonfante A, Ferlini A, Cifuentes F, Van Esch H, Backx L, Schinzel A, Vermeesch JR, Zuffardi O. Cryptic deletions are a common finding in “balanced” reciprocal and complex chromosome rearrangements: a study of 59 patients. J Med Genet. 2007; 44:750–762. REFERENCES [Chromosome 5q-syndrome-ENT pathologies]. Laryngorhinootologie. 2002; 81:565–567. [PubMed: 12189572] Higgins AW, Alkuraya FS, Bosco AF, Brown KK, Bruns GA, Donovan DJ, Eisenman R, Fan Y, Farra CG, Ferguson HL, Gusella JF, Harris DJ, Herrick SR, Kelly C, Kim H-G, Kishikawa S, Korf BR, Kulkarni S, Lally E, Leach NT, Lemyre E, Lewis J, Ligon AH, Lu W, Maas RL, MacDonald ME, Moore SD, Peters RE, Quade BJ, Quintero-Rivera F, Saadi I, Shen Y, Shendure J, Williamson RE, Morton CC. Characterization of apparently balanced chromosomal rearrangements from the developmental genome anatomy project. Am J Hum Genet. 2008; 82:712–722. [PubMed: 18319076] Jerome LA, Papaioannou VE. DiGeorge syndrome phenotype in mice mutant for the T-box gene, Tbx1. Nat Genet. 2001; 27:286–291. [PubMed: 11242110] Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Brown et al. Page 9 NIH-PA Author Manuscript Jonk LJ, de Jonge ME, Pals CE, Wissink S, Vervaart JM, Schoorlemmer J, Kruijer W. Cloning and expression during development of three murine members of the COUP family of nuclear orphan receptors. Mech Dev. 1994; 47:81–97. [PubMed: 7947324] NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Auth Nelson PA, Sutcliffe JG, Thomas EA. A new UDP-GalNAc:polypeptide N- acetylgalactosaminyltransferase mRNA exhibits predominant expression in the hypothalamus, thalamus and amygdala of mouse forebrain. Brain Res Gene Expr Patterns. 2002; 1:95–99. [PubMed: 15018805] Ney PA, Andrews NC, Jane SM, Safer B, Purucker ME, Weremowicz S, Morton CC, Goff SC, Orkin SH, Nienhuis AW. Purification of the human NF-E2 complex: cDNA cloning of the hematopoietic cell-specific subunit and evidence for an associated partner. Mol Cell Biol. 1993; 13:5604–5612. [PubMed: 8355703] Pereira FA, Qiu Y, Tsai MJ, Tsai SY. Chicken ovalbumin upstream promoter transcription factor (COUP-TF): expression during mouse embryogenesis. J Steroid Biochem Mol Biol. 1995; 53:503–508. [PubMed: 7626501] Pereira, FA.; Xu, J.; Price, SD.; Shope, CD.; Brownell, WE.; Lysakowski, A.; Tsai, M-J. Vestibular and auditory system defects in COUP-TFI mutants. Paper presented at the 24th Annual Mid- Winter Research Meeting of the Association for Research in Otolaryngology; St. Petersburg, FL. 2001. Qiu Y, Pereira FA, DeMayo FJ, Lydon JP, Tsai SY, Tsai MJ. Null mutation of mCOUPTFI results in defects in morphogenesis of the glossopharyngeal ganglion, axonal projection, and arborization. Genes Dev. 1997; 11:1925–1937. [PubMed: 9271116] Riley P, Anson-Cartwright L, Cross JC. The Hand1 bHLH transcription factor is essential for placentation and cardiac morphogenesis. Nat Genet. 1998; 18:271–275. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. REFERENCES [PubMed: 9500551] Simpson JC, Wellenreuther R, Poustka A, Pepperkok R, Wiemann S. Systematic subcellular localization of novel proteins identified by large-scale cDNA sequencing. EMBO Rep. 2000; 1:287–292. [PubMed: 11256614] Skvorak AB, Weng Z, Yee AJ, Robertson NG, Morton CC. Human cochlear expressed sequence tags provide insight into cochlear gene expression and identify candidate genes for deafness. Hum Mol Genet. 1999; 8:439–452. [PubMed: 9949203] Tang LS, Alger HM, Pereira FA. COUP-TFI controls Notch regulation of hair cell and support cell differentiation. Development. 2006; 133:3683–3693. [PubMed: 16914494] Uyttendaele H, Panteleyev AA, de Berker D, Tobin DT, Christiano AM. Activation of Notch1 in the hair follicle leads to cell-fate switch and Mohawk alopecia. Differentiation. 2004; 72:396–409. [PubMed: 15606499] Vauclair S, Nicolas M, Barrandon Y, Radtke F. Notch1 is essential for postnatal hair follicle development and homeostasis. Dev Biol. 2005; 284:184–193. [PubMed: 15978571] NIH-PA Author Manuscript Viiri KM, Korkeamaki H, Kukkonen MK, Nieminen LK, Lindfors K, Peterson P, Maki M, Kainulainen H, Lohi O. SAP30L interacts with members of the Sin3A corepressor complex and targets Sin3A to the nucleolus. Nucleic Acids Res. 2006; 34:3288–3298. [PubMed: 16820529] Warburton D. De novo balanced chromosome rearrangements and extra marker chromosomes identified at prenatal diagnosis: clinical significance and distribution of breakpoints. Am J Hum Genet. 1991; 49:995–1013. [PubMed: 1928105] Wenzel, GI.; Pereira, FA.; Oghalai, JS. Cochlear Dysfunction in COUP-TFI Mutant Mice. Paper presented at The 27th Annual Mid-Winter Research Meeting of the Association for Research in Otolaryngology; Daytona Beach, FL. 2004. Page 10 Page 10 Brown et al. NIH-PA Author Manuscript Figure 1. Front and side views of DGAP169 at 11 weeks of age. Note the low-set, small, malformed ears, unusual pattern of hair growth, upslanting palpebral fissures, broad nasal tip, anteverted nares, and micrognathia. The face is also asymmetric with the eyes positioned at slightly different levels. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] g NIH-PA Author Manuscript Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Figure 1. Figure 1. Front and side views of DGAP169 at 11 weeks of age. Note the low-set, small, malformed ears, unusual pattern of hair growth, upslanting palpebral fissures, broad nasal tip, anteverted nares, and micrognathia. The face is also asymmetric with the eyes positioned at slightly different levels. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] NIH-PA Author Manuscript NIH-PA Author Manuscript Brown et al. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. REFERENCES Page 11 NIH-PA Author Manuscript Figure 2. CT images of DGAP169 temporal bone. a) The malformed stapes is indicated by the arrow. b) The malformed malleus is indicated by the arrow. c) The asterisk marks the underdeveloped middle and apical turns of the cochlea, while the arrow marks the abnormal path of the facial nerve through the mastoid bone. d) The abnormal course of the facial nerve through the middle ear is indicated by the arrow. NIH-PA Author Manuscript Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Figure 2. Figure 2. CT images of DGAP169 temporal bone. a) The malformed stapes is indicated by the arrow. b) The malformed malleus is indicated by the arrow. c) The asterisk marks the underdeveloped middle and apical turns of the cochlea, while the arrow marks the abnormal path of the facial nerve through the mastoid bone. d) The abnormal course of the facial nerve through the middle ear is indicated by the arrow. NIH-PA Author Manuscript Page 12 Brown et al. Page 12 Figure 3. FISH results on DGAP169 metaphase chromosomes using chromosome 5 BAC probes. a) BAC RP11-108e6 (red) maps to 5q15 and spans the 5q15 inversion breakpoint. b) BAC CTD-3236h3 (red) maps to 5q15 and is deleted from the inv(5). BAC RP11-790o1 (green) at 5q33.2 marks chromosome 5. C) BAC RP11-635k22 (green) maps to 5q33.2 and is deleted from the inv(5). BAC RP11-521h6 (red) at 5q15 marks chromosome 5. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] NIH-PA Author Manuscript NIH-PA Author Manuscript N Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Figure 3. FISH results on DGAP169 metaphase chromosomes using chromosome 5 BAC probes. a) BAC RP11-108e6 (red) maps to 5q15 and spans the 5q15 inversion breakpoint. b) BAC CTD-3236h3 (red) maps to 5q15 and is deleted from the inv(5). BAC RP11-790o1 (green) at 5q33.2 marks chromosome 5. C) BAC RP11-635k22 (green) maps to 5q33.2 and is deleted from the inv(5). BAC RP11-521h6 (red) at 5q15 marks chromosome 5. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] NIH-PA Author Manuscript Page 13 Page 13 Brown et al. NIH-PA Author Manuscript Figure 4. Maps of the inv(5)(q15q33.2) breakpoint regions from the UCSC genome browser. a) Map of 5q15 showing the breakpoint region and the deleted region centromeric to the breakpoint. b) Map of 5q33.2 showing the region deleted by the breakpoint. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. REFERENCES The red portion of the line at the bottom of each section shows the minimal deleted regions. The grey portions of the line show the maximal deleted regions. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] Page 13 NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Au Figure 4. Am J Med Genet A. Author manuscript; available in PMC 2010 May 1. Figure 4. Figure 4. Maps of the inv(5)(q15q33.2) breakpoint regions from the UCSC genome browser. a) Map of 5q15 showing the breakpoint region and the deleted region centromeric to the breakpoint. b) Map of 5q33.2 showing the region deleted by the breakpoint. The red portion of the line at the bottom of each section shows the minimal deleted regions. The grey portions of the line show the maximal deleted regions. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] Figure 4. Maps of the inv(5)(q15q33.2) breakpoint regions from the UCSC genome browser. a) Map of 5q15 showing the breakpoint region and the deleted region centromeric to the breakpoint. b) Map of 5q33.2 showing the region deleted by the breakpoint. The red portion of the line at the bottom of each section shows the minimal deleted regions. The grey portions of the line show the maximal deleted regions. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Page 14 Page 14 Brown et al. *BAC position based on March 2006 genome assembly *BAC position based on March 2006 genome assembly Table I FISH results for BACs mapping to the 5q15 and 5q33 breakpoint regions 5q15 BAC BAC position on chromosome 5* FISH result for inv(5) RP11-60g8 90,227,590-90,372,004 5q15 RP11-92c19 92,593,206-92,749,220 5q15 CTD-3236h3 92,612,533-92,691,223 Deleted RP11-829m2 92,620,518-92,801,910 Deleted RP11-608g16 92,784,673-92,972,632 Deleted RP11-65f13 92,894,088-93,074,143 5q15 RP11-521h6 93,023,218-93,207,639 5q15 RP11-213a8 93,168,124-93,308,711 5q15 RP11-108e6 93,196,412-93,376,816 5q15 and 5q33 RP11-176p18 93,241,892-93,406,803 5q15 and 5q33 RP11-297g19 93,288,262-93,463,665 5q33 RP11-33a7 93,746,809-93,916,228 5q33 RP11-86c20 152,618,661-152,788,737 5q15 RP11-343m7 153,138,586-153,319,561 5q15 CTD-2270b22 153,311,373-153,523,768 5q15 CTD-2338o23 153,390,771-153,495,446 Deleted CTD-2336d9 153,409,619-153,523,782 Deleted RP11-635k22 153,436,850-153,605,891 Deleted CTD-2314a16 153,557,982-153,723,774 Deleted RP11-314p8 153,647,251-153,820,842 Deleted RP11-108c12 153,647,252-153,844,258 Deleted RP11-790o1 153,693,987-153,872,038 5q33 RP11-960p1 153,880,910-154,059,745 5q33 RP11-1149b7 154,371,494-154,551,424 5q33 * FISH results for BACs mapping to the 5q15 and 5q33 breakpoint regions script NIH-PA Author Manuscript NIH-PA Author Manuscript

https://openalex.org/W3018168732

https://europepmc.org/articles/pmc7191355?pdf=render

English

Current Scenario of Postcholecystectomy Bile Leak and Bile Duct Injury at a Tertiary Care Referral Centre of Nepal

Minimally invasive surgery

cc-by

Hindawi Minimally Invasive Surgery Volume 2020, Article ID 4382307, 6 pages https://doi.org/10.1155/2020/4382307 Hindawi Minimally Invasive Surgery Volume 2020, Article ID 4382307, 6 pages https://doi.org/10.1155/2020/4382307 Narendra Pandit ,1 Tek Narayan Yadav,1 Laligen Awale,1 Kunal Bikram Deo ,1 Yogesh Dhakal,2 and Shailesh Adhikary1 1Surgical Gastroenterology Division, Department of Surgery, B. P. Koirala Institute of Health Sciences (BPKIHS), Dharan, Nepal 2Department of Anaesthesiology and Critical Care, B. P. Koirala Institute of Health Sciences (BPKIHS), Dharan, Nepal Correspondence should be addressed to Narendra Pandit; [email protected] Received 28 December 2019; Revised 17 March 2020; Accepted 4 April 2020; Published 21 April 2020 Academic Editor: Peng Hui Wang Academic Editor: Peng Hui Wang Copyright © 2020 Narendra Pandit et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Objective. With the adoption of safe cholecystectomy principles at an academic institute, the risk of major bile duct injury has decreased. Tis study aims at evaluating the present status of bile duct injury, compared to the study published in 2013 by index centre. Methods. Tis is a retrospective review of a prospectively maintained database of bile leak and bile duct injury from 2014 to 2019. Patients who completed postcholecystectomy bile leak or bile duct injury treatment and were on regular follow-up were included. Results. Eighteen patients (0.78%) among 2,300 consecutive cholecystectomies presented with bile duct injury, including 8 (0.35%) major bile duct injuries and 10 (0.43%) bile leaks compared to major bile duct injury rate of 0.68% (92/11,345 cholecystectomies) between 2001 and 2010. Injuries were classified as Strasberg’s type A (52.9%), type D (5.9%), and type E (41.1%). Eight patients (47%) of bile leak were managed conservatively with drains, while two required laparotomy and lavage. Te mean time for spontaneous closure of bile leak was 11 days. Intraoperative repair was done in three cases: Roux en Y hep- aticojejunostomy in 2 and end-to-end repair over T-tube in 1 for sharp transection of the duct. Delayed repair (Roux-en-Y hepaticojejunostomy) was done in five patients. Te median postcholecystectomy hospital stay was 8 days, with no mortality. Tere was no restricture at a median follow-up of 13 months. Conclusion. With the adoption of a safe culture of cholecystectomy, the major bile duct injury rate has decreased currently. Repair of bile duct injury by experienced hepatobiliary surgeon results in excellent outcome. recent days [4, 6]. In 2013, the rate of bile duct injury, management, and its outcome (2001–2010) was studied from our institute [7]. Narendra Pandit ,1 Tek Narayan Yadav,1 Laligen Awale,1 Kunal Bikram Deo ,1 Yogesh Dhakal,2 and Shailesh Adhikary1 In the present study, we aimed at studying the present status of the rate of bile leak and (major) bile duct injury, its management, and outcome, a decade later at our centre. 1. Introduction Cholecystectomy is the most common operation performed worldwide. It is commonly performed by laparoscopic or by open method in case of difficult anatomy or pathology [1]. Both the procedures are associated with the risk of bile duct injury, with the risk being 0.4–0.6% and 0.2-0.3% for lap- aroscopic and open cholecystectomy, respectively [2]. Te bile duct injury significantly increases the healthcare cost, morbidity, and mortality and decreases survival [3, 4]. However, the data of risk stratification of bile duct injury are a decade old. Nowadays, with the adoption of principles of safe cholecystectomy and extravigilance at the academic institute, the risk of major bile duct injury has somehow decreased [5]. On the contrary, the incidence of bile leak following laparoscopic cholecystectomy has increased in 2. Materials and Methods Tis series represents a retrospective review of a prospec- tively maintained database of all patients who developed bile leak and bile duct injury from April 2014 to May 2019 at our tertiary referral centre. Te institute is a 750-bedded, aca- demic centre with a separate HPB unit. Inclusion criteria were patients who completed the treatment for bile leak and bile duct injury due to laparoscopic/open cholecystectomy 2 Minimally Invasive Surgery Figure 1: Magnetic resonance cholangiopancreatography (MRCP) showing Strasberg’s type E2 injury following open cholecystectomy. and were on regular follow-up. Tose patients who refused for intervention, surgery, incomplete treatment, lost to follow-up, and cholecystectomy combined with other ab- dominal procedures were excluded (n ˆ 4). Te study was approved by the Institutional Ethical Board. pp y Bile leak was defined as leak from the cystic duct stump or the aberrant bile ducts with maintained continuity of the extrahepatic duct and appearance of bile from the surgical or percutaneous drainage. It was later confirmed by normal ultrasound, liver function tests, or magnetic resonance cholangiography (MRCP). Similarly, (major) bile duct injury was defined as all transaction, segment loss, or stenosis of the extrahepatic bile duct or major segmental ducts requiring hepaticojejunostomy or end-to-end bile duct anastomosis, or undergoing more than 1 endoscopic retrograde chol- angiopancreatography (ERCP) within a year of cholecys- tectomy. Imaging investigations performed to diagnose the bile duct injury and leak were ultrasound, contrast com- puted tomography (CT), and MRCP depending on the presentation and severity of the injury. Figure 1: Magnetic resonance cholangiopancreatography (MRCP) showing Strasberg’s type E2 injury following open cholecystectomy. drains (one difficult cholecystectomy and the other following intraoperative closure of iatrogenic common hepatic duct rent) following laparoscopic cholecystectomy, which pre- sented with controlled external biliary fistula, low output (<200 ml), and ceased spontaneously within a week. Te remaining 6 patients presented with bilioma with sepsis, required image-guided percutaneous catheter drainage (PCD) (1 PCD-2, 2 PCD-1, and 3 PCD-1), antibiotics, source control, and achieving controlled external biliary fistula. Te leak subsided spontaneously at a mean time of 11 days (range: 4–34 days). None required endoscopic intervention (stenting/sphincterotomy), as the fistula volume was low- output, decreasing trend, improving the general condition of the patients and an unavailability of service at our centre at the time of writing of this paper. 2. Materials and Methods y j y Medical records were examined individually to extract data on demographics, type of cholecystectomy, indications for cholecystectomy, number of cases referred from other centre vs. injuries at index hospital, mode of presentation, timing of detection of injury, type of injury as per Strasberg’s classification [2], and type of intervention (conservative vs. surgery). Type of surgery, timing of repair, postoperative morbidity, mortality, total length of hospital stay, and fol- low-up were also recorded. Whenever possible, the data were presented as a standard tabular reporting format specific for biliary injury as proposed by Cho et al. [8]. y j y y Statistical analysis was performed with SPSS v 17.0 software for the descriptive statistical analysis by calculating mean, median, standard deviation, and percentage where appropriate. To see the trend of bile duct injury at our centre, compared to the study published in 2013, a Z-test for two sample proportions was used. A P value <0.05 was considered to be significant. g p p Ten (55.5%) among 18 patients required surgical in- tervention. Tere were seven Roux-en-Y-hep- aticojejunostomy (Hepp-Couinaud approach) by the experienced hepatobiliary surgeon, five delayed and two performed intraoperatively. One patient required end-to- end common bile duct repair over the T-tube (detected intraoperatively) for complete transection without segment loss during open cholecystectomy. Te remaining 2 patients required emergency laparotomy, peritoneal lavage, and drain placement for peritonitis due to the class A Strasberg injury (confirmed by postoperative MRCP) (Figure 2). Postoperatively, two patients developed superficial surgical site infections (SSIs). Tere was no mortality in our series of patients (Table 2). Te median length of hospital stay postcholecystectomy was 8 days (range: 5–28 days). Tere was no restricture at a median follow-up of 13 months (range: 8–36 months), as confirmed by history and clinical examination, liver function tests, and ultrasonography (Table 3). When the trend of injuries was compared, there was a significant decrease in incidence (0.21% vs. 0.68%; P ˆ 0.007) of major bile duct injury at index hospital (Table 4). 3. Results Eighteen patients presented with bile duct injury among 2,300 consecutive cholecystectomies (0.78%), including 8 (0.35%) major bile duct injuries and 10 (0.43%) bile leaks. After excluding five injuries that were referred from other centre, the current incidence of bile duct injury and major injury was 0.56% (13/2,300 cholecystectomies) and 0.21% (5/ 2,300 cholecystectomies) respectively. Tese injuries were seen in 11 females and 7 males, with a mean age of 40 years. Cholecystectomy was performed laparoscopically (4-port) in 15 (83.3%) and by open in 3 (16.7%) patients. Four (22.2%) injuries were detected intraoperatively, while the remaining 14 (77.8%) were detected in the postoperative period. As per Strasberg’s classification system, the injuries were classified as type A in 9(50%), type D in 1(5.5%), and type E in 8(44.5%) patients (Figure 1). Among E class, E1-1, E2-1, and E3 in 2 patients were seen. Five patients sustained type E injury at index hospital. Tere were no associated vascular injuries. Eight patients (44.4%) were managed conservatively (Tables 1 and 2). Among them, 2 had indwelling surgical 4. Discussion In the present study, the rate of bile leak and major duct injury was 0.43% and 0.35%, respectively, which is in line with the published international standard [5]. In a study published by Gupta et al. [7], from our in- stitute (study period 2001 to 2010) there was a high rate of Table 1: Demographics, details of index operation, presentation, and staging of injury. Parameters Total patients, n ˆ 18 Age (years), mean (range) 40 (16–55) Male : female (M : F) 7 :11 Place of injury Outside hospital 5 (27.8%) In hospital 13 (72.2%) Indication for cholecystectomy Biliary colic 9 (50%) Acute cholecystitis 4 (22.2%) Mucocele 4 (22.2%) Xanthogranulomatous cholecystitis 1 (5.5%) Surgery started as Open 3 (16.7%) Laparoscopic 15 (83.3%) Any conversion from laparoscopy to open 0 Methods of cholecystectomy Dissection of hepatocystic triangle first 13 (72.2%) Unknown 5 (27.8%) Methods of cystic duct identification Critical view of safety 11(61.1%) Infundibular approach 1(5.5%) Top-down 1(5.5%) Unknown 5(27.8%) Detection of injury Intraoperative 4 (22.2%) Postoperative 14 (77.8%) Injury occurred during which part of procedure Open 3 (16.7%) Laparoscopic 15 (83.3%) Management of recognized injury End-to-end anastomosis + T-tube 1(5.5%) Hepaticojejunostomy 2 (11.1%) Suture closure of rent in common hepatic duct 1 (5.5%) Indication for referral (n ˆ 15) Jaundice 5 (29.4%) Intra-abdominal sepsis 8(35.3%) Biliary fistula 2 (11.8%) Time from index operation to referral for surgical repair Intraoperative 3 (16.7%) 0–3 days 3(16.7%) 4–7 days 3(16.7%) 8 days–6 weeks 5 (27.8%) 6 weeks–3 months 4 (22.2%) Staging of injury A 9 (50%) D 1 (5.5%) E1 5(27.8%) E2 1 (5.5%) E3 2 (11.1%) Vasculobiliary injury 0 Other organs injured 0 able 1: Demographics, details of index operation, presentation, and staging of injury. leak has increased in recent decades (1.5% to 3%) [4, 11]. In the present study, the rate of bile leak and major duct injury was 0.43% and 0.35%, respectively, which is in line with the published international standard [5]. It significantly increases the morbidity, mortality, and costs of hospital stay and decreases the quality of life [5, 9]. Furthermore, it is associated with litigations. Despite in- creasing experience and familiarity, incidence of common bile duct injury following laparoscopic and open chole- cystectomy still continues to be 0.4% to 0.6% and 0.2% to 0.3%, respectively [2, 10]. On the contrary, the rate of bile In a study published by Gupta et al. 4. Discussion Bile leak and major bile duct injury are the most feared complications after open and laparoscopic cholecystectomy. Minimally Invasive Surgery 3 It significantly increases the morbidity, mortality, and costs of hospital stay and decreases the quality of life [5, 9]. Furthermore, it is associated with litigations. Despite in- leak has increased in recent decades (1.5% to 3%) [4, 11]. In the present study, the rate of bile leak and major duct injury was 0.43% and 0.35%, respectively, which is in line with the Table 1: Demographics, details of index operation, presentation, and staging of injury. Parameters Total patients, n ˆ 18 Age (years), mean (range) 40 (16–55) Male : female (M : F) 7 :11 Place of injury Outside hospital 5 (27.8%) In hospital 13 (72.2%) Indication for cholecystectomy Biliary colic 9 (50%) Acute cholecystitis 4 (22.2%) Mucocele 4 (22.2%) Xanthogranulomatous cholecystitis 1 (5.5%) Surgery started as Open 3 (16.7%) Laparoscopic 15 (83.3%) Any conversion from laparoscopy to open 0 Methods of cholecystectomy Dissection of hepatocystic triangle first 13 (72.2%) Unknown 5 (27.8%) Methods of cystic duct identification Critical view of safety 11(61.1%) Infundibular approach 1(5.5%) Top-down 1(5.5%) Unknown 5(27.8%) Detection of injury Intraoperative 4 (22.2%) Postoperative 14 (77.8%) Injury occurred during which part of procedure Open 3 (16.7%) Laparoscopic 15 (83.3%) Management of recognized injury End-to-end anastomosis + T-tube 1(5.5%) Hepaticojejunostomy 2 (11.1%) Suture closure of rent in common hepatic duct 1 (5.5%) Indication for referral (n ˆ 15) Jaundice 5 (29.4%) Intra-abdominal sepsis 8(35.3%) Biliary fistula 2 (11.8%) Time from index operation to referral for surgical repair Intraoperative 3 (16.7%) 0–3 days 3(16.7%) 4–7 days 3(16.7%) 8 days–6 weeks 5 (27.8%) 6 weeks–3 months 4 (22.2%) Staging of injury A 9 (50%) D 1 (5.5%) E1 5(27.8%) E2 1 (5.5%) E3 2 (11.1%) Vasculobiliary injury 0 Other organs injured 0 Minimally Invasive Surgery 3 It significantly increases the morbidity, mortality, and costs of hospital stay and decreases the quality of life [5, 9]. Furthermore, it is associated with litigations. Despite in- creasing experience and familiarity, incidence of common bile duct injury following laparoscopic and open chole- cystectomy still continues to be 0 4% to 0 6% and 0 2% to leak has increased in recent decades (1.5% to 3%) [4, 11]. Minimally Invasive Surgery Minimally Invasive Surgery Table 2: Preoperative risk assessment, laboratory values, intraoperative events, and outcomes of operative group. Parameters Results (n ˆ 10) Type II diabetes mellitus 2 (20%) Current smoker within 1 year 5 (50%) Hypertension requiring medication 3 (30%) Preoperative blood transfusions (red blood cells within 72 hr before surgery) 1 (10%) Sepsis within 48 hr before surgery 0 Cirrhosis 0 Mean hemoglobin (g/dl) 11.6 ± 1.8 Median total serum bilirubin, mg/dl (range) 2.0 (0.8–18.0) Albumin (g/dl) 3.9 ± 0.25 Timing of repair (time from index surgery) <24 hr 3 (30%) >7 days–<6 weeks 3 (30%) 8 to 12 weeks 4 (40%) Procedure done Hepaticojejunostomy 7 (70%) End-to-end anastomosis + T-tube 1 (10%) Laparotomy + lavage + drainage 2 (20%) End-to-side hepaticojejunostomy 2 (20%) Side-to-side hepaticojejunostomy 5 (50%) Any form of liver resection 0 Superficial SSI 2 (20%) Mortality 0 Table 2: Preoperative risk assessment, laboratory values, intraoperative events, and outcomes of operative group. Parameters Results (n ˆ 10) Type II diabetes mellitus 2 (20%) Current smoker within 1 year 5 (50%) Hypertension requiring medication 3 (30%) Preoperative blood transfusions (red blood cells within 72 hr before surgery) 1 (10%) Sepsis within 48 hr before surgery 0 Cirrhosis 0 Mean hemoglobin (g/dl) 11.6 ± 1.8 Median total serum bilirubin, mg/dl (range) 2.0 (0.8–18.0) Albumin (g/dl) 3.9 ± 0.25 Timing of repair (time from index surgery) <24 hr 3 (30%) >7 days–<6 weeks 3 (30%) 8 to 12 weeks 4 (40%) Procedure done Hepaticojejunostomy 7 (70%) End-to-end anastomosis + T-tube 1 (10%) Laparotomy + lavage + drainage 2 (20%) End-to-side hepaticojejunostomy 2 (20%) Side-to-side hepaticojejunostomy 5 (50%) Any form of liver resection 0 Superficial SSI 2 (20%) Mortality 0 hospital and mortality from it have decreased significantly. Tis can be attributed to the “culture of safe cholecystec- tomy,” by adopting critical view of safety, Rouviere’s sulcus as a landmark for initiation of dissection, bail-out options or early conversion in difficult cholecystectomy, time-out and in vicinity “colleaguography” before clipping, and extra- vigilance in an academic centre for the risk and conse- quences of bile duct injury [4, 12–14]. Figure 2: Follow-up magnetic resonance cholangiopancreatog- raphy (MRCP) showing normal extrahepatic bile duct with bil- ioenteric continuity, following sealed cystic duct stump leak (Strasberg’s type A injury). 4. Discussion [7], from our in- stitute (study period: 2001 to 2010), there was a high rate of major bile duct injury (0.68%). Te major bile duct injury Minimally Invasive Surgery 4 Minimally Invasive Surgery q j y [ ] Similarly, the rate of bile leak from the cystic duct stump or the aberrant/duct of Luschka (Strasberg class A injury) has increased by laparoscopic approach. Te bile leak is equally dangerous, if presented late with sepsis or the di- agnosis is delayed [15]. In a study by Viste et al. [1], the rate of bile leak was 0.9%, all attributed from the cystic duct or assumed ducts from liver bed, which compromised 52% of total bile duct injury. One out of four deaths was from peritonitis due to the leak from the cystic duct stump. In our study, nine patients had bile leak from the assumed cystic duct or duct of Luschka and one from common hepatic duct rent closure leak. Among them, two required laparotomy and lavage for peritonitis, while the remaining were man- aged conservatively with drainage of bilioma. All developed controlled external biliary fistula, which closed spontane- ously, with a mean fistula closure time of 11 days. None underwent ERCP, sphincterotomy, or stenting, which nowadays is the treatment modality of choice, as the facilities were not available at our centre [16]. Moreover, due to the logistic reason (financial constraints), lack of health insur- ance, and geographic status of the country, patient refused to move to other higher centre (700 km) for ERCP endoscopic intervention. Tere is no debate that early ERCP endoscopic intervention is safe, effective, and considered the first line of therapy in bile leak. It improves the clinical outcome, de- creases bilioma formation, decreases the rate of laparotomy Figure 2: Follow-up magnetic resonance cholangiopancreatog- raphy (MRCP) showing normal extrahepatic bile duct with bil- ioenteric continuity, following sealed cystic duct stump leak (Strasberg’s type A injury). Table 3: Results of bile duct reconstruction. Parameters Length of follow-up, median (months) 13 (8–36) Lost to follow-up 0 Any evidence of restricture or recurrent cholangitis 0 Any postoperative interventions for anastomotic problems 0 over the decade, where the laparoscopic approach was flourishing, was seen in 92 patients (among 11,345 chole- cystectomies), with 83 patients requiring bilioenteric anas- tomosis. Te mortality rate from the injury was 3.3%. Certainly, a decade later at the same institute, the rate of major bile duct injury (0.21% vs. 0.68%; P ˆ 0.007) at index Minimally Invasive Surgery 5 Table 4: Comparison of the present study with the previous study from our centre. Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the Institutional Research Committee (IRC-BPKIHS) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Consent Informed consent was obtained from all individual partic- ipants included in the study. Conflicts of Interest Te authors declare that they have no conflicts of interest. Te study is limited by its retrospective design, short- time frame, small sample size, lack of state-of-the-art ERCP for the management of bile leak at our centre, and lack of long-term follow-up to detect restricture. Despite this, the study beautifully shows the decreased rates of major bile duct injury at our centre, with good short-term outcomes. Minimally Invasive Surgery Table 4: Comparison of the present study with the previous study from our centre. Parameters Current study (2014–2019) Previous study (2001–2010) Gupta et al. [7] Total cholecystectomies (in-hospital) 2,300 11,345 Mean age, years, (range) 40 (16–55) 46.5 (23–68) Referred 5 (27.8%) 15 (16.3%) Index hospital 13 (72.2%) 77 (83.7%) Overall incidence Major bile duct injury 8 (0.35%) In-hospital major bile duct injury 5 (0.21%) 77 (0.68%) (Z ˆ 2.659 and P ˆ 0.007) Bile leak 10 (0.43%) NA Morbidity 2 (11.1%) NA Mortality (30 and 90 days) 0 3 (3.3%) Follow-up (n ˆ 18) (n ˆ 75) Median follow-up (mo) 13 31 Restricture 0 3 (4%) Good outcome 18 (100%) 62 (82%) Previous study (2001–2010) and experienced hepatobiliary surgeon results in excellent outcome. and number of percutaneous interventions, removes the missed common bile duct stones as a cause for leak, and decreases time to fistula closure and even deaths [15, 16]. In the present study, fortunately, other than the increased number of percutaneous interventions and increased time to fistula closure, there were no increased laparotomy rates or death due to the conservative management of bile leak. References [1] A. Viste, A. Horn, K. Øvrebø, B. Christensen, J.-H. Angelsen, and D. Hoem, “Bile duct injuries following laparoscopic cholecystectomy,” Scandinavian Journal of Surgery, vol. 104, no. 4, pp. 233–237, 2015. [1] A. Viste, A. Horn, K. Øvrebø, B. Christensen, J.-H. Angelsen, and D. Hoem, “Bile duct injuries following laparoscopic cholecystectomy,” Scandinavian Journal of Surgery, vol. 104, no. 4, pp. 233–237, 2015. [2] J. T. Cohen, K. P. Charpentier, and R. E. Beard, “An update on iatrogenic biliary injuries: identification, classification, and management,” Surgical Clinics of North America, vol. 99, no. 2, pp. 283–299, 2019. Data Availability Te data used to support the findings of the study are available from the corresponding author upon request. Te important finding of this study is the excellent outcome of patients undergoing repair for major bile duct injury. It has been well described for major bile duct injury that the first repair should be the best repair, not by the primary surgeon, but by the expert hepatobiliary surgeon at the experienced centre [2, 4]. We had a policy of performing delayed repair (>6 weeks), with control of intra-abdominal sepsis and nutritional stabilization of the patients, or an on- table repair if diagnosed intraoperatively. Bilioenteric drainage (Roux-en-Y hepaticojejunostomy in 7 patients) is the preferred treatment option for major duct injury; however, end-to-end common bile duct repair over the T-tube (1 patient) is also the safe option for sharp transaction injuries without segment loss and a nonelectrocautery injury [2, 3]. Tere was no restricture in those eight patients who required on-table and delayed repair at a median follow-up of 13 months. 5. Conclusion Bile leak and major duct injury are the most feared com- plications of cholecystectomy and results in significant morbidity, mortality, and healthcare costs. With the adoption of a safe culture of cholecystectomy, the rate of major bile duct injury has decreased compared to the results of a decade back at our centre. Similarly, the bile leak from the cystic duct/duct of Luschka has been increasingly de- tected. Repair of major bile duct injury by the nonprimary [3] S. Ray, S. Sanyal, S. Das, K. Jana, A. K. Das, and S. Khamrui, “Outcomes of surgery for post-cholecystectomy bile duct injuries: an audit from a tertiary referral center,” Journal of Visceral Surgery, vol. 157, no. 1, pp. 3–11, 2019. [4] B. W. Renz, F. B¨osch, and M. K. Angele, “Bile duct injury after cholecystectomy: surgical therapy,” Visceral Medicine, vol. 33, no. 3, pp. 184–190, 2017. 6 Minimally Invasive Surgery [5] Z. V. Fong, H. A. Pitt, S. M. Strasberg et al., “Diminished survival in patients with bile leak and ductal injury: man- agement strategy and outcomes,” Journal of the American College of Surgeons, vol. 226, no. 4, pp. 568–576, 2018. [6] J. H. Kim, W. H. Kim, J. H. Kim, B. M. Yoo, and M. W. Kim, “Management of patients who return to the hospital with a bile leak after laparoscopic cholecystectomy,” Journal of Laparoendoscopic & Advanced Surgical Techniques, vol. 20, no. 4, pp. 317–322, 2010. [7] R. K. Gupta, C. S. Agrawal, S. Sah, S. Sapkota, O. P. Pathania, and P. L. Sah, “Bile duct injuries during open and laparoscopic cholecystectomy: management and outcome,” Journal of Nepal Health Research Council, vol. 11, no. 11, pp. 187–193, 2013. [8] J. Y. Cho, A. R. Jaeger, D. E. Sanford, R. C. Fields, and S. M. Strasberg, “Proposal for standardized tabular reporting of observational surgical studies illustrated in a study on primary repair of bile duct injuries,” Journal of the American College of Surgeons, vol. 221, no. 3, pp. 678–688, 2015. [9] D. Martin, E. Uldry, N. Demartines, and N. Halkic, “Bile duct injuries after laparoscopic cholecystectomy: 11-year experi- ence in a tertiary center,” BioScience Trends, vol. 10, no. 3, pp. 197–201, 2016. [10] S. S. Sikora, A. Kumar, N. R. Das, A. Sarkari, R. Saxena, and V. K. Kapoor, “Laparoscopic bile duct injuries: spectrum at a tertiary-care center,” Journal of Laparoendoscopic & Advanced Surgical Techniques, vol. 11, no. 2, pp. 63–68, 2001. 5. Conclusion [11] G. Tzovaras, P. Peyser, L. Kow, T. Wilson, R. Padbury, and J. Toouli, “Minimally invasive management of bile leak after laparoscopic cholecystectomy,” HPB, vol. 3, no. 2, pp. 165– 168, 2001. [12] S. M. Strasberg, “A perspective on the critical view of safety in laparoscopic cholecystectomy,” Annals of Laparoscopic and Endoscopic Surgery, vol. 2, no. 5, p. 91, 2017. [13] V. K. Kapoor, “Colleaguography” in place of cholangiography, to prevent bile duct injury during laparoscopic cholecystec- tomy,” Journal of Minimal Access Surgery, vol.15, no. 3, p. 273, 2019. [14] S. M. Strasberg, “A three-step conceptual roadmap for avoiding bile duct injury in laparoscopic cholecystectomy: an invited perspective review,” Journal of Hepato-Biliary-Pan- creatic Sciences, vol. 26, no. 4, pp. 123–127, 2019. [15] F. Ahmad, R. Saunders, G. Lloyd, D. Lloyd, G. Robertson, and G. S. M. Robertson, “An algorithm for the management of bile leak following laparoscopic cholecystectomy,” Te Annals of Te Royal College of Surgeons of England, vol. 89, no. 1, pp. 51–56, 2007. [16] A. Abbas, S. Sethi, P. Brady, and P. Taunk, “Endoscopic management of postcholecystectomy biliary leak: when and how? A nationwide study,” Gastrointestinal Endoscopy, vol. 90, no. 2, pp. 233–241, 2019.

https://openalex.org/W4312220673

https://dergipark.org.tr/tr/download/article-file/2857374

Turkish

Evaluation of the Relationship Between Pain Beliefs and Socio-Demographic Characteristics in Older Individuals

Yaşlı sorunları araştırma dergisi

cc-by

1 Çalışma “Yaşlılarda Ağrı İnançlarının Sosyo-Demografik Özelliklerle İlişkisinin Değerlendirilmesi” başlıklı Lisans tezinden üretilmiştir. 2 Sorumlu yazar e-mail: [email protected] Makalenin Kaynak Olarak Gösterimi (Cite as): Korkmaz Yaylagül, N., & İmancıoğlu, H. (2022). Yaşlılarda ağrı inançlarının sosyo- demografik özelliklerle ilişkisinin değerlendirilmesi. Yaşlı Sorunları Araştırma Dergisi (YSAD) / Elderly Issues Research Journal (EIRJ), 15(2), 74-82. doi:10.46414/yasad.1225392 YAŞLI SORUNLARI ARAŞTIRMA Elderly Issues Research Journal 15(2), 74-82, (2022) ISSN 1308-5816 http://dergipark.gov.tr/yasad ARAŞTIRMA MAKALE Yaşlılarda Ağrı İnançlarının Sosyo-Demografik Özelliklerle İlişkisinin Değerlendirilmesi1 Nilüfer KORKMAZ YAYLAGÜL2* & Halil İMANCIOĞLU** *Akdeniz Üniversitesi, Sağlık Bilimleri Fakültesi, Gerontoloji Bölümü, Antalya / Türkiye ORCID NO: https://orcid.org/0000-0001-9918-7968 **Akdeniz Üniversitesi, Sağlık Bilimleri Enstitüsü, Antalya / Türkiye ORCID NO: https://orcid.org/0000-0002-3979-5854 Nilüfer KORKMAZ YAYLAGÜL2* & Halil İMANCIOĞLU** *Akdeniz Üniversitesi, Sağlık Bilimleri Fakültesi, Gerontoloji Bölümü, Antalya / Türkiye ORCID NO: https://orcid.org/0000-0001-9918-7968 **Akdeniz Üniversitesi, Sağlık Bilimleri Enstitüsü, Antalya / Türkiye ORCID NO: https://orcid.org/0000-0002-3979-5854 ANAHTAR KELİMELER Yaşlılık, ağrı, ağrı inançları, organik inançlar, psikolojik inançlar ABSTRACT Pain is one of the important problems of old age and is a factor affecting the quality of life. How an individual experiences pain is affected by demographic variables such as income, education, age and gender, as well as many factors such as the individual's emotional state and pain experience. The aim of the study, which was conducted with 100 people who are 65 and over, living in Tavas, Denizli, is to reveal their beliefs about pain. The data obtained with the demographic questionnaire and the pain beliefs scale. The obtained data were analyzed with the PASW Statistics program according to the measurement levels of the variables. The pain beliefs, organic beliefs and psychological beliefs of the participants were found to be 53.59, 33.65 and 19.94, respectively. A statistically significant difference was found between the gender variable and the "Organic Beliefs" sub-dimension (p<0.05). It is seen that the "Organic Beliefs" sub-dimension average score of female participants (X=34.60) is higher than the average of "Organic Beliefs" sub-dimension score of male participants (X=32.43). A statistically significant difference was found between the gender variable and the “Pain Beliefs Scale” (p<0.05). It is seen that the mean score of “Pain Beliefs Scale” of female 1 Çalışma “Yaşlılarda Ağrı İnançlarının Sosyo-Demografik Özelliklerle İlişkisinin Değerlendirilmesi” başlıklı Lisans tezinden üretilmiştir. 2 S l il k k il f @h t il ÖZET MAKALE GEÇMİŞİ Geliş 02 Mart 2022 Kabul 27 Aralık 2022 Ağrı, yaşlılığın önemli sorunlarından biri olup yaşam kalitesini etkileyen bir faktördür. Ağrıyı bireyin nasıl deneyimlediği gelir, eğitim, yaş ve cinsiyet gibi demografik değişkenlerin yanı sıra bireyin duygusal durumu ve ağrı deneyimi gibi birçok faktörden etkilenmektedir. Denizli ili Tavas ilçesinde yaşayan 65 yaş ve üstü 100 kişi ile yapılan araştırmada ağrıya ilişkin inançlarının ortaya konması amaçlanmıştır. Demografik bilgi anketi ve ağrı inançları ölçeği ile elde edilen veriler, PASW Statistics programında değişkenlerin ölçüm düzeyine uygun olarak analiz edilmiştir. Katılımcıların ağrı inançları, organik inançları ve psikolojik inanç ortalamaları sırasıyla 53.59, 33.65, 19,94 olarak bulunmuştur. Cinsiyet değişkeni ile “Organik İnançlar” alt boyutu arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Kadın katılımcıların (X=34,60) “Organik İnançlar” alt boyut puan ortalaması erkek katılımcıların (X=32,43) “Organik İnançlar” alt boyut puan ortalamasından yüksek olduğu görülmektedir. Cinsiyet değişkeni ile “Ağrı İnançları Ölçeği” arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Kadın katılımcıların (X=55,12) “Ağrı İnançları Ölçeği” puan ortalaması erkek katılımcıların (X=51,63) “Ağrı İnançları Ölçeği” puan ortalamasından yüksek olduğu görülmektedir. Yaşamı boyunca ağrı yaşama sıklığı ile “Ağrı İnançları Ölçeği” ve “Psikolojik İnançlar” alt boyutu arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Yaşamı boyunca her zaman ağrı yaşadığını bildiren katılımcıların (X=20,43) “Psikolojik İnançlar” alt boyut puan ortalaması, yaşamı boyunca nadiren ağrı yaşadığını bildiren katılımcıların (X=18,53) “Psikolojik İnançlar” alt boyutu puan ortalamasından yüksek olduğu görülmektedir. Yaşamı boyunca her zaman ağrı yaşadığını bildiren katılımcıların (X=54,48) “Ağrı İnançları Ölçeği” puan ortalamasının, yaşamı boyunca nadiren ağrı yaşadığını bildiren katılımcıların (X=51,03) “Ağrı İnançları Ölçeği” puan ortalamasından yüksek olduğu ortaya çıkmıştır. Yapılan bu araştırma genellenebilir nitelikte değildir, ancak yaşlılıkta ağrı konusunda yapılmış sınırlı sayıdaki çalışma nedeniyle literatüre katkı sağlamaktadır. Yapılacak olan çalışmalarda kapsamlı tarama türündeki araştırmaların yanı sıra konunun derinlemesine incelenmesini sağlayacak yorumlayıcı nitel araştırmaların da yapılması önerilmektedir. ANAHTAR KELİMELER Yaşlılık, ağrı, ağrı inançları, organik inançlar, psikolojik inançlar 1.GİRİŞ Uluslararası Ağrı Araştırmaları Derneği (International Association for the Study of Pain-IASP)’nin kabul gören tanımına göre ağrı “gerçek veya olası doku hasarıyla ilişkili veya buna benzer hoş olmayan bir duyusal ve duygusal deneyim”dir (IASP International Association for the Study of Pain, Erişim Tarihi: 04.10.22). Ağrı; yaşamı tehdit eden, bireyin yerine getirmesi gereken günlük yaşam aktivitelerini olumsuz etkileyerek yaşam kalitesini bozabilen bir durumdur. Giderilemeyen ağrı, bireylerin fizyolojik ve psikolojik sağlık ve iyilik halini olumsuz yönde etkilemektedir. Doku hasarına bağlı olarak fiziksel bir tepki şeklinde ortaya çıkan ağrının gerçekte genetik, duygusal, kültürel ve bireysel faktörler ile ilgili farklı boyutları içinde barındırdığı ve bu nedenle benzer fizyolojik etkenler ile oluşmuş olsa bile, ağrı deneyimi ve şiddetinde bireysel farklılıkların gözlendiği ifade edilmektedir (Koçoğlu & Özdemir, 2011). 2. YÖNTEM Araştırma betimleyici araştırma niteliği taşımaktadır. Betimleyici araştırma, araştırma konusu üzerine genel bir görüş elde etmek ve konuyu sistematik olarak betimlemek amacıyla yapılır (Aydoğdu, Karamustafaoğlu & Bülbül, 2017). Bu yüzden yaşlılarda ağrı inancı ile sosyo-demografik özelliklerin ilişkisini değerlendirmek için betimleyici nicel araştırma deseni benimsenmiştir ve veri toplama tekniği olarak anket tekniğinden yararlanılmıştır. Katılımcılara Ağrı İnançları Ölçeği ve Sosyo- demografik bilgi formu uygulanmıştır. Çalışma Denizli’nin Tavas ilçesinde gerçekleştirilmiştir. Anket görüşmeleri 28.01.2018-15.03.2018 tarihleri arasında ilçe merkezindeki parklarda, bahçelerde katılımcıların kendi evlerinde yüz yüze yapılmıştır. Ağrı, yaşlılıkta karşılaşılan en önemli sağlık sorunlarından biridir. Ortaya çıkan ağrı, yaşlı bireylerin fiziksel fonksiyonlarını, günlük yaşam aktivitelerini, yaşam kalitesini ve iyilik durumunu olumsuz yönde etkilemekte, çoklu ilaç kullanımı, kognitif azalma, yürüme anomalileri ve kazalara sebep olabilmektedir (Çilingir & Bulut, 2017). Türkiye’de yaşlılıkta ağrı üzerine çeşitli çalışmalar bulunmaktadır. Bu çalışmalar doğrultusunda ağrı ile ilgili çalışmalar ağırlıklı olarak ağrının yeri (Şimşek, Yumin, Öztürk, Sertel, & Yumin, 2011), ağrı ile başa çıkma yöntemleri (Özel, Yıldırım, & Fadıoğlu, 2014; Karslı & Akdeniz, 2011), yaşlının ağrıya yaklaşımı (Çilingir & Bulut, 2017; Öztürk & Karan, 2008), ağrı ve yaşam kalitesi (Tavşanlı, Özçelik, & Karadakovan, 2013), ağrı ve depresyon (Yıldız, Erol & Ergün, 2009; Saltan, 2017), biliş düzeyi, ağrı inançları ve vücut farkındalığı (Tıglı, Günebakan, Toslalı, & Aytar, 2019), ağrı inançları ve depresyon düzeylerini etkileyen faktörler (Demirci & Özer, 2020) konuları üzerinde yoğunlaşmaktadır. Yapılan çalışmalarda yaşlı bireylerin daha az ağrı bildiriminde bulundukları ortaya konmuştur. Yaşlıların diğer yaş gruplarına göre daha az bildirimde bulunmaları, yaşlılarda ağrı değerlendirmesini ve kontrol altına alınmasını zorlaştırmaktadır (Tavşanlı, Özçelik & Karadakovan, 2013). Yaşlı bireylerin maruz kaldığı ağrıların giderilmesi, risk grubunda bulunan bireylerin saptanması, gereken desteğin sunulması gerek aile ARTICLE HISTORY ARTICLE HISTORY Received 02 March 2022 Accepted 27 December 2022 KEYWORDS Old Age, pain, pain beliefs, organic beliefs, psychological beliefs Received 02 March 2022 Accepted 27 December 2022 Old Age, pain, pain beliefs, organic beliefs, psychological beliefs “Yaşlılarda Ağrı İnançlarının Sosyo-Demografik Özelliklerle İlişkisinin Değerlendirilmesi” başlıklı Lisans tezinden 75 75 Yaşlılarda ağrı inançlarının sosyo-demografik özelliklerle ilişkisi participants (X=55.12) is higher than the mean score of “Pain Beliefs Scale” of male participants (X=51.63). A statistically significant difference was found between the frequency of experiencing pain throughout life and the “Pain Beliefs Scale” and “Psychological Beliefs” sub-dimensions (p<0.05). It is seen that the "Psychological Beliefs" sub-dimension score average of the participants who reported that they always experienced pain throughout their life (X=20.43) was higher than the "Psychological Beliefs" sub-dimension average score of the participants who reported that they had rarely experienced pain throughout their life (X=18.53). It was found that the "Pain Beliefs Scale" mean score of the participants who reported that they always experienced pain throughout their life (X=54.48) was higher than the "Pain Beliefs Scale" mean score of the participants who reported that they had rarely experienced pain throughout their life (X=51.03). This research is not generalizable, but it contributes to the literature due to the limited number of studies on pain in old age. In future studies, it is recommended to carry out interpretive qualitative studies that will provide an in-depth examination of the subject, as well as comprehensive survey research. gerekse toplumsal farkındalığın geliştirilmesinin yaşlı bireylerin yaşam kalitesini arttıracağı düşünülmek- tedir (Demirci & Özer, 2020). Bu doğrultuda ileri yaşta yaşam kalitesini etkileyen önemli bir sorun olan ağrıya yönelik müdahalelerde, bireylerin ağrıya yönelik inançlarını etkileyen faktörlerin bilinmesi gereklidir. Buna karşılık literatürde ağrıya yönelik çalışmalarda spesifik olarak ağrı inancını etkileyen sosyo- demografik özellikleri ortaya koyan bir çalışmaya rastlanmamıştır. Bu bağlamda bu araştırmada yaşlılarda ağrı inancını etkileyen sosyo-demografik faktörlerin ortaya konması hedeflenmektedir. 2.2. Veri Toplama Aracı 2.2. Veri Toplama Aracı Veriler katılımcılardan; cinsiyet, yaş, eğitim durumu, gelir düzeyi, kronik ağrı gibi değişkenleri içeren sosyo- demografik soru formu ve 1992 yılında Edwards ve arkadaşları tarafından geliştirilen Organik İnançlar alt testi (8 madde) ve Psikolojik İnançlar alt testi (4 madde) olmak üzere iki alt testten oluşan Ağrı İnançları Ölçeği (AİÖ) (The Pain Beliefs Questionnaire- PBQ) kullanılarak toplanmıştır (Edwards, Pearce, Stokes, & Jones, 1992). Ağrı İnançları Ölçeğinin orijinal formu, Edwards ve arkadaşları (1992) tarafından ağrının kaynakları ve tedavi şekli ile ilgili inançları anlamak amacıyla geliştirilmiştir. Toplam 20 maddeden oluşan ölçek ilk olarak, 100 kronik ağrı hastası ve ağrısız kontrol grubunda yer alan 194 üniversite öğrenicisi olmak üzere toplam 294 kişi üzerinde uygulanmıştır, sonuç olarak yapılan faktör analizinde toplam değişimin %68,15’ini açıklayan 2 faktör ortaya çıkmıştır. Buna göre, 8 maddenin yer aldığı Organik İnançlar alt testi ve 4 maddenin yer aldığı Psikolojik İnançlar alt testi oluşturulmuştur (Berk, 2006). Araştırma sorusundan yola çıkılarak ağrı inancı değişkeni ve eğitim düzeyi, yaş, cinsiyet, ortalama gelir, medeni durum, kronik ağrı ve ağrı yaşama sıklığı değişkenlerine araştırma tasarımında yer verilmiştir. Yaşlılık: “yaşam sürecinde gelişme ve olgunlaşmanın ardından kendine özgü fizyolojik ve psikolojik değişimlerin ortaya çıktığı son evre” olarak görülür (Arslan, 2015, s. 67). Yaş değişkeni, WHO’nun yaş sınıflaması referans alınarak geç yaşlılık (65-74), ileri yaşlılık (75-84) ve çok ileri yaşlılık (85+) şeklinde kategorize edilmiştir (WHO, 2007). Ölçeğin uygulanacağı katılımcılardan, 1. “hiçbir zaman” ile 6. “her zaman” arasında yer alan 6 seçenekten oluşan ifadelerden kendileri için uygun olanını işaretlemeleri istenmektedir. Her bir madde için 1 ile 6 arasında puanlama uygulanmaktadır. Her alt test için toplam puan, o alt teste ait maddelerden elde edilen toplam puanın o alt teste ait madde sayısına bölünmesiyle hesaplanmaktadır. Ölçek puanları için bir kesme noktası bulunmamaktadır ve ölçeğin alt puanından alınan puanın artması o teste ilişkin ağrı inançlarının yüksek olduğunu, puanın azalması ise o teste ilişkin ağrı inançlarının yüksek olduğunu göstermektedir. Ölçeğin güvenirlik çalışmasında iç tutarlılık skorları Cronbach Alpha katsayısı ile değerlendirilmiş ve Organik İnançlar alt testi için bu katsayı 0.71 ve Psikolojik İnançlar alt testi için 0.73 hesaplanmıştır (Berk, 2006, s.50). Ağrı İnançları Ölçeğinin Türkçe geçerlik ve güvenirlik çalışması 2006 yılında HÖ Sertel Berk tarafından yapılmıştır. Gelir değişkeni; düşük, orta ve yüksek gelir şeklinde, gelir değişkeninin mod ve medyanı esas alınarak üç grup halinde kategorize edilmiştir. Araştırmanın temel hipotezi; 65 yaş üstü bireylerin ağrı inançları puan ortalaması demografik faktörlere göre farklılık göstermektedir. 3. BULGULAR Yapılan araştırma kapsamında katılımcıların %56’sı kadın, %44’ü erkek, medeni durumlarına bakıldığında %81’i evli bireylerden oluşmaktadır. Araştırmaya katılanların en fazla %73 oranla genç yaşlılık, ardından %23 oranla ileri yaşlılık ve %4 oranla çok ileri yaşlılık dönemlerinde oldukları belirlenmiştir. Katılımcılar %58 oranla düşük gelir kategorisinde %15 oranla orta gelir kategorisinde ve %27 oranla yüksek gelir kategorisinde oldukları belirlenmiştir. Eğitim durumlarına bakıldığında %26’sı okuryazar değil, %58’i ilkokul, %7’si ortaokul, %6’sı lise ve %3’ünün üniversite mezunu olduğu belirlenmiştir. Katılımcıların yaş ortalaması 71,9’dur. Yaşı en büyük olan katılımcı 90 yaşındayken, en genç katılımcının yaşı 65’dir. Ağırlıklı olarak %11 oranında 68 yaşında katılımcılar bulunmaktadır. Araştırmada katılımcıların gelirlerinin ortalaması 1141,80 TL’dir. En çok tekrar eden gelir 1000 TL olup, en düşük 0 TL iken en yüksek gelir 2000 TL olmuştur. Katılımcıların ağrıya ilişkin bazı özellikleri tanısı konmuş kronik ağrı olma durumu %70 oranında olup, kronik ağrı olmama durumu %30’dur. Yaşamı boyunca ağrı yaşama sıklığı sorusuna %70 oranla “her zaman” cevabı verildiği tespit edilmiştir. Son bir yılda ağrı yaşama sıklığı ele alındığında %88 oranla “evet” cevabı, %12 oranla “hayır” cevabı verildiği tespit edilmiştir (Tablo 1). 2.3. Etik Boyut Araştırma için Akdeniz Üniversitesi Klinik Araştırmalar Etik Kurulu’ndan (55578142-050.01.04- E.37959) onay alınmıştır. Araştırmaya katılan bireyler gönüllülük esasına göre belirlenmiştir. Katılmayı kabul edenlere araştırmanın amacı ve önemi hakkında bilgi verilmiştir. Araştırmada her katılımcıdan aydınlatılmış onam alınmıştır. Elde edilen veriler anonimliğe dikkat edilerek kullanılmıştır. Katılımcılardan kendi istekleri olmadığı sürece isim, iletişim adresi vs. bilgileri alınmamıştır. 2.1. Çalışma Grubu Araştırmanın örneklemi, katılımcılara kolay erişilebilmesi açısından araştırmacılardan birinin ikamet ettiği Denizli ili Tavas ilçesinde ikamet eden 65 yaş ve üstü bireylerden seçilmiş, zaman ve maliyet kısıtlılığı nedeniyle amaçlı örneklem yöntemiyle örneklem büyüklüğü 100 kişi olarak kararlaştırıl- mıştır. Araştırmada dahil edilme kriterleri, katılımcının 65 yaş ve üzerinde olması, veri toplama araçlarını cevaplayabilecek bilişsel yeterliliği olması, iletişim probleminin olmaması (işitme, dil, anlama vb.) ve çalışmaya katılmaya gönüllü olmasıdır. Araştırmadan dışlama kriterleri ise, bireylerin ankete katılmaktan vazgeçmesi ya da görüşme sırasında görüşmeyi sonlandırmasıdır. YSAD-EIRJ 2022 / 15(2), 74-82 Korkmaz Yaylagül,İmancıoğlu 76 yapana göre farklı anlam taşıdığı durumlarda kullanılan araştırmanın bakış açısına göre tutarlı bir değerlendirme yapabilmek için ölçekte bulunan ifadeler i: 6 puan, ii: 5puan, iii: 4 puan, iv: 3 puan, v: 2 puan, vi: 1 puan şeklinde puanlandırılmıştır (Seçer, 2018). Araştırmanın bağımlı değişkeni, ağrı inançları puan ortalaması, organik inançlar alt test ortalaması ve psikolojik inançlar alt test ortalaması; bağımsız değişkenler ise; eğitim düzeyi, yaş, cinsiyet, ortalama gelir, medeni durum, kronik ağrı ve ağrı yaşama sıklığıdır. Araştırmaya katılan katılımcıların sosyo- demografik özelliklerini incelemek üzere parametrik testlerden ikili bağımsız değişkenlerde analiz için t- testi ve ikiden fazla bağımsız değişkenlerde ANOVA testi kullanılmıştır. 2.4. İşlem 65 yaş ve üstü 100 katılımcıyla yapılan bu araştırmanın anket sonuçları incelenerek PASW Statistics 22 programına tek tek girilerek veriler elektronik ortama aktarılmıştır. Verilerin elektronik ortama aktarılması sürecinde, ölçülen değere ait bir özelliğin ölçüm YSAD-EIRJ 2022 / 15(2), 74-82 77 Yaşlılarda ağrı inançlarının sosyo-demografik özelliklerle ilişkisi 77 Tablo 1. Yaşlı bireylerin sosyo demografik özellikleri ve ağrıya ilişkin bilgileri (n=100) Değişken Frekans Yüzde Yaş Genç yaşlılık İleri yaşlılık Çok ileri yaşlılık 73 73 23 23 4 4 Cinsiyet Kadın Erkek 56 56 44 44 Medeni Durum Evli Evli olmayan 81 81 19 19 Eğitim Durumu Okur-Yazar Değil İlköğretim Ortaöğretim Lise Üniversite 26 26 58 58 7 7 6 6 3 3 Gelir Dağılımı Düşük gelir Orta gelir Yüksek gelir 58 58 15 15 27 27 Tanısı konmuş herhangi bir uzun süreli (kronik) ağrınız var mı? Var Yok 70 70 30 30 Yaşamınız boyunca ağrı yaşama sıklığı Nadiren Her zaman 26 26 74 74 Son bir yıl içinde ağrı yaşama durumu Evet Hayır 88 88 12 12 Tablo 1. Yaşlı bireylerin sosyo demografik özellikleri ve ağrıya ilişkin bilgileri (n=100) olarak anlamlı bir fark bulunamamıştır (p>0,05). Gelir değişkeni ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır (p>0,05). Tanısı konmuş herhangi bir uzun süreli (kronik) ağrısı olma durumu ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır (p>0,05). Son bir yıl içinde ağrı yaşama durumu ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır (p>0,05). Katılımcıların “Ağrı İnançları Ölçeği” puan ortalamasının 53,59± 6,79; “Organik İnançlar” alt boyut puan ortalamasının 33,65±4,87; “Psikolojik İnançlar” alt boyut puan ortalamasının 19,94±3,51 olduğu belirlenmiştir (Tablo 2). Tablo 2. Yaşlı bireylerin ağrı inançları ölçeği ve alt boyutları puan ortalamalarının dağılımı (n=100) Tablo 2. Yaşlı bireylerin ağrı inançları ölçeği ve alt boyutları puan ortalamalarının dağılımı (n=100) Ort±ss Min-Max (Alınabilecek puanlar) Ağrı inançları ölçeği 53,59±6,79 36-66 (12-72) Organik İnançlar 33,65±4,87 23-44 (8-48) Psikolojik İnançlar 19,94±3,51 10-24 (4-24) Buna karşılık cinsiyet değişkeni ile “Organik İnançlar” alt boyutu arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Kadın katılımcıların (X=34,60) “Organik İnançlar” alt boyut puan ortalaması erkek katılımcıların (X=32,43) “Organik İnançlar” alt boyut puan ortalamasından yüksek olduğu görülmektedir. Cinsiyet değişkeni ile “Ağrı İnançları Ölçeği” arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Kadın katılımcıların (X=55,12) “Ağrı İnançları Ölçeği” puan ortalaması erkek katılımcıların (X=51,63) “Ağrı İnançları Ölçeği” puan ortalamasından yüksek olduğu görülmektedir. 2.4. İşlem Yaşamı boyunca ağrı yaşama sıklığı ile “Ağrı İnançları Ölçeği” ve “Psikolojik İnançlar” alt boyutu arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Yaşamı boyunca her zaman ağrı yaşadığını bildiren katılımcıların (X=20,43) “Psikolojik İnançlar” Araştırmada yaş değişkeni ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır (p>0,05). Medeni durum değişkeni ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır (p>0,05). Eğitim değişkeni ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel YSAD-EIRJ 2022 / 15(2), 74-82 Korkmaz Yaylagül,İmancıoğlu 78 (X=54,48) “Ağrı İnançları Ölçeği” puan ortalamasının, yaşamı boyunca nadiren ağrı yaşadığını bildiren katılımcıların (X=51,03) “Ağrı İnançları Ölçeği” puan ortalamasından yüksek olduğu görülmektedir. alt boyut puan ortalamasının, yaşamı boyunca nadiren ağrı yaşadığını bildiren katılımcıların (X=18,53) “Psikolojik İnançlar” alt boyutu puan ortalamasından yüksek olduğu görülmektedir. Yaşamı boyunca her zaman ağrı yaşadığını bildiren katılımcıların Tablo 3. Demografik bilgilere göre ağrı inançları ölçeği ve alt boyutları puan dağılımları (n=100) Değişken Organik İnançlar Psikolojik İnançlar Ağrı İnançları Ort±Ss İstatiksel analiz Ort±Ss İstatiksel analiz Ort±Ss İstatiksel analiz Yaş Genç yaşlılık İleri yaşlılık Çok ileri yaşlılık 33,42±4,76 p=,356 20,15±3,41 p=,264 53,57±6,30 p=,301 34,73±5,30 19,73±3,96 54,47±8,36 31,50±3,87 17,25±1,25 48,75±4,57 Cinsiyet Kadın Erkek 34,60±4,89 p=,026* 20,51±3,30 p=,063 55,12±6,37 p=,010** 32,43±4,61 19,20±3,66 51,63±6,88 Medeni Durum Evli Evli olmayan 33,65±4,70 p=,986 19,87±3,44 p=,711 53,53±6,54 p=,858 33,63±5,66 20,21±3,86 53,84±7,98 Eğitim Durumu Okur-Yazar Değil İlköğretim Ortaöğretim Lise Üniversite 34,96±5,15 p=,256 19,76±3,51 p=,929 54,73±6,94 p=,615 33,36±4,61 20,15±3,53 53,51±6,65 34,57±4,54 19,42±3,69 54,00±5,97 31,66±5,68 19,00±4,56 50,66±9,37 29,66±5,50 20,33±1,15 50,00±5,56 Gelir Dağılımı Düşük gelir Orta gelir Yüksek gelir 34,15±4,88 p=,458 20,46±3,37 p=,060 54,62±6,78 p=,172 33,26±4,47 18,06±3,49 51,33±6,81 32,77±5,09 19,85±3,57 52,62±6,61 Tanısı konmuş herhangi bir uzun süreli (kronik) ağrınız var mı? Var Yok 33,97±4,89 p=,316 20,32±3,33 p=,091 54,30±6,42 p=,111 32,90±4,82 19,03±3,80 51,93±7,44 Yaşamınız boyunca ağrı yaşama sıklığı Nadiren Her zaman 32,50±4,57 p=,163 18,53±3,76 p=,017* 51,03±7,43 p=,025* 34,05±4,94 20,43±3,30 54,48±6,36 Son bir yıl içinde ağrı yaşama durumu Evet Hayır 33,67±4,83 p=,910 20,07±3,38 p=,284 53,75±6,51 p=,526 33,50±5,33 18,91±4,35 52,41±8,86 *p<0,05 **p<0,01 bilgilere göre ağrı inançları ölçeği ve alt boyutları puan dağılımları (n=100) Organik İnançlar Psikolojik İnançlar Ağrı İnançları Tablo 3. Demografik bilgilere göre ağrı inançları ölçeği ve alt boyutları puan dağılımları (n=100) Organik İnançlar Psikolojik İnançlar Ağrı İnançları Psikolojik inançlar puanında; ağrının temelinde psikolojik faktörlerin (anksiyete, depresyon vb.) etkili olduğu; yönetiminde ise bilişsel-davranışsal müdahaleye dayalı multidisipliner bir ağrı yönetimi programlarına katılım, rahatlama, gevşeme ve dikkati başka yöne çekme gibi yöntemlerden yararlanıldığı ortaya konmaktadır (Walsh & Radcliffe, 2002). 4. TARTIŞMA Ayrıca Demirci ve Özer (2020)’in Aile Sağlığı Merkezi’ne başvuran yaşlı bireylerin ağrı inançlarını ve depresyon düzeylerini etkileyen faktörleri saptamak amacıyla yaptıkları çalışmada erkeklerin organik inançlar alt boyut puan ortalamasının kadınların organik inançlar alt boyut puan ortalamasından yüksek olması istatistiksel olarak anlamlı bulunmuştur. İster (2012), kök hücre nakli uygulanan hematolojik kanser hastaları üzerine yaptığı çalışma, Koçoğlu ve Özdemir (2011) de yetişkin nüfustaki ağrı ve ağrı inançları ile sosyo-demografik ve ekonomik özellikler arasındaki ilişkinin değerlendirilmesi üzerine yaptıkları çalışmayla benzerlik göstererek organik ve psikolojik inançların cinsiyete göre değişmediğini ortaya koymuşlardır. Babadağ ve arkadaşları (2015)’nın yapmış olduğu çalışmada algoloji hastalarının psikolojik inanç ve organik inanç puanı arasında cinsiyete göre anlamlı bir fark bulamamıştır. Ayrıca kadın hastaların psikolojik inançlar puanının erkek hastalarınkine göre daha yüksek olduğu görülürken, aralarında anlamlı bir fark bulunamamıştır. Literatürdeki cinsiyet ve ağrı inançları alt boyutları ile yapılan çalışmalar değerlendirildiğinde bu çalışmaya benzer cinsiyet ve alt boyutlar arasında fark olduğu çalışmalar yer aldığı gibi (Berk, 2006; Kılıçarslan, 2019), cinsiyet ve alt boyutlar arasında fark olmadığını bulan çalışmalar da tt (İ t 2012 K ğl & Ö d i 2011 Babadağ, Balcı Alparslan ve Güleç, 2015 ). Bu durumda cinsiyet ve ağrı inançları konusunda literatürde görüş birliği olmadığından farklı örneklemlerde sonuçların değişebildiği görülmektedir. Bu çalışmada yaş değişkeni ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır. Bu çalışmaya benzer şekilde İster (2012), çalışmasında organik ve psikolojik inançları puan ortalaması yaşa göre değişmediğini bulmuştur. Gagliese ve Melzack (1997)’ın, ağrıda psikolojik ve organik faktörlere ilişkin inançlar üzerinde yaşın etkisini araştırmak amacıyla yaptıkları çalışmada; kronik ağrı deneyimindeki yaş farklılıklarını değerlendirmek için bağımsız faktör olarak yaş grubu oluşturulmuş ve ağrısı olan ve olmayan grupların her ikisinde de ağrı inançlarında yaşın etkin olmadığı ortaya konmuştur. Buna karşılık Cornally ve McCarthy (2011) tarafından yaşlı hastalar ile yapılan çalışmada, katılımcılarda ağrının psikolojik inançlardan ziyade yaşa bağlı olduğu inancının ön planda olduğu ve bununla birlikte orta yaşa bağlı olması inancının daha fazla olduğu belirtilmiştir. Ulus ve arkadaşlarının (2014) ileri yaştaki kişilerde ağrı inançlarının belirlenmesine yönelik araştırmasında, katılımcıların yaklaşık yarısının “ağrı organın hasar görmesi sonucu oluşur” ve yarıdan fazlasının da “ağrı bir hastalığın işaretidir” cevabını verdiği belirlenmiştir. Bu çalışmada ağrının geriatrik bireylerde yaşla ilişkilendirildiği ve yaşlıların ağrıyı hastalık gibi gördükleri ortaya çıkmıştır, böylece geriatrik yaşlılarda organik inançların daha yüksek olduğunu bulgulamışlardır (Ulus, ve diğerleri, 2014). 4. TARTIŞMA İster (2012), kök hücre nakli uygulanan hematolojik kanser hastaları üzerine yaptığı çalışma, Koçoğlu ve Özdemir (2011) de yetişkin nüfustaki ağrı ve ağrı inançları ile sosyo-demografik ve ekonomik özellikler arasındaki ilişkinin değerlendirilmesi üzerine yaptıkları çalışmayla benzerlik göstererek organik ve psikolojik inançların cinsiyete göre değişmediğini ortaya koymuşlardır. Babadağ ve arkadaşları (2015)’nın yapmış olduğu çalışmada algoloji hastalarının psikolojik inanç ve organik inanç puanı arasında cinsiyete göre anlamlı bir fark bulamamıştır. Ayrıca kadın hastaların psikolojik inançlar puanının erkek hastalarınkine göre daha yüksek olduğu görülürken, aralarında anlamlı bir fark bulunamamıştır. Literatürdeki cinsiyet ve ağrı inançları alt boyutları ile yapılan çalışmalar değerlendirildiğinde bu çalışmaya benzer cinsiyet ve alt boyutlar arasında fark olduğu çalışmalar yer aldığı gibi (Berk, 2006; Kılıçarslan, 2019), cinsiyet ve alt boyutlar arasında fark olmadığını bulan çalışmalar da mevcuttur (İster, 2012; Koçoğlu & Özdemir, 2011; Yaşlılarda ağrı inançlarının sosyo-demografik özelliklerle ilişkisi 79 ağrı yaşama sıklığında kadın olmanın risk faktörü olduğu bulunmuştur. Berkley (1997) çalışmasında deneysel olarak verilen somatik uyaranlar için, kadınların erkeklere göre daha düşük eşiklere, daha fazla ayırt etme yeteneğine, daha yüksek ağrı derecelerine ve zararlı uyaranlara daha az toleransa sahip oldukları bulmuş ve kadınların erkeklere göre daha sık, uzun süreli ve şiddetli ağrıdan şikayet ettiklerini, daha fazla vücut bölgesinde ağrı hissettiklerini ortaya koymuştur. Türkiye’de de Erdine (2005)’in yaptığı çalışmada, kadınların erkeklere göre ağrı sıklığının ve ağrı oranlarının daha fazla olduğu görülmüştür (kadın=%70 ve erkek=%55). Yapılan bu çalışmaların sonuçları göz önüne alındığında kadınların ağrıyı yaşama/algılama, ağrı şiddeti ve ağrıyı daha sık yaşamalarının ağrı inançları puanlarını arttırdığı düşünülmektedir. Araştırmada kadın katılımcıların psikolojik inançlar ve organik inançlar alt test puan ortalamalarının erkeklerin puan ortalamalarından yüksek olduğu bulunmuştur. Bu konuda yapılan çalışmalarda da cinsiyet faktörü ağrı inançları ölçeğinin alt boyutlarıyla ele alınmıştır. Sertel Berk (2006) tarafından kronik ağrı yaşayan hastalarda yapılan çalışmada, kadınların organik inançlar puan ortalamasının (4,6 ±0,65), erkeklerin (4,27 ±0,97) puan ortalamasından istatistiksel olarak anlamlı düzeyde yüksek olduğu belirtilmiştir. Kılıçarslan (2019)’ın hemşirelik öğrencilerinde ağrı inançları ve ağrı korkuları arasındaki ilişkiyi incelediği çalışmasında kadın öğrencilerin organik inançlar alt test puan ortalamalarının (3,71±0,58), erkek öğrencilerin organik inançlar alt test puan ortalamalarından (3,48±0,59) istatistiksel olarak anlamlı düzeyde yüksek olduğu ortaya koymuştur. Sertel Berk (2006) ve Kılıçarslan (2019) çalışmalarının sonuçları bu çalışmanın sonuçları ile paralellik göstermekte olup, bulguları destekler niteliktedir. 4. TARTIŞMA Bu araştırmada 65 yaş ve üstü bireylerin ağrı inançları puan ortalamalarını etkileyen sosyo-demografik faktörlerin tespit edilmesi amaçlanmıştır. Araştırmada 65 yaş ve üstü bireylerin ağrı inançlarını ölçmek amacı ile yaş, cinsiyet, eğitim durumu, gelir, medeni durum gibi sosyo-demografik özellikler ve kronik ağrı olma durumu, ağrı yaşama sıklığı gibi ağrıya ilişkin özellikler açısından değerlendirmeler yapılmıştır. Bu çalışmada cinsiyet açısından “Organik İnançlar” alt boyutu ve “Ağrı İnançları Ölçeği” arasında istatistiksel olarak anlamlı bir fark bulunmuştur ve kadınların “Organik İnançlar” alt boyutu ve “Ağrı İnançları Ölçeği” puan ortalamasının erkeklerden yüksek olduğu görülmüştür. Benzer şekilde Koçoğlu ve Özdemir (2011)’in ve Yağcı ve Telci (2016)’nın çalışmasında Organik inançlar puanında; ağrı “vücutta yaralanma, vücudun zarar görmesinden” kaynaklandığına dayanmaktadır. Katılımcılar, yaralanma ile ağrının doğru orantılı bir şekilde arttığı inancına sahiptir. YSAD-EIRJ 2022 / 15(2), 74-82 Yaşlılarda ağrı inançlarının sosyo-demografik özellikler ağrı yaşama sıklığında kadın olmanın risk faktörü olduğu bulunmuştur. Berkley (1997) çalışmasında deneysel olarak verilen somatik uyaranlar için, kadınların erkeklere göre daha düşük eşiklere, daha fazla ayırt etme yeteneğine, daha yüksek ağrı derecelerine ve zararlı uyaranlara daha az toleransa sahip oldukları bulmuş ve kadınların erkeklere göre daha sık, uzun süreli ve şiddetli ağrıdan şikayet ettiklerini, daha fazla vücut bölgesinde ağrı hissettiklerini ortaya koymuştur. Türkiye’de de Erdine (2005)’in yaptığı çalışmada, kadınların erkeklere göre ağrı sıklığının ve ağrı oranlarının daha fazla olduğu görülmüştür (kadın=%70 ve erkek=%55). Yapılan bu çalışmaların sonuçları göz önüne alındığında kadınların ağrıyı yaşama/algılama, ağrı şiddeti ve ağrıyı daha sık yaşamalarının ağrı inançları puanlarını arttırdığı düşünülmektedir. Araştırmada kadın katılımcıların psikolojik inançlar ve organik inançlar alt test puan ortalamalarının erkeklerin puan ortalamalarından yüksek olduğu bulunmuştur. Bu konuda yapılan çalışmalarda da cinsiyet faktörü ağrı inançları ölçeğinin alt boyutlarıyla ele alınmıştır. Sertel Berk (2006) tarafından kronik ağrı yaşayan hastalarda yapılan çalışmada, kadınların organik inançlar puan ortalamasının (4,6 ±0,65), erkeklerin (4,27 ±0,97) puan ortalamasından istatistiksel olarak anlamlı düzeyde yüksek olduğu belirtilmiştir. Kılıçarslan (2019)’ın hemşirelik öğrencilerinde ağrı inançları ve ağrı korkuları arasındaki ilişkiyi incelediği çalışmasında kadın öğrencilerin organik inançlar alt test puan ortalamalarının (3,71±0,58), erkek öğrencilerin organik inançlar alt test puan ortalamalarından (3,48±0,59) istatistiksel olarak anlamlı düzeyde yüksek olduğu ortaya koymuştur. Sertel Berk (2006) ve Kılıçarslan (2019) çalışmalarının sonuçları bu çalışmanın sonuçları ile paralellik göstermekte olup, bulguları destekler niteliktedir. Ayrıca Demirci ve Özer (2020)’in Aile Sağlığı Merkezi’ne başvuran yaşlı bireylerin ağrı inançlarını ve depresyon düzeylerini etkileyen faktörleri saptamak amacıyla yaptıkları çalışmada erkeklerin organik inançlar alt boyut puan ortalamasının kadınların organik inançlar alt boyut puan ortalamasından yüksek olması istatistiksel olarak anlamlı bulunmuştur. 4. TARTIŞMA (2001) kronik ağrılı bir hasta popülasyonunda hem bildirilen klinik ağrıda hem de deneysel ağrı toleransında etnik farklılıkları araştırmak amacıyla yaptıkları çalışmada sosyo- ekonomik düzey ile ilişkili ağrıyı sağlık hizmetlerine erişim kapsamında ele almış ve bu hastaların sürekli ağrı yaşama nedeninin uygun tıbbi tedaviden yararlanamamak olduğunu belirtmişlerdir (Edwards, Doleys, Fillingim, & Lowery, 2001). Bu çalışmada tanısı konmuş herhangi bir uzun süreli (kronik) ağrının olma durumu ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır. Edwards’ın (1992) araştırmasında, kronik ağrı hastalarının organik faktörleri daha çok ön plana çıkardığı, ayrıca ağrı ile ilgili sorunların psikolojik faktörlerle de ilişkilendirildiği belirtilmiştir (Edwards, Pearce, Stokes, & Jones, 1992). Baird ve meslektaşlarının (2013) araştırmasında klinik dışı grup ile klinik olarak takip edilen kronik bel ağrısı olan grubun ağrı inançları ele alınmıştır. Buna göre; kronik bel ağrısı yaşayanların organik inançlar puanının daha yüksek olduğu, klinik olmayan grupta ise sık ağrı yaşayanlarda organik inançlar puanının yüksek olduğu saptanmıştır (Baird & Haslam, 2013). Edwards ve ark. (1992) tarafından yapılan çalışmada, ağrısı olmayan öğrencilerin psikolojik inanç puanlarının kronik ağrılılardan daha fazla, organik inanç puanlarının ise kronik ağrılılardan daha az olduğu bildirilmiştir (Edwards, Pearce, Stokes, & Jones, 1992). Sertel Berk (2006), yapmış olduğu çalışmada ise kronik ağrılı bireylerin organik inançları puan ortalamasının düşük olduğunu bulmuştur. Bu çalışmada medeni durum ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır. Koçoğlu ve Özdemir (2011) yapmış olduğu çalışmada organik ve psikolojik inançların medeni duruma göre değişmediğini belirtmiştir. Babadağ (2014) yapmış olduğu çalışmada ağrı inançları ile medeni durum arasında ilişki olmadığını saptamıştır. Buna karşılık İster (2012)’in kök hücre nakli uygulanan hematolojik hastalar üzerinde yaptığı çalışmada organik ve psikolojik inançların hastaların evli veya evli olmama durumlarına göre değişiminin istatistiksel açıdan anlamlı bulunmuştur, buna göre evli hastaların organik inançlar puan ortalaması (3,9±0,9), evli olmayan hastalarınkinden; (3,5±0,7); evli olmayan hastaların (4,8±0,9) psikolojik inançlar puan ortalaması evli hastalarınkinden (4,1±1,1) yüksek bulunmuştur. Sertel Berk (2006) yapmış olduğu çalışmada evli olmayan hastaların psikolojik inançlar puan ortalamasının, evli hastalarınkinden yüksek olduğunu ortaya koymuştur. Aynı çalışmada psikolojik inançların cinsiyete göre değişmediği bildirilmiştir. Literatürdeki medeni durum ve ağrı inançları alt boyutları ile yapılan çalışmalar değerlendirildiğinde bu çalışmaya benzer medeni durum; ağrı inançları ölçeği ve alt boyutlar arasında fark olmadığını bulan çalışmalar yer aldığı gibi fark olduğunu bulan çalışmalar da mevcuttur. Bu durumda medeni durum ve ağrı inançları konusunda literatürde görüş birliği olmadığı, farklı örneklemlerde sonuçların değişebildiği görülmektedir. 4. TARTIŞMA Bu doğrultuda literatürde bu çalışmaya benzer şekilde yaş; ağrı inançları ve alt boyutlar arasında fark olmadığını bulan çalışmalar (İster, 2012; Gagliese & Melzack, 1997) yer aldığı gibi (Cornally & McCarthy, 2011; Ulus, ve diğerleri, 2014) yaş; ağrı inançları ve alt boyutlar arasında fark olduğunu bulan çalışmaların da olduğu görülmektedir. Bu çalışmada eğitim değişkeni okuryazar olmayan, ilkokul, ortaokul, lise ve üniversite mezunu olmak üzere 5 kategoriye ayrılmış, eğitim değişkeni ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır. Literatürde eğitim değişkeni ve ağrıya değinen çalışmalara bakıldığında; Babadağ (2014), algoloji hastalarının ağrı inançları ve ağrıyla başa çıkma düzeyleri arasındaki ilişkiyi ele aldığı çalışmasında eğitim durumunun organik ve psikolojik ağrı inançları puanları üzerinde anlamlı bir ilişki bulamamıştır. Sertel Berk (2006) tarafından kronik ağrı yaşayan hastalarda yapılan çalışmada yüksek eğitimlilerin düşük eğitimlilere göre psikolojik inanç atıf skorlarının daha yüksek olduğu bulunmuştur. Birge ve Mollaoğlu (2018), hastaların ağrı inançları ile ağrıyı yönetmede kullandıkları ilaç dışı yöntemler üzerine yaptıkları araştırmada eğitim durumunun ağrı inançları üzerinde istatistiksel olarak anlamlı bir fark olduğunu, okuryazar olan hastaların psikolojik alt test puan ortalamalarının eğitim düzeyi daha yüksek hastalara göre düşük olduğunu belirtmektedirler. Erciyas (2019) cerrahi ağrı deneyimi olmayan hastaların ameliyat sonrası ağrı inançları üzerine yapılan çalışmada psikolojik inançlar alt test ameliyat öncesi puanları açısından okuryazar veya okuryazar YSAD-EIRJ 2022 / 15(2), 74-82 Korkmaz Yaylagül,İmancıoğlu 80 olmayanlar ve lisans veya lisans üstü olanlar arasında istatistiki olarak anlamlı bir fark olduğunu ortaya koymuştur. Sertel Berk (1998) baş ve bel ağrısı olan bir grup kronik ağrı hastası ile yürüttüğü çalışmasında, düşük eğitimli olanların her iki grupta da daha sık rastlandığını bildirerek, kronik ağrı yaşayan hastalarda yapmış olduğu çalışmada da bel ağrılarının psikolojik inanç atıflarının daha düşük olduğu bulgusuna vardığı için eğitim düzeyinin etkisinin daha kapsamlı araştırılmasına vurgu yapmaktadır. Genel olarak kronik ağrı ve eğitim düzeyleri arasındaki ilişki ele alındığında Pietri-Taleb ve arkadaşları (1995) tarafından beyaz yakalılar ve mavi yakalılar üzerinde yapılmış olan çalışmada bel ağrısı yaşayan katılımcıların düşük eğitim düzeyine sahip oldukları belirtilmektedir. Bu doğrultuda literatürde bu çalışmaya benzer olarak eğitim durumu; ağrı inançları ölçeği ve alt boyutlar arasında farkın olmadığını bulan çalışmalar (Babadağ, Balcı Alparslan ve Güleç, 2015 ) yer aldığı gibi eğitim durumu; ağrı inançları ölçeği ve alt boyutlar arasında farkın olduğu bulan çalışmaların (Berk, 1998; Berk, 2006; Birge & Mollaoğlu, 2018; Erciyas, 2019; Pietri-Taleb, Riihimaki, Eir, Lindstrom, & Moneta, 1995) olduğu görülmektedir. belirlenmiştir (Tanrıverdi, ve diğerleri, 2009). Edwards ve ark. 4. TARTIŞMA Yaşamı boyunca ağrı yaşama sıklığı ile “Ağrı İnançları Ölçeği” ve “Psikolojik İnançlar” alt boyutu arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Demirci ve Özer (2020)’in Aile Sağlığı Merkezi’ne başvuran yaşlı bireylerin ağrı inançları ve depresyon düzeyleri ve arasındaki ilişkiyi değerlendirmek amacıyla yaptığı çalışmada Aile Sağlığı Merkezine başvuruda bulunan yaşlı bireylerden erkeklerin; genel sağlık durumunu iyi hissedenlerin ve son altı ay içerisinde sürekli ağrı yaşamadığını ifade edenlerin organik inançları puanlarının daha yüksek olduğu istatistiksel olarak anlamlı bulunmuştur. Bu çalışmanın bulgularına benzer bir şekilde kişinin sağlık durumunun ağrı inançlarını olumsuz etkilediği belirtilmiştir (Jia & Jackson, 2016). 5.SONUÇ İleri yaşta bireylerin ağrı inançlarını etkileyen sosyo- demografik özellikleri ortaya koymayı amaçlayan bu çalışma, amaca yönelik örnekleme tekniği ile Denizli, Tavas’ta yaşayan 65 yaş üzeri 100 kişi ile gerçekleştirilmiştir. Araştırmada elde edilen veriler cinsiyet, yaş, eğitim, medeni durum, gelir gibi sosyo- demografik özelliklerle kıyaslanmış ve daha önce yapılmış araştırma bulguları ile karşılaştırılmıştır. Çalışmada düşük, orta ve yüksek düzeyde gelire sahip kişilerin ağrı inançları durumu kıyaslandığında “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır. Literatürde yapılan çalışmalara bakıldığında Babadağ (2014), yapmış olduğu çalışmada gelir düzeyi ile organik ve psikolojik inançlar puanları arasında anlamlı ilişki bulamadığını belirtmiştir. Tanrıverdi ve ark. (2009)’nın çalışmasında, ekonomik durumun kötüleşmesinin ağrı sıklığını arttırdığı, ev hanımları ve işçilerin ağrıyı en fazla deneyimleyen grup olduğu Bu çalışmanın sonucuna göre araştırmaya katılanların ağrı inançları puan ortalaması 53,59; organik inançlar alt test puan ortalaması 33,65; psikolojik inançlar alt testine bakıldığında ortalama 19,94 olarak bulunmuştur. Cinsiyet değişkeni ile “Organik İnançlar” alt boyutu ve “Ağrı İnançları Ölçeği” arasında YSAD-EIRJ 2022 / 15(2), 74-82 Yaşlılarda ağrı inançlarının sosyo-demografik özelliklerle ilişkisi 81 Babadağ B, Balcı Alparslan G, Güleç S. (2015). The relationship between pain beliefs and coping with pain of Algology patients' pain management. Nursing, 16(6), 910–919. istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Kadın katılımcıların puan ortalaması her iki boyutta da erkeklerden yüksek bulunmuştur. Çalışmada yaş, eğitim, medeni durum, gelir ve tanısı konmuş herhangi bir uzun süreli (kronik) ağrı olma durumu değişkenleri ile “Ağrı İnançları Ölçeği” ve alt boyutları arasında istatistiksel olarak anlamlı bir fark bulunamamıştır (p>0,05). Buna karşılık yaşamı boyunca ağrı yaşama sıklığı ile “Ağrı İnançları Ölçeği” ve “Psikolojik İnançlar” alt boyutu arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Sonuç olarak çalışmada; kadın olma ile “Organik İnançlar” alt boyutu ve “Ağrı İnançları Ölçeği”; yaşamı boyunca her zaman ağrı yaşama ile “Psikolojik İnançlar” alt boyutu ve “Ağrı İnançları Ölçeği” arasında istatistiksel olarak anlamlı bir fark bulunmuştur (p<0,05). Baird, A. J., & Haslam, R. A. (2013). Exploring differences in pain beliefs within and between a large nonclinical (workplace) population and a clinical (chronic low back pain) population using the pain beliefs questionnaire. Physical Therapy, 1615–1624. Berk, H. Ö. (1998). Headache vs. Non-Headache Turkish Chronic Pain Patients: A Descriptive Study About Their Demographic and Psychosocial Characteristics . İstanbul: Boğaziçi Üniversitesi Sosyal Bilimler Enstitüsü . Berk, H. Ö. (2006). Kronik Ağrı Yaşantısı ve Ağrı İnançları: Ağrı İnançları Ölçeğinin Türkçe Geçerlilik ve Güvenirlik Çalışması. İstanbul: T.C. Çıkar Çatışması Çıkar Çatışması Erciyas, A. (2019). Cerrahi Ağrı Deneyimi Olmayan Hastaların Ameliyat Sonrası Ağrı İnançları . Ankara: Ankara Yıldırım Beyazıt Üniversitesi Sağlık Bilimleri Enstitüsü. Yazarların herhangi bir çıkara dayalı ilişkisi yoktur. Yazarların herhangi bir çıkara dayalı ilişkisi yoktur. 5.SONUÇ İstanbul Üniversitesi Sosyal Bilimler Enstitüsü Psikoloji Anabilim Dalı . Çalışma bütçe ve zaman kısıtlılığı nedeniyle, bir tek ilçede küçük bir örneklemle gerçekleştirilmiştir. Bu nedenle çalışmanın genellenebilir nitelikte olmaması en önemli sınırlılığıdır. Ancak yaşlılıkta ağrı konusunda yapılmış sınırlı sayıdaki çalışma nedeniyle literatüre katkı sağlamaktadır. Yaşlı bireylerin hayatında önemli bir yer tutan ağrı, ağrının şiddeti, yaşama sıklığı ile organik ve psikolojik inançlar birlikte değerlendirilmelidir. Ağrı inançlarının yaşlı bireylerin ağrıya yaklaşımlarını etkileyebileceği düşünülerek, ağrı inanç düzeylerinin düzenli olarak değerlendirilmesi önemlidir. Yaşlıların ve aile üyelerinin bu konuya ilişkin farkındalıklarını arttırmaya yönelik farkındalık eğitimlerinin düzenlenmesi yararlı olacaktır. Yapılacak bilimsel çalışmalarda kapsamlı tarama türündeki araştırmaların yanı sıra yorumlayıcı nitel araştırmaların da yapılması konunun derinlemesine incelenmesini sağlayacaktır. Bunun yanı sıra daha büyük ve farklı örneklem gruplarıyla ağrı inançları düzeylerini etkileyen faktörlerin belirlenmesi önerilmektedir. Berkley, K. J. (1997). Sex differences in pain. Behavioral and Brain, 371-380. Birge, A. Ö., & Mollaoğlu, M. (2018). Hastaların ağrı inançları ve ağrıyı yönetmede kullandıkları ilaç dışı yöntemler. Ağrı Dergisi, 84-92. Cornally, N., & McCarthy, G. (2011). Chronic pain: The help-seeking behavior, attitudes and beliefs of older adults living in the community. Pain Management Nursing, 206-217. Çilingir, D., & Bulut, E. (2017). Yaşlı bireylerde ağrıya yaklaşım. Anadolu Hemşirelik ve Sağlık Bilimleri Dergisi, 144-153. Demirci, K., & Özer, Z. (2020). Yaşlı bireylerin ağrı inançları ve depresyon düzeylerini etkileyen faktörler. Gevher Nesibe Journal of Medical & Health Sciences, 91-103. Edwards, L. C., Pearce, S. A., Stokes, L. T., & Jones, A. (1992). The pain beliefs questionnaire: An investigation of beliefs in the causes and consequences of pain. Pain, 267-272. Araştırma Desteği Edwards, R. R., Doleys, D. M., Fillingim, R. B., & Lowery, D. (2001). Ethnic differences in pain tolerance: Clinical implications in a chronic pain population. Psychosom Med, 16-23. Bu çalışma için hiçbir kurumdan destek alınmamıştır. Etik İzin Bu çalışma için Akdeniz Üniversitesi Klinik Araştırmalar Etik Kurulu’ndan (55578142-050.01.04- E.37959) onay alınmıştır. Araştırmaya katılan bireyler gönüllülük esasına göre belirlenmiştir. Erdine, S. (2005). Ağrılar. İstanbul: Asır Yayıncılık. Gagliese, L., & Melzack, R. (1997). Age differences in the quality of chronic pain: A preliminary study. Pain Research and Management, 157-162. KAYNAKÇA International Association for the Study of Pain, I. (Erişim Tarihi: 04.08.21). KAYNAKÇA Arslan, N. (2015). Derin Yaşamak. İstanbul: Sokak Kitapları Yayıncılık. İster, E. D. (2012). Kök Hücre Nakli Uygulanan Hematolojik Kanser Hastalarının Ağrı İnançları . Ankara: Hacettepe Üniversitesi Sağlık Bilimleri Enstitüsü. Aydoğdu, Ü. R., Karamustafaoğlu, O., & Bülbül, M. Ş. (2017). Akademik araştırmalarda araştırma yöntemleri ile örneklem ilişkisi: Doğrulayıcı doküman analizi örneği. Dicle Üniversitesi Ziya Gökalp Eğitim Fakültesi Dergisi, 556-565. Jia, X., & Jackson, T. (2016). Pain beliefs and problems in functioning among people with arthritis: a meta-analytic review. Journal of behavioral medicine, 735-756. Babadağ, B. (2014). Algoloji Hastalarının Ağrı İnançları ile Ağrıyla Başa Çıkma Durumları Arasındaki İlişki. Eskişehir: T.C. Eskişehir Osmangazi Üniversitesi Sağlık Bilimleri Enstitüsü. Karslı, B., & Akdeniz, M. (2011). Birinci basamakta yaşlı hastalarda ağrı yönetimi. Türkiye Klinikleri J Fam Med-Special Topics, 48-56. YSAD-EIRJ 2022 / 15(2), 74-82 Korkmaz Yaylagül,İmancıoğlu 82 Kılıçarslan, F. N. (2019). Hemşirelik Öğrencilerinin Ağrı İnançları ve Ağrı Korkuları Arasındaki İlişkinin İncelenmesi . Ankara: Ankara Yıldırım Beyazıt Üniversitesi Sağlık Bilimleri Enstitüsü. Tanrıverdi, G., Okanlı, A., Çetin, H., Özyazıcıoğlu, N., Sezgin, H., Karaman, H. Ö., & Korkut, M. (2009). Yaşlı popülasyonunda ağrı. Turk J Geriatr, 12(4),190-197. ( ) Tavşanlı, N. G., Özçelik, H., & Karadakovan, A. (2013). Ağrısı olan yaşlı bireylerin yaşam kalitesinin incelenmesi. Ağrı Dergisi, 93-100. Koçoğlu, D., & Özdemir, L. (2011). Yetişkin nüfusta ağrı ve ağrı inançlarının sosyo-demografik ekonomik özelliklerle ilişkisi. Ağrı Dergisi, 64-70. Özel, F., Yıldırım, Y., & Fadıoğlu, Ç. (2014). Huzurevinde yaşayan yaşlılarda ağrı yönetimi. Ağrı Dergisi, 57- 64. Tıglı, A., Günebakan, Ö., Toslalı, Z., & Aytar, A. (2019). Geriatrik Kişilerde Biliş Düzeyi, Ağrı İnançları ve Vücut Farkındalığı’nın İlişkisi. Sağlık ve Toplum, 86-91. Öztürk, G. B., & Karan, M. A. (2008, Kasım 24). Yaşlıda Ağrıya Yaklaşım. Akadgeriatri Web Sitesi: http://www.akadgeriatri.org/managete/fu_fold er/2009-01/html/2009-1-1-031-044.htm adresinden alındı Ulus, B., İrban, A., Bakırcı, N., Yılmaz, E., Uslu, Y., Yücel, N., & Aslan, F. E. (2014). Huzurevinde yaşayan yaşlılarda ağrı özellikleri ağrı inançları ve depresyon riskinin belirlenmesi. Turk J Geriatr, 17(2), 180-187. Pietri-Taleb, F., Riihimaki, H., Eir, Lindstrom, K., & Moneta, G. B. (1995). The role of psychological distress and personality in the incidence of sciatic pain among working men. American Journal of Public Health, 541-545. Walsh, D. A., & Radcliffe, J. C. (2002). Pain beliefs and perceived physical disability of patients with chronic low back pain. Pain, 23-31. WHO. (2007). Global Age-friendly Cities: A Guide. Geneva : World Health Organization . Saltan, A. (2017). YSAD-EIRJ 2022 / 15(2), 74-82 KAYNAKÇA Yaşlılarda depresyon, ağrı ve sosyodemografik özellikler arasındaki ilişkinin incelenmesi . Düzce Üniversitesi Sağlık Bilimleri Enstitüsü Dergisi, 67-72. Yağcı, N., & Telci, E. A. (2016). Yaşlılarda cinsiyetin genel sağlık durumuna, kognitif fonksiyona, depresif semptomlara, bel ve boyun ağrısına etkisinin incelenmesi: Denizli örneği. Yaşlı Sorunları Araştırma Dergisi/Elderly Issues Research Journal, 9(1), 53-60. Seçer, İ. (2018). Psikolojik Test Geliştirme ve Uyarlama Süreci; SPSS ve LISREL Uygulamaları. Ankara: Anı Yayıncılık. Yıldız, A., Erol, S., & Ergün, A. (2009). Bir huzurevinde kalan yaşlılarda ağrı ve depresyon riski. Turk J Geriatr, 12(3), 156-164. Şimşek, T. T., Yumin, E. T., Öztürk, A., Sertel, M., & Yumin, M. (2011). Ev ortamında yaşayan yaşlı bireylerde ağrı ile sağlık durumu, mobilite ve yaşam aktivite düzeyi arasındaki ilişki . Türkiye Fiziksel Tıp ve Rehabilitasyon Dergisi , 216-220. YSAD-EIRJ 2022 / 15(2), 74-82 YSAD-EIRJ 2022 / 15(2), 74-82

https://openalex.org/W2055353197

https://revistas.ucm.es/index.php/HICS/article/download/43990/41596

Spanish; Castilian

La importancia del teléfono móvil para la comunicación publicitaria

Historia y comunicación social

cc-by

Referencia normalizada: Niño González, J. I. (2013) La importancia del teléfono móvil para la comunicación publicitaria. Histo­ ria y Comunicación Social. Vol. 18 Nº Especial Octubre. Págs. 581-590. Sumario: 1. Introducción. 2. Metodología. 3. Análisis e interpretación de datos. 4. Conclusiones. 5. Bibliografía. Resumen La publicidad, en su constante búsqueda de vías alternativas para evitar los saturados medios tradiciona­ les, no puede perder de vista las posibilidades que le ofrece el teléfono móvil como medio de comunica­ ción persuasiva. Con objeto de obtener el máximo rendimiento de estas innegables condiciones, hay que conocer cuáles son las líneas de explotación publicitaria que ofrecen mejores opciones para desarrollar acciones eficaces y entender cómo se ejecutan. Palabras claves: Teléfono móvil, publicidad, tecnología, comunicación Abstract Advertising, in its constant search for alternative ways to avoid saturated traditional media cannot lose sight of the possibilities offered by mobile phone as a means of persuasive communication. In order to get the most out of these undeniable conditions, we must know which advertising business lines offer better options to develop effective actions and understand how they run. Key Words: Mobile phone, advertising, technology, communication La importancia del teléfono móvil para la comunicación publicitaria José Ignacio Niño González Universidad Complutense de Madrid [email protected] Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 581 ISSN: 1137-0734 http://dx.doi.org/10.5209/rev_HICS.2013.v18.43990 1. Introducción En las últimas décadas la sociedad ha experimentado una profunda transformación que se fundamenta en el desarrollo tecnológico y que tiene una especial significación en el ámbito de la comunicación. Para Duderstadt (2002) el cambio en la forma en la que los individuos se interrelacionan con el entorno le lleva a afirmar que la sociedad está experimentando una evolución que es el origen de la Sociedad de la Información. Dicha transformación alcanza su punto culminante con la irrupción de la tecnolo­ gía inalámbrica, un avance científico que ha modificado aspectos fundamentales del José Ignacio Niño González La importancia del teléfono móvil La importancia del teléfono móvil comportamiento social. La conocida como “revolución móvil”, rompe los estánda­ res vigentes y propicia un nuevo modo de interrelación basado en la movilidad que permite al ser humano estar conectado en cualquier lugar. Estamos en presencia de un cambio de escenario que tiene un protagonista esencial: el teléfono móvil. Hablamos de un dispositivo con una presencia tan fuerte en nuestra vida cotidiana que sorprende por la velocidad con la que se ha introducido de forma tan categórica en la sociedad: Gráfico 1: fuentes CMT e INE En el grafico 1 podemos observar que el informe elaborado por la CMT (Comisión del Mercado de las Telecomunicaciones) registra 50.662.531 de líneas de teléfonos móviles en España durante el 2012, número que supera con creces a la población censada, que de acuerdo con la información suministrada por el INE (Instituto Nacio­ nal de Estadística) para el mismo periodo se cifraba en 46.116.779 de individuos. Las cifras son contundentes y exponen el elevado grado de penetración del teléfono móvil, pero debemos ir más allá del mero dato numérico para entender cuáles son los factores que dimensionan esta realidad. En el marco de las sociedades modernas, cuanto mayor es el nivel de información, de conocimiento y de aprendizaje, mayor es su capacidad para desarrollar el grado de bienestar. Si en consecuencia con estos requerimientos, las herramientas e instrumentos que resuelven estas necesidades se desarrollan tecnológicamente, es fácil comprender el elevado nivel de aceptación que estos tendrán. “La tecnología de las comunicaciones, e Internet a la cabeza, es un bien para el hombre y para la humanidad entera; es un elemento de progreso y de cultura, de enriquecedor intercambio y de universal comunicación.” (Méndiz, 2002: 14). Historia y Comunicación Social Vol. 18. Nº Esp. 1. Introducción Octubre (2013) 581-590 Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 582 José Ignacio Niño González La importancia del teléfono móvil La importancia del teléfono móvil La importancia del teléfono móvil El impacto de la tecnología en nuestra sociedad de la información es muy grande y cada vez tiene más presencia en nuestros usos cotidianos. En su “Teoría de la Acep­ tación de la Tecnología” Davis, F. & Venkatesh, V. (1989) explican que cualquier nueva tecnología supone un reto para el ser humano y por lo tanto puede provocar un rechazo, pero si el individuo entiende que el esfuerzo que le va a llevar familiarizarse con esa tecnología viene justificado por una recompensa en forma de beneficios que le faciliten la vida, no tardará en aceptar esa novedad. En la misma línea, Schiffman y Kanuk (1997) establecen cinco factores que influyen sobre el grado de aceptación de nuevos productos: • La ventaja relativa: el individuo compara el nuevo producto con los existentes para descubrir que ventajas aporta la novedad. • La ventaja relativa: el individuo compara el nuevo producto con los existentes para descubrir que ventajas aporta la novedad. • La compatibilidad: si el producto es coherente con las necesidades actuales. • La complejidad de su uso: cuando el individuo entiende que un nuevo producto es fácil de usar se produce una adopción inmediata. Sin embargo, aunque el individuo perciba que es un producto que le ofrece muchas ventajas, si se ve incapaz de usarlo con eficacia, lo rechazará. • La facilidad para ser probado: si el nuevo producto ha podido ser probado con anterioridad. • La facilidad para ser probado: si el nuevo producto ha podido ser probado con anterioridad. • La observabilidad: cuando las ventajas del nuevo producto se observan de forma evidente. • La observabilidad: cuando las ventajas del nuevo producto se observan de forma evidente. Como hemos visto anteriormente, podemos concluir que en España todo el mundo tiene teléfono móvil (exceptuando tramos de edades muy jóvenes y personas de mucha edad) y sin duda es un claro ejemplo de instrumento tecnológico cuyo uso se adopta de forma inmediata porque los usuarios lo perciben como un producto novedoso que aporta grandes ventajas. Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 1. Introducción Estamos ante un dispositivo personal, único y portátil que está con nosotros en cualquier momento y que utilizamos de modo habi­ tual porque nos ayuda a hacernos más fácil muchas tareas de nuestra vida cotidiana. Pero el terminal móvil no es tan solo una herramienta de comunicación con muchas funcionalidades, también es un dispositivo con incuestionables capacidades mediáticas. Y es en esta sentido, en su condición de medio de comunicación, en el que la publicidad tiene que trabajar para explotar sus posibilidades. Como sistema de comunicación persuasiva, la publicidad tiene que crear vínculos con cualquier medio innovador que le ofrezca la opción de llevar con eficacia mensajes publicitarios a potenciales grandes audiencias. Y esta relación es aún más importante si pensamos que hablamos de un medio que se fundamenta en una tecnología de última generación que tiene una presencia especialmente grande en el segmento formado por los “nati­ vos digitales”, pilares básicos de la sociedad digital. Por tanto, para la publicidad es esencial identificar qué puede aportarle este nuevo medio tecnológico, y así poder realizar acciones de comunicación que aprovechan al máximo todo su potencial. Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 583 José Ignacio Niño González La importancia del teléfono móvil 2. Metodología En este trabajo el objeto de estudio se analiza desde una perspectiva cualitativa, realizándose entrevistas en profundidad a un grupo de individuos seleccionados por su experiencia y relación con el ámbito de referencia con el fin de recoger y valorar sus aportaciones. Mediante esta técnica se persigue obtener datos relevantes cuantificados que sirvan al investigador para aislar conceptos y poder establecer un esquema que dé respuestas a las cuestiones planteadas. Es un proceso de análisis no matemático que se basa en el rigor de la cuantificación de las respuestas obtenidas y en la interpretación objetiva por parte del investigador. “Con el término “investigación cualitativa”, entendemos cualquier tipo de inves­ tigación que produce hallazgos a los que no se llega por medio de procedimientos estadísticos u otros medios de cuantificación” (Strauss & Corbin, 2002: 11). Las entrevistas en profundidad se realizan sobre un colectivo de expertos con un perfil profesional que en conjunto cubre todas las áreas de actuación en las que se desarrolla la actividad publicitaria en el dispositivo móvil: anunciante, publicitario, experto en medios, experto en contenidos, consultor tecnológico y empresa de tecno­ logía. La entrevista comienza con unas preguntas que buscan centrar el objeto de estudio y establecer un hilo conductor que sirva para conocer en detalle las opiniones de los entrevistados tanto desde el plano personal como profesional. Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 2.1 Perfil de los entrevistados 2.1 Perfil de los entrevistados e los entrevistados Perfil Empresa Director General Glodimark Director de Marketing Sennheiser Director Quality Procedures Barclays Bank Consejero Delegado ZenithOptimedia Director de Operaciones Amper Director General Novavoz Director de Innovación Ba-aba Software Director de Nuevos negocios Presentalia Profesor Universidad Rey Juan Carlos Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 584 José Ignacio Niño González La importancia del teléfono móvil 2.2 Cuestionario de la entrevista Las preguntas propuestas como hilo conductor son: 1/ ¿Qué aportan y qué no aportan las nuevas tecnologías a la publicidad? 1/ ¿Qué aportan y qué no aportan las nuevas tecnologías a la publicidad? 2/ ¿Tiene proyección de futuro la publicidad en el móvil y por qué? 3/ ¿Cómo se relaciona el móvil en su vertiente publicitaria con las redes sociales? 2.3 Objeto formal 3/ ¿Cómo se relaciona el móvil en su vertiente publicitaria con las redes sociales? 2.3 Objeto formal Este trabajo realiza un análisis que busca determinar cuáles son para los diferentes profesionales las principales características que entienden aporta el teléfono móvil a la actividad publicitaria y como se deben aplicar estas innovaciones con objeto de conseguir campañas cada vez más eficaces y rentables. 2.4 Preguntas de investigación 2.4 Preguntas de investigación En esta investigación las preguntas fundamentales son: En esta investigación las preguntas fundamentales son: ¿Qué aportan las nuevas tecnologías a la actividad publicitaria en el teléfono móvil? ¿Qué piensan de la combinación publicidad-redes sociales en el teléfono móvil? 2.5 Objetivos ¿Qué piensan de la combinación publicidad-redes sociales en el teléfono móvil? 2.5 Objetivos Identificar que variables diferenciadoras aportan las nuevas tecnologías a la publi­ cidad. Identificar que variables diferenciadoras aportan las nuevas tecnologías a la publi­ cidad. Establecer que contenidos publicitarios pueden ser los que funcionan en el ámbito de las redes sociales a través del terminal móvil. Establecer que contenidos publicitarios pueden ser los que funcionan en el ámbito de las redes sociales a través del terminal móvil. 2.6 Hipótesis Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 2.6 Hipótesis Hipótesis particular 1: El teléfono móvil es un medio con alto nivel de interactivi­ dad que permite medir las acciones publicitarias con precisión. Hipótesis particular 2: La inversión en publicidad en redes sociales se va a desarro­ llar con especial intensidad en el ámbito de los dispositivos móviles. Hipótesis nuclear: El teléfono móvil posee características que le permiten postu­ larse como medio de gran potencial para la publicidad. 585 585 José Ignacio Niño González La importancia del teléfono móvil 3. Análisis e interpretación de datos Se realiza el análisis considerando 4 niveles desde diferentes planos de investiga­ ción: horizontal, transversal, relacional y global. El plano horizontal busca identificar en cada pregunta los conceptos relevantes. El análisis transversal establece la interre­ lación de estos conceptos entre todas las preguntas. El plano relacional identifica y relaciona los conceptos que se han expresado de forma diferente pero con una misma intención de significado. El nivel de análisis global da una perspectiva integrada de los resultados obtenidos en los anteriores niveles. 3.1 Análisis horizontal, transversal y relacional La primera cuestión planteada al grupo de entrevistados persigue conocer cuál es en su opinión la principal aportación de esta nueva tecnología a la publicidad y nos encontramos con un concepto dominante: hay una coincidencia generalizada en que el aspecto básico que aporta el teléfono móvil es la gran capacidad que tiene este medio para medir los resultados de las acciones publicitarias que se realizan en él. Para la actividad publicitaria la medición es esencial máxime si entendemos que en la actualidad las campañas se articulan en torno a un parámetro básico: el retorno de la inversión de las mismas. En este sentido es básico disponer de medios que permitan medir con exactitud los resultados obtenidos. También vemos que aparecen dos conceptos con intensidad. En primer lugar para los entrevistados el teléfono móvil permite que los receptores sean creadores de contenido. Por tanto, si se conoce que tipos de contenidos genera cada usuario, la publicidad estará en disposición de establecer un perfil de ese individuo e impactarle con contenidos persuasivos muy afines con sus gustos y necesidades. En segundo lugar nuestro panel de expertos señala la interactividad como otra de las grandes aportaciones de la telefonía móvil para la actividad publicitaria. Gracias a esta condi­ ción, la relación con el receptor se convierte en un dialogo que permite identificar con claridad sus necesidades y en consecuencia la publicidad puede actuar con mensajes muy eficaces. En el siguiente gráfico podemos ver como se concretan numéricamente las expresiones de conceptos realizadas por los entrevistados: Respecto a la segunda pregunta planteada, los entrevistados coinciden al señalar que las empresas son conscientes de la proyección de futuro de esta publicidad, sin embargo matizan que es una actividad que está todavía en fase de desarrollo pero en la que hay que estar ya que los próximos años vendrán marcados por una predomi­ nancia de la actividad publicitaria digital y de un perfil de potencial consumidor que será un “nativo digital”. Con la tercera pregunta los entrevistados opinan sobre cómo ven la publicidad en las redes sociales y la valoración es muy positiva. Establecen un paralelismo con el auge que están experimentando las redes sociales a través del ordenador personal y consideran que si en este dispositivo las perspectivas son muy atractivas, más poten­ cial existe en un dispositivo que reúne todas las características del ordenador pero que Historia y Comunicación Social Vol. 18. Nº Esp. Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 3.1 Análisis horizontal, transversal y relacional Octubre (2013) 581-590 586 La importancia del teléfono móvil José Ignacio Niño González además suma su portabilidad y capacidad para conectarse a las redes sociales desde cualquier lugar en cualquier momento. Fuente: Propia Fuente: Propia. Fuente: Propia. Fuente: Propia. Las posibilidades que ofrecen las redes sociales para la publicidad generan un gran número de comentarios como se constata en el gráfico que observamos a continua­ ción: Fuente: Propia. 3.2 Análisis global e interpretación de datos 3.2 Análisis global e interpretación de datos Los expertos entrevistados consideran que la publicidad tiene en el terminal móvil un medio con unas posibilidades de desarrollo innovadoras, hasta ahora ningún medio podía ofrecer a la comunicación publicitaria las opciones que pone a su alcance la Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 587 José Ignacio Niño González La importancia del teléfono móvil La importancia del teléfono móvil La importancia del teléfono móvil telefonía móvil. Como aportación destacada aparece la capacidad de medición que ofrece este tipo de publicidad, una prestación presente en todas las modalidades de publicidad basada en las nuevas tecnologías pero que tiene especial relevancia en el teléfono móvil. Las acciones se pueden medir con mucho rigor y los perfiles de los potenciales usuarios están identificados. 4. Conclusiones Los expertos coinciden en que las nuevas tecnologías dotan a la publicidad de capacidades únicas como es la interactividad o identificación de perfiles de usua­ rios, pero por encima destaca la capacidad de medición, la de mayor rigor conocida hasta por la publicidad. Y cuanto más cerca está el entrevistado de la planificación de medios, más relevante es esta afirmación ya que es donde esta capacidad tiene más significación y representa una de las variables más importantes en esta actividad. Por tanto podemos decir que la hipótesis particular 1 se cumple y que la publicidad en el móvil es una actividad que se puede medir con gran rigor y que el teléfono móvil está dotado de una gran capacidad interactiva. En cuanto a la segunda cuestión en estudio, los entrevistados no tienen dudas e identifican la publicidad en redes sociales a través del móvil como un campo de gran desarrollo y enorme potencial. De hecho, los expertos explican que en sus empre­ sas se están destinando partidas presupuestarias para desarrollar acciones en redes sociales y que la tendencia es incrementar los esfuerzos en éste área aun viviéndose momentos de reducción de inversión en el resto de actividades publicitarias. También reflejan como dato que apoya estas opiniones el que muchos fabricantes de termi­ nales móviles ofrezcan aparatos desarrollados específicamente para interactuar con las redes sociales en movilidad. Este aspecto, unido a la mejora de contenidos de la publicidad desplegada en las redes sociales dan muestra del gran futuro que espera a esta forma de hacer comunicación persuasiva. Con estas afirmaciones, podemos concluir que la hipótesis particular 2 se confirma y afirmar que en las redes sociales a través del teléfono móvil la publicidad tiene un campo de desarrollo de gran potencial. Por tanto, la confirmación de las hipótesis particulares permiten afirmar que el telé­ fono móvil aporta a la publicidad condiciones esenciales para ser un medio esencial para desarrollar campañas publicitarias en los próximos años y por tanto verificamos la hipótesis nuclear. 588 Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 La importancia del teléfono móvil La importancia del teléfono móvil José Ignacio Niño González Libros Octubre (2013) 581-590 La importancia del teléfono móvil La importancia del teléfono móvil La importancia del teléfono móvil José Ignacio Niño González José Ignacio Niño González Libros AGUADO, J.M. & MARTÍNEZ, I. (2008): “Sociedad móvil: tecnología, identidad y cultura”. Biblioteca Nueva. Madrid. Ó Ó FUNDACIÓN DE LA INNOVACIÓN BANKINTER (2007): “Web 2.0: el negocio de las redes sociales”. Madrid. FUNDACIÓN DE LA INNOVACIÓN BANKINTER (2007): “Web 2.0: el negocio de las redes sociales”. Madrid. IAB SPAIN (2012): “Estudio de Inversión en Publicidad Digital”. Madrid IAB SPAIN (2012): “Estudio de Inversión en Publicidad Digital”. Madrid ITO, M. & OKABE, D. & MATSUDA, M. (2005): “Personal, Portable, Pedestrian: mobile phones in Japanese life”. MIT Press. Cambridge (Mass). É ITO, M. & OKABE, D. & MATSUDA, M. (2005): “Personal, Portable, Pedestrian: mobile phones in Japanese life”. MIT Press. Cambridge (Mass). MÉNDIZ, A. & VICTORIA, J. S. (2002): “Publicidad, Comunicación y Marketing en Internet”. Área de Cultura y Educación. Diputación Provincial de Málaga. MÉNDIZ, A. & VICTORIA, J. S. (2002): “Publicidad, Comunicación y Marketing en Internet”. Área de Cultura y Educación. Diputación Provincial de Málaga. ONTSI (2011): “Informe anual de los contenidos digitales en España. Industria y hábitos de consumo”. Ministerio de Industria, Turismo y Comercio. Madrid. ONTSI (2011): “Las redes sociales en internet”. Ministerio de Industria, Turismo y Comercio. Madrid.i YOUNG, A. & AITKEN, L. (2007): “Profitable Marketing Communications: a guide to marketing return on investment”. Kogan Page Limited. London. g g g ZENITHOPTIMEDIA (2012): “Estudio sobre los medios digitales”. Madrid. Artículos en publicaciones web Artículos en publicaciones web CMT (2012): “Informe Anual”. http://cmtdata.cmt.es/cmtdata/jsp/inf_anual.jsp?­ tipo=1 CMT (2012): “Informe Anual”. http://cmtdata.cmt.es/cmtdata/jsp/inf_anual.jsp?­ tipo=1 p COMPUTERWORLD (2013): http://www.computerworld.es/movilidad/gartner-el- 50-de-las-aplicaciones-moviles-seran-hibridas-en-2016. COMPUTERWORLD (2013): http://www.computerworld.es/movilidad/gartner-el- 50-de-las-aplicaciones-moviles-seran-hibridas-en-2016. GARTNER (2013): http://www.gartner.com/imagesrv/digital-marketing/pdfs/ Recognize-the-Importance-of-Digital-Marketing.pdf GLOBALWEBINDEX (2013): https://www.globalwebindex.net/social-platfor­ ms-gwi-8-update-decline-of-local-social-media-platforms/. i GOYAL, S. (2013): “Advertising on Social Media”. Scientific Journal of Pure and Applied Sciences. https://www.sjournals.com/index.php/SJPAS/article/down­ load/447/pdf2(5) 220-223. ISSN 2322-2956. IAB (Interactive Advertising Bureau) (2013): “Glosario”. http://www.iabspain.net/ glosario/ IAB (Interactive Advertising Bureau) (2013): “Glosario”. http://www.iabspain.net/ glosario/ IAB (2013): “IV Estudio anual sobre Redes Sociales”. http://www.iabspain.net/ wp-content/uploads/downloads/2013/01/IV-estudio-anual-RRSS_reducida.pdf. IAB (2013): “IV Estudio anual sobre Redes Sociales”. http://www.iabspain.net/ wp-content/uploads/downloads/2013/01/IV-estudio-anual-RRSS_reducida.pdf. INE (2012): “Anuario Estadístico de España 2012”. http://www.ine.es/prodyser/ pubweb/anuario12/anu12_02demog.pdf INE (2012): “Anuario Estadístico de España 2012”. http://www.ine.es/prodyser/ pubweb/anuario12/anu12_02demog.pdf ZELDMANN, J. (2006): “Web 3.0”. Published in JavaScript, Business, Industry, Usability ∙ 95 Comments. ZELDMANN, J. (2006): “Web 3.0”. Published in JavaScript, Business, Industry, Usability ∙ 95 Comments. Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 589 Historia y Comunicación Social Vol. 18. Nº Esp. Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590 El autor José Ignacio Niño González es Doctor en Publicidad y Relaciones Públicas por la Universidad Complutense de Madrid 590 Historia y Comunicación Social Vol. 18. Nº Esp. Octubre (2013) 581-590

https://openalex.org/W4317436269

https://www.researchsquare.com/article/rs-2476562/latest.pdf

English

Contrasting patterns in growth attributes, root traits plasticity, and defence mechanism of maize under deficit moisture and phosphorus supply

Research Square (Research Square)

cc-by

Contrasting patterns in growth attributes, root traits plasticity, and defence mechanism of maize under deficit moisture and phosphorus supply Hafiz Athar Hussain  (  [email protected] ) Hafiz Athar Hussain  (  [email protected] ) Chinese Academy of Agricultural Sciences https://orcid.org/0000-0001-5239-6172 Hafiz Athar Hussain  (  [email protected] ) Chinese Academy of Agricultural Sciences https://orcid.org/0000-0001-5239-6172 Introduction Climate changes have multifaceted effects on the consequences of abiotic stress, threatening the productivity and sustainability of agricultural systems (Hussain et al. 2018). During their life spam, plants generally encounter a wide range of abiotic stresses such as drought, salinity, temperature extremes, and nutrient deficiencies. Generally, the occurrence of single stress factor in the natural condition generally leads to other stresses in plants, e.g., water deficiency not only induces osmotic stress but also causes P-starvation (Zhang et al. 2020, Asensio et al. 2021, Chen et al. 2022). It has been estimated that more than 60% of the global population will inhabit areas with water deficiencies by 2025 (FAO 2018). Moreover, > 30% of the global cropland is phosphorus deficient (MacDonald et al. 2011), so the P input into croplands is expected to increase by 51–86% by the year 2050 (Mogollón et al. 2018). The relationship between P and soil water is well known— soil P movement occurs through mass flow and diffusion and depends on pores filled with water (Oliveira et al. 2010). Most of the studies indicated that the plant P uptake is decreased under low soil moisture (Cramer et al. 2009, He and Dijkstra 2014, Suriyagoda et al. 2014) as P supply is reduced to root that affects water relations factors in the stressed plant (Sardans and Peñuelas 2004, Hussain et al. 2019), and decreased P supply through mineralization and by reduced P diffusion and mass flow in the soil (Sardans and Peñuelas 2012, Suriyagoda et al. 2014). Drought stress affects plant growth by altering ion uptake and enzyme activity, decreasing soil water potential and water uptake by roots (Hussain et al. 2020, Ali et al. 2022), and alterations in the metabolism of plants through overproduction of ROS like hydroxyl ion, hydrogen peroxide and superoxide (Zia et al. 2021). However, these radicals affect plant structural and functional stability by oxidizing proteins, lipids, and nucleic acids. Nevertheless, plants are fortified by various physiological and molecular processes to overcome these effects amongst which eliciting of antioxidant enzymes has a vital role in plants stress tolerance (Shemi et al. 2021a, Shemi et al. 2021b). In addition, inorganic phosphate (Pi) inadequacy is common problem in natural ecosystems due to its relatively high immobility and uneven distribution in soil. Like as drought, P- deficiency exhibited the series of complex morphological, physiological, and developmental adaptations in the roots and shoots (Li et al. 2022). Research Article License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/20 Abstract Background: The phosphorus uptake in plants is decreased due to soil water deficiency which negatively affects plant growth. How the interaction of soil moisture and P deficiency affects the morpho-physiological and biochemical responses of plants is still little known. Methods: This study investigates physiological and biochemical responses of maize hybrids under interaction of moisture and P deficit conditions. Three levels of water (well-watered, moderate drought, severe drought) and P-supply (P100, P50, and P0) were used for two maize hybrids. Results: The interaction of water and P-deficiency (60% and 40% FC along with P50 or P0) reduced the shoot and root traits, and P uptake in both maize cultivars. The activities of SOD, total protein, free proline, and total reducing sugar were increased with decreasing water and P- supply but response of POD and CAT remained variable. The significant upregulation of ZmNAC111 was noted under 40%FC with P100 treatment, but no such effects were recorded under 100%FC at all P treatments in both cultivars. The expression of ZmPHR1.1 and ZmPHR1.2 was increased with reducing P supply, but higher expression was observed at P0 at 60%FC and 40%FC, which indicates these are key genes for P-deficiency tolerance. Conclusions: Overall, P0 with 60% and 40%FC caused severe reductions in growth traits among P levels but 40%FC was most destructive at all levels of P-supply. These findings would be helpful to understand the drought and low P tolerance mechanism of maize and provide future directions for regulation of responsive genes for developing tolerant maize cultivars. Material And Methods The experiment was conducted in the growth room with controlled condition at the National Institute for Biotechnology and Genetic Engineering (NIBGE), Faisalabad, Pakistan. The seeds of hybrid maize cv. Zhengdan958 (ZD958) and Xianyu335 (XY335) were obtained from the Institute of Crop Science, Chinese Academy of Agricultural Sciences, Beijing, China. The soil was collected from the 0–20-cm layer of an experimental field of NIBGE, and sieved through 2-mm mesh. The soil texture class was sandy loam with pH 7.46, 12.3 g kg− 1 organic matter, 13.05 mg kg− 1 total nitrogen, 6.52 mg kg− 1 available P, 98.72 mg kg− 1 exchangeable potassium. Seeds were surface sterilized in 70% ethanol for 1 min followed by 5% sodium hypochlorite for 5 min, then washed four times in deionized water and planted in plastic pots (20 cm in-depth, 16 cm in diameter) filled with 2 kg soil. In each pot, four seeds were initially sown and after emergence, one plant was maintained in each pot. Recommended fertilizer was applied in the soil before sowing including the levels of Phosphorus, P1 (100 mg P kg− 1 soil), P2 (50 mg P kg− 1 soil) and P3 (no phosphorus). The plants were allowed to be grow for 18 days at normal condition and then subjected to three different soil moisture regimes as, W1: well-watered (> 80%FC), W2: moderate drought stress (60% FC), W3: severe drought stress (40% FC) for 10 days. The average day-night temperature during the experiment period was 18–32°C and relative humidity was 60–70%, respectively. The plants were harvested at 28 DAS and growth parameters were recorded and fresh samples were taken for physiological and molecular parameters. All the treatments were arranged in a completely randomized design (CRD) with three replications for each treatment. The experimental treatments were re-positioned weekly to minimize the environmental effects. Introduction The findings of this study would be helpful to understand the drought and low P tolerance mechanism of plants and will provide future directions for characterization of function and regulation of responsive genes for developing stress tolerant maize cultivars. Introduction At the molecular level, Arabidopsis ATAF1/2 and CUC2 (NAC) transcription factors play important roles in the regulation of various gene expression in response to stress conditions, and their molecular engineering is proposed as a potential strategy for the genetic improvement of stress tolerance in crops (Tran et al. 2010). Previously it was noted that the expression levels of 14 different NAC genes were increased in maize after drought treatments, suggesting that these genes might play important role in stress regulation, particularly those showing strong response to the drought stress whereas 5 NAC genes were involved in the communication between different signal transduction pathways (Peng et al. 2015). However, the enhanced expression of ZmNAC111 conferred drought tolerance in maize seedlings, improved plant water-use efficiency (WUE), modulated stomatal closure and enhanced the expression of stress-responsive genes under stress condition (Mao et al. 2015). For P-deficient plants, phosphate transporter 1 (PHR1) transcription factor is viewed as a positive regulator for inorganic phosphate (Pi) starvation signalling that up-regulates the “induced by phosphate starvation1” (IPS1) genes (Rubio et al. 2001). However, ZmPHR1 led to the upregulation of multiple genes that regulate metabolism during Pi-starvation, which in turn resulted in an elevation of Pi content in plants (Wang et al. 2013). Moreover, relative expression of ZmPHR1.1, and ZmPHR1.2, was significantly up-regulated in response to low phosphate conditions in maize (Wang et al. 2021b). Previously, most of the studies have focused on the single stress factor of drought or P-deficiency, but the synergistic interaction of drought and P deficiency for maize responses and underlying mechanisms need further investigation. Moreover, the response pattern of ZmNAC and ZmPHR1 transcription factor genes to the combination of water and P-deficiency, remains a particularly interesting question. Here, we studied the response of maize hybrids to drought and P-deficiency at physiological, biochemical, and gene levels. The aims of this study were to determine the interactive effects of drought and P-deficiency on the (a) morpho-physiological and biochemical attributes (b) root system architecture (c) P uptake and translocation in plant tissues, and (d) expression pattern of ZmNAC and ZmPHR1 transcription factor genes. We hypothesized that the negative effects of combined water and P-deficiency will be variable and severe than that of individual stress factor. Measurement Of Plant Growth Traits At harvest, shoot length of maize seedlings was recorded with a meter scale and seedlings were further separated immediately into roots and shoots for the measurement of fresh weights using digital electric balance. Subsamples of shoots were stored at − 40°C for biochemical analysis. The root samples were carefully scanned using the Epson PerfectionV700 Photo Flatbed scanner (No. B11B178023, PT Epson, Jakarta, Indonesia) and root functional traits: total root length, root surface area, average diameter, and root volume, were analyzed using the WinRHIZO software (Regent Instructions, Quebec, Canada). Furthermore, the biomass allocation such as root mass fraction (the proportion of root dry weight to the total plant dry weight), the root/shoot ratio (ratio of the belowground biomass to the aboveground biomass), specific root length (length per unit dry weight of total root system), root mass density (mass per unit root volume) were analysed (Wen et al. 2020). Later, the samples were oven-dried for recording shoot and root dry weights. Introduction Under P deficiency, shoot growth was more reduced than root growth and therefore resulted in a significant increase of biomass allocation to roots (Mollier and Pellerin, 1999). P deficiency also reduce root hydraulic conductance and whole plant water potential, probably by lowering the activity of water channel protein. Zribi et al. (2011) indicated that leaf water content was correlated with leaf osmotic potential under P deficiency. The soluble sugars usually accumulate in P-deficient plants, since utilization of the photo assimilate was largely restricted (Wissuwa et al., 2005). Maize (Zea mays L.) is sensitive to drought (Hussain et al. 2019, Hussain et al. 2020, Shemi et al. 2021b) and phosphorus deficit condition (Nadeem et al. 2011, Klamer et al. 2019, Xia et al. 2021) particularly during early growth stages. Single or combined impact of drought and P deficiency considerably decrease maize growth by affecting Fv/Fm, chlorophyll content, leaf relative water content and leaf water potential (Kaya et al. 2020). Concurrent effects of drought and P deficiency triggered the morpho-physiological, biochemical and structural responses (Suriyagoda et al. 2014, Chtouki et al. 2022) by reducing leaf area, leaf water potential, photosynthetic activity, soluble protein contents, nutrient uptake, and metabolic enzyme activities of plants (Shubhra, Garg et al. 2004, Gunes et al. 2006, Goufo et al. 2017, Xia et al. 2020). Water and P deficiency, both individually and combined, significantly decrease seed weight and aboveground biomass upto ∼80% (Smith et al. 2022), indicating that both these stressors are harmful for the physiological and metabolic functioning of the plants. Page 2/20 Page 2/20 At the molecular level, Arabidopsis ATAF1/2 and CUC2 (NAC) transcription factors play important roles in the regulation of various gene expression in response to stress conditions, and their molecular engineering is proposed as a potential strategy for the genetic improvement of stress tolerance in crops (Tran et al. 2010). Previously it was noted that the expression levels of 14 different NAC genes were increased in maize after drought treatments, suggesting that these genes might play important role in stress regulation, particularly those showing strong response to the drought stress whereas 5 NAC genes were involved in the communication between different signal transduction pathways (Peng et al. 2015). However, the enhanced expression of ZmNAC111 conferred drought tolerance in maize seedlings, improved plant water-use efficiency (WUE), modulated stomatal closure and enhanced the expression of stress-responsive genes under stress condition (Mao et al. 2015). Measurement Of Enzymatic Antioxidants Leaf samples were harvested and kept in a disposable zipper bag and stored at -80°C for the determination of antioxidant enzyme activity. The activity of SOD was measured by following procedure of Dhindsa et al. (1981). For this, 200 mg leaf sample was homogenized in 2 mL of extraction buffer (0.5 mM EDTA + 0.1 M phosphate pH 7.5) with precooled mortar and pestle. The homogenate was centrifuged at 10,000 rpm at 4°C and supernatant was stored at 4°C. SOD activity in the supernatant was assayed by its ability to inhibit photochemical reduction of nitro blue tetrazolium. A 3 mL assay mixture containing (0.2 mL of 200 mM methionine + 1.5 M sodium carbonate + 0.1 mL 3 mM EDTA + 0.1 mL 2.25 mM NBT + 0.1 mL riboflavin (60 µM) + 1.5 mL 100 mM potassium phosphate buffer + 1 mL distilled water and 0.1 mL of enzyme) was incubated under two 15 W inflorescent lamps for 15 min; illuminated and nonilluminated reactions without supernatant served as calibration. Absorbance of the samples along with the blank was recorded at 560 nm wavelength in a spectrophotometer (UV-4000, ORI, Germany). One unit of SOD enzyme activity was defined as the quantity of enzyme that reduced the absorbance reading of samples to 50% in comparison with tubes lacking enzymes (supernatant). A 0.5 g of fresh leaves was ground in a 5 mL of 50 mM phosphate buffer (pH 7.8) with the help of pestle and mortar. The homogenates were centrifuged at 15000 rpm for 20 min at 4°C. The supernatant was used to assess peroxidase (POD) and catalase (CAT) activity. POD activity was determined by following procedure of Putter (1974) with slight modifications. The reaction mixture contained 10 mM guaiacol + 5 mM H2O2 and 50 mM phosphate buffer (pH 7.0). The reaction mixture was preheated at 20°C in a water bath. Then, 2.8 mL reaction solution + 0.2 ML enzyme was added in 10 mL centrifuged tube and mix thoroughly. Absorbance was recorded with blank and with reaction mixture. Four absorbance readings were recorded at 470 nm wavelength with 1 minute time interval using spectrophotometer (UV-4000, ORI, Germany). CAT activity in fresh leaves was measured by the method of Aebei (1984) with slight modification. 100 mM H2O2 + 50 mM phosphate buffer (pH 7.8) solution was preheated in water bath at 25°C. Measurement Of Enzymatic Antioxidants In 10 mL tube, 0.2 mL phosphate buffer + 0.2 mL enzyme solution was added and preheated in water bath for 3 min. Then, 0.3 mL (100 mM H2O2) solution was added in 10 mL tube. The control tube was heated in boiling water bath for 5 min to kill the enzyme solution. After mixing, the absorbance at the 240 nm wavelength was calculated at interval of 1 min. Continuous determination for 4 min. 1 Unit of enzyme activity (U) was the decrease of 0.1 of A240 within 1 min. Estimation Of Phosphorus Contents Page 3/20 Page 3/20 The P contents from maize shoot and root samples was determined by vanadate molybdate method with some modifications using a UV/visible spectrophotometer as suggested by Chapman and Pratt (1962). One g of ground plant sample was taken for analysis and placed in a 100-mL volumetric flask, and 10 mL of tri-acid mixture (HNO3, H2SO4 and HClO with the ratio of 9:4:1) was added and the contents were mixed by swirling. The flask was placed on a hotplate in the fumehood and start heating at 80–90°C and then the temperature was raised to about 150–200°C. Continues heating were applied until the production of red NO2 fumes ceases. The contents were further heated until the volume was reduced to 3–4 mL and became colourless. After cooling the contents, the volume was made up with the distilled water and filtered through No. 1 filter paper. Then, 5 mL of digested solution was taken in a 50-mL volumetric flask, and 10 mL of vanadomolybdate reagent and make up the volume with distilled water. The absorbance of samples was observed by spectrophotometer to calculate the P content from the standard curve. Measurement Of Ho And Lipid Peroxidation The H2O2 contents were determined using the procedure described by Velikova et al. (2000). 500 mg of fresh leaves was homogenized with 5 mL of 0.1% (w/v) trichloro-acetic acid in pre-chilled pestle and mortar. The homogenate was centrifuged at 12,000 rpm for 15 min. Then, 0.5 mL supernatant was mixed in 0.5 mL of 0.05 M phosphate buffer (pH 7.0) + 1 mL of 1 M potassium iodide, and absorbance was recorded at 390 nm by using water as blank. The same steps were followed for making a standard curve by preparing different dilution of hydrogen peroxide. Lipid peroxidation was quantified by the estimation of malondialdehyde (MDA) content using thiobarbituric acid assays (De Vos et al.1991). For analysis, fresh leaves (0.2 g) were placed on ice bath, grinded with 5.0 mL of 5% (w/v) TCA, centrifuged, and the MDA content was recorded at 532 and 600 nm spectrophotometrically. Estimation Of Osmolyte Accumulation Free proline contents were assessed by following the acid ninhydrin method (Shan et al. 2007). Fresh leaf material (0.5 g) was extracted using 5 mL of 3% sulfosalicylic acid for 10 min with shaking at 100°C. The 2mL of filtered aqueous extract was mixed with glacial acetic acid (2mL) and acid ninhydrin reagent (2 mL), and heated (100°C) for 30 min. The reaction mixture after cooling was segregated against toluene (4 mL) and the absorbance of the organic phase was recorded at 520nm. The resulting values were related with a standard curve plotted using known amounts of proline (Sigma, St Louis, MO, USA). Page 4/20 Page 4/20 Total soluble sugar was estimated by anthracene ketone method as described by Zong and Wang. The fresh leaf sample (0.2 g) was homogenized with 25 mL distilled water and centrifuged (4000 rpm) for 20 minutes. Anthracene (0.1 g) was dissolved in 100mL diluted sulfuric acid to prepare anthracene sulfuric acid reagent. One mL extract and 5 mL anthracene sulfuric acid reagent were taken in a tube, shaken and put in boiling bath for 10 minutes. After 2 h stability, the sample was transferred in cuvette and the absorbance was read at 620 nm. Total soluble sugar was estimated by anthracene ketone method as described by Zong and Wang. The fresh leaf sample (0.2 g) was homogenized with 25 mL distilled water and centrifuged (4000 rpm) for 20 minutes. Anthracene (0.1 g) was dissolved in 100mL diluted sulfuric acid to prepare anthracene sulfuric acid reagent. One mL extract and 5 mL anthracene sulfuric acid reagent were taken in a tube, shaken and put in boiling bath for 10 minutes. After 2 h stability, the sample was transferred in cuvette and the absorbance was read at 620 nm. Total soluble proteins from the fresh leaf material were measured according to the Bradford (1976) method. 0.25 g fresh leaf material was grinding in 5 mL of phosphate buffer in chilled pestle and mortar, centrifuge at 12,000 rpm for 10 minutes and separated the supernatant. Then take 0.3 mL of supernatant and added 3 mL of Bradford reagent and gently mixed the solution and place it at room temperature for 10 min and absorbance of reading was taken at 595 nm using spectrophotometer. Rna Isolation And Qrt- Pcr RNA was extracted from maize tissue by using Trizol method with some modification. 1 µg RNA was reverse transcribed into cDNA using RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). PowerUp™ SYBR™ Green Master Mix was used for the qPCR reaction using CFX96 qPCR thermocycler with the 100 ng of cDNA and 10uM/µl primers. α-tubulin gene primers were used as an internal control. Quantitative real-time PCR primer sequences for ZmTUB4, ZmPHR1.1, ZmPHR1.2 and ZmNAC111, are listed here in Table 1. Table 1 List of qPCR primer sequences Primer Name   Primer sequence (5’-3’) ZmTUB4 Forward GCTATCCTGTGATCTGCCCTGA Reverse CGCCAAACTTAATAACCCAGTA ZmPHR1.1 Forward GCCACAGGCGACAGATCTAA Reverse CTCACCAATGGACTCACGGA ZmPHR1.2 Forward AAGGGCATTGGACACTGGAG Reverse GTGAGGTGGTAGTGGAGTGC ZmNAC111 Forward CCAACGGTGTGAACAAGAGG Reverse CCATGCCTCGAATCACTTGG Table 1 List of qPCR primer sequences Primer Name   Primer sequence (5’-3’) ZmTUB4 Forward GCTATCCTGTGATCTGCCCTGA Reverse CGCCAAACTTAATAACCCAGTA ZmPHR1.1 Forward GCCACAGGCGACAGATCTAA Reverse CTCACCAATGGACTCACGGA ZmPHR1.2 Forward AAGGGCATTGGACACTGGAG Reverse GTGAGGTGGTAGTGGAGTGC ZmNAC111 Forward CCAACGGTGTGAACAAGAGG Reverse CCATGCCTCGAATCACTTGG Data analysis The collected data were statistically analysed through analysis of variance technique using statistical package Statistix 8.1 (Analytical Software, Tallahassee, FL, USA). Mean variances were compared through Tukey’s HSD test (p < 0.05). Graphical presentation was done through Sigmaplot 10.0. and OriginPro 2021. The root growth of both maize cultivars was affected under the individual and interactive levels of water and P deficiency (Table 2). Total root length of ZD958 was significantly (p < 0.05) reduced under all P levels with 60% and 40%FC but non significantly (p < 0.05) affected by different P levels at 100%FC. Likewise, total root length of XY335 was significantly (p < 0.05) decreased under all levels of the water and P supply compared with P100 + 100%FC (Table 2). Effects of drought and P-deficiency on shoot growth Different alphabetical letters above error bars represent the significant difference among treatments at p  < 0 05 Effects of water and P- deficiency on the total root length, root fresh weight, root dry weight, specific root length, root tissue density, root mass fraction, root surface area, root average diameter, root volume and root/shoot ratio of the ZD958 and XY335 Maize cultivars. The error bar indicates the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p  < 0 05 fraction, root surface area, root average diameter, root volume and root/shoot ratio of the ZD958 and XY335 Maize cultivars. The error bar indicates the ± SE of three replicates. Effects of drought and P-deficiency on shoot growth Both water and P-deficiency had profound negative effect on the shoot growth traits viz., shoot length, shoot fresh weight, shoot dry weight and number of leaves of both maize cultivars (Fig. 1 and Fig. 2). Significant reductions in the shoot length and number of leaves in both cultivars were observed under all P treatments with moderate (60% FC) and severe (40%FC) drought conditions, nevertheless,shoot length and number of leaves of both cultivars were non-significantly (p < 0.05) affected by all levels of P under well-water supply (100%FC). Same as shoot length, the number of leaves of both cultivars were negatively affected at the 40%FC with different levels of P-supply (Fig. 2). Moreover, the shoot fresh weight and shoot dry weight of the both cultivars were significantly (p < 0.05) affected by all interactive levels of water and P- supply (Fig. 1). These results indicates that the drought at 40%FC with different levels of P was drastic to the shoot fresh and dry weight of the maize (Fig. 1c-f), as water deficiency restrict the uptake of P-uptake in the upper parts of the plants. Furthermore, shoot length was affected due to water deficiency but P-deficiency didn’t affect the shoot length significantly y y g g y Influence of drought and P-deficiency on root system architecture Page 5/20 Page 5/20 The root growth of both maize cultivars was affected under the individual and interactive levels of water and P deficiency (Table 2). Total root length of ZD958 was significantly (p < 0.05) reduced under all P levels with 60% and 40%FC but non significantly (p < 0.05) affected by different P levels at 100%FC. Likewise, total root length of XY335 was significantly (p < 0.05) decreased under all levels of the water and P supply compared with P100 + 100%FC (Table 2). Page 6/20 Table 2 Effects of water and P- deficiency on the total root length, root fresh weight, root dry weight, specific root length, root tissue density, root mass fraction, root surface area, root average diameter, root volume and root/shoot ratio of the ZD958 and XY335 Maize cultivars. The error bar indicates the ± SE of three replicates. Effects of drought and P-deficiency on shoot growth Mean square values are present with significant level: *p < 0.05, **p < 0.01., ***p < 0.001 and not significant: ns, W: water levels, P: phosphorus levels, W×P: water× phosphorus levels; Trait notations: MDA, malondialdehyde; H2O2, hydrogen peroxide; SOD, superoxide dismutase; POD, peroxidase; CAT, catalase. Traits ZD958   XY335 W P W×P   W P W×P Shoot length 925.44*** 25.44ns 2.18ns   1849.48*** 45.56ns 1.70ns Shoot fresh weight 149.37*** 125.81*** 1.20ns   48.78*** 59.53*** 5.73ns Shoot dry weight 4.02*** 3.07*** 0.08ns   2.89*** 1.73*** 0.13ns No. of leaves 10.56*** 2.12* 0.07ns   13.59*** 0.44ns 0.09ns Total root length 921.66*** 768.39*** 95.21**   501.90*** 1046.69*** 44.66** Root fresh weight 53.47*** 15.92*** 2.41**   38.20*** 10.68*** 1.45** Root dry weight 1.04*** 0.59*** 0.04**   1.31*** 0.46*** 0.06** Specific root length 9272533*** 2352509** 2363286***   3026071*** 3194868** 1170267** Root mass density 0.01056*** 0.00535** 0.00847***   0.00017ns 0.00066ns 0.00105ns Root mass fraction 0.00550* 0.00106ns 0.00228ns   0.00873* 0.00459ns 0.00707* Root surface area 9027154*** 1852858*** 160235*   2224347*** 3063651*** 113333* Root area diameter 13.16*** 2.74*** 0.43**   3.58*** 5.01*** 0.69*** Root volume 13900.9*** 2855.7*** 270.6**   5532.61*** 4748.27*** 355.06** Root/Shoot ratio 0.03863*** 0.06176*** 0.03207***   0.25844*** 0.04351* 0.01430ns Shoot P 1.49*** 1.38*** 0.04ns   1.24*** 0.97*** 0.04ns Root P 0.38*** 0.62*** 0.01ns   0.96*** 0.16* 0.01ns Malondialdehyde 299.26*** 27.03*** 2.70ns   196.49*** 33.91*** 2.52ns Hydrogen peroxide 805.91*** 85.25*** 6.01ns   780.37*** 57.20*** 7.54ns Superoxide dismutase 5064.80ns 2990.57** 33.98ns   8479.32*** 1682.83ns 30.97ns Cultivars Treatments Total root length (m) Root fresh weight (g) Root dry weight (g) Specific root length (m g− 1) Root mass density (g/ cm3) Root mass fraction (g g− 1) Root surface area (cm2) Root average diameter (mm) Root volume (cm3) Root/Shoot ratio (g g−  1) 40%FC +  P0 27.14  ±  0.00c 2.59 ±  0.03d 0.40 ±  0.03e 68.98 ±  5.52ab 0.16 ±  0.02a 0.31 ±  0.02ab 416.85  ± 25.66d 0.47 ±  0.02e 24.87  ± 0.48e 0.48 ±  0.05bc POD, peroxidase; CAT, catalase. Traits ZD958   XY335 W P W×P   W P W×P Shoot length 925.44*** 25.44ns 2.18ns   1849.48*** 45.56ns 1.70ns Shoot fresh weight 149.37*** 125.81*** 1.20ns   48.78*** 59.53*** 5.73ns Shoot dry weight 4.02*** 3.07*** 0.08ns   2.89*** 1.73*** 0.13ns No. Effects of drought and P-deficiency on shoot growth Different alphabetical letters above error bars represent the significant difference among treatments at p  < 0.05 Cultivars Treatments Total root length (m) Root fresh weight (g) Root dry weight (g) Specific root length (m g− 1) Root mass density (g/ cm3) Root mass fraction (g g− 1) Root surface area (cm2) Root average diameter (mm) Root volume (cm3) Root/Shoot ratio (g g−  1) ZD958 100% FC +  P100 60.57  ±  0.01a 8.97 ±  0.43a 1.55 ±  0.02a 38.99 ±  0.39c 0.18 ±  0.02b 0.31 ±  0.01a 3178.27  ±  265.00a 4.10 ±  0.27a 124.25  ± 4.49a 0.47 ±  0.03bc 100% FC +  P50 57.92  ±  0.04a 8.09 ±  0.31a 1.42 ±  0.10ab 41.15 ±  4.11bc 0.18 ±  0.02b 0.31 ±  0.02a 2813.69  ±  202.40a 3.35 ±  0.11ab 109.15  ±  3.18ab 0.48 ±  0.04bc 100% FC +  P0 50.69  ±  0.03ab 7.37 ±  0.36ab 0.92 ±  0.03c 55.36 ±  3.64b 0.12 ±  0.01b 0.31 ±  0.01a 2598.99  ±  18.12ab 3.01 ±  0.02b 98.30  ± 7.88b 0.45 ±  0.02cd 60% FC +  P100 49.11  ±  0.02ab 7.56 ±  0.54a 1.27 ±  0.04b 38.80 ±  2.20c 0.17 ±  0.02b 0.35 ±  0.01a 2613.12  ±  32.46ab 2.92 ±  0.20b 96.36  ± 6.43b 0.53 ±  0.03abc 60% FC +  P50 42.49  ±  0.02bc 7.27 ±  0.56ab 1.23 ±  0.07b 34.90 ±  3.62c 0.17 ±  0.02b 0.36 ±  0.03a 2198.76  ±  138.32b 2.89 ±  0.27b 89.78  ± 3.76b 0.62 ±  0.04ab 60% FC +  P0 30.42  ± 0.01 d 5.05 ±  0.23c 0.71 ±  0.01cde 42.65 ±  1.92bc 0.14 ±  0.01b 0.33 ±  0.04a 1255.87  ± 31.11c 1.39 ±  0.07c 45.68  ± 2.67c 0.64 ±  0.03a 40%FC +  P100 53.75  ±  0.03ab 5.78 ±  0.25bc 0.75 ±  0.04cd 71.96 ±  1.14a 0.13 ±  0.01b 0.26 ±  0.02a 1284.38  ± 12.21c 1.49 ±  0.10c 46.49  ± 2.91c 0.30 ±  0.01d 40%FC +  P50 31.67  ±  0.02cd 2.42 ±  0.06d 0.65 ±  0.03de 48.96 ±  0.72bc 0.27 ±  0.02a 0.30 ±  0.02a 820.16  ±  40.74cd 0.92 ±  0.07c 32.29  ±  0.77cd 0.66 ±  0.04a 40%FC +  P0 27.05  ±  0.02d 1.93 ±  0.14d 0.49 ±  0.02e 55.08 ±  5.31b 0.26 ±  0.01a 0.34 ±  0.02a 501.83  ± 38.88d 0.81 ±  0.07c 17.99  ± 0.92d 0.61 ±  0.03ab XY335 100% FC +  P100 65.06  ±  0.03a 8.04 ±  0.52a 1.58 ±  0.13a 45.63 ±  3.44d 0.20 ±  0.02a 0.35 ±  0.01a 2599.38  ± 13.96a 3.97 ±  0.20a 135.82  ± 8.23a 0.67 ±  0.08abc 100% FC +  P50 46.21  ±  0.01b 7.22 ±  0.21ab 1.26 ±  0.06b 37.09 ±  2.90d 0.17 ±  0.02a 0.33 ±  0.02a 2153.35  ±  173.76b 2.70 ±  0.26bc 101.79  ± 5.24b 0.92 ±  0.05a 100% FC +  P0 34.28  ±  0.00c 5.69 ±  0.26c 0.89 ±  0.05cd 38.62 ±  2.04d 0.16 ±  0.01a 0.31 ±  0.02ab 1815.29  ± 94.94b 2.21 ±  0.16bcd 73.94  ±  6.40cd 0.74 ±  0.11ab 60% FC +  P100 45.44  ±  0.02b 6.65 ±  0.27bc 1.13 ±  0.09bc 40.99 ±  4.43d 0.17 ±  0.02a 0.35 ±  0.02a 2623.65  ± 32.57a 2.49 ±  0.10bc 88.76  ±  5.62bc 0.58 ±  0.06bc 60% FC +  P50 33.13  ±  0.01c 5.40 ±  0.06c 0.97 ±  0.02bc 34.37 ±  1.97d 0.18 ±  0.02a 0.37 ±  0.03a 1763.24  ±  140.38b 2.16 ±  0.12bcd 70.22  ±  2.19cd 0.72 ±  0.03ab 60% FC +  P0 28.81  ±  0.00c 3.28 ±  0.28d 0.60 ±  0.03de 48.40 ±  2.90cd 0.18 ±  0.01a 0.27 ±  0.05ab 1315.59  ± 10.35c 2.02 ±  0.09cd 66.59  ±  3.48cd 0.59 ±  0.03bc 40%FC +  P100 45.17  ±  0.03b 3.39 ±  0.08d 0.53 ±  0.02e 84.64 ±  2.19a 0.16 ±  0.01a 0.21 ±  0.01b 1924.65  ± 97.88b 2.83 ±  0.17b 82.17  ±  4.16bc 0.41 ±  0.02c 40%FC +  P50 32.72  ±  0.00c 2.62 ±  0.02d 0.51 ±  0.02e 64.19 ±  2.74bc 0.20 ±  0.01a 0.31 ±  0.02ab 1207.42  ± 16.65c 1.69 ±  0.05d 51.16  ± 2.08d 0.43 ±  0.02c Page 7/20 Page 7/20 rs Treatments Total root length (m) Root fresh weight (g) Root dry weight (g) Specific root length (m g− 1) Root mass density (g/ cm3) Root mass fraction (g g− 1) Root surface area (cm2) Root average diameter (mm) Root volume (cm3) Root/Shoot ratio (g g−  1) 40%FC +  P0 27.14  ±  0.00c 2.59 ±  0.03d 0.40 ±  0.03e 68.98 ±  5.52ab 0.16 ±  0.02a 0.31 ±  0.02ab 416.85  ± 25.66d 0.47 ±  0.02e 24.87  ± 0.48e 0.48 ±  0.05bc Table 3 Summery of two-way analysis of variance (ANOVA) regarding the effect of different water and P levels on morpho- physiological growth and biochemical attributes of two different maize hybrids. Furthermore, significant (p < 0.05) variations in specific root length of both maize cultivars were noted under the all levels of water and P- supply; the highest specific root length of both cultivars was recorded under severe drought + optimum P-supply (40%FC + P100). Effects of drought and P-deficiency on shoot growth of leaves 10.56*** 2.12* 0.07ns   13.59*** 0.44ns 0.09ns Total root length 921.66*** 768.39*** 95.21**   501.90*** 1046.69*** 44.66** Root fresh weight 53.47*** 15.92*** 2.41**   38.20*** 10.68*** 1.45** Root dry weight 1.04*** 0.59*** 0.04**   1.31*** 0.46*** 0.06** Specific root length 9272533*** 2352509** 2363286***   3026071*** 3194868** 1170267** Root mass density 0.01056*** 0.00535** 0.00847***   0.00017ns 0.00066ns 0.00105ns Root mass fraction 0.00550* 0.00106ns 0.00228ns   0.00873* 0.00459ns 0.00707* Root surface area 9027154*** 1852858*** 160235*   2224347*** 3063651*** 113333* Root area diameter 13.16*** 2.74*** 0.43**   3.58*** 5.01*** 0.69*** Root volume 13900.9*** 2855.7*** 270.6**   5532.61*** 4748.27*** 355.06** Root/Shoot ratio 0.03863*** 0.06176*** 0.03207***   0.25844*** 0.04351* 0.01430ns Shoot P 1.49*** 1.38*** 0.04ns   1.24*** 0.97*** 0.04ns Root P 0.38*** 0.62*** 0.01ns   0.96*** 0.16* 0.01ns Malondialdehyde 299.26*** 27.03*** 2.70ns   196.49*** 33.91*** 2.52ns Hydrogen peroxide 805.91*** 85.25*** 6.01ns   780.37*** 57.20*** 7.54ns Superoxide dismutase 5064.80ns 2990.57** 33.98ns   8479.32*** 1682.83ns 30.97ns Peroxidase 43.07* 7.78ns 19.71ns   18.54ns 0.09ns 17.82ns Catalase 420.70*** 54.01** 7.49ns   336.93*** 31.47ns 18.43ns Total protein content 4.13*** 0.31* 0.01ns   6.99*** 0.45** 0.02ns Free Proline 164.94*** 58.89*** 8.37*   387.33*** 111.17*** 7.84* Total soluble sugar 236.55*** 2.29ns 12.03**   226.19*** 14.42ns 12.02ns the root fresh weight and root dry weight of the both maize cultivars were significantly reduced by decreasing water and P-su Furthermore, significant (p < 0.05) variations in specific root length of both maize cultivars were noted under the all levels of water and P- supply; the highest specific root length of both cultivars was recorded under severe drought + optimum P-supply (40%FC + P100). Page 8/20 Page 8/20 In addition, the significantly (p < 0.05) and maximum root tissue density in ZD958 was noted under 40%FC with different P-levels but the highest root tissue density of XD335 was recorded under 100%FC + P100 and 40%FC + P100. However, there was no statistically difference (p >  0.05) in root tissue density under all treatments of water and P-supply in XD335. In addition, the significantly (p < 0.05) and maximum root tissue density in ZD958 was noted under 40%FC with different P-levels but the highest root tissue density of XD335 was recorded under 100%FC + P100 and 40%FC + P100. However, there was no statistically difference (p >  0.05) in root tissue density under all treatments of water and P-supply in XD335. Similarly, there was non-significant (p > 0.05) variation in the root mass fraction of ZD958 under all treatments of water and P-supply. Effects of drought and P-deficiency on shoot growth But, significant (p < 0.05) variation in the root mass fraction of XD335 was noted under all treatments of water and P-supply. The root surface area, root average diameter and root volume of both maize cultivars were significantly (p < 0.05) reduced by of the decrease in water and P-supply compared with 100%FC + P100. There was significant (p < 0.05) variation in the root/shoot ratio of the both maize cultivars under the all treatments of water and P application. Overall, root growth traits were affected by decreasing the water and P-supply but the influence of water deficiency was worse than the P-deficiency (Table 2). Effects of drought and P-deficiency on the oxidative stress The MDA and H2O2 contents were significantly (p < 0.05) increased with decreasing the water and P-supply in both maize cultivars (Fig. 4). The MDA and H2O2 contents of both cultivars were significantly increased at 40%FC + P100, 40%FC + P50, and 40%FC + P0 as compared to 100%FC + P100.. Both the 60%FC and 40%FC increased the MDA and H2O2 contents of ZD958 and XY335, but such effects were more severe at 40%FC (Fig. 4). These results showed that the overaccumulation of MDA and H2O2 were increased that caused oxidative stress under the influence of water deficiency as compared to all P-levels in both maize cultivars (Fig. 4). Effects of drought and P-deficiency on the antioxidant enzymes Effects of drought and P-deficiency on the antioxidant enzymes The activities of antioxidative enzymatic were affected by reducing the water and P-supply levels in ZD958 and XY335 (Fig. 5). According to the results, SOD was significantly (p < 0.05) increased by decreasing the water and P-supply levels but the maximum SOD values were noted under severe water and P-deficiency (40%FC + P0) in both cultivars, as compared to other levels of water and P application. However, non-significant (p > 0.05) variation in POD values were recorded under the influence of water and P application levels. Interestingly, the highest values of POD were noted under severe water deficiency and optimum P-supply (40%FC + P100) in ZD958 and XY335, compared with other levels of water and P, respectively. The trend of CAT was opposite to SOD, which was significantly (p < 0.05) reduced by decreasing the water and P-supply levels. The higher reduction of CAT was not at severe water deficiency (40%FC) with all P application levels (P100, P50 and P0) in both maize cultivars, indicating that the CAT was not triggered in maize in response to drought. Effects of drought and P-deficiency on the compatible solutes Effects of drought and P-deficiency on the compatible solutes Effects of drought and P-deficiency on the compatible solutes The levels of compatible solutes viz., total protein contents, free proline, and total soluble sugars in ZD958 and XY335 were significantly (p <  0.05) increased under the interactive levels of water and P-supply (Fig. 6). The data indicated that the total protein contents and free proline were increased at 60%FC and 40%FC at all levels of the P-application in both cultivars. However, the higher improvement of total soluble sugar was recorded by severe water deficiency with optimum P-supply (40%FC + P100) in both cultivars, as compared to other levels of water and P- supply. Overall, interaction of 40%FC and P0 significantly up-regulated the compatible solutes in ZD958 and XY335 cultivars. Total protein, free proline and total soluble sugar were less affected by reducing P application, compared with water levels (Fig. 6). Relative expression of drought and P-starvation responsive genes nfluence of drought and P-deficiency on phosphorus (P) up The P uptake in both maize cultivars was significantly (p < 0.05) reduced by decreasing the water and P-supply treatments (Fig. 3). Compared to root-P, the shoot-P concentrations were higher in both, ZD958 and XY335 cultivars. At 40%FC + P0, the shoot-P was decreased by 81.7% and 77.4% in ZD958 and XY335, respectively compared with 100%FC + P100. Likewise, the root-P of ZD958 and XY335 was declined by 72.4% and 62.5%, respectively, compared with 100%FC + P100. It was observed that the optimum P-supply (P100) helps the maize plants in P uptake under water deficit conditions (60%FC and 40FC). Likewise, optimum water supply (100%FC) helps the maize plants in P uptake under low P- supply (P50 and P0). Overall, P uptake in maize plants was most affected by the severe water and P-deficiency (40%FC and P0) (Fig. 3). Discussion Drought and P-deficiency are two of the most important factors that limit the productivity of plants in natural environment (Ho et al. 2005, Falalou et al. 2018, Smith et al. 2022) and plants responses to these stress conditions are highly complex which involve the changes at the morpho-physiological, biochemical and gene levels. This study explains how interactive levels of water and P supply disturb the morpho- physiological, biochemical and candidate gene responses in maize plants. Our results showed that the interaction of water and P-deficiency levels has a profound negative effect (p<0.05) on the shoot length, shoot fresh weight, shoot dry weight, and the number of leaves of ZD958 and XY335 (Fig.1 and Fig.2). As previously, it was stated that the interactive impacts of water deficit condition and low P application reduces the aboveground biomass of common bean (Smith et al. 2022) and chickpea (Chtouki et al. 2022). Moreover, the maize plants root and shoot growth was additively impaired under the drought and P deficiency (Klamer et al. 2019). Furthermore, water and P-deficiency alter the activities, morphology and architecture system of the plant roots. Previously, it has been reported that the low phosphorus significantly increased the root surface area, root volume and total root length but decreased the average root diameter in maize (Xia et al. 2021). Moreover, the maize root surface area and root volume were not affected by drought, but total root length was increased under drought condition (Hussain et al. 2020). The present study showed that the total root length, root fresh weight, root dry weight, root surface area, root average diameter, and root volume of the ZD958 and XY335 were reduced by the interactive levels of water and P-deficiency (Table 2). Specific root length was increased by decreasing the water and P-supply but root tissue density and root mass fraction showed variations in both maize cultivars under the interactive level water and P-deficiency (Table 2). The root/shoot ratio was most affected by severe water deficiency with optimum P application. Interestingly, total root length, surface area, average diameter, root volume was less affected under severe water deficiency when P level was optimum (Table 2). Drought can majorly impact the root function by altering cell water permeability and influencing the growth and architecture of the plant root system. Correlations matrix revealed association among various traits The Pearson correlation analysis revealed significant positive and negative correlations among the studied traits under the interactive levels of water and P-supply in both maize cultivars (Fig. 8). The results showed the positive association of shoot traits with different root traits of both maize cultivars (ZD958 and XD335), but SRL, RTD and R/S ratio of ZD958 were negatively correlated with the shoot traits. The root and shoot P contents were strongly positively correlated with the shoot and root traits except SRL in XD335, and SRL, RTD and RMF. The shoot and root traits of both cultivars were negatively correlated with MDA, H2O2,. Moreover, the SRL had positive correlation with MDA, H2O2, SOD, POD, TPC, TSS, FP but negatively correlation with CAT. Furthermore, H2O2 and MDA had strong positive correlation with SOD, TPC, FP and TSS. Relative expression of drought and P-starvation responsiv To understand the underlying mechanism on how maize plants were affected by interactive levels of water and P-deficiency, the expression of some drought and P-starvation responsive genes were analyzed. Results showed that the ZmNAC111 was upregulated under 60% and 40% FC, as compared to 100% FC with the application of P100 in both maize cultivars. Comparatively, it was less regulated at both water deficit levels with P50 and P0 treatments. Overall, the relative expression trend of ZmNAC111 was P100 > P50 > P0 with 100%FC < 60%FC < 40%FC. However, the higher expression level of ZmPHR1.1 was recorded under P0 with all treatments of water, compared with P100 and P50 but higher level was noted under the Page 9/20 Page 9/20 interactive level of 40%FC with P0 in both maize cultivars. Moreover, ZmPHR1.1 was more increased in ZD958, as compared to XY335. Similarly, the ZmPHR1.2 was the most upregulated under P0 with all levels of water treatment as compared to P100 and P50 but higher expression was recorded under the interactive level of 40%FC with P0. Overall, the expression trend of ZmPHR1 genes was P100 < P50 < P0 and 100%FC < 60%FC < 40%FC. Discussion Page 10/20 Page 10/20 Similar to drought, P deficiency inevitably causes increased production of reactive oxygen species (ROS) as by product of photosynthesis (Meng et al. 2021). Our results showed that the rate of lipid peroxidation and accumulation of ROS were increased in the leaves of maize under different levels of water and P-supply regimes (Fig. 4-7). The higher production of ROS was found under the interactive level of 40%FC and P0. As, water and P deficiency induced higher ROS bursts which are scavenged by antioxidant systems which have been reported in maize previously (Zhang et al. 2014; Hussain et al. 2020). Similar to drought, P deficiency inevitably causes increased production of reactive oxygen species (ROS) as by product of photosynthesis (Meng et al. 2021). Our results showed that the rate of lipid peroxidation and accumulation of ROS were increased in the leaves of maize under different levels of water and P-supply regimes (Fig. 4-7). The higher production of ROS was found under the interactive level of 40%FC and P0. As, water and P deficiency induced higher ROS bursts which are scavenged by antioxidant systems which have been reported in maize previously (Zhang et al. 2014; Hussain et al. 2020). We observed the increased activities of SOD and POD which may help to overcome the negative effect of ROS in the leaves of maize cultivars under the interactive influence of drought stress and P- levels (Fig. 4-8, 4-9, 4-10 and 4-11). The enzymatic components may directly scavenge the ROS or produce the non-enzymatic antioxidants. Specifically, SOD is involved in the dismutation of O2•−into H2O2 in the mitochondrion, chloroplast, peroxisome and cytoplasm, while POD scavenges for H2O2 produced through the dismutation of O2•− catalysed by SOD. In addition, the higher levels of compatible solutes viz., total protein contents, free proline and total soluble sugar were noted in ZD958 and XY335 under interactive impact of drought and P-deficiency (Fig. 4-12), which indicated that the plants have self-defence mechanism to combat the stress condition. Previously, also the higher accumulation of osmolytes were noted to lower oxidative stress damage in maize under P- deficiency (Tang et al. 2019) and drought (Hussain et al. 2018, Hussain et al. 2020) To understand the mechanism on how maize plants respond to interactive levels of the water and P-deficiency, we tested the relative expression of selected stress responsive genes viz. ZmNAC111, ZmPHR1.1 and ZmPHR1.2. Competing Interests The authors declare no competing interests. Discussion Previously, it has been reported that the over expression of ZmNAC111 confer drought tolerance in maize, improves plant water-use efficiency and enhances the expression of stress- responsive genes under drought condition (Mao et al. 2015). However, the role of ZmNAC111 against interactive levels of water and P- deficiency has not yet been reported. In this study, we observed a significant upregulation in the expression of ZmNAC111 in both maize cultivars under severe water deficiency (40%FC) with sufficient P-supply (P100) treatment, and weak expression was noted under sufficient water level (100%FC) at all P-treatments (Fig. 7B). These findings suggest that the upregulation of ZmNAC111 indicate tolerance against drought condition, as optimum P-supply (P100) helps the plants to improve drought tolerance by higher expression of ZmNAC111. Moreover, ZmPHR1 is the well-described phosphate starvation-responsive transcription factor of the MYB family that enhance the phosphate absorption when it was overexpressed under low-phosphate conditions (Wang et al. 2013, Wang et al. 2021a, Wang et al. 2021b). But its role under the interactive levels of drought and P-deficiency has not yet been reported. Our results indicated the expression of ZmPHR1.1 and ZmPHR1.2 was weak at P100, but the expression level increased with decreasing the P supply. Interestingly, the higher expression of ZmPHR1.1 and ZmPHR1.2 was noted at P-deficiency (P0) treatment with moderate water deficiency (60%FC) and severe water deficiency (40%FC) which indicated that the water deficiency restricted the P movement to upper parts of plants, and P deficiency with drought was more lethal than P deficiency at normal moisture level. Discussion Drought stress and associated reduction in soil moisture can reduce plant nutrient uptake by reducing nutrient supply through mineralization (Sanaullah et al. 2012). In addition, effect of drought stress on nutrient supply through mineralization may depend on the duration and intensity or severity of dry periods (Farooq et al. 2009). Generally, negative effects of drought on plant P uptake become larger with increased dry period with others stressor. In support of our results, previous findings also reported that sufficient supply of Pi could reduce the negative impact of water deficit condition. Under water-deficit conditions, P application enhanced root growth, nutrient uptake, and water use efficiency, ultimately increasing yield and ameliorating the adverse effects of drought (Waraich et al. 2011). Phosphorus supply could enhance the plant resistance capacity to water stress by enhancement of roots system and improving accessibility of a plant to a large spectrum of water and nutrients sources (Razaq et al., 2017). Water deficiency in soil results in significant decrease in uptake of P at any soil-P levels. Sufficient soil-P levels were observed to reduce the water deficiency damage on uptake of P in comparison with moderate or low P supply in both cultivars. Under well water conditions, positive effects of available soil-Pi levels on uptake of P were noticed , which is in agreement with previous findings reported for various crops (Vance et al., 2003; Jones et al., 2004; Jemo et al., 2006). Under water-deficit conditions, several cowpea varieties showed higher increase in uptake of P following the high soil-Pi supply as compared to the moderate or low soil-Pi supply. Likewise, internal P concentrations declined with reduced water and low P supply, whereas micronutrients were little affected by drought and P-deficiency (Klamer et al. 2019). The present study demonstrated that the drought stress significantly (p<0.05) reduced the shoot-P and root-P concentrations by decreasing the water and P-supply in ZD958 and XY335 (Fig. 3). So, it is noted that even a mild drought stopped the P uptake roots, due to unavailability of P after the soil dried. Drought and P-deficiency collectively affects many physiological and biochemical processes and thus reduces the plant growth. It is reported that water stress induced by PEG significantly increased the ROS production in the maize cultivars (Hussain et al. 2020). Funding Financial support for this paper was provided by the National Natural Science Foundation of China (41977072) and Science and Technology Innovation Project of Chinese Academy of Agricultural Sciences. Competing Interests Conclusions Interaction of drought and P-deficiency affected the shoot and root traits, oxidative status, nutrients uptake and osmolytes accumulation in both hybrid maize cultivars. The shoot length, fresh and dry weight of root and shoot, root architecture and P uptake were significantly (p <  0.05) reduced by moderate and severe drought with P50 and P0. However, he H2O2 and MDA contents were increased with decreasing the moisture and P-supply. Moreover, SOD, total protein, free proline and total reducing sugar were increased with decreasing the moisture and P- supply, but POD and CAT were fluctuated to with oxidative stress caused by water and P supply levels. We found the upregulation of ZmNAC111 under severe drought with sufficient P supply, but no such effects were recorded under 100%FC with all P levels which indicates its vital role in drought tolerance and positive coloration with P availability. Moreover, the higher expression of ZmPHR1.1 and ZmPHR1.2 was observed at P-deficiency at moderate and severe drought, which shows these are key genes for P-deficiency tolerance. Further molecular studies are required to determine the mechanistic relationships of water deficit and P deficiency based on the root system architecture, uptake and transport of nutrients and defence mechanism to design strategies for crop breeding. References Page 11/20 Page 11/20 Page 11/20 1. Ali S, Tyagi A, Park S, Mir RA, Mushtaq M, Bhat B, Al-Mahmoudi H, Bae H (2022) Deciphering the Plant Microbiome to Improve Drought Tolerance: Mechanisms and Perspectives. Environmental and Experimental Botany: 104933 2. Asensio D, Zuccarini P, Ogaya R, Marañón-Jiménez S, Sardans J, Peñuelas J (2021) Simulated climate change and seasonal drought increase carbon and phosphorus demand in Mediterranean forest soils. Soil Biol Biochem 163:108424 arbon and phosphorus demand in Mediterranean forest soils. Soil Biol Biochem 163:108424 3. Chen N, Qin J, Tong S, Wang W, Jiang Y (2022) One AP2/ERF Transcription Factor Positively Regulates Pi Uptake and Drought Tolerance in Poplar. Int J Mol Sci 23(9):5241 4. Chtouki M, Laaziz F, Naciri R, Garré S, Nguyen F, Oukarroum A (2022) Interactive effect of soil moisture content and phosphorus fertilizer form on chickpea growth, photosynthesis, and nutrient uptake. Sci Rep 12(1):1–13 5. Cramer MD, Hawkins H-J, Verboom GA (2009) The importance of nutritional regulation of plant wat 6. Falalou H, Mariama H, Achirou FB, Oumarou H, Vadez V (2018) Genotypic variation for root development, water extraction and yield components in groundnut under low phosphorus and drought stresses. Am J Agric Forestry 6(5):122–131 7. FAO (2018) FUTURE OF FOOD AND AGRICULTURE 2018: Alternative Pathways to 2050. FOOD & AG 7. FAO (2018) FUTURE OF FOOD AND AGRICULTURE 2018: Alternative Pathways to 2050. FOOD & AGRICULTURE ORG 8. Garg BK, Burman U, Kathju S (2004) The influence of phosphorus nutrition on the physiological resp drought. J Plant Nutr Soil Sci 167(4):503–508 8. Garg BK, Burman U, Kathju S (2004) The influence of phosphorus nutrition on the physiological response of moth bean genotypes to drought. J Plant Nutr Soil Sci 167(4):503–508 9. Goufo P, Moutinho-Pereira JM, Jorge TF, Correia CM, Oliveira MR, Rosa EA, António C, Trindade H (2017) Cowpea (Vigna unguiculata L. Walp.) metabolomics: osmoprotection as a physiological strategy for drought stress resistance and improved yield. Front Plant Sci 8:586 10. Gunes A, Cicek N, Inal A, Alpaslan M, Eraslan F, Guneri E, Guzelordu T (2006) Genotypic response of chickpea (Cicer arietinum L.) cultivars to drought stress implemented at pre-and post-anthesis stages and its relations with nutrient uptake and efficiency. Plant Soil and Environment 52(8):368 1. He M, Dijkstra FA (2014) Drought effect on plant nitrogen and phosphorus: a meta-analysis. New Ph 12. References Ho MD, Rosas JC, Brown KM, Lynch JP (2005) Root architectural tradeoffs for water and phosphorus acquisition. Funct Plant Biol 32(8):737–748 13. Hussain HA, Hussain S, Khaliq A, Ashraf U, Anjum SA, Men S, Wang L (2018) Chilling and drought stresses in crop plants: implications, cross talk, and potential management opportunities. Front Plant Sci 9:393 14. Hussain HA, Men S, Hussain S, Zhang Q, Ashraf U, Anjum SA, Ali I, Wang L (2020) Maize tolerance against drought and chilling stresses varied with root morphology and antioxidative defense system. Plants 9(6):720 15. Hussain HA, Men S, Hussain S, Chen Y, Ali S, Zhang S, Zhang K, Li Y, Xu Q, Liao C (2019) Interactive effects of drought and heat stresses on morpho-physiological attributes, yield, nutrient uptake and oxidative status in maize hybrids. Sci Rep 9(1):1–12 16. Kaya C, Şenbayram M, Akram NA, Ashraf M, Alyemeni MN, Ahmad P (2020) Sulfur-enriched leonardite and humic acid soil amendments enhance tolerance to drought and phosphorus deficiency stress in maize (Zea mays L.). Sci Rep 10(1):1–13 17. Klamer F, Vogel F, Li X, Bremer H, Neumann G, Neuhäuser B, Hochholdinger F, Ludewig U (2019) Estimating the importance of maize root hairs in low phosphorus conditions and under drought. Ann Botany 124(6):961–968 17. Klamer F, Vogel F, Li X, Bremer H, Neumann G, Neuhäuser B, Hochholdinger F, Ludewig U (2019) Estimating the importance of maize root hairs in low phosphorus conditions and under drought. Ann Botany 124(6):961–968 18. Li P, Weng J, Rehman A, Niu Q (2022) Root Morphological and Physiological Adaptations to Low Phosphate Enhance Phosphorus Efficiency at Melon (Cucumis melo L.) Seedling Stage. Horticulturae 8(7):636 18. Li P, Weng J, Rehman A, Niu Q (2022) Root Morphological and Physiological Adaptations to Low Phosphate Enhance Phosphorus Efficiency at Melon (Cucumis melo L.) Seedling Stage. Horticulturae 8(7):636 19. MacDonald GK, Bennett EM, Potter PA, Ramankutty N (2011) Agronomic phosphorus imbalances across the world's croplands. Proceedings of the National Academy of Sciences 108(7): 3086–3091 9. MacDonald GK, Bennett EM, Potter PA, Ramankutty N (2011) Agronomic phosphorus imbalances ac Proceedings of the National Academy of Sciences 108(7): 3086–3091 19. MacDonald GK, Bennett EM, Potter PA, Ramankutty N (2011) Agronomic phosphorus imbalances across the world's croplands. Proceedings of the National Academy of Sciences 108(7): 3086–3091 20. References Mao H, Wang H, Liu S, Li Z, Yang X, Yan J, Li J, Tran L-SP, Qin F (2015) A transposable element in a NAC gene is associated with drought tolerance in maize seedlings. Nat Commun 6(1):1–13 20. Mao H, Wang H, Liu S, Li Z, Yang X, Yan J, Li J, Tran L-SP, Qin F (2015) A transposable element in a NAC gene is associated with drought tolerance in maize seedlings. Nat Commun 6(1):1–13 21. Mogollón J, Beusen A, Van Grinsven H, Westhoek H, Bouwman A (2018) Future agricultural phosphorus demand according to the shared socioeconomic pathways. Glob Environ Change 50:149–163 21. Mogollón J, Beusen A, Van Grinsven H, Westhoek H, Bouwman A (2018) Future agricultural phosphorus demand according to the shared socioeconomic pathways. Glob Environ Change 50:149–163 22. Nadeem M, Mollier A, Morel C, Vives A, Prud’homme L, Pellerin S (2011) Relative contribution of seed phosphorus reserves and exogenou phosphorus uptake to maize (Zea mays L.) nutrition during early growth stages. Plant Soil 346(1):231–244 22. Nadeem M, Mollier A, Morel C, Vives A, Prud’homme L, Pellerin S (2011) Relative contribution of seed phosphorus reserves and exogenous phosphorus uptake to maize (Zea mays L.) nutrition during early growth stages. Plant Soil 346(1):231–244 23. Oliveira EMM, Ruiz HA, Alvarez VVH, Ferreira PA, Costa FO, Almeida ICC (2010) Nutrient supply by mass flow and diffusion to maize plant in response to soil aggregate size and water potential. Revista Brasileira de Ciência do Solo 34:317–328 23. Oliveira EMM, Ruiz HA, Alvarez VVH, Ferreira PA, Costa FO, Almeida ICC (2010) Nutrient supply by mass flow and diffusion to maize plants in response to soil aggregate size and water potential. Revista Brasileira de Ciência do Solo 34:317–328 24. Peng X, Zhao Y, Li X, Wu M, Chai W, Sheng L, Wang Y, Dong Q, Jiang H, Cheng B (2015) Genomewide identification, classification and analysis of NAC type gene family in maize. J Genet 94(3):377–390 24. Peng X, Zhao Y, Li X, Wu M, Chai W, Sheng L, Wang Y, Dong Q, Jiang H, Cheng B (2015) Genomewide identification, classification and analysis of NAC type gene family in maize. J Genet 94(3):377–390 25. Rubio V, Linhares F, Solano R, Martín AC, Iglesias J, Leyva A, Paz-Ares J (2001) A conserved MYB transcription factor involved in 25. References Suriyagoda L, De Costa W, Lambers H (2014) Growth and phosphorus nutrition of rice when inorganic fertiliser application is partly replaced by straw under varying moisture availability in sandy and clay soils. Plant Soil 384(1):53–68 33. Tang H, Niu L, Wei J, Chen X, Chen Y (2019) Phosphorus limitation improved salt tolerance in maize through tissue mass density increase, osmolytes accumulation, and Na + uptake inhibition. Front Plant Sci 10:856 33. Tang H, Niu L, Wei J, Chen X, Chen Y (2019) Phosphorus limitation improved salt tolerance in maize through tissue mass density increase, osmolytes accumulation, and Na + uptake inhibition. Front Plant Sci 10:856 34. Tran L-SP, Nishiyama R, Yamaguchi-Shinozaki K, Shinozaki K (2010) Potential utilization of NAC transcription factors to enhance abiotic stress tolerance in plants by biotechnological approach. GM crops 1(1):32–39 34. Tran L-SP, Nishiyama R, Yamaguchi-Shinozaki K, Shinozaki K (2010) Potential utilization of NAC transcription factors to enhance abiotic stress tolerance in plants by biotechnological approach. GM crops 1(1):32–39 35. Wang J, Pan W, Nikiforov A, King W, Hong W, Li W, Han Y, Patton-Vogt J, Shen J, Cheng L (2021a) Identification of two glycerophosphodiester phosphodiesterase genes in maize leaf phosphorus remobilization. Crop J 9(1):95–108 35. Wang J, Pan W, Nikiforov A, King W, Hong W, Li W, Han Y, Patton-Vogt J, Shen J, Cheng L (2021a) Identification of two glycerophosphodiester phosphodiesterase genes in maize leaf phosphorus remobilization. Crop J 9(1):95–108 36. Wang R, Zhong Y, Liu X, Zhao C, Zhao J, Li M, Ul Hassan M, Yang B, Li D, Liu R (2021b) Cis-regulation of the amino acid transporter genes ZmAAP2 and ZmLHT1 by ZmPHR1 transcription factors in maize ear under phosphate limitation. J Exp Bot 72(10):3846–3863 36. Wang R, Zhong Y, Liu X, Zhao C, Zhao J, Li M, Ul Hassan M, Yang B, Li D, Liu R (2021b) Cis-regulation of the amino acid transporter genes ZmAAP2 and ZmLHT1 by ZmPHR1 transcription factors in maize ear under phosphate limitation. J Exp Bot 72(10):3846–3863 37. Wang X, Bai J, Liu H, Sun Y, Shi X, Ren Z (2013) Overexpression of a maize transcription factor ZmPHR1 improves shoot inorganic phosphate content and growth of Arabidopsis under low-phosphate conditions. Plant Mol Biology Report 31(3):665–677 38. References Rubio V, Linhares F, Solano R, Martín AC, Iglesias J, Leyva A, Paz-Ares J (2001) A conserved MYB transcription factor involved in phosphate starvation signaling both in vascular plants and in unicellular algae. Genes Dev 15(16):2122–2133 25. Rubio V, Linhares F, Solano R, Martín AC, Iglesias J, Leyva A, Paz-Ares J (2001) A conserved MYB transcription factor inv phosphate starvation signaling both in vascular plants and in unicellular algae. Genes Dev 15(16):2122–2133 g y ( ) p phosphate starvation signaling both in vascular plants and in unicellular algae. Genes Dev 15(16):2122–2133 tion signaling both in vascular plants and in unicellular algae. Genes Dev 15(16):2122–2133 26. Sardans J, Peñuelas J (2004) Increasing drought decreases phosphorus availability in an evergreen Mediterranean forest. Plant Soil 267(1):367–377 26. Sardans J, Peñuelas J (2004) Increasing drought decreases phosphorus availability in an evergreen Mediterranean forest. Plant Soil 267(1):367–377 Page 12/20 Page 12/20 27. Sardans J, Peñuelas J (2012) The role of plants in the effects of global change on nutrient availability and stoichiometry in the plant-soil system. Plant Physiol 160(4):1741–1761 28. Shemi R, Wang R, Gheith E-S, Hussain HA, Cholidah L, Zhang K, Zhang S, Wang L (2021a) Role of exogenous-applied salicylic acid, zinc and glycine betaine to improve drought-tolerance in wheat during reproductive growth stages. BMC Plant Biol 21(1):1–15 29. Shemi R, Wang R, Gheith E-S, Hussain HA, Hussain S, Irfan M, Cholidah L, Zhang K, Zhang S, Wang L (2021b) Effects of salicylic acid, zinc and glycine betaine on morpho-physiological growth and yield of maize under drought stress. Sci Rep 11(1):1–14 30. Shubhra DJ, Goswami CL, Munjal R (2004) Influence of phosphorus application on water relations, biochemical parameters and gum content in cluster bean under water deficit. Biol Plant 48(3):445–448 31. Smith MR, Dinglasan E, Veneklaas E, Polania J, Rao IM, Beebe SE, Merchant A (2022) Effect of Drought and Low P on Yield and Nutritional Content in Common Bean. Frontiers in Plant Science 13 31. Smith MR, Dinglasan E, Veneklaas E, Polania J, Rao IM, Beebe SE, Merchant A (2022) Effect of Drought and Low P on Yield and Nutritional Content in Common Bean. Frontiers in Plant Science 13 32. Suriyagoda L, De Costa W, Lambers H (2014) Growth and phosphorus nutrition of rice when inorganic fertiliser application is partly replaced by straw under varying moisture availability in sandy and clay soils. Plant Soil 384(1):53–68 32. References Wen Z, Pang J, Tueux G, Liu Y, Shen J, Ryan MH, Lambers H, Siddique KH (2020) Contrasting patterns in biomass allocation, root morphology and mycorrhizal symbiosis for phosphorus acquisition among 20 chickpea genotypes with different amounts of rhizosheath carboxylates. Funct Ecol 34(7):1311–1324 39. Xia Z, He Y, Zhou B, Korpelainen H, Li C (2020) Sex-related responses in rhizosphere processes of dioecious Populus cathayana exposed to drought and low phosphorus stress. Environ Exp Bot 175:104049 40. Xia Z, Zhang S, Wang Q, Zhang G, Fu Y, Lu H (2021) Effects of root zone warming on maize seedling growth and photosynthetic characteristics under different phosphorus levels.Frontiers in Plant Science12 41. Zhang H, Shi L, Lu H, Shao Y, Liu S, Fu S (2020) Drought promotes soil phosphorus transformation and reduces phosphorus bioavailability in a temperate forest. Sci Total Environ 732:139295 42. Zia R, Nawaz MS, Siddique MJ, Hakim S, Imran A (2021) Plant survival under drought stress: Implications, adaptive responses, and integrated rhizosphere management strategy for stress mitigation. Microbiol Res 242:126626 Figures Page 13/20 Figure 1 Effects of water and P application levels on the shoot length, shoot fresh weight (SFW), and shoot dry weight (SDW) of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Figure 1 Figure 1 Effects of water and P application levels on the shoot length, shoot fresh weight (SFW), and shoot dry weight (SDW) of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Page 14/20 Page 14/20 Figure 2 Effects of water and P application levels on the number of leaves of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Effects of water and P application levels on the number of leaves of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Figure 3 Effects of water and P application levels on the shoot phosphorus contents and root phosphorus contents of the ZD958 and XY335 Maiz cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05 Figure 3 Effects of water and P application levels on the shoot phosphorus contents and root phosphorus contents of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05 Page 15/20 Figure 4 Effects of water and P application levels on the malondialdehyde (MDA) content and Hydrogen peroxide (H2O2) production of the ZD XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the sig difference among treatments at p< 0.05. Figure 4 Figure 4 Effects of water and P application levels on the malondialdehyde (MDA) content and Hydrogen peroxide (H2O2) production of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Page 16/20 Figure 5 Effects of water and P application levels on the superoxide dismutase (SOD), peroxidase (POD), catalase (CAT) activit XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars re difference among treatments at p< 0.05. Figure 5 Effects of water and P application levels on the superoxide dismutase (SOD), peroxidase (POD), catalase (CAT) act XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bar difference among treatments at p< 0.05. Figure 5 Effects of water and P application levels on the superoxide dismutase (SOD), peroxidase (POD), catalase (CAT) activities of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Page 17/20 Page 17/20 Page 17/20 Figure 6 Effects of water and P application levels on the total protein contents, free proline and total soluble sugar content of the Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent t difference among treatments at p< 0.05. Figure 6 Effects of water and P application levels on the total protein contents, free proline and total soluble sugar content o Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars repres Figure 6 Effects of water and P application levels on the total protein contents, free proline and total soluble sugar content of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant Fi 6 Figure 6 Figure 6 Effects of water and P application levels on the total protein contents, free proline and total soluble sugar content of the ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Page 18/20 Figure 7 Effects of water and P application levels on the relative expression of ZmNAC and ZmPHR1genes in ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Figure 7 Figure 7 Effects of water and P application levels on the relative expression of ZmNAC and ZmPHR1genes in ZD958 and XY335 Maize cultivars. Error bars indicate the ± SE of three replicates. Different alphabetical letters above error bars represent the significant difference among treatments at p< 0.05. Page 19/20 Figure 8 Pearson’s correlation among traits of maize hybrids cultivars (ZD958 at low-left diagonal in round shape, XD335 at upper-right diagonal with digits) under interactive levels of water and P application. The scale bar on the right indicates the intensity of the correlation from 1 (highest positive in red) to –1 (highest negative in dark blue). Trait notations: SL, shoot length; SFW, shoot fresh weight; SDW, shoot dry weight; NL, no. of leaves; TRL, total root length; RFW, root fresh weight; RDW, root dry weight; SRL, specific root length; RTD, root tissue density; RMF, root mass fraction, RSA, root surface area; RAD, root average diameter; RV, root volume; R/S ratio, root/shoot ratio; SP, shoot phosphorus; RP, root phosphorus; MDA, malondialdehyde; H2O2, hydrogen peroxide; SOD,  superoxide dismutase; POD, peroxidase; CAT, catalase; TPC, total protein contents; FP, free proline; TSS, total soluble sugar. Figure 8 Pearson’s correlation among traits of maize hybrids cultivars (ZD958 at low-left diagonal in round shape, XD335 at upper-right diagonal with digits) under interactive levels of water and P application. The scale bar on the right indicates the intensity of the correlation from 1 (highest positive in red) to –1 (highest negative in dark blue). Trait notations: SL, shoot length; SFW, shoot fresh weight; SDW, shoot dry weight; NL, no. of leaves; TRL, total root length; RFW, root fresh weight; RDW, root dry weight; SRL, specific root length; RTD, root tissue density; RMF, root mass fraction, RSA, root surface area; RAD, root average diameter; RV, root volume; R/S ratio, root/shoot ratio; SP, shoot phosphorus; RP, root phosphorus; MDA, malondialdehyde; H2O2, hydrogen peroxide; SOD,  superoxide dismutase; POD, peroxidase; CAT, catalase; TPC, total protein contents; FP, free proline; TSS, total soluble sugar. Page 20/20

https://openalex.org/W2024387180

https://orca.cardiff.ac.uk/id/eprint/58575/1/PLoS%20One%202014.pdf

English

Dynamic Visualization of Dendritic Cell-Antigen Interactions in the Skin Following Transcutaneous Immunization

PloS one

cc-by

Introduction [11]. Cubosomes are a dispersion of the cubic liquid crystalline phase that retain the unique nanostructure of the parent cubic phase; a highly twisted, continuous lipid bilayer and two congruent non-intersecting water channels [12]. It has been reported that the penetration enhancing effect of cubosomes is due to the lipids of the particles forming a mixture with the lipids of the SC due to their similar cubic phase structure and thereby fluidizing the SC [13,14]. Cubosomes have been used successfully for the transder- mal delivery of several drugs including indomethacin [14] and hydrophilic plant extracts from Berberis koreana [15]. Transcutaneous immunization (TCI) is a novel approach to deliver peptide, proteins, DNA or viral particulate vaccines into the skin to the abundant resident antigen presenting cells (APC) such as Langerhans’s cells (LC) in the epidermis and dermal dendritic cells (DDC) in the dermis. LC are distributed throughout the epidermis and make up approximately 3–5% of the total cells present [1] and have been reported to be regulators of both tolerance and immunity [2]. DDC reside in the dermis and are assumed to be key regulators of cutaneous adaptive immune responses [3]. LC have traditionally been discriminated from DDC by their location, the presence of Birbeck granules and their high langerin (CD207+) expression [4]. However CD207+ expression has also been observed on DDC populations [5]. It has been reported that these CD207+ DC are involved in cross- presentation of antigens and can potentially stimulate antigen- specific cytotoxic T lymphocyte (CTL) and T helper 1 cell (Th1) immune responses [6,7,8] whereas CD2072 DDC were found to induce a primarily CD4+ T cell response [9]. Another approach to overcoming the barrier provided by the SC is to transiently perforate skin using microneedles (MN). The short needles (less than 1 mm) in microneedle arrays disrupt the SC, forming transient micro-channels that facilitate the delivery of drugs [16], macromolecules [17,18] or vaccines [18,19] to the deeper tissue. MN possess the potential to revolutionise the field of vaccine delivery due to low manufacturing and product distribu- tion costs [20], the fact they may not require skilled vaccine- administration expertise, are painless to use [21] and will likely have wide public acceptance [22]. Abstract Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] Dynamic Visualization of Dendritic Cell-Antigen Interactions in the Skin Following Transcutaneous Immunization Teerawan Rattanapak1, James C. Birchall2, Katherine Young1, Atsuko Kubo3, Sayumi Fujimori3, Masaru Ishii4, Sarah Hook1* 1 School of Pharmacy, University of Otago, Dunedin, New Zealand, 2 School of Pharmacy and Pharmaceutical Sciences, Cardiff University, Cardiff, United Kingdom, 3 Laboratory of Cellular Dynamics, Immunology Frontier Research Center, Osaka University, Osaka, Japan, 4 Department of Immunology and Cell Biology, Graduate School of Medicine and Frontier Biosciences and Laboratory of Cellular Dynamics, Immunology Frontier Research Center, Osaka University, Osaka, Japan February 2014 | Volume 9 | Issue 2 | e89503 Abstract Delivery of vaccines into the skin provides many advantages over traditional parenteral vaccination and is a promising approach due to the abundance of antigen presenting cells (APC) residing in the skin including Langerhans cells (LC) and dermal dendritic cells (DDC). However, the main obstacle for transcutaneous immunization (TCI) is the effective delivery of the vaccine through the stratum corneum (SC) barrier to the APC in the deeper skin layers. This study therefore utilized microneedles (MN) and a lipid-based colloidal delivery system (cubosomes) as a synergistic approach for the delivery of vaccines to APC in the skin. The process of vaccine uptake and recruitment by specific types of skin APC was investigated in real-time over 4 hours in B6.Cg-Tg (Itgax-EYFP) 1 Mnz/J mice by two-photon microscopy. Incorporation of the vaccine into a particulate delivery system and the use of MN preferentially increased vaccine antigen uptake by a highly motile subpopulation of skin APC known as CD207+ DC. No uptake of antigen or any response to immunisation by LC could be detected. Citation: Rattanapak T, Birchall JC, Young K, Kubo A, Fujimori S, et al. (2014) Dynamic Visualization of Dendritic Cell-Antigen Interactions in the Skin Following Transcutaneous Immunization. PLoS ONE 9(2): e89503. doi:10.1371/journal.pone.0089503 Editor: Susan Kovats, Oklahoma Medical Research Foundation, United States of America Editor: Susan Kovats, Oklahoma Medical Research Foundation, United States of America Received November 27, 2013; Accepted January 21, 2014; Published February 24, 2014 Received November 27, 2013; Accepted January 21, 2014; Published February 24, 2014 tanapak et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits tion, and reproduction in any medium, provided the original author and source are credited. Copyright:
2014 Rattanapak et al. This is an open-access article distributed under the terms of the Creative Commons Attribu unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This research was supported by a University of Otago Research Grant, the Maurice Wilkins Centre for Molecular Biodiscovery, a NZ-JSPS Bilateral Partnerships & Scientist Exchange; Grants-in-Aid for Encouragement of Young Scientists (A) (22689030), for Scientific Research on Innovative Areas (22113007), and by the FIRST Program from the Ministry of Education, Science, Sports and Culture of Japan. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Materials and Methods [24,28]. In this study, a combination of MN and cubosomes was therefore employed to overcome the barrier imposed by the SC and enhance vaccine delivery into the skin. Materials Phytantriol was purchased from A & E Connock (Hampshire, England). Poloxamer 407 (LutrolH F127) was obtained from BASF (Ludwigshafen, Germany) and 1,2-propandiol (Propylene Glycol, purity$99.5%) from Merck (Darmstadt, Germany). CD8 Oval- bumin peptide (SIINFEKL, OVA257–264) and CD4 Ovalbumin peptide (ISQAVHAAHAEINEAGR, OVA323–339) and 5(6)-tetra- methylcarboxy rhodamine-labelled OVA257–264 (TMR-SIIN- FEKL) was purchased from Mimotopes (Clayton, Australia). Monophosphoryl Lipid A from Salmonella Minnesota RE 595 was purchased from Sigma-Aldrich (Missouri, USA). Purified saponin (Quil-A) was purchased from Brenntag Biosector (Frederikssund, Denmark). All other chemicals were of analytical grade. Ethics Statement All experiments were approved by the Animal Ethics Commit- tee, University of Otago or by the Animal Experimental Committee of Osaka University. While TCI has been investigated by a number of different research groups, no studies have been carried out investigating the dynamics of interactions between vaccines and APC at the cellular level in real-time and in living animals. As it has been reported that the different APC subsets in the skin may promote qualitatively different immune responses, such interactions may be crucial in determining the outcome of TCI and by understanding these interactions it may be possible to design more effective vaccines. This lack of such data is due to the limitations of traditional imaging techniques such as fluorescent microscopy and confocal laser scanning microscopy (CLSM) which have a reduced ability to visualize into deep turbid tissue [29,30]. However, a combined approach utilizing advanced fluorescent labelling techniques and two-photon microscopy (2 PM) has been widely used in structural and functional studies of deep tissue [31]. Another key advantage of 2 PM over CLSM is reduced photo-breaching at out-of-focal plane regions [32]. Additionally, second harmonic generation (SHG) occurs concur- rently with 2 PM [33]. SHG is not generated from absorption but from hyper-Rayleigh scattering by the two low energy photons [34]. SGH occurs when light interacts with asymmetric macro- molecular arrangements such as collagen fibres [35]. Consequent- ly, the overall purpose of this study was to visualize TCI in real- time with particular emphasis on interactions with different APC populations in the skin using 2 PM and also to examine the ability of a combined approach of cubosomes and MN to induce immune responses. Introduction Numerous studies have investigated the ability of MN to deliver vaccines into the skin [23,24,25] demonstrating the effectiveness of this approach as compared to subcutaneous [26,27] or intramuscular vaccination The major barrier for transcutaneous delivery of vaccines is the outer protective layer of skin; the stratum corneum (SC). One of the most common methods for enhancing vaccine permeation into skin is the utilization of lipid vesicles [10] for example, cubosomes February 2014 | Volume 9 | Issue 2 | e89503 February 2014 | Volume 9 | Issue 2 | e89503 1 PLOS ONE | www.plosone.org Imaging Transcutaneous Immunization Figure 1. Two-photon microscopy (2 PM) visualization of Langerhans and dermal dendritic EYFP-CD11c+ cells in mouse skin. (A) XZ cross-sectional image of intact skin. XY images of (B) skin surface autofluorescence, (C) Langerhans cells (LC) localised in the epidermis and (D) dermal dendritic cells (DDC) in the dermal layer. Scale bar is 50 mm. doi:10.1371/journal.pone.0089503.g001 Figure 1. Two-photon microscopy (2 PM) visualization of Langerhans and dermal dendritic EYFP-CD11c+ cells in mouse skin. (A) XZ cross-sectional image of intact skin. XY images of (B) skin surface autofluorescence, (C) Langerhans cells (LC) localised in the epidermis and (D) dermal dendritic cells (DDC) in the dermal layer. Scale bar is 50 mm. doi:10.1371/journal.pone.0089503.g001 Microneedle Arrays Microneedle arrays fabricated from biocompatible polycarbon- ate polymers using an injection moulding process were supplied by Professor James Birchall (School of Pharmacy and Pharmaceutical Sciences, Cardiff University, UK). The 13-needles (670-mm/6 needles, 520-mm/6 needles and 335-mm/1 needle) were arranged in concentric circles [36]. February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org 2 Imaging Transcutaneous Immunization Figure 2. Impact of TCI on LC. Two-photon microscopy (2 PM) visualization of the distribution of LC (green) and formulation (red) of unimmunized mice (A, B) or following application of (C, D) TMR-SIINFEKL-loaded cubosomes for 40 min (A, C) and 240 min (B representative of groups of three mice. Scale bar is 100 mm. (E) The number of LC/mm2 in the epidermis in untreated controls (con the application of aqueous TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosom Figure 2. Impact of TCI on LC. Two-photon microscopy (2 PM) visualization of the distribution of LC (green) and formulation (red) in the epidermis of unimmunized mice (A, B) or following application of (C, D) TMR-SIINFEKL-loaded cubosomes for 40 min (A, C) and 240 min (B, D). Images are representative of groups of three mice. Scale bar is 100 mm. (E) The number of LC/mm2 in the epidermis in untreated controls (control) or following the application of aqueous TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes (cubosomes) February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org 3 Imaging Transcutaneous Immunization in the presence and absence of MN for 40 and 240 min. Cells were quantified from three randomised images from one mouse per group in three independent experiments. Data presented are the mean+SD. doi:10.1371/journal.pone.0089503.g002 Transcutaneous Immunization Adoptive transfer was performed as described previously [39]. Briefly 46106 lymphocytes from OT-I and OT-II were injected into the tail vein of each recipient C57BL/6 mouse a day prior to TCI. The following day the C57BL/6 mice were anaesthetised and were close clipped. Self-adhering foam base (RestonTM, 3 M), was attached on the skin of mouse to form a reservoir. Vaccine formulations, 100 mL containing 100 mg of CD4+ peptide, 100 mg of CD8+ peptide, 40 mg of QA and 20 mg of MPL formulated in cubosomes (2 mg of lipid) or in Milli-Q water or cubosomes containing adjuvants only (control) were applied on the skin inside the reservoir which was then covered with an occlusive dressing (Opsite-FlexifixTM, Smith & Nephew Medical Limited and BlendermTM, 3 M). For experiments utilizing MN, the micronee- dle array was applied manually to the prepared skin before application of formulations. The reservoirs were removed after 6 hr. The immunization procedure was repeated at day 14 and then at day 18 the mice were sacrificed and regional lymph nodes taken. Single cell suspensions were then prepared from all harvested lymphoid tissues. Aliquots of single cell suspensions of lymphocytes were incubated with anti-CD16/CD32 before being stained with anti-CD4-V500, anti-CD8-PE-Cy7, anti-Va2-PE and anti-Vb5-biotin. This was followed by staining with SA-Percp- Cy5.5 and propidium iodide. Antibodies and propidium iodide were purchased from BD Biosciences Pharmingen (California, USA). Cells were analysed on a FACSCantoTM II (BD Biosci- ences) and data was analysed using FlowJo 8.6 (Tree Star Inc). Mice For 2 PM experiments, 6–10 week old male and female B6.Cg- Tg (Itgax- Enhanced Yellow Fluorescent Protein) 1 Mnz/J mice originally purchased from The Jackson Laboratory (Maine, USA) were maintained under specific pathogen-free conditions at the Osaka University animal facility under an approved protocol from the Animal Experimental Committee of Osaka University. For other studies, male C57BL/6, 6–8 weeks old were utilized, bred and maintained under specific pathogen-free condition at the HTRU, Dunedin, New Zealand. Two-photon Microscopy Prior to performing the imaging experiments, the hair on the lower back of mice was removed with an electronic hair clipper. Any remaining hair was removed using depilatory cream (NairH, Carter Products, NY). The skin was wiped with moist cotton to remove all the depilatory cream. The imaging system was an A1RMP multiphoton microscope (Nikon, Japan) driven by a Chameleon Vision II Ti:Sapphire laser (Coherent, CA, USA) tuned to 920 nm with an inverted microscope equipped with a 206 oil immersion objective (CFI- Plan-Fluor, N.A. 0.75, Nikon). Laser power at the acquisition area was 25 mW. Fluorescent probes were detected through bandpass emission filters. Formulations were detected using a 561/25 nm filter and collagen using a 440/25 nm filter. Images were acquired from 20 to 45 z-planes, spaced 5 mm apart every 5 min for up to 240 min. Snapshot images were acquired using scanning averages of 2 scans. Raw imaging data was processed with Imaris software (Bitplane AG, Zu¨rich, Switzerland). To assess vaccine penetration, 2 PM images were captured in the XZ-plane using the ortho slicer function in the Imaris software. The microscope system was enclosed in an environmental compartment in which anesthetised mice were warmed by heated air. In vivo Uptake of Formulations by Skin LC and DC The immunization procedures were similar to that described in the previous section. However, the testing area was protected by paper and then covered with an occlusive dressing and the cage was covered to protect from light. After euthanasia, a section of the immunization site was rinsed with PBS to remove residual formulation and then harvested. The skin was macerated and incubated at 37uC for 1 hr in collagenase type IA (C9891, Sigma- Aldrich, St. Louis, MO) digestion buffer. A single cell suspension was fixed in 8% paraformaldehyde at room temperature. Cells were then permeabilized using saponin (S-7900, Sigma-Aldrich, St. Louis, Mo) buffer and incubated with anti-CD16/CD32 before being stained with anti-CD207-Alexa Flour 488 (eBioscience, San Diego, CA) for 30 minutes followed by washing and staining with anti-CD11c-APC, anti-CD11b-PE-Cy7 (BD PharMingen, San Diego, CA) and anti-F4/80- Brilliant Violet 421 (Biolegend, San Diego, CA). Cells were then analysed using a FACSCantoTM II flow cytometer (BD Biosciences, San Jose, CA). Data was analysed using FlowJo 7.6 analysis software (Tree Star, Inc., Oregon, USA). Antigen/adjuvant-loaded cubosomes with a size of 158 to182 nm were formulated using the lipid precursor method [37]. Briefly, phytantriol, poloxamer 407, propylene glycol and MPL were weighed into a 20-mL scintillation vial and dissolved in 5 mL chloroform. Rhodamine-labeled cubosomes were prepared by adding octadecyl rhodamine B chloride (R-18, Invitrogen Molecular Probes, OR, USA) into the scintillation vial prior to preparation of the cubosomes with a mass ratio of dye to lipid of 1:2000. Chloroform was evaporated under a stream of nitrogen at 45uC. The concentrated actives solution (0.8 mg/20 mL Quil-A (QA) and 2 mg/20 mL peptide) and glass beads were added and mixed by shaking until visually homogenous. Milli-Q water (45uC) was gradually added into the scintillation vial to a final volume of 1 mL while vortexing for 10 min. Peptide entrapment was approximately 23% as previously reported [38]. Interferon-c Assay Cells were quantified from three randomised images from one mouse per group in three independent experiments. Data presented are the mean+SD. ***P,0.001 compared to the control group (two-tailed unpaired Student’s t-test). doi:10.1371/journal.pone.0089503.g003 TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes (cubosomes) in the presence and absence of MN at 40 and 240 min. Cells were quantified from three randomised images from one mouse per group in three independent experiments. Data presented are the mean+SD. ***P,0.001 compared to the control group (two-tailed unpaired Student’s t-test). doi:10.1371/journal.pone.0089503.g003 TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes (cubosomes) in the presence and absence of MN at 40 and 240 min. Cells were quantified from three randomised images from one mouse per group in three independent experiments. Data presented are the mean+SD. ***P,0.001 compared to the control group (two-tailed unpaired Student’s t-test). doi:10.1371/journal.pone.0089503.g003 Figure 4. Formulation-DDC colocalization. Two-photon microscopy (2 PM) visualization of collagen (blue), DDC (green) and vaccine (red) in the dermis of untreated mice (A) or 40 min following application of an aqueous TMR-SIINFEKL mixture (B, E), rhodamine-labelled cubosomes (C, F), or TMR-SIINFEKL-loaded cubosomes (D, G) to intact (B-D) or MN pretreated (E-G) skin. Cross sections (XZ, far left) and 3-D images (XYZ) are shown as well as merged images and scatter 2D plots of voxel intensities (far right panel) in the red and green channels for selected DC. Scale bar is 100 mm. doi:10.1371/journal.pone.0089503.g004 Figure 4. Formulation-DDC colocalization. Two-photon microscopy (2 PM) visualization of collagen (blue), DDC (green) and vaccine (red) in the dermis of untreated mice (A) or 40 min following application of an aqueous TMR-SIINFEKL mixture (B, E), rhodamine-labelled cubosomes (C, F), or TMR-SIINFEKL-loaded cubosomes (D, G) to intact (B-D) or MN pretreated (E-G) skin. Cross sections (XZ, far left) and 3-D images (XYZ) are shown as well as merged images and scatter 2D plots of voxel intensities (far right panel) in the red and green channels for selected DC. Scale bar is 100 mm. doi:10.1371/journal.pone.0089503.g004 Figure 4. Formulation-DDC colocalization. Two-photon microscopy (2 PM) visualization of collagen (blue), DDC (green) and vaccine (red) in the dermis of untreated mice (A) or 40 min following application of an aqueous TMR-SIINFEKL mixture (B, E), rhodamine-labelled cubosomes (C, F), or TMR-SIINFEKL-loaded cubosomes (D, G) to intact (B-D) or MN pretreated (E-G) skin. Interferon-c Assay c y A 100 mL aliquot of a single cell suspension of 26106 lymphocytes/mL was plated in triplicate a 96-well round bottomed plate for restimulation with anti-CD3 (10 mg/mL), OVA (20 mg/mL) and IL-2 (2 ng/mL) or IL-2 alone (2 ng/mL). The plates were incubated for 3 days in a CO2 incubator (5% CO2 at 37uC) (HERAcell incubator, Heraeus, Hanau, Germany). On day 4, 100 mL of the cell culture supernatant was removed and stored at 220uC until analysis. Interferon c was measured using a BDTM Cytometric Bead Array (CBA) Mouse T Cell Cytokine Kit according to the manufacturer’s instructions. Formulations (diluted 1:20 in sterile MilliQ-water) were applied onto the skin with excess formulation being removed after 2 min by capillary action using a tissue. For experiments utilizing MN, the MN were applied manually to the prepared skin before application of formulations. A glass slide was attached to the skin using tissue glue (Vetbond Tissue Adhesive, 3 M Animal Care Products, MN, USA). Mice were then placed on a custom- designed stage. Mice were anesthetised with isoflurane (Escain; 2% vaporized in 100% oxygen) throughout the experimental period. February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org 4 Imaging Transcutaneous Immunization Figure 3. Impact of TCI on DDC. Two-photon microscopy (2 PM) visualization of DDC (green) and collagen (blue) in the dermis i mice (A, B) or following application of MN and TMR-SIINFEKL-loaded cubosomes (C, D) for 40 min (A, C) and 240 min (B, D). Images ar of groups of three mice. Scale bar is 200 mm. (E) The number of DDC/mm2 in untreated controls (control) or following the applicat PLOS ONE | www.plosone.org 5 February 2014 | Volume 9 | Is Figure 3. Impact of TCI on DDC. Two-photon microscopy (2 PM) visualization of DDC (green) and collagen (blue) in the dermis in unimmunized mice (A, B) or following application of MN and TMR-SIINFEKL-loaded cubosomes (C, D) for 40 min (A, C) and 240 min (B, D). Images are representative of groups of three mice. Scale bar is 200 mm. (E) The number of DDC/mm2 in untreated controls (control) or following the application of aqueous February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org 5 Imaging Transcutaneous Immunization TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes (cubosomes) in the presence and absence of MN at 40 and 240 min. DDC, not LC, Respond Initially to Transcutaneously Applied Antigen The two main populations of skin CD11c+ APC, the LC and DDC, reside in distinct locations (Fig. 1. and Movie S1). Two criteria were used to distinguish LC from DDC; location within the skin and physical morphology. LC were widespread in the epidermis [40] which could be clearly identified by the absence of collagen fibres (Fig. 1). Figure 1 B, C and D are XY images taken at the skin surface, through the epidermis and through the dermis, respectively. The unique morphology of LC, with their dendrites extending from the cell body, is clearly visible in Fig. 1 C. DDC had no visible dendrites and were found in the dermis (Fig. 1 D). Interestingly, DDC could also be found extending into the epidermis in XZ cross-sectional images (Fig. 6 and Movie S4) capturing and taking up formulations from the upper layers of the skin. After MN insertion, the epidermis and the dermis were transiently disrupted (Movie S2). Due to the technical require- ments for application of formulations and set-up of the microscope, the first images were taken 30–40 minutes after TCI. Formulation was observed to penetrate from the micro- channel over time and DDC were found in close proximity to areas of formulation (Movie S2 and [38]). Imaging Transcutaneous Immunization Data presented are the mean+SEM (n = 9). doi:10.1371/journal.pone.0089503.g005 doi:10.1371/journal.pone.0089503.g005 Imaging Transcutaneous Immunization Imaging Transcutaneous Immunization Figure 5. DDC-formulation colocalization. Pearson’s correlation coefficient calculated from skin treated with: an aqueous peptide mixture (N), rhodamine-labeled cubosomes (m) and TMR peptide- loaded cubosomes (&). The black and red colour represent intact and MN pretreated skin respectively. Unimmunized intact skin served as control (#). Cells were quantified from three randomised images from one mouse in three independent experiments. Data presented are the mean+SEM (n = 9). doi:10.1371/journal.pone.0089503.g005 reduced in all samples (Fig. 2 B and D) due to photobleaching. Pretreatment of skin with MN before immunization similarly did not impact on LC density (Fig. 2 E.). Subsequently, the ability of LC to capture formulation was examined. Detailed analysis of images using ImarisColoc (Bit- plane) was carried out to quantify LC-antigen colocalization. Pearson’s coefficient was used to describe the correlation of the intensity distribution of the green fluorescence of the LC and the red fluorescence from the vaccine formulations. A Pearson’s coefficient of 1 indicates a significant positive correlation; 0 indicates no significant correlation and 21 indicates complete negative correlation. Values between 0.5 and 1 would indicate formulation-cell colocalization [41]. The calculated Pearson’s coefficient of LC from all vaccine experiments was zero throughout the experimental period. It was therefore established that no detectable formulation-LC colocalization occurred. DDC were easily distinguishable in the dermis of CD11c+- EYFP transgenic mice. The size of DDC could not be precisely measured in these highly motile cells as they were constantly changing shape and moving both horizontally and vertically through the skin, squeezing between other cells and making contact with other DDC (Movie S3). An increase in the number of DDC was found in the skin of all mice treated with formulations when compared to unimmunized skin (control) at 40 min and 240 min following TCI (Fig. 3). However, this was significant only for mice immunized using TMR-peptide in cubosomes 240 min following immunization (P,0.001). MN pretreatment resulted in a significant increase in DDC at both 40 and 240 min post- application of TMR-peptide in cubosomes (P,0.001). Figure 5. DDC-formulation colocalization. Pearson’s correlation coefficient calculated from skin treated with: an aqueous peptide mixture (N), rhodamine-labeled cubosomes (m) and TMR peptide- loaded cubosomes (&). The black and red colour represent intact and MN pretreated skin respectively. Unimmunized intact skin served as control (#). Cells were quantified from three randomised images from one mouse in three independent experiments. Data and Statistical Analysis Data analysis was carried out using Microsoft Excel, Version 2010. The results are expressed as mean 6 SD and mean 6 SEM. GraphPad Prism Version 5.0 was used to examine the statistically significant differences using the two-tailed unpaired Student’s t-test or the one-way analysis of variance (ANOVA) followed by Tukey’s pairwise comparison. Formulation-DDC colocalization was then examined by 2 PM. XYZ sections of the dermis (Fig. 4) were examined in detail 40 min after TCI. Formulation of the vaccine into cubosomes resulted in the highest levels of antigen being detected in dermis. Antigen-DDC colocalization was examined in selected cells (Fig. 4 far right column) through 2D plots of voxel intensities. Areas of high fluorescence intensity (due to autofluorescence of skin appendages) were avoided. This analysis confirmed antigen (red channel)-DDC (green channel) colocalization. Quantitative anal- ysis of DDC-formulation colocalization was performed and expressed in terms of a Pearson’s correlation coefficient (Fig. 5) against time (40–240 min). The TMR-peptide in cubosome formulation achieved a Pearson’s coefficient above 0.5 after application to both intact and MN pretreated skin at all time points assessed. Interferon-c Assay Cross sections (XZ, far left) and 3-D images (XYZ) are shown as well as merged images and scatter 2D plots of voxel intensities (far right panel) in the red and green channels for selected DC. Scale bar is 100 mm. doi:10.1371/journal.pone.0089503.g004 February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org 6 Minor Population of Langerhin+DDC Takes up Antigen Following TCI APC were isolated from skin immunized transcutaneously in the presence and absence of MN pretreatment or from unimmunized control mice. Skin DC were identified based on CD11c expression and were then gated based on CD207 and CD11b expression (Fig. 7 A) into LC (CD207hi, CD11bhi), langerin+ DDC (CD207hi, CD11blo [1]) and langerin2 DDC (CD11b hi/lo CD207 lo [7,42]). LC comprised approximately 8–10% of the isolated skin DC (Fig. 7 C). A small population of langerin+ DC was also found in the full- thickness skin (3–4% of total DC in the skin) in agreement with previously published studies [43]. The largest population of DC in the skin were the Langerin2 DDC population (,70%). Following formulation application, while there was no significant change in the number of any of the DC subsets (Fig. 7 C), uptake of The physical morphology and number of the LC in intact and MN pretreated skin before and after immunisation was examined (Fig. 2). The cells were non-motile (determined from time series analysis over 4 hours) and had 4–5 dendrites projecting from the body of the cell. LC density in untreated skin was determined to be approximately 980 cells/mm2. Upon application of the vaccine to the skin, no change in LC morphology or density could be seen (Fig. 2 D and E) over the time period investigated (40–240 minutes). By 240 min fluorescence of the LC in the skin was PLOS ONE | www.plosone.org February 2014 | Volume 9 | Issue 2 | e89503 7 Imaging Transcutaneous Immunization Figure 6. Two-photon microscopy (2 PM) visualization of DDC uptake of antigen. (A) Orthogonal views (XY, XZ, YZ) through a DDC illustrating colocalization of DDC (green), TMR-SIINFEKL (red) and collagen (blue). Colocalization is visible as a yellow colour and orange represents the leakage of intense red fluorescence into the green channel. (B–C) 3D images of the DDC from (A). All channels are shown in (B), blue has been dropped out in (C) and both blue and green in (D). TMR-SIINFEKL uptake by the DDC is shown in (D) indicated by arrow. Scale bar is 20 mm. doi:10.1371/journal.pone.0089503.g006 Figure 6. Two-photon microscopy (2 PM) visualization of DDC uptake of antigen. (A) Orthogonal views (XY, XZ, YZ) through a DDC illustrating colocalization of DDC (green), TMR-SIINFEKL (red) and collagen (blue). Colocalization is visible as a yellow colour and orange represents the leakage of intense red fluorescence into the green channel. Minor Population of Langerhin+DDC Takes up Antigen Following TCI The ability of TCI with the OVA CD4 and CD8 minimal peptide epitopes either delivered in water or in cubosomes, with or without MN pretreatment to induce immune responses following a prime-boost regime was assessed (Fig. 8). The number of antigen- specific CD4+ T cell in the lymph nodes (Fig. 8 A) was low and not above that found in control mice immunised with a control cubosome formulation containing the adjuvants but no antigen (P.0.05). Mice immunized transcutaneously with peptide in water or peptide formulated in cubosomes had a measureable antigen- specific CD8+ T cell expansion as compared to the control cubosome formulation (Fig. 8 B). The combination of pretreat- ment with MN before the application of the formulation did not significantly enhance the antigen specific expansion of CD8+ T cells as compared to formulation applied to intact skin. Similar results were found in a functional assay where the ability of lymph node cells to produce IFN-c upon restimulation with antigen was Minor Population of Langerhin+DDC Takes up Antigen Following TCI (B–C) 3D images of the DDC from (A). All channels are shown in (B), blue has been dropped out in (C) and both blue and green in (D). TMR-SIINFEKL uptake by the DDC is shown in (D) indicated by arrow. Scale bar is 20 mm. doi:10.1371/journal.pone.0089503.g006 examined. Lymphocytes from mice immunized using MN and either cubosome or peptide in water produced significant levels of antigen specific IFN-c (Fig. 8 C). However when the formulations were applied to intact skin only lymphocytes from mice immunised with cubosomes produced significant levels of IFN-c (P,0.05). fluorescent peptide by skin DC (Fig. 7 B and D) was detected in the minor langerin+ DDC subset in mice treated with MN and peptide-loaded cubosomes (P,0.05). Consistent with the 2 PM antigen-DC colocalization studies, no uptake by LC was detected. fluorescent peptide by skin DC (Fig. 7 B and D) was detected in the minor langerin+ DDC subset in mice treated with MN and peptide-loaded cubosomes (P,0.05). Consistent with the 2 PM antigen-DC colocalization studies, no uptake by LC was detected. The ability of TCI with the OVA CD4 and CD8 minimal peptide epitopes either delivered in water or in cubosomes, with or without MN pretreatment to induce immune responses following a prime-boost regime was assessed (Fig. 8). The number of antigen- specific CD4+ T cell in the lymph nodes (Fig. 8 A) was low and not above that found in control mice immunised with a control cubosome formulation containing the adjuvants but no antigen (P.0.05). Mice immunized transcutaneously with peptide in water or peptide formulated in cubosomes had a measureable antigen- specific CD8+ T cell expansion as compared to the control cubosome formulation (Fig. 8 B). The combination of pretreat- ment with MN before the application of the formulation did not significantly enhance the antigen specific expansion of CD8+ T cells as compared to formulation applied to intact skin. Similar results were found in a functional assay where the ability of lymph node cells to produce IFN-c upon restimulation with antigen was fluorescent peptide by skin DC (Fig. 7 B and D) was detected in the minor langerin+ DDC subset in mice treated with MN and peptide-loaded cubosomes (P,0.05). Consistent with the 2 PM antigen-DC colocalization studies, no uptake by LC was detected. Imaging Transcutaneous Immunization (cubosomes) in the presence and absence of MN pretreatment. The results of the antigen uptake are given as. Data are mean+SEM and are representative of three independent experiments with 5 mice per group. *P,0.05 (two-tailed unpaired Student’s t-test). doi:10.1371/journal.pone.0089503.g007 Figure 7. Antigen uptake by skin APC. (A) Representative dot plots showing gating for flow cytometric analysis. CD11chi DC cells were selected and further gates determined on the basis of CD207 and CD11b expression. (B) Representative histograms of the uptake of vaccine; TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (Rho- damine) or TMR peptide-loaded cubosomes (cubosomes) by Langerin+ DDC following TCI. The solid grey histograms are Langerin+ DDC from unimmunised controls and the open histograms are Langerin+ DDC from immunised mice. (C) Number of LC, Langerin+ DDC and Langerin2 DDC at 240 min in untreated control mice (control) or in mice immunised with aqueous TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes (cubo- somes) in the presence and absence of MN pretreatment. (D) Uptake of antigen (median fluorescence intensity) by LC, Langerin+ DDC and Langerin2 DDC at 240 min in untreated control mice (control) or in mice immunised with aqueous TMR-SIINFEKL (peptide), rhodamine- by utilizing MN to deliver vaccine-loaded poly-D-L-lactide-co- glycolide (PGLA) nanoparticles to the skin APC. by utilizing MN to deliver vaccine-loaded poly-D-L-lactide-co- glycolide (PGLA) nanoparticles to the skin APC. LC present in the epidermis are the first APC that transcuta- neously applied vaccine antigens should encounter. However in the early stages of immunization examined here, LC were not found to be motile, the number of LC was not altered in response to the application of formulation and no antigen-LC colocalization could be detected either by 2 PM or flow cytometry. Sen et al. [46] similarly found that LC were completely immotile until 48 hr after the injection of adjuvants such as CpG and LPS and even then only a subpopulation of LC (approximately 15%) moved at a very low velocity. Studies of TCI using MN in human skin explants have also shown that LC responses occur over 24 to 48 hours [18,47]. The role of LC in the initiation of immune responses is conflicting and incompletely understood. Imaging Transcutaneous Immunization While LC have been shown to be able to extend their dendrites into the SC layer to capture antigen [48] and to be able to efficiently take up soluble antigen [49] other studies have shown they are not involved in priming immune responses in skin but are perhaps involved in tolerance [2]. The fact that LC do not appear to be involved in the immediate response in interesting and is perhaps suggestive of a more regulatory role. TCI using peptide loaded-cubosome formulations was however found to have an impact on DDC recruitment and this occurred earlier if MN pretreatment was employed. Two adjuvants with different immune activatory activities were included in the formulation in order to optimize immune stimulation. QA is a hydrophilic TLR-independent adjuvant that has recently been reported to activate the NLRP3-inflammasome [50] while MPL is a lipophilic TLR-4 ligand [51,52,53]. Cubosomes were chosen as the delivery system as their biphasic structure facilitates entrapp- ment of both adjuvants and the antigen. In addition particulate delivery systems have been shown themselves to activate the inflammasome [54]. Activation of the NALP3-inflammasome leads to release of IL-1b, a potent proinflammatory cytokine, which results in the recruitment of immature DC [55,56]. TLR-4 agonists activate NF-kB, a pro-inflammatory gene transcription regulator, which results in the up-regulation of the co-stimulatory molecules CD80 and CD86 [57] and the production of IP-10/ CXCL10 chemokines responsible for Th1 recruitment [58]. In this study we observed DC recruitment occurring 40 min after immunization with TMR-peptide in cubosomes (when MN were utilized) and within 4 hr when the vaccine was applied to the intact skin. This only occurred if the vaccine antigen and adjuvant were loaded into cubosomes, not when they were applied as an aqueous mixture. Previous investigations into the mechanism of vaccine permeation into skin revealed that the lipophilic cubo- somes are retained in the microchannels formed by MN creating an immunestimulatory depot and also facilitating the movement of unentrapped peptide antigen and adjuvant into the deeper layers of the skin to interact with APC [38]. The kinetics of the response is similar to that found following topical application of measles virus nucleoprotein onto mucosal tissue where DC were found to be recruited rapidly, peaking at 2 hr, while DC migration to the draining lymph node took place over 24 hr [59]. Discussion The aim of this study was to investigate the interface between TCI and the immune system using 2 PM and flow cytometry. TCI was accomplished using a lipid nanocarrier in combination with MN. Cubosomes were chosen as the carrier due to their ability to promote penetration of vaccine into the skin in vitro [11]. MN were used to temporarily disrupt the skin barrier and improve vaccine delivery to the epidermis and the dermis where LC and DDC reside. The synergistic effect of combining lipid particles and MN on immunological responses has been previously reported by Hirschberg et al. [44]. They indicated that the induction of antibody responses occurred only when mice were immunized with a combined approach using both vesicles and MN. Zaric et al. [45] also reported that robust immune responses were generated February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org 8 Imaging Transcutaneous Immunization Imaging Transcutaneous Immunization Interestingly, rhodamine labelled cubosomes containing the adjuvants MPL and QA but no antigen did not significantly effect DDC recruitment. This requirement for antigen suggests that DC-T cell interactions are required for the observed DDC recruitment. The skin contains Figure 7. Antigen uptake by skin APC. (A) Representative dot plots showing gating for flow cytometric analysis. CD11chi DC cells were selected and further gates determined on the basis of CD207 and CD11b expression. (B) Representative histograms of the uptake of vaccine; TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (Rho- damine) or TMR peptide-loaded cubosomes (cubosomes) by Langerin+ DDC following TCI. The solid grey histograms are Langerin+ DDC from unimmunised controls and the open histograms are Langerin+ DDC from immunised mice. (C) Number of LC, Langerin+ DDC and Langerin2 DDC at 240 min in untreated control mice (control) or in mice immunised with aqueous TMR-SIINFEKL (peptide), rhodamine-labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes (cubo- somes) in the presence and absence of MN pretreatment. (D) Uptake of antigen (median fluorescence intensity) by LC, Langerin+ DDC and Langerin2 DDC at 240 min in untreated control mice (control) or in mice immunised with aqueous TMR-SIINFEKL (peptide), rhodamine- labeled cubosomes (rhodamine) and TMR-SIINFEKL-loaded cubosomes February 2014 | Volume 9 | Issue 2 | e89503 9 PLOS ONE | www.plosone.org Imaging Transcutaneous Immunization Figure 8. Immune responses stimulated by TCI. Number of transgenic Va2+Vb5+ CD4+ (A) and Va2+Vb5+ CD mice vaccinated with 100 mg of CD4+ peptide, 100 mg of CD8+ peptide, 40 mg of QA and 20 mg of MPL formulated i Milli-Q water (Water) or with cubosomes containing only MPL and QA (Control) in the presence and absence of MN PLOS ONE | www.plosone.org 10 February 2014 Figure 8. Immune responses stimulated by TCI. Number of transgenic Va2+Vb5+ CD4+ (A) and Va2+Vb5+ CD8+ (B) T cells in lymph nodes of mice vaccinated with 100 mg of CD4+ peptide, 100 mg of CD8+ peptide, 40 mg of QA and 20 mg of MPL formulated in cubosomes (Cubosomes) or in Milli-Q water (Water) or with cubosomes containing only MPL and QA (Control) in the presence and absence of MN pretreatment. (C) Production of PLOS ONE | www plosone org 10 February 2014 | Volume 9 | Issue 2 | e89503 Figure 8. Immune responses stimulated by TCI. Number of transgenic Va2+Vb5+ CD4+ (A) and Va2+Vb5+ CD8+ (B) T cells in lymph Figure 8. Immune responses stimulated by TCI. Imaging Transcutaneous Immunization Number of transgenic Va2+Vb5+ CD4+ (A) and Va2+Vb5+ CD8+ (B) T cells in lymph nodes of mice vaccinated with 100 mg of CD4+ peptide, 100 mg of CD8+ peptide, 40 mg of QA and 20 mg of MPL formulated in cubosomes (Cubosomes) or in Milli-Q water (Water) or with cubosomes containing only MPL and QA (Control) in the presence and absence of MN pretreatment. (C) Production of PLOS ONE | www.plosone.org February 2014 | Volume 9 | Issue 2 | e89503 10 Imaging Transcutaneous Immunization interferon-c (IFN-c), expressed as fold increase over background, by cells isolated from lymph nodes of mice and re-stimulated in vitro with OVA or media (as a negative control). Data shown are the mean+SEM from three independent experiments with 3–4 mice in each experiment. *P,0.05, **P, 0.01 and ***P,0.001 (two-tailed unpaired Student’s t-test). doi:10.1371/journal.pone.0089503.g008 large numbers of skin resident memory T cells [60] and naı¨ve T cells [61]. It would be of interest to examine the interactions occurring between DC and T cell subsets in the skin using 2 PM. In conclusion the use of a lipid based delivery system in combination with MN can potentiate immune responses to peptide antigens delivered transcutaneously by facilitating uptake of vaccine by dermal dendritic cells. Further preclinical studies in more appropriate animal models (for example miniature pigs) or clinical studies in man will be necessary in order to determine the full potential of such vaccine formulations. cells [61]. It would be of interest to examine the interactions occurring between DC and T cell subsets in the skin using 2 PM. Antigen-DDC colocalization could be visualized by 2 PM and confirmed by calculation of Pearson’s correlation coefficient, in skin treated with TMR-peptide in cubosomes in the presence and absence of MN pretreatment. Current dogma suggests that DDC are unable to penetrate into the epidermis because of the barrier provided by the epidermis-dermis tight junction [62], although it has been reported that DC cluster around hair follicles and that dendrites can penetrate the basement membrane to sample antigen present in the epidermis [63]. Flow cytometry suggested that it was the langerin+ DDC subset that was taking up antigen. As the vaccine was not actively targeting these cells the preferential uptake of the vaccine must be related to the phenotype or physical location of this population of cells. p (MP4) Movie S4 Uptake of vaccine by DDC. A highly motile DDC can be seen taking up vaccine from the epidermis. Peptide is visible as red fluorescence, collagen is shown as blue fluorescence and green fluorescence represents EYFP-CD11c+ cells. Areas of antigen/DDC colocalization are yellow. (MP4) Imaging Transcutaneous Immunization Bursch et al [63] reported that langerin+ DDC have been identified penetrating into the epidermis therefore it may be that the physical location of these cells results in the preferential uptake. This population of cells has additionally been reported to trigger Th1 immune responses and induce predominantly cellular immunity, including antigen- specific CD8+ T cell [6]. It is possible that by utilizing MN the vaccine is targeted to a different subset of APC than it would be if applied to intact skin and that a qualitatively different immune response is produced. Author Contributions Conceived and designed the experiments: TR JCB MI SH. Performed the experiments: TR KY SF AK. Analyzed the data: TR JCB KY SF AK MI SH. Contributed reagents/materials/analysis tools: JCB MI SH. Wrote the paper: TR JCB MI SH. Conceived and designed the experiments: TR JCB MI SH. Performed the experiments: TR KY SF AK. Analyzed the data: TR JCB KY SF AK MI SH. Contributed reagents/materials/analysis tools: JCB MI SH. Wrote the paper: TR JCB MI SH. 6. Igya´rto´ Botond Z, Haley K, Ortner D, Bobr A, Gerami-Nejad M, et al. (2011) Skin-Resident Murine Dendritic Cell Subsets Promote Distinct and Opposing Antigen-Specific T Helper Cell Responses. Immunity 35: 260–272. 2. Romani N, Clausen BE, Stoitzner P (2010) Langerhans cells and more: langerin- expressing dendritic cell subsets in the skin. Immunol Rev 234: 120–141. 7. Henri S, Poulin LF, Tamoutounour S, Ardouin L, Guilliams M, et al. (2010) CD207+ CD103+ dermal dendritic cells cross-present keratinocyte-derived References antigens irrespective of the presence of Langerhans cells. J Exp Med 207: 189– 206. antigens irrespective of the presence of Langerhans cells. J Exp Med 207: 189– 206. 1. Merad M, Ginhoux F, Collin M (2008) Origin, homeostasis and function of Langerhans cells and other langerin-expressing dendritic cells. Nat Rev Immunology 8: 935–947. 8. Elnekave M, Furmanov K, Nudel I, Arizon M, Clausen BE, et al. (2010) Directly Transfected Langerin+ Dermal Dendritic Cells Potentiate CD8+ T Cell Responses following Intradermal Plasmid DNA Immunization. J Immunol 185: 3463–3471. 2. Romani N, Clausen BE, Stoitzner P (2010) Langerhans cells and more: langerin- expressing dendritic cell subsets in the skin. Immunol Rev 234: 120–141. 3. Dupasquier M, Stoitzner P, van Oudenaren A, Romani N, Leenen PJM (2004) Macrophages and Dendritic Cells Constitute a Major Subpopulation of Cells in the Mouse Dermis. J Investig Dermatol 123: 876–879. 9. Stoecklinger A, Eticha TD, Mesdaghi M, Kissenpfennig A, Malissen B, et al. (2011) Langerin+ Dermal Dendritic Cells Are Critical for CD8+ T Cell Activation and IgH c-1 Class Switching in Response to Gene Gun Vaccines. J Immunol 186: 1377–1383. 4. Valladeau J, Ravel O, Dezutter-Dambuyant C, Moore K, Kleijmeer M, et al. (2000) Langerin, a novel C-type lectin specific to Langerhans cells, is an endocytic receptor that induces the formation of Birbeck granules. Immunity 12: 71–81. 10. Bouwstra JA, Honeywell-Nguyen PL, Gooris GS, Ponec M (2003) Structure of the skin barrier and its modulation by vesicular formulations. Prog Lipid Res 42: 1–36. 5. Poulin LF, Henri S, de Bovis B, Devilard E, Kissenpfennig A, et al. (2007) The dermis contains langerin+ dendritic cells that develop and function indepen- dently of epidermal Langerhans cells. J Exp Med 204: 3119–3131. 11. Rattanapak T, Young K, Rades T, Hook S (2012) Comparative study of liposomes, transfersomes, ethosomes and cubosomes for transcutaneous immunisation: characterisation and in vitro skin penetration. J Pharm Pharma- col 64: 1560–1569. 6. Igya´rto´ Botond Z, Haley K, Ortner D, Bobr A, Gerami-Nejad M, et al. (2011) Skin-Resident Murine Dendritic Cell Subsets Promote Distinct and Opposing Antigen-Specific T Helper Cell Responses. Immunity 35: 260–272. 12. Rizwan SB, Hanley T, Boyd BJ, Rades T, Hook S (2009) Liquid crystalline systems of phytantriol and glyceryl monooleate containing a hydrophilic protein: Characterisation, swelling and release kinetics. J Pharm Sci 98: 4191–4204. 7. Henri S, Poulin LF, Tamoutounour S, Ardouin L, Guilliams M, et al. Supporting Information Movie S1 Two-photon microscopy (2 PM) visualization of EYFP-CD11c+ in mouse skin. Collagen is shown as blue fluorescence and green fluorescence represents stratum corneum or hair follicle autofluorescence and EYFP-CD11c+ cells. (MP4) Movie S2 Transient formation of a microchannel and penetration of vaccine following application of peptide loaded cubosomes (shown in red) to MN pretreated skin. A 3D image (top) and XZ cross sections (bottom) are shown. Collagen is visible as blue fluorescence and green fluorescence represents EYFP-CD11c+ cells. Bright yellow florescence is a result of autofluorescence of hair follicles. EYFP-CD11c+ DDC can be seen in close proximity to the vaccine formulation. (MP4) Movie S3 Rapid movement of DDC in normal skin. Collagen is shown as blue fluorescence and green fluorescence represents EYFP-CD11c+ cells. (MP4) Immune responses to vaccines applied transcutaneously to mice were examined and while there were perhaps some minor differences in the responses induced by the vaccines, all vaccines were able to stimulate weak but detectable responses. This is a limitation of using mice to examine immune responses to vaccines applied transcutaneously, as vaccination is much more effective due to the thin SC (approximately 9 mm in mice and human 20 mm in humans [64]) and high number of hair follicles (approximately 660/cm2 in mice and 11/cm2 in humans [65]). An additional issue is the removal of hair by methods such as plucking, shaving or through the use of a depillatory cream. All these methods, including the depillatory cream used here, can act to enhance vaccine penetration [66,67]. Therefore while mice are an invaluable tool for investigating cellular interactions occurring in the skin they are less useful for evaluating the immunological activity of potential vaccine formulations. Imaging Transcutaneous Immunization 13. Norlen L, Al-Amoudi A (2004) Stratum Corneum Keratin Structure, Function, and Formation: The Cubic Rod-Packing and Membrane Templating Model. J Investig Dermatol 123: 715–732. 41. Zinchuk V, Grossenbacher-Zinchuk O (2001) Quantitative Colocalization Analysis of Confocal Fluorescence Microscopy Images. Current Protocols in Cell Biology: John Wiley & Sons, Inc. J g 14. Esposito E, Cortesi R, Drechsler M, Paccamiccio L, Mariani P, et al. (2005) Cubosome Dispersions as Delivery Systems for Percutaneous Administration of Indomethacin. Pharml Res 22: 2163–2173. 42. Shklovskaya E, Roediger B, Fazekas de St. Groth B (2008) Epidermal and Dermal Dendritic Cells Display Differential Activation and Migratory Behavior While Sharing the Ability to Stimulate CD4+ T Cell Proliferation In Vivo. J Immunol 181: 418–430. 15. Kwon T, Lee H, Kim J, Shin W, Park S, et al. (2010) In vitro skin permeation of cubosomes containing water soluble extracts of Korean barberry. Colloid Journal 72: 205–210. 43. Nagao K, Ginhoux F, Leitner WW, Motegi S-I, Bennett CL, et al. (2009) Murine epidermal Langerhans cells and langerin-expressing dermal dendritic cells are unrelated and exhibit distinct functions. PNAS USA 106: 3312–3317. J 16. Banks SL, Pinninti RR, Gill HS, Paudel KS, Crooks PA, et al. (2010) Transdermal delivery of naltrexol and skin permeability lifetime after microneedle treatment in hairless guinea pigs. J Pharm Sci 99: 3072–3080. 44. Hirschberg H, van Kuijk S, Loch J, Jiskoot W, Bouwstra J, et al. (2012) A combined approach of vesicle formulations and microneedle arrays for transcutaneous immunization against hepatitis B virus. Eur J Pharm Sci 46: 1–7. 17. Coulman SA, Barrow D, Anstey A, Gateley C, Morrissey A, et al. (2006) Minimally invasive cutaneous delivery of macromolecules and plasmid DNA via microneedles. Current Drug Delivery 3: 65–75. 45. Zaric M, Lyubomska O, Touzelet O, Poux C, Al-Zahrani S, et al. (2013) Skin Dendritic Cell Targeting via Microneedle Arrays Laden with Antigen Encapsulated Poly-D-L-Lactide-co-Glycolide Nanoparticles Induces Efficient Anti-Tumour and Anti-Viral Immune Responses. ACS Nano. 7: 2042–2055. 18. Pearton M, Kang SM, Song JM, Kim YC, Quan FS, et al. (2010) Influenza virus-like particles coated onto microneedles can elicit stimulatory effects on Langerhans cells in human skin. Vaccine 28: 6104–6113. Anti-Tumour and Anti-Viral Immune Responses. ACS Nano. 7: 204 46. Sen D, Forrest L, Kepler TB, Parker I, Cahalan MD (2010) Selective and site- specific mobilization of dermal dendritic cells and Langerhans cells by Th1- and Th2-polarizing adjuvants. PNAS USA 107: 8334–8339. 19. Imaging Transcutaneous Immunization Song JM, Kim YC, Lipatov AS, Pearton M, Davis CT, et al. (2010) Microneedle delivery of H5N1 influenza virus-like particles to the skin induces long-lasting B- and T-cell responses in mice. Clin Vaccine Immunol 17: 1381–1389. 47. Pearton M, Kang S-M, Song J-M, Anstey AV, Ivory M, et al. (2010) Changes in Human Langerhans Cells Following Intradermal Injection of Influenza Virus- Like Particle Vaccines. PLoS ONE DOI: 10.1371/journal.pone.0012410. 20. Prausnitz MR, Mikszta JA, Cormier M, Andrianov AK (2009) Microneedle- Based Vaccines Vaccines for Pandemic Influenza. Springer Berlin Heidelberg. 369–393. 48. Kubo A, Nagao K, Yokouchi M, Sasaki H, Amagai M (2009) External antigen uptake by Langerhans cells with reorganization of epidermal tight junction barriers. J Exp Med 206: 2937–2946. 21. Kaushik S, Hord AH, Denson DD, McAllister DV, Smitra S, et al. (2001) Lack of Pain Associated with Microfabricated Microneedles. Anesthesia & Analgesia 92: 502–504. J p 49. Stoitzner P, Tripp CH, Eberhart A, Price KM, Jung JY, et al. (2006) Langerhans cells cross-present antigen derived from skin. PNAS USA 103: 7783–7788. 22. Birchall J, Clemo R, Anstey A, John D (2011) Microneedles in Clinical Practice– An Exploratory Study Into the Opinions of Healthcare Professionals and the Public. Pharm Res 28: 95–106. 50. Li H, Willingham SB, Ting JP-Y, Re F (2008) Cutting Edge: Inflammasome Activation by Alum and Alum’s Adjuvant Effect Are Mediated by NLRP3. J Immunol 181: 17–21. 23. Mikszta JA, Sullivan VJ, Dean C, Waterston AM, Alarcon JB, et al. (2005) Protective Immunization against Inhalational Anthrax: A Comparison of Minimally Invasive Delivery Platforms. J Infect Dis 191: 278–288. J 51. Ryan EJ, Daly LM, Mills KHG (2001) Immunomodulators and delivery systems for vaccination by mucosal routes. Trends Biotechnol 19: 293–304. 24. Widera G, Johnson J, Kim L, Libiran L, Nyam K, et al. (2006) Effect of delivery parameters on immunization to ovalbumin following intracutaneous adminis- tration by a coated microneedle array patch system. Vaccine 24: 1653–1664. 52. Poltorak A, He X, Smirnova I, Liu M-Y, Huffel CV, et al. (1998) Defective LPS Signaling in C3H/HeJ and C57BL/10ScCr Mice: Mutations in Tlr4 Gene. Science 282: 2085–2088. 25. Mikszta JA, Dekker JP, Harvey NG, Dean CH, Brittingham JM, et al. (2006) Microneedle-Based Intradermal Delivery of the Anthrax Recombinant Protec- tive Antigen Vaccine. Infect Immunity 74: 6806–6810. 53. Ulrich JT, Myers KR (1995) Monophosphoryl lipid A as an adjuvant. Past experiences and new directions. Pharm Biotechnol 6: 495–524. 54. Imaging Transcutaneous Immunization Demento SL, Eisenbarth SC, Foellmer HG, Platt C, Caplan MJ, et al. (2009) Inflammasome-activating nanoparticles as modular systems for optimizing vaccine efficacy. Vaccine 27: 3013–3021. 26. Dean CH, Alarcon JB, Waterston AM, Draper K, Early R, et al. (2005) Cutaneous Delivery of a Live, Attenuated Chimeric Flavivirus Vaccines against Japanese Encephalitis (ChimeriVaxTM-JE) in Non-Human Primates. Human Vaccines 1: 106–111. y 55. Chen M, Wang H, Chen W, Meng G (2011) Regulation of adaptive immunity by the NLRP3 inflammasome. Int Immunopharmacol 11: 549–554. 27. Mikszta JA, Alarcon JB, Brittingham JM, Sutter DE, Pettis RJ, et al. (2002) Improved genetic immunization via micromechanical disruption of skin-barrier function and targeted epidermal delivery. Nat Med 8: 415–419. 56. Sharp FA, Ruane D, Claass B, Creagh E, Harris J, et al. (2009) Uptake of particulate vaccine adjuvants by dendritic cells activates the NALP3 inflamma- some. PNAS USA 106: 870–875. 28. Matriano JA, Cormier M, Johnson J, Young WA, Buttery M, et al. (2002) Macroflux microprojection array patch technology: a new and efficient approach for intracutaneous immunization. Pharm Res 19: 63–70. 57. Martin M, Michalek SM, Katz J (2003) Role of innate immune factors in the adjuvant activity of monophosphoryl lipid A. Infect Immunity 71: 2498–2507. 58. Re F, Strominger JL (2001) Toll-like Receptor 2 (TLR2) and TLR4 Differentially Activate Human Dendritic Cells. J Biol Chem 276: 37692–37699. pp 29. Cahalan MD, Parker I, Wei SH, Miller MJ (2002) Two-photon tissue imaging: seeing the immune system in a fresh light. Nat Rev Immunol 2: 872–880. 59. Etchart N, Desmoulins P-O, Chemin K, Maliszewski C, Dubois B, et al. (2001) Dendritic Cells Recruitment and In Vivo Priming of CD8+ CTL Induced by a Single Topical or Transepithelial Immunization Via the Buccal Mucosa with Measles Virus Nucleoprotein. J Immunol 167: 384–391. 30. Masters B, So P (2001) Confocal microscopy and multi-photon excitation microscopy of human skin in vivo. Opt Express 8: 2–10. py p p 31. Helmchen F, Denk W (2005) Deep tissue two-photon microscopy. Nat Meth 2: 932–940. 60. Mackay LK, Stock AT, Ma JZ, Jones CM, Kent SJ, et al. (2012) Long-lived epithelial immunity by tissue-resident memory T (TRM) cells in the absence of persisting local antigen presentation. PNAS USA 109: 7037–7042. 32. Hadjantonakis A-K, Dickinson ME, Fraser SE, Papaioannou VE (2003) Technicolour transgenics: imaging tools for functional genomics in the mouse. Nat Rev Genet 4: 613–625. p g g p 61. Imaging Transcutaneous Immunization Clark RA, Chong B, Mirchandani N, Brinster NK, Yamanaka K-i, et al. (2006) The Vast Majority of CLA+ T Cells Are Resident in Normal Skin. J Immunol 176: 4431–4439. 33. Gauderon R, Lukins PB, Sheppard CJR (2001) Optimization of second- harmonic generation microscopy. Micron 32: 691–700. 34. Murzina TV, Kolmychek IA, Maidykovski AI, Nikulin AA, Sychev FY, et al. (2008) Second- and third-harmonic generation and hyper-Rayleigh scattering in porous-silicon-based photonic microcavities. Opt Lett 33: 2581–2583. 62. Ratzinger G, Stoitzner P, Ebner S, Lutz MB, Layton GT, et al. (2002) Matrix Metalloproteinases 9 and 2 Are Necessary for the Migration of Langerhans Cells and Dermal Dendritic Cells from Human and Murine Skin. J Immunol 168: 4361–4371. 35. Mohler W, Millard AC, Campagnola PJ (2003) Second harmonic generation imaging of endogenous structural proteins. Methods 29: 97–109. 63. Bursch LS, Wang L, Igyarto B, Kissenpfennig A, Malissen B, et al. (2007) Identification of a novel population of Langerin+ dendritic cells. J Exp Med 204: 3147–3156. 36. Coulman SA, Birchall JC, Alex A, Pearton M, Hofer B, et al. (2011) In Vivo, In Situ Imaging of Microneedle Insertion into the Skin of Human Volunteers Using Optical Coherence Tomography. Pharm Res 28: 66–81. 64. Wester RC, Maibach HI (2001) In vivo methods for percutaneous absorption measurements (Reprinted from Percutaneous Adsorption, pg 215–228, 1999). J Toxicol-Cut Ocular Toxicol 20: 411–422. 37. Rizwan SB, Assmus D, Boehnke A, Hanley T, Boyd BJ, et al. (2011) Preparation of phytantriol cubosomes by solvent precursor dilution for the delivery of protein vaccines. Eur J Pharm Biopharm 79: 15–22. 65. Bronaugh RL, Stewart RF, Congdon ER (1982) Methods for in vitro percutaneous absorption studies. II. Animal models for human skin. Toxicol Applied Pharmacol 62: 481–488. 38. Rattanapak T, Birchall J, Young K, Ishii M, Meglinski I, et al. (2013) Transcutaneous immunization using microneedles and cubosomes: Mechanistic investigations using Optical Coherence Tomography and Two-Photon Micros- copy. J Control Rel 172: 894–903. 66. Shiozuka M, Wagatsuma A, Kawamoto T, Sasaki H, Shimada K, et al. (2010) Transdermal delivery of a readthrough-inducing drug: a new approach of gentamicin administration for the treatment of nonsense mutation-mediated disorders. J Biochem 147: 463–470. 39. Myschik J, McBurney WT, Hennessy T, Phipps-Green A, Rades T, et al. (2008) Immunostimulatory biodegradable implants containing the adjuvant Quil-A– Part II: In vivo evaluation. J Drug Target 16: 224–232. Part II: In vivo evaluation. J Drug Target 16: 224–232. 67. References (2010) CD207+ CD103+ dermal dendritic cells cross-present keratinocyte-derived 7. Henri S, Poulin LF, Tamoutounour S, Ardouin L, Guilliams M, et al. (2010) CD207+ CD103+ dermal dendritic cells cross-present keratinocyte-derived PLOS ONE | www.plosone.org 11 February 2014 | Volume 9 | Issue 2 | e89503 February 2014 | Volume 9 | Issue 2 | e89503 Imaging Transcutaneous Immunization Yu Z, Chung WG, Sloat BR, Lohr CV, Weiss R, et al. (2011) The extent of the uptake of plasmid into the skin determines the immune responses induced by a DNA vaccine applied topically onto the skin. J Pharm Pharmacol 63: 199–205. 40. Valladeau J, Saeland S (2005) Cutaneous dendritic cells. Sem Immunol 17: 273– 283. February 2014 | Volume 9 | Issue 2 | e89503 PLOS ONE | www.plosone.org February 2014 | Volume 9 | Issue 2 | e89503 12

https://openalex.org/W4232254645

https://europepmc.org/articles/pmc6879124?pdf=render

English

Retraction: Pathological Roles of Interleukin-22 in the Development of Recurrent Hepatitis C after Liver Transplantation

PloS one

cc-by

RETRACTION Retraction: Pathological Roles of Interleukin- 22 in the Development of Recurrent Hepatitis C after Liver Transplantation The PLOS ONE Editors Concerns have been raised that the transplants performed in the local context of procedures reported in this article [1] may have involved organs/tissues procured from prisoners [2]. International ethics standards call for transparency in organ donor and transplantation pro- grams and clear informed consent procedures including considerations to ensure that donors are not subject to coercion. Details as to the transplant donor sources and methods of obtaining informed consent from donors were not reported in [1]. The authors did not clarify these issues or the cause(s) of donor death in response to journal queries, although they provided ethics approval docu- mentation and informed consent forms for organ donors. The first author claimed that the authors were not involved with organ donation or surgical procedures and noted that this study focused on transplant recipients. The Ethics Statement of [1] declared that patients pro- vided informed written consent, and stated, “No conflict of interest during all organ transplan- tations. No organ trafficking involved.” The ethics approval document provided (S1 Fig in [1]) was dated after the onset of participant recruitment and it indicated that the ethics review and approval did not include consideration of an informed consent form. Furthermore, while the ethics document indicated that no organ trafficking was involved it did not clarify donor sources or whether organs had been procured from prisoners. The authors did not provide the primary data underlying this article’s results and as such the article does not comply with the PLOS Data Availability Policy. Owing to the above concerns, and in line with international ethics standards for organ/tis- sue donation and transplantation, the PLOS ONE Editors retract this article. YG did not agree with retraction. HR, FM, JL, EC, HL, JZ, HL, ZL, and MZ did not respond or could not be reached. 1. Gao Y, Ren H, Meng F, Li J, Cheung E, Li H, et al. (2016) Pathological Roles of Interleukin-22 in the Development of Recurrent Hepatitis C after Liver Transplantation. PLoS ONE 11(4): e0154419. https:// doi.org/10.1371/journal.pone.0154419 PMID: 27123854 2. Rogers W, Robertson MP, Ballantyne A, et al Compliance with ethical standards in the reporting of donor sources and ethics review in peer-reviewed publications involving organ transplantation in China: a scoping review BMJ Open 2019; 9:e024473. https://doi.org/10.1136/bmjopen-2018-024473 PMID: 30723071 OPEN ACCESS Citation: The PLOS ONE Editors (2019) Retraction: Pathological Roles of Interleukin-22 in the Development of Recurrent Hepatitis C after Liver Transplantation. PLoS ONE 14(11): e0225971. https://doi.org/10.1371/journal.pone.0225971 Published: November 26, 2019 Published: November 26, 2019 Copyright: © 2019 The PLOS ONE Editors. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 1 / 1 PLOS ONE | https://doi.org/10.1371/journal.pone.0225971 November 26, 2019

https://openalex.org/W4387467682

https://melatijournal.com/index.php/jisma/article/download/217/245

Indonesian

ANALISIS KUALITAS PRODUK, HARGA, DAN PROMOSI DALAM KEPUTUSAN PEMBELIAN ULANG PRODUK

Jurnal Ilmu Sosial, Manajemen, dan Akuntansi

cc-by-sa

ABSTRAK ABSTRAK Penelitian ini bertujuan untuk menganalisa kualitas produk, harga dan promosi dalam keputusan pembelian ulang. Penelitian ini menggunakan metode Kuantitatif. Analisis data secara deskriptif. Berdasarkan hasil analisis deskriptif interpretasi nilai rata-rata dari varibel bebas yang menentukan pelanggan melakukan pembelian produk, skor tertinggi adalah variabel promosi yaitu 4,04 dalam pernyataan tertinggi pada “Scarlett Whitening melakukan promosi penjualan di Shopee, Instagram, Tiktok, Facebook dll” (4,17). Kedua adalah variabel harga dengan skor 3,81 dalam pernyataan tertinggi pada “kesesuaian harga” (3,88). Ketiga pada variabel kualitas produk dengan skor 3,80 dengan pernyataan tertinggi pada “manfaat produk” (3,61). Sedangkan dalam variabel pembelian ulang, pernyataan tertinggi pada dimensi rekomendasi produk dengan skor rata-rata 3,95. Kata Kunci : Harga Keputusan Pembelian Ulang Kualitas Produk Promosi This is an open access article under the CC BY-SA license. Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 No.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 No.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) 717 Journal homepage: https://melatijournal.com/index.php/JISMA ANALISIS KUALITAS PRODUK, HARGA, DAN PROMOSI DALAM KEPUTUSAN PEMBELIAN ULANG PRODUK ANALISIS KUALITAS PRODUK, HARGA, DAN PROMOSI DALAM KEPUTUSAN PEMBELIAN ULANG PRODUK Erlin Ade Wiryani 1, Farida Farida2 1,2 Fakultas Ekonomi Dan Bisnis Universitas Sahid Jakarta Email: [email protected], [email protected] 1. PENDAHULUAN Journal homepage: https://melatijournal.com/index.php/JISMA 718 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 Issue.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) Selain itu Ecommerce juga di definisikan sebagai siklus bisnis yang memanfaatkan inovasi tekhnologi dan perdagangan atau penawaran produk, servis dan data secara elektronik (Kurniadi, Y U., 2020). Menurut Hasan (2013), minat beli adalah kecondongan konsumen untuk memilih brand atau melakukan aktifitas yang berkaitan dengan pembelian seperti yang diperkirakan kemungkinan konsumen akan melakukan pembelian sebagai prediksi perilaku konsumen di masa yang akan datang. Jika merasa puas kemungkinan besar konsumen akan kembali membeli produk itu lagi. Tetapi jika konsumen merasa tidak puas, kemungkinan besar tidak akan membeli produk lagi. (Muthmainnah et al., 2019) g Scarlett adalah produk skincare lokal yang luar biasa yang menawarkan macam-macam produk perawatan kulit secara online di marketplace. Marketplace adalah sebuah pasar virtual yang merupakan tempat bertemunya penjual dan pembeli untuk melakukan transaksi secara online (B. D. Hutauruk, J. F. Naibaho, 2017) Marketplace memiliki kemampuan yang hampir sama dengan pasar tradisional, yang membedakannya adalah sebuah jaringan yang digunakan untuk bantuan pasar atau marketplkace yang dapat dilangsungkan secara efektif maupun secara efisien dalam memberitahukan informasi terbaru dan layanan jasa untuk penjual maupun untuk pembeli ( Ipandarmanto, 2005). Scarlett Whitening produk lokal yang dikenali oleh banyak orang, serta memiliki banyak followers, Jumlah followers Scarlett Whitening pada tahun 2022 di akun instagram sebesar 5,5 juta followers sedangkan jumlah followers di Shopee sebesar 2.1 Juta followers. Berdasarkan penelusuran internal tim Compas, pada bulan Juni 2021, penjualan produk serum wajah di official store shopee mencapai Rp148,02 miliar penjualan. Berdasarkan total penjualan tersebut, 6 merek serum lokal berhasil masuk dalam kategori 10 Top Brand Serum wajah terlaris penjualannya di Shopee. Gambar 1.1 Merek Serum Wajah Terlaris Pada Juni tahun 2021 (10 Merek Produk Serum Wajah) Sumber : https://compas.co.id/article/penjualan-serum/ (Hasil olahan oleh penulis) Perbandingan kualitas produk serum wajah Scarlett dengan Somethinc adalah sama. Berdasarkan ulasan ntang lima Somethinc mencapai 112.000 sedangkan ulasan bintang lima Scarlett Whiteng mencapai 101.000. Ulasan ngan bintang lima berarti pelanggan memerikan feedback sangat baik. Perbandingan harga Scarlett Whitening yaitu .000 atau lebih murah dibanding harga Somethinc mencapai 89.000 di Shopee. Sedangkan berdasarkan hasil giatan promosi yang dilakukan di Shopee, Scarlett Whitening mencapai 2,1 Juta pengikut sedangkan Somethinc nya 1,7 Juta pengikut di Shopee. 1. PENDAHULUAN Somethinc; 7,81 Bening's; 4,25 Scarlett; 3,76 L'Oreal; 3,15 Garnier; 2,21 Bio Beauty Lab; 2,12 Avoskin; 2,03 Whitelab; 1,96 Trueve; 1,65 Breylee; 0,912 0 2 4 6 8 10 1 Pembelian (Milyar) 10 Top Brand Serum Wajah Di Shopee Somethinc; 7,81 Bening's; 4,25 Scarlett; 3,76 L'Oreal; 3,15 Garnier; 2,21 Bio Beauty Lab; 2,12 Avoskin; 2,03 Whitelab; 1,96 Trueve; 1,65 Breylee; 0,912 0 2 4 6 8 10 1 Pembelian (Milyar) 10 Top Brand Serum Wajah Di Shopee Garnier; 2,21 Bio Beauty Lab; 2,12 Avoskin; 2,03 Whitelab; 1,96 1 4 6 Pembelian (Milyar) 8 Gambar 1.1 Merek Serum Wajah Terlaris Pada Juni tahun 2021 (10 Merek Produk Serum Wajah) Sumber : https://compas.co.id/article/penjualan-serum/ (Hasil olahan oleh penulis) Perbandingan kualitas produk serum wajah Scarlett dengan Somethinc adalah sama. Berdasarkan ulasan bintang lima Somethinc mencapai 112.000 sedangkan ulasan bintang lima Scarlett Whiteng mencapai 101.000. Ulasan dengan bintang lima berarti pelanggan memerikan feedback sangat baik. Perbandingan harga Scarlett Whitening yaitu 75.000 atau lebih murah dibanding harga Somethinc mencapai 89.000 di Shopee. Sedangkan berdasarkan hasil kegiatan promosi yang dilakukan di Shopee, Scarlett Whitening mencapai 2,1 Juta pengikut sedangkan Somethinc hanya 1,7 Juta pengikut di Shopee. Perbandingan kualitas produk serum wajah Scarlett dengan Somethinc adalah sama. Berdasarkan ulasan bintang lima Somethinc mencapai 112.000 sedangkan ulasan bintang lima Scarlett Whiteng mencapai 101.000. Ulasan dengan bintang lima berarti pelanggan memerikan feedback sangat baik. Perbandingan harga Scarlett Whitening yaitu 75.000 atau lebih murah dibanding harga Somethinc mencapai 89.000 di Shopee. Sedangkan berdasarkan hasil kegiatan promosi yang dilakukan di Shopee, Scarlett Whitening mencapai 2,1 Juta pengikut sedangkan Somethinc hanya 1,7 Juta pengikut di Shopee. 2. METODE PENELITIAN Dalam penelitian karya ilmiah ini jenis penelitian yang digunakan yaitu menggunakan metode deskriptif kuantitatif. Metode deskriptif kuantitatif untuk meneliti masalah hubungan variabel satu dengan variabel lain. Penelitian ini bermaksud untuk menganalisis variabel independen yang terdiri dari kualitas produk, harga, dan promosi dengan variabel dependen yaitu keputusan pembelian ulang pelanggan serum wajah Scarlett Whitening di marketplace Shopee. Analisa data dengan menghitung mean (skor rata-rata) dimensi variabel, dan dideskriptifkan berdasarkan hasil menggunakan skala interval. Journal homepage: https://melatijournal.com/index.php/JISMA 1. PENDAHULUAN Dalam dunia persaingan bisnis, bisnis dan perdagangan harus selalu memperhatikan kepuasan konsumen, karena setiap produk yang ditawarkan memiliki dampak yang signifikan terhadap calon konsumen dan diharapkan dapat menggunakan kembali produk yang ditawarkan. Perusahaan juga harus dapat memuaskan konsumen dengan mendistribusikan produk yang berkualitas baik dan memberikan pelayanan yang terbaik kepada konsumen sehingga semua konsumen dapat melakukan pembelian ulang dengan nyaman. Pada umumnya orang selalu memperhitungkan segala sesuatunya saat melakukan aktivitas, seperti mengambil keputusan saat membeli barang atau akan menggunakan jasa. gg j Didalam memutuskan harga tenaga kerja dan produk, tujuannya adalah meningkatkan suatu penjualan, mempertahankan dan mengembangkan saham, kemantapan harga, mencapai tujuan keuntungan dari usaha, dan untuk mendapatkan nilai laba yang besar (Mawarsari, 2018). Harga adalah komponen utama yang ada di macam-macam unsur bauran pemasaran yang akan menghasilkan laba bagi penjual atau perusahaan. Perusahaan harus memperhatikan mengenai faktor biaya, besar kecilnya biaya yang ditentukan perusahaan sangat berpengaruh dalam persaingan antar perusahaan (Devi Marta Dwi .L, Theresia Pradiani, 2021) Kualitas produk sesuatu yang harus pertama kali dilihat oleh perusahaan jika keinginan perusahaan mampu menghasilkan produk yang dapat bersaing dengan pasaran. Pembeli pada umumnya perlu mendapatkan produk dengan kualitas yang baik. Jika perusahaan dapat memberikan produk yang berkualitas baik, maka perusahaan bisa memenuhi keinginan para pembeli sehingga dapat meningkatkan jumlah pembeli dan pelanggan (Dwijantoro et al., 2022). Menurut Kotler (2007), produk adalah barang yang terdapat pada perusahaan yang merupakan kebutuhan konsumen. Sementara menurut Tjiptono (2006), "produk tersebut berasal dari kebutuhan pasar yang signifikan atau sebagai pemenuhan pertimbangan pasar, mencari, membeli, memanfaatkan atau mengkonsumsi semua yang dapat ditawarkan oleh produsen (Hartaroe et al., 2016). Promosi bukan hanya sekedar alat komunikasi khusus antara perusahaan dan pembeli, tetapi juga sebagai alat untuk memengaruhi pelanggan sehingga pelanggan melakukan pembelian atau penggunaan layanan sesuai kebutuhan dan keinginan mereka. Promosi adalah perspektif penting dalam bauran pemasaran dan sering kali dikaitkan sebagai tindakan yang berkelanjutan, dimana dinyatakan oleh beberapa pendefinisi sebelumnya. Dikarenakan promosi bisa membuat perusahaan menciptakan banyak produk baru. (Sani et al., 2022). Promosi atau disebut komunikasi pemasaran merupakan metode penyampaian informasi yang memberikan manfaat mengenai suatu perusahaan atau produk yang mempengaruhi potensial pembeli (Mulyana, 2019). Promosi dan pembelian produk atau barang yang efektif di era modern ini adalah promosi melalaui media digital sosial media, dan pembelian produk secara online seperti melalui Ecommerce. E-commerce secara umum dapat diuraikan perdagangan secara elektronik melalui web. Journal homepage: https://melatijournal.com/index.php/JISMA Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 No.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) 719 Penelitian karya ilmiah ini memanfaatkan dua sumber informasi untuk mencari data dan mengumpulkan data yang diperlukan. Sumber data yang akan digunakan dalam penelitian ini yaitu: 1. Sumber data primer Untuk mendapatkan data ini, peneliti akan menggunakan kuesioner yang disebarkan kepada responden, perolehan data primer dalam penelitian ini yaitu hasil kuesioner yang diisi oleh para responden yang merupakan pelanggan serum wajah Scarlett Whitening atau pelanggan yang telah membeli produk melalui Official Store di Marketplace Shopee lebih dari satu kali pembelian. 2. Sumber data sekunder Data sekunder ialah data yang sudah tersedia di dalam penelitian lain. Analisis data sekunder yang dilakukan peneliti mencari data dan mengumpulkan data yang berasal dari buku-buku, jurnal, internet dan sebagainya menjadi referensi peneliti dalam melakukan penelitian ini. 2. Sumber data sekunder Data sekunder ialah data yang sudah tersedia di dalam penelitian lain. Analisis data sekunder yang dilakukan peneliti mencari data dan mengumpulkan data yang berasal dari buku-buku, jurnal, internet dan sebagainya menjadi referensi peneliti dalam melakukan penelitian ini. Penelitian ini menggunakan populasi dari para followers di marketplace official shopee dengan nama akun @scarlett_whitening yang berjumlah 2,1 juta followers. Mengenai data informasi yang telah didapatkan tentang jumlah populasi sebanyak 2,1 juta orang, dikarenakan adanya keterbatasan peneliti dari segi waktu serta dari segi biaya, oleh sebab itu peneliti menetapkan penggunaan rumus Slovin untuk menentukan banyaknya jumlah sampel yang akan digunakan pada penelitian ini. g p p Hasil dari perhitungan menggunakan rumus slovin, didapatkan hasil sampel yang akan digunakan dalam penelitian ini berjumlah 99,9 atau digenapkan menjadi 100 orang. Teknik pengumpulan data dengan menyebar kuisoner secara online kepada para followers dan pelanggan yang telah mengetahui produk serum wajah Scarlett Whitening dan pernah melakukan pembelian minimal dua kali pembelian dengan bentuk instrumen google form. Peneliti menggunakan lima jenis tanggapan responden yang digunakan yaitu: gg j gg p p y g g y Tabel 1 Skala Likert Sangat Setuju Skor 5 Setuju Skor 4 Cukup Setuju Skor 3 Tidak Setuju Skor 2 Sangat Tidak Setuju Skor 1 Tabel 1 Skala Likert Sangat Setuju Skor 5 Setuju Skor 4 Cukup Setuju Skor 3 Tidak Setuju Skor 2 Sangat Tidak Setuju Skor 1 Skor tiga sampai lima artinya responden mendukung pernyataan yang diajukan sedangkan skor satu dan dua artinya responden tidak mendukung pernyataan yang diajukan. Journal homepage: https://melatijournal.com/index.php/JISMA Dalam penelitian ini, tidak tercantum penggunaan skala likert dengan jawaban “ragu-ragu", untuk mengatasi jawaban agar berada pada jawaban yang aman. Oleh karena itu, peneliti memilih jenis jawaban yang digunakan yaitu SS, S, CS, TS, STS untuk menghasilkan jawaban yang jelas dari responden setuju atau tidak setuju. Hasil skala likert yang berupa angka-angka yang merupakan hasil dari data penyebaran kuesioner diolah dengan mencari mean (rata-rata) dari tiap variabel penelitian lalu diolah menggunakan SPSS. Hasilnya dijelaskan berdasarkan versi yang sudah ditentukan untuk mengetahui efektivitas dari keputusan pembelian ulang. kriteria skor rata-rata menggunakan rumus : Journal homepage: https://melatijournal.com/index.php/JISMA skor maksimal – skor minimal Interval = 3 (tiga) kategori 5 - 1 Interval = 3 Interval = 1,33 Tabel 2 Interprestasi Skor Rata-Rata Rata-Rata Skor Kriteria 1,00 – 2,33 Rendah / Buruk 2,34 – 3,67 Cukup / Sedang 3,68 – 5,00 Tinggi / Baik 3. HASIL DAN ANALISIS Berdasarkan analisa data diperoleh hasil sebagai berikut : Jawaban dari responden yang sudah dicari nilai mean dari tiap variabel diintepretasikan sebagai berikut : skor maksimal – skor minimal Journal homepage: https://melatijournal.com/index.php/JISMA skor maksimal – skor minimal Interval = 3 (tiga) kategori 5 - 1 Interval = 3 Interval = 1,33 Tabel 2 Interprestasi Skor Rata-Rata Rata-Rata Skor Kriteria 1,00 – 2,33 Rendah / Buruk 2,34 – 3,67 Cukup / Sedang 3,68 – 5,00 Tinggi / Baik 3. HASIL DAN ANALISIS Berdasarkan analisa data diperoleh hasil sebagai berikut : Jawaban dari responden yang sudah dicari nilai mean dari tiap variabel diintepretasikan sebagai berikut : Journal homepage: https://melatijournal.com/index.php/JISMA Interval 1,33 Tabel 2 Interprestasi Skor Rata-Rata Rata-Rata Skor Kriteria 1,00 – 2,33 Rendah / Buruk 2,34 – 3,67 Cukup / Sedang 3,68 – 5,00 Tinggi / Baik 3. HASIL DAN ANALISIS Berdasarkan analisa data diperoleh hasil sebagai berikut : Jawaban dari responden yang sudah dicari nilai mean dari tiap variabel diintepretasikan sebagai berikut : 3. Journal homepage: https://melatijournal.com/index.php/JISMA HASIL DAN ANALISIS 720 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 Issue.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) Tabel 3 Hasil Interpretasi Standar Kualitas Produk Indikator Kategori Jawaban Responden Rata- rata Kriteria Rata-Rata Kualitas Produk 1 2 3 4 5 X1.P1 0 3 36 49 12 3,70 Baik 3,80 X1.P2 0 7 42 45 6 3,50 Cukup X1.P3 0 5 39 45 11 3,62 Cukup Dimensi Rata – Rata 3,61 Cukup Manfaat Produk X1.P4 0 1 30 47 22 3,90 Baik X1.P5 1 3 36 40 20 3,75 Baik X1.P6 0 2 23 45 30 4,03 Baik Dimensi Rata – Rata 3,89 Baik Keistimewaan Produk X1.P7 0 2 28 53 17 3,85 Baik X1.P8 0 1 31 41 27 3,94 Baik Dimensi Rata – Rata 3,90 Baik Features Produk Sumber: Olah data Peneliti (2022) Berdasarkan hasil tabel diatas, rata-rata dari kualitas produk adalah 3,80 ada pada interval 3,68 – 5,00 artinya kualitas produk baik. Dimensi fitur produk menghasilkan nilai tertinggi yaitu 3,90 Sedangkan nilai tertinggi dari pernyataan yaitu 3,94 pernyataan bahwa “pelanggan tidak merasakan adanya efek samping setelah menggunakan produk Scarlett Whitening karena produk Scarlett tidak terbuat dari bahan bahan kimia yang berbahaya” sedangkan dimensi terendah yaitu pada manfaat produk dengan rata-rata 3,61 dan nilai terendah dari pernyataan sebesar 3,50 pernyataan bahwa “Produk Scarlett Whitening memiliki daya tahan lebih dari 8 jam penggunaan” dapat ditarik kesimpulan dari hasil rata rata tertinggi maka yang paling berpengaruh yaitu kualitas produk, karena produk scarlett aman digunakan. Didukung dengan penelitian (Erdalina & Evanita, 2015) yang menyatakan kualitas produk menghasilkan skor rata-rata terendah. Journal homepage: https://melatijournal.com/index.php/JISMA Tabel 5 Hasil Interpretasi Standar Promosi Indikator Kategori Jawaban Responden Rata- rata Kriteria Rata – rata promosi 1 2 3 4 5 X3.P1 0 2 17 43 38 4,17 Baik 4,04 X3.P2 1 2 20 46 31 4,04 Baik Dimensi Rata – rata 4,11 Baik Frekuensi Promosi X3.P3 0 3 22 45 30 4,02 Baik X3.P4 0 3 30 44 23 3,87 Baik Dimensi Rata – rata 3,95 Baik Daya Tarik Promosi X3.P5 0 1 26 44 29 4,01 Baik X3.P6 0 0 19 52 29 4,10 Baik Dimensi Rata – rata 4,06 Baik Ketepatan Promosi Sumber: Olah data Peneliti (2022) Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 No.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) 721 harga memiliki skor rata rata terendah namun berbanding balik pada penelitian (Savitri & Wardana, 2018) yang menyatakan harga menghasilkan skor rata – rata tertinggi. T b l 5 harga memiliki skor rata rata terendah namun berbanding balik pada penelitian (Savitri & Wardana, 2018) yang menyatakan harga menghasilkan skor rata – rata tertinggi. Tabel 5 Hasil Interpretasi Standar Promosi Sumber: Olah data Peneliti (2022) Berdasarkan hasil tabel diatas, rata-rata promosi adalah 4,04 ada pada interval 3,68 – 5,00 artinya promosi baik. Dimensi frekuensi promosi menghasilkan nilai tertinggi yaitu 4,11 Sedangkan nilai tertinggi dari pernyataan yaitu 4,17 pernyataan bahwa “Scarlett Whitening melakukan promosi penjualan di Shopee, Instagram, Tiktok, Facebook dll”. sedangkan dimensi terendah yaitu pada daya tarik promosi dengan rata-rata 3,95 dan nilai terendah pernyataan yaitu 3,87 pernyataan bahwa “Promosi iklan yang dilakukan Scarlett Whitening yang membuat saya tertarik untuk melakukan pembelian ulang” dapat ditarik kesimpulan dari hasil nilai rata rata tertinggi maka yang paling mendorong pada promosi yaitu karena Scarlett Whitening melakukan banyak promosi diberbagai media. Didukung dengan penelitian (Farida & Anggraeni, 2020) yang menyatakan promosi menghasilkan skor rata-rata tinggi. Journal homepage: https://melatijournal.com/index.php/JISMA Tabel 4 Hasil Interpretasi Standar Harga Indikator Kategori Jawaban Responden Rata- rata Kriteria Rata-Rata Harga 1 2 3 4 5 X2.P1 0 6 31 40 23 3,80 Baik 3,81 X2.P2 0 1 30 50 19 3,87 Baik X2.P3 0 2 24 48 26 3,98 Baik Dimensi Rata – rata 3,88 Baik Kesesuaian Harga X2.P4 0 7 28 49 16 3,74 Baik X2.P5 1 6 29 47 17 3,73 Baik Dimensi Rata – rata 3,74 Baik Harga Bersaing Sumber: Olah data Peneliti (2022) Tabel 4 Hasil Interpretasi Standar Harga i b Tabel 4 Hasil Interpretasi Standar Harga Berdasarkan hasil tabel diatas, rata-rata harga adalah 3,81 ada pada interval 3,68 – 5,00 artin i k i h h ilk il i i i i 3 88 S d k il i i i d i Berdasarkan hasil tabel diatas, rata-rata harga adalah 3,81 ada pada interval 3,68 – 5,00 artinya harga baik. Dimensi kesesuaian harga menghasilkan nilai tertinggi yaitu 3,88 Sedangkan nilai tertinggi dari pernyataan yaitu 3,98 pernyataan bahwa “Harga Scarlett Whitening bervariasi sesuai dengan ragam dan ukuran produk” sedangkan dimensi terendah yaitu pada harga bersaing dengan rata-rata 3,74 dan nilai terendah dari pernyataan yaitu 3,73 pernyataan bahwa “Harga Scarlett Whitening di Official Shopee lebih ekonomis dibanding dengan harga di marketplace lain” dapat ditarik kesimpulan mengapa harga belum mendorong adanya keputusan pembelian ulang jika dilihat dari nilai rata rata terendah artinya ada yang menjual produk scarlett whitening lebih ekonomis atau lebih murah dibanding di shopee / lebih murah di marketplace lain. Didukung oleh penelitian (Ashari et al., 2020) yang menyatakan variabel Journal homepage: https://melatijournal.com/index.php/JISMA Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 No.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) 721 harga memiliki skor rata rata terendah namun berbanding balik pada penelitian (Savitri & Wardana, 2018) yang menyatakan harga menghasilkan skor rata – rata tertinggi. Sumber: Olah data Peneliti (2022) 4. KESIMPULAN ESIMPULAN Berdasarkan hasil penelitian yang telah dilakukan, maka dapat ditarik kesimpulan sebagai berikut: 1. Dari keempat variabel jika dilihat dari hasil intepretasi yang paling besar yaitu pada variabel promosi keseluruhan dimensi menghasilkan skor rata-rata 4,04 yang kedua pada variabel pembelian ulang keseluruhan dimensi menghasilkan skor rata-rata 3,82. Yang ketiga pada variabel harga keseluruhan dimensi menghasilkan skor rata-rata 3,81 dan yang terendah nilainya pada variabel kualitas produk keseluruhan dimensi menghasilkan skor rata-rata 3,80. 2. Hasil interpretasi nilai rata rata tertinggi pada variabel kualitas produk yaitu pada dimensi keistimewaan produk, pernyataan bahwa pelanggan tidak merasakan adanya efek samping setelah menggunakan produk Scarlett Whitening karena produk Scarlett tidak terbuat dari bahan bahan kimia yang berbahaya atau karena produk aman digunakan. 2. Hasil interpretasi nilai rata rata tertinggi pada variabel kualitas produk yaitu pada dimensi keistimewaan produk, pernyataan bahwa pelanggan tidak merasakan adanya efek samping setelah menggunakan produk Scarlett Whitening karena produk Scarlett tidak terbuat dari bahan bahan kimia yang berbahaya atau karena produk aman digunakan. 3. Hasil interpretasi nilai rata rata tetinggi pada variabel harga yaitu pada dimensi kesesuaian harga dengan pernyataan bahwa harga Scarlett Whitening bervariasi sesuai dengan ragam dan ukuran produk. Jika harga yang ditetapkan sesuai dengan kualitas yang didapatkan maka akan semakin banyak konsumen yang merasakan puas sehingga besar kemungkinan untuk melakukan pembelian ulang. 4. Hasil interpretasi nilai rata rata tetinggi pada variabel promosi yaitu frekuensi promosi dengan pernyataan bahwa Scarlett Whitening melakukan banyak promosi penjualan di Shopee, Instagram, Tiktok, Facebook dll. Semakin bagus dan banyaknya promosi yang ditampilkan semakin banyak pelanggan yang melihat iklan dan tertarik untuk melakukan pembelian ulang suatu produk. 5. Hasil interpretasi nilai rata rata tetinggi pada variabel pembelian ulang yaitu rekomendasi produk dengan pernyataan “saya terus mencari informasi tentang produk Scarlett Whitening baik dari banyaknya ulasan bintang lima di Shopee maupun dari teman yang sudah menggunakannya yang membuat pelanggan yakin untuk melakukan pembelian ulang produk” semakin banyaknya pembelian produk di Shopee maka semakin banyak pula pelanggan yang memberikan penilaian produk sehingga semakin meningkat pula penjualan. Journal homepage: https://melatijournal.com/index.php/JISMA Tabel 6 Tabel 6 Hasil Interpretasi Standar Keputusan Pembelian Ulang Tabel 6 Hasil Interpretasi Standar Keputusan Pembelian Ulang Indikator Kategori Jawaban Responden Rata- rata Kriteria Rata – rata keputusan pembelian ulang 1 2 3 4 5 Y1 0 2 17 43 38 3,90 Baik 3,82 Y2 1 2 20 46 31 3,99 Baik Dimensi Rata – rata 3,95 Baik Rekomendasi Produk Y3 0 3 22 45 30 3,83 Baik Y4 0 3 30 44 23 3,72 Baik Dimensi Rata – rata 3,78 Baik Kecenderungan Membeli Ulang Y5 0 1 26 44 29 3,76 Baik Y6 0 0 19 52 29 3,69 Baik Dimensi Rata – rata 3,73 Baik Perilaku Setelah Pembelian Sumber: Olah data Peneliti (2022) Journal homepage: https://melatijournal.com/index.php/JISMA 722 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 Issue.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) Berdasarkan hasil tabel diatas, rata-rata keputusan pembelian ulang adalah 3,82 ada pada interval 3,68 – 5,00 artinya keputusan pembelian ulang baik. Dimensi tertinggi rekomendasi produk dengan nilai yaitu 3,95 Sedangkan nilai tertinggi dari pernyataan yaitu 3,99 pernyataan bahwa “Saya terus mencari informasi tentang produk Scarlett Whitening baik dari banyaknya ulasan bintang lima di Shopee maupun dari teman yang sudah menggunakannya yang membuat pelanggan yakin untuk melakukan pembelian ulang produk” sedangkan dimensi terendah yaitu pada perilaku setelah pembelian dengan rata-rata 3,73 dan nilai terendah dari pernyataan yaitu 3,69 pernyataan bahwa “Kulit saya cocok menggunakan produk Scarlett Whitening sehingga saya tidak beralih ke produk lainnya” dapat ditarik kesimpulan dari hasil nilai rata-rata tertinggi maka yang paling mendorong adanya keputusan pembelian ulang yaitu karena pelanggan mencari informasi melihat dari komentar rating bintang lima serta informasi dari orang orang yang telah menggunakan produk scarlett whitening. Didukung oleh penelitian (Ashari et al., 2020) yang menyatakan keputusan pembelian ulang menghasilkan skor rata-rata yang tinggi. DAFTAR PUSTAKA [1] Ashari, P. A., Hardiyono, H., & Aswar, N. F. (2020). Analisis Pengaruh Citra Merek, Kualitas Produk, Harga dan Promosi Terhadap Keputusan PEmbelian Smartphone Samsung di Makasar. Management Development and Applied Research Journal, 3(1), 18–32. pp , ( ), [2] B. D. Hutauruk, J. F. Naibaho, B. R. (2017). Analisis Dan Perancangan Aplikasi Marketplace Cinderamata Khas Batak Berbasis Android. Jurnal Methodika, 3(1), 242–246. [3] Devi Marta Dwi .L, Theresia Pradiani, I. K. R. (2021). Pengaruh Kualitas Produk, Harga, Diskon, Dan Keputusan Pembelian Online Maharani Prima Skin Care Di Blitar. Angewandte Chemie International Edition, 6(11), 951– 952., 2013–2015. [4] Dwijantoro, R., Dwi, B., & Syarief, N. (2022). Pengaruh Harga, Kualitas Produk, Dan Promosi Terhadap Keputusan Pembelian Marketplace Shopee. Jurnal Riset Manajemen Dan Bisnis, 16(2), 63. https://doi.org/10.21460/jrmb.2021.162.392 5] Erdalina, W., & Evanita, S. (2015). Pengaruh kualitas produk, harga dan iklan televisi ter pembelian produk kosmetik merek citra. Jurnal Riset Manajemen Bisnis Dan Publik, 3(1), 1–1 p p j [6] Farida, F., & Anggraeni, A. (2020). Pengaruh Efektifitas Promosi Terhadap Niat Beli Konsumen Produk Skincare E rto ’ s Beauty Care Melalui Media Sosial Tiktok. 11(1). y ( ) [7] Hartaroe, B. P., Mardani, R. M., & Abs, M. K. (2016). Prodi manajemen. 82–94. [8] Kurniadi, Y U., et al. (2020). Nusantara ( Jurnal Ilmu Pengetahuan Sosial ). Nusantara: Jurnal Ilmu Pengetahuan Sosial, 7(2), 408–420. Journal homepage: https://melatijournal.com/index.php/JISMA 723 [9] Mawarsari, I. (2018). Pengaruh Kualitas Produk, Harga, Dan Promosi Terhadap Minat Beli Ulang Konsumen Olahan Buah Carica Di Daerah Wonosobo. Jurnal Ilmiah Ekonomi Bisnis, 10(2), 9–12. [9] Mawarsari, I. (2018). Pengaruh Kualitas Produk, Harga, Dan Promosi Terhadap Minat Beli Ulang Konsumen Olahan Buah Carica Di Daerah Wonosobo. Jurnal Ilmiah Ekonomi Bisnis, 10(2), 9–12. , ( ), [10] Mulyana, M. V. S. P. dan K. (2019). Inisiasi VIII Strategi Promosi dan Komunikasi. Materi Tutorial Online Ekma 4216 Manajemen Pemasaran, 57–63. [11] Muthmainnah, M., Madiawati, P. N., Telkom, U., Ulang, M. B., & Sampling, P. (2019). Pengaruh Bauran Pemasaran Terhadap Minat Beli Ulang Pada Transmart Carrefour Buah Batu Bandung the Effect of Marketing Mix To Repurchase Intention on Transmart Carrefour Buah Batu Bandung. E-Proceeding of Management, 6(1), 965–974. [12] Sani, S. A., Batubara, M., Silalahi, P. R., Syahputri, R. R., & Liana, V. (2022). Pengaruh Promosi, Kualitas Produk dan Harga terhadap Keputusan Pembelian Konsumen pada Produk Scarlett Whitening. Al-Kharaj : Jurnal Ekonomi, Keuangan & Bisnis Syariah, 4(5), 1327–1342. p g j p [14] https://compas.co.id/article/penjualan-serum/ diakses pada bulan april 2022 g y p g j [13] Savitri, I. A. P. D., & Wardana, I. M. (2018). Pengaruh Citra Merek, Kualitas Produk Dan Persepsi Harga Terhadap Kepuasan Dan Niat Beli Ulang. E-Jurnal Manajemen Universitas Udayana, 7(10), 5748. https://doi.org/10.24843/ejmunud.2018.v07.i10.p19 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 No.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) Journal homepage: https://melatijournal.com/index.php/JISMA DAFTAR PUSTAKA https://doi.org/10.47467/alkharaj.v4i5.931 g y p g j [13] Savitri, I. A. P. D., & Wardana, I. M. (2018). Pengaruh Citra Merek, Kualitas Produk Dan Persepsi Harga Terhadap Kepuasan Dan Niat Beli Ulang. E-Jurnal Manajemen Universitas Udayana, 7(10), 5748. https://doi.org/10.24843/ejmunud.2018.v07.i10.p19 Journal homepage: https://melatijournal.com/index.php/JISMA 724 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 Issue.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) HALAMAN INI SENGAJA DIKOSONGKAN 724 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 Issue.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) HALAMAN INI SENGAJA DIKOSONGKAN 724 Jurnal Ilmu Sosial, Manajemen, dan Akuntansi (JISMA) Vol.1 Issue.5 Desember 2022, pp: 717-724 ISSN: 2830-2605 (Online) HALAMAN INI SENGAJA DIKOSONGKAN Journal homepage: https://melatijournal.com/index.php/JISMA

https://openalex.org/W4287978806

https://zenodo.org/record/4308285/files/WJARR-2018-0009.pdf

English

Handbook for instructors of agriculture teacher trainees based on self-regulated learning framework

Zenodo (CERN European Organization for Nuclear Research)

cc-by

Effectiveness of lemon juice and honey in shelf life enhancement of kunun zaki Article DOI: https://doi.org/10.30574/wjarr.2019.1.1.0009 Article DOI: https://doi.org/10.30574/wjarr.2019.1.1.0009 Abstract Kunun zaki, a refreshing non-alcoholic local beverage commonly consumed in Nigeria, has a short shelf life. This study was therefore carried out to determine the abilities of lemon juice and honey to improve the shelf life of the product. Traditionally produced kunun zaki was treated with lemon juice and honey in singles and in combinations at concentrations of 1.5, 2.0, 2.5 and 3.0 ml per 100 ml of kunun zaki. The treated kunun zaki samples together with the control (kunun zaki without preservative) were kept at room temperature (25±2 oC) for a period of five days during which physico-chemical characteristics and sensory qualities of the samples were monitored every 24 hours. Titratable acidity (TA) and pH were observed to increase and decrease respectively as keeping period increased. The observed changes became statistically significant (p< .05) after three days of keeping for most of the samples. At the end of the 5- day preservation period, the sample treated with 3 ml of honey recorded the lowest TA of 2.05±0.02 and the highest pH of 5.36±0.04. This sample also was the most accepted in terms of overall acceptability and other sensory parameters. It is concluded that the shelf life of kunun zaki can be improved using appropriate concentrations of lemon juice and honey but especially with honey (3 ml per 100 ml kunun zaki) which was able to preserve the product throughout the five-day period. Keywords: Honey; Kunun zaki; Lemon juice; Shelf life; Sensory quality  Corresponding author E-mail address: [email protected] Copyright © 2019 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 World Journal of Advanced Research and Reviews Cross Ref DOI: 10.30574/wjarr Journal homepage: https://www.wjarr.com (RESEARCH ARTICLE) World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 World Journal of Advanced Research and Reviews Cross Ref DOI: 10.30574/wjarr Journal homepage: https://www.wjarr.com (RESEARCH ARTICLE) or(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. E-mail address: [email protected] Copyright © 2019 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 pathogens in the product. The aim of the study was to determine the effectiveness of the use of honey and lemon juice in improving the shelf life of kunun zaki. 2.4. Effects of lemon juice and honey on pH and titratable acidity of kunun zaki [4] With the aid of sterile syringes, 1.5, 2.0, 2.5 and 3.0 ml volumes of lemon juice, honey, and a 1:1 mixture of lemon juice and honey were separately introduced into 100 ml volumes of kunun zaki in clean plastic screw-capped bottles. The bottles were hand-shaken to properly mix their contents. Plastic bottles containing 100 ml of kunun zaki without lemon juice, honey or their combinations were also provided and these served as the control. The bottles that contained the experimental kunun zaki samples were allowed to stand on the laboratory bench at room temperature (25±2 oC) for five days. pH of the samples were determined in triplicates at the beginning of the experiment and at 24 h intervals for 5 days using a pH meter (Horiba, M-8) that had been standardized with pH 4 and pH 7 buffers. Titratable acidity was also determined in triplicates for the variously treated kunun zaki samples at 24-hour intervals. For the determination of titratable acidity, 10 ml of sample was measured into a conical flask and four drops of phenolphthalein indicator was added. This was titrated with standard 0.1N sodium hydroxide to a distinct faint pink point. The titratable acidity, expressed as % lactic acid, was calculated for each sample as described by [6]. 2.2. Production of kunun zaki [2] One kilogramme of guinea corn (Sorghum bicolor) was washed thoroughly with clean tap water and steeped in 2 L of tap water (1:2 w/v) for a period of 18 h at room temperature (25±2 oC). After the steeping period, the grains were washed again and then wet-milled together with pre-washed spices – black pepper (Piper sp., 2 g), cloves (Eugenia caryophyllata, 2 g), ginger (Zingiber officinale, 50 g) and dried sweet potato (Ipomoea batatas, 20 g) using a properly washed grinding machine. The obtained paste was divided into two unequal portions (1:3 v/v). The larger portion was gelatinised by adding boiling water (1:1 v/v) in a plastic container. It was stirred vigorously for about 3 minutes after which 100 g of ground malted sorghum grains was added. The pastes mixture was stirred vigorously for 4 minutes to aid liquefaction and saccharification of the gelatinized starch and left for a period of 18 h to ferment. One litre of clean water was added to the fermented product which was then stirred and sieved using a clean muslin cloth. Two hundred and fifty grammes of sugar (sucrose) was added to the filtrate (kunun zaki) as a sweetening agent. 2.3. Preparation of preservatives and preservation experiment procedure [3] The test preservatives used in this experiment were lemon juice, honey, and a 1:1 mixture of lemon juice and honey. Lemon fruits were washed with clean tap water, sliced into halves with the aid of a sterile knife, and fruit juice was squeezed out into a clean sterile beaker. The juice was sieved to remove seeds and chaff. Using a sterile syringe, 1.5, 2.0, 2.5 and 3.0 ml volumes of lemon juice were separately introduced into 100 ml volumes of kunun zaki in clean plastic screw-capped bottles. The kunu-lemon juice mixtures were hand-shaken to ensure proper mixing. The procedure was repeated with honey and also with the 1:1 mixture of lemon juice and honey. A total of 12 kunun zaki samples were thus treated in three groups of four samples each. Each group was treated with a different preservative at concentrations of 1.5, 2.0, 2.5 and 3.0 ml each in 100 ml of kunun zaki. The whole experiment was replicated thrice. 2.1. Sample collection and preparation [1] Raw materials used in the work (i.e. guinea corn, ginger, cloves, black pepper, dried sweet potato chips, sugar, lemon fruits and honey) were purchased from Terminus market in Jos metropolis of Plateau State, Nigeria. The raw materials were transported to the laboratory in clean polythene bags. The materials were sorted to remove unwanted particles including sand. 1. Introduction Kunun zaki is a non-alcoholic beverage mostly consumed by people of the Northern parts of Nigeria. It is a product of fermentation of grains. Kunun zaki is a Hausa phrase which means “sweet beverage”. Among the group of non-alcoholic beverage called kunu, kunun zaki is the most consumed in Nigeria with about 73% of Nigerians consuming it daily and 26% consuming it occasionally [1]. Osuntogun and Aboabo [2] reported that on dry matter basis, kunun zaki contains 76.3% starch, 11.6% protein, 3.3% fats, 1.9% fiber and 1.3% ash, showing it to be a nutritious product of fermentation. Kunun zaki, however, has been reported to have a short shelf life ranging from 18 – 36 hours without pasteurization and refrigeration [3]. The short shelf life of the product is attributable to high moisture content among other factors. Akoma et al. [4] reported a moisture content of 78.8% for kunun zaki. The high moisture content of kunun zaki predisposes the product to rapid microbial spoilage. Due to the short shelf life of the product, large scale production is often avoided to avoid economic losses, due to spoilage. This reduces the economic value of the product [4]. In an attempt to solve the problem of short shelf life associated with kunun zaki, preservatives of both natural and chemical sources have been used, resulting in some level of improvement in shelf life [5]. There is however, need to search further for more preservatives especially from natural sources, these being more acceptable than chemical preservatives. Enhancement of the shelf life of kunun zaki through the use of natural preservatives would reduce cost of preservation by pasteurization and refrigeration methods. It will also allow production of kunun zaki in a larger scale without fear of easy spoilage thus meeting market demands for the product. This will enhance the economic value of the product, retain its nutritional quality and also reduce the risk of health related issues due to development of Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 3.1. Physico-chemical characteristics of kunun zaki samples [7] pH values of the kunun zaki samples decreased with increase in keeping time for both control and for samples treated with lemon juice and honey. The daily pH values for the kunun zaki samples under preservation ranged from 5.35±0.01 to 6.57±0.03 for Day 1, 5.27±0.02 to 6.90±0.07 for Day 2, 5.21±0.05 to 6.27±0.03 for Day 3, 5.14±0.02 to 5.90±0.02 for Day 4 and 5.08±0.02 to 5.36±0.04 for Day 5. The lowest pH value was thus observed on day 5 for most of the samples with few exceptions. The observed decreases in pH values of the kunun zaki preserved with 3 ml of honey remained statistically non-significant (p> .05) until Day 4 while that of the kunun zaki sample preserved with 3 ml of lemon juice became significant by Day 3. The effects of the preservatives on the pH of the kunun zaki samples under preservation are presented in Table 1. Table 1 Effects of preservation with lemon juice and honey on pH of kunun zaki Samples (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 6.23±0.01a 6.41±0.02a 6.27±0.03a 5.90±0.02b 5.32±0.01b LJ(2.0) 6.28±0.04a 6.16±0.03ab 6.00±0.05b 5.23±0.04c 5.29±0.02c LJ(2.5) 6.33±0.03a 6.14±0.02ab 6.10±0.05b 5.20±0.01c 5.26±0.01c LJ(3.0) 6.38±0.04a 6.11±0.06b 6.01±0.07b 5.23±0.02c 5.30±0.01c H(1.5) 6.51±0.04a 6.90±0.07ab 6.09±0.04b 5.23±0.02c 5.24±0.01c H(2.0) 6.57±0.03a 6.40±0.03a 6.19±0.02a 5.32±0.02b 5.25±0.01b H(2.5) 6.36±0.05ab 6.25±0.05a 5.79±0.08b 5.30±0.01c 5.27±0.01c H(3.0) 6.29±0.06a 6.17±0.06ab 5.89±0.01ab 5.58±0.01ab 5.36±0.04b H+LJ(1.5) 6.31±0.04a 6.22±0.03ab 5.97±0.08b 5.29±0.03b 5.35±0.01b H+LJ(2.0) 6.30±0.07a 6.14±0.02a 5.83±0.03b 5.26±0.01b 5.35±0.01b H+LJ(2.5) 6.24±0.03a 6.11±0.03a 5.91±0.03b 5.26±0.03c 5.24±0.07c H+LJ(3.0) 6.28±0.03ab 6.11±0.03a 6.10±0.06a 5.25±0.03b 5.23±0.03b Control 5.35±0.01a 5.27±0.02a 5.21±0.05b 5.14±0.02c 5.08±0.02c LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml kunun zaki kept at room temperature without any preservative Table 1 Effects of preservation with lemon juice and honey on pH of kunun zaki LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). 2.6. Statistical analysis [6] The effect of lemon juice, honey and their combinations on pH, titratable acidity, and sensory quality were analyzed statistically with the aid of SPSS software version 16 using One-Way Analysis of Variance (ANOVA). P value less than .05 were considered statistically significant. Means with significant differences were separated using LSD. Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 2.5. Effects of lemon juice and honey on sensory quality of kunun zaki [5] Daily sensory analysis was conducted on the treated kununn zaki samples using a panel of four University of Jos students who were familiar with the product. Parameters analysed included aroma, taste, colour, texture and overall acceptability. The kunun zaki samples were served the panelists in coded transparent plastic cups. Each panelist was provided with water for mouth rinsing before and after assessment of each kunun zaki sample. The samples were assessed on a five point hedonic scale where 1 = dislike very much, 2 = dislike slightly, 3 = indifferent, 4 = like slightly, and 5 = like very much, a slight modification of the scale described by [7]. 2 3.2. Sensory quality of experimental kunun zaki samples under preservation [8] Table 3 Effects of lemon juice and honey preservatives on the aroma acceptability of kunun zaki Samples (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 4.75±0.50 a 3.25±0.50ac 2.75±1.50bc 2.00±0.82c 1.75±0.50c LJ(2.0) 4.75±0.50 a 3.50±1.29ab 2.50±0.58b 2.00±0.82 b 1.75±0.50 b LJ(2.5) 4.00±0.82a 3.50±1.00ab 2.50±0.58ab 2.00±0.82b 1.75±0.50b LJ(3.0) 3.75±0.96a 3.50±1.00 a 2.75±0.50a 2.00±0.00 b 1.25±0.25b H(1.5) 4.75±0.50 a 3.75±0.50ab 2.75±0.50bc 2.50±1.00bc 1.75±0.50c H(2.0) 4.50±0.58 a 3.00±0.00b 3.00±0.00b 2.50±1.00b 1.75±0.50b H(2.5) 4.75±0.50a 3.50±0.58b 3.00±0.00b 2.75±0.50bc 1.75±0.50c H(3.0) 4.25±0.96a 3.50±0.58a 3.25±0.50ab 3.00±0.82ab 1.75±0.50b H+LJ(1.5) 3.25±0.50a 2.75±0.50ab 2.50±0.58a 1.50±0.58b 1.25±0.96b H+LJ(2.0) 3.25±0.96a 3.25±0.50a 2.50±0.58a 1.50±0.96b 1.25±0.50b H+LJ(2.5) 3.50±0.58a 3.00±0.82ab 2.50±0.58ab 1.50±0.96b 1.00±0.58b H+LJ(3.0) 3.00±0.82a 3.00±0.00a 2.50±0.58a 1.50±0.82b 1.25±0.50b Control 3.50±0.58a 3.00±0.00ab 2.75±0.50a 1.75±0.96b 1.25±0.50b LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml kunun zaki kept at room temperature without any preservative. able 3 Effects of lemon juice and honey preservatives on the aroma acceptability of kunun zaki ; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row significantly different (p < .05). Control: 100 ml kunun zaki kept at room temperature without any preservative. Sensory quality evaluation revealed that all the kunun zaki products were acceptable on the day of production, but acceptability values gradually declined for all the sensory parameters (i.e. aroma, texture, color, taste and overall acceptability) as keeping time increased. The aroma of most of the samples including the control was acceptable up to Day 3. By Day 4, the aroma quality of the control and most of the other samples had significantly decreased (p < .05). Though there were gradual decreases in the aroma scores for the kunun zaki sample preserved with 3 ml of honey, the decreases were not statistically significant even up to the Day 5. Details of the effects of lemon juice and honey on the aroma of kunun zaki are given in Table 3. 3.1. Physico-chemical characteristics of kunun zaki samples [7] Control: 100 ml kunun zaki kept at room temperature without any preservative Table 2 Effects of preservation with lemon juice and honey on titratable acidity of kunun zaki Sample (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 1.23±0.15a 1.61±0.01ab 1.98±0.16b 2.03±0.05b 3.74±0.04c LJ(2.0) 1.12±0.07a 1.60±0.05b 1.90±0.05c 2.17±0.05d 3.04±0.03e LJ(2.5) 1.04±0.03a 1.54±0.08b 2.08±0.06c 2.06±0.06c 2.46±0.06d LJ(3.0) 1.46±0.22a 1.86±0.04ab 1.99±0.05ab 2.93±0.42b 2.51±0.06b H(1.5) 0.99±0.09a 1.32±0.06b 1.59±0.06bc 2.04±0.01cd 2.25±0.05d H(2.0) 0.84±0.29a 1.31±0.00ab 1.85±0.01bc 2.04±0.02c 2.41±0.03c H(2.5) 1.30±0.12a 1.46±0.03ab 1.63±0.1b 2.08±0.04c 2.36±0.04c H(3.0) 0.97±0.03a 1.02±0.02a 1.42±0.05a 1.64±0.05a 2.05±0.02b H+LJ(1.5) 1.30±0.21a 1.48±0.18ab 1.85±0.08bc 2.07±0.08bc 2.15±0.04c H+LJ(2.0) 1.31±0.18a 1.54±0.07ab 1.76±0.07bc 2.17±0.02cd 2.32±0.03d H+LJ(2.5) 1.21±0.05a 1.79±0.03b 1.99±0.05c 2.21±0.03d 2.35±0.03c H+LJ(3.0) 1.10±0.34ac 1.59±0.05ab 1.69±0.04b 2.33±0.03bc 2.36.0.08c Control 0.76±0.13a 1.48±0.13a 1.51±0.06a 1.57±0.01b 2.57±0.06c LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml kunun zaki kept at room temperature without any preservative LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml kunun zaki kept at room temperature without any preservative 3 Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 Table 2 shows the effects of preservation with lemon juice and honey on titratable acidity (TA) of the kunun zaki samples under preservation. The TA of all the kunun zaki samples increased with increase in keeping time. Highest TA was observed on day 5 for all the samples. TA values for the different samples under preservation ranged from 0.76±0.13 to 1.23±0.15 for Day 1, 1.48±0.13 to 1.61±0.01 for Day 2, 1.51±0.06 to 1.98±0.16 for Day 3, 1.57±0.01 to 2.03±0.05 for Day 4 and 2.57±0.06 to 3.74±0.04 for Day 5. The observed increases in TA values were statistically significant (P< .05). LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 The taste acceptability ratings of the kunun zaki samples under preservation are presented in Table 4. The most accepted sample was again the sample preserved with 3 ml of honey. The sample did not show any significant difference in the taste scores up to Day 5. The samples preserved with 2.5 ml and 3 ml of lemon juice were the next best in terms of taste acceptability. They had no significant changes in their taste scores up to Day 4. The control sample exhibited a significant decrease in taste acceptability after two days. able 5 Colour acceptability ratings of kunun zaki samples preserved with lemon juice and honey Table 5 Colour acceptability ratings of kunun zaki samples preserved with lemon juice and honey Samples (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 4.50±0.58a 4.25±0.50a 3.50±0.58ab 3.00±0.00bc 2.25±0.50c LJ(2.0) 5.00±0.00a 4.00±0.00b 3.25±0.50bc 2.75±0.50cd 2.25±0.50d L J(2.5) 4.50±0.58a 4.00±0.00ab 3.25±0.50abc 2.75±0.50bc 2.00±0.82c LJ(3.0) 4.75±0.50a 4.00±0.00ab 3.50±1.00abc 2.75±0.50bc 2.25±0.50c H(1.5) 4.75±0.50 a 4.00±0.00ab 3.25±1.26abc 2.75±0.50bc 2.25±0.50c H(2.0) 4.50±0.58 a 4.00±0.00ab 3.50±1.00abc 2.50±1.00bc 2.25±0.50c H(2.5) 4.50±0.58 a 3.75±0.50ab 3.50±1.00abc 2.75±0.50bc 2.00±0.82c H(3.0) 4.75±0.50a 4.00±.0.00ab 3.75±0.50a 3.25±0.50ab 2.25±0.50b H+LJ(1.5) 4.25±0.50a 4.00±0.00ab 3.00±0.00bc 2.25±0.96c 2.25±0.50c H+LJ(2.0) 4.50±0.58a 3.75±0.50ab 3.25±0.50abc 2.25±0.96bc 2.00±0.82 c H+LJ(2.5) 4.25±0.50a 3.75±0.50ab 3.25±0.50abc 2.25±0.96bc 1.75±0.96c H+LJ(3.0) 3.75±0.96a 3.75±0.50a 3.25±0.50ab 2.25±0.96ab 1.50±1.00b Control 4.75±1.00a 3.75±0.50a 3.00±0.82b 2.00±1.16b 1.50±1.00b LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative. Table 5 Colour acceptability ratings of kunun zaki samples preserved with lemon juice and honey LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative. In terms of colour acceptability, the scores for the control kunun zaki decreased significantly after two days while there were no significant changes for most of the other samples up to Day 3. The samples preserved with 3 ml of honey and 3 ml of the mixture of honey and lemon exhibited non-significant changes in colour up to Day 4. By Day 5 the decrease in colour acceptability values had become significant for all the samples. The texture quality of the control sample was not acceptable after two days while that of five other samples were acceptable up to Day 3. Only the sample preserved with 3 ml of honey remained acceptable up to Day 4. Details of the texture acceptability of kunun zaki samples preserved with lemon juice and honey are presented in Table 6. 3.2. Sensory quality of experimental kunun zaki samples under preservation [8] Table 4 Effects of preservation with lemon juice and honey on taste acceptability of kunun zaki samples Samples (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 5.00±0.00 a 3.25±0.50bc 2.25±0.96cd 1.75±0.50d 1.25±0.50d LJ(2.0) 4.50±0.58a 3.00±0.82b 2.50±0.58bc 2.00±0.00bc 1.50±0.58c LJ(2.5) 3.50±1.00a 2.75±0.96ab 2.00±0.00ab 2.00±0.00ab 1.50±0.58b LJ(3.0) 3.25±0.50a 2.75±0.96a 2.50±0.58a 2.00±0.00a 1.50±0.58b H(1.5) 4.50±0.58a 3.75±0.96ab 3.00±0.00abc 2.50±1.00bc 1.50±0.58c H(2.0) 4.50±0.58a 3.50±0.58ab 3.00±0.00abc 2.75±1.26bc 1.50±0.58c H(2.5) 4.50±1.00a 3.25±0.50ab 3.25±0.50ab 2.50±1.00bc 1.25±0.50c H(3.0) 4.00±0.82a 3.25±0.50a 3.25±0.50a 2.75±1.26a 2.00±0.60a H+LJ(1.5) 4.00±0.82a 3.25±0.82a 2.00±0.00b 1.75±0.82bc 1.00±0.58c H+LJ(2.0) 2.75±0.50a 3.00±0.00ab 2.25±0.50abc 2.00±0.82bc 1.25±0.50c H+LJ(2.5) 3.00±0.00a 2.75±0.50a 2.50±0.00a 1.50±0.82b 1.25±0.58b H+LJ(3.0) 2.75±0.50a 2.75±0.96a 1.50±0.50b 1.25±0.96b 1.00±0.58b Control 3.75±0.50a 3.50±0.58ab 3.00±0.82bc 1.75±1.50c 1.50±0.58cb LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative Effects of preservation with lemon juice and honey on taste acceptability of kunun zaki samples Table 4 Effects of preservation with lemon juice and honey on taste acceptability of kun 4 Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 4.1. Physico-chemical characteristics of kunun zaki [9] The pH and titratable acidity of the samples (treated and control samples) was observed to decrease and increase respectively as the period of keeping increased which is consistent with the findings of [8, 9, 10]. This increases in TA and decrease in pH is an indication of fermentation activity by microorganisms taking place in the samples resulting in acid production and accumulation which manifests as increase in titratable acidity and decrease in pH. This is typical of carbohydrate rich foods such as Kunun zaki. Therefore, effectiveness of a preservative is dependent on the ability of the preservative to inhibit the activity of microorganisms in the food sample thereby slowing down changes in TA and pH. The fact that changes in the TA and pH of the kunun zaki sample treated with honey (3ml) were not statistically significant up to Day 4 and only became significant on Day 5, added to the fact that the same sample recorded the lowest TA and highest pH values among all the other samples indicate that the 3 ml honey treatment must have had an inhibitory effect on the microorganisms present in the sample [11, 12]. Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 After Day 2, the control sample was no longer acceptable while seven out of the twelve samples preserved were found to be acceptable up to Day 3. Details of the overall acceptability of the samples are presented in Table 7. Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 Table 7 Overall acceptability scores of kunun zaki samples preserved with lemon juice and honey Samples (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 5.00±0.50a 4.00±0.50ab 3.00±0.83bc 3.00±0.58bc 2.00±0.50c LJ(2.0) 5.00±0.50a 4.00±0.58b 3.00±0.00bc 3.00±0.58bc 2.00±0.50c LJ(2.5) 4.00±0.00a 4.00±0.58ab 3.00±0.00ab 3.00±0.58bc 2.00±0.58c LJ(3.0) 3.80±0.50a 3.50±0.58ab 2.50±0.58bc 2.30±0.50c 2.00±0.00c H(1.5) 5.00±0.50a 4.00±0.50ab 3.00±0.50b 3.00±0.50bc 2.00±0.50c H(2.0) 5.00±0.58a 4.00±0.82ab 3.00±0.50abc 2.00±0.96bc 2.00±0.58c H(2.5) 4.00±0.82a 4.00±0.58ab 3.00±0.50ab 2.00±0.96bc 1.00±0.50c H(3.0) 5.00±0.58a 4.00±0.50a 4.00±0.50a 3.00±0.50a 3.00±0.58a H+LJ(1.5) 4.00±0.58a 3.00±0.50a 2.75±0.50ab 1.25±0.96b 1.00±0.50b H+LJ(2.0) 5.00±0.58a 4.00±0.50ab 3.00±0.50bc 3.00±1.00c 2.00±0.50c H+LJ(2.5) 4.00±0.82a 4.00±0.58ab 3.00±0.50abc 2.00±0.96bc 2.00±0.58c H+LJ(3.0) 4.00±0.50a 4.00±0.50ab 3.00±0.50abc 3.00±0.50bc 2.00±0.50c Control 4.75±0.58a 4.00±0.58ab 3.00±0.58bc 2.00±0.96c 2.00±0.58c LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative. Table 7 Overall acceptability scores of kunun zaki samples preserved with lemon juice and honey LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative. In terms of overall acceptability, the most acceptable kunun zaki sample was the sample treated with 3 ml of honey. It was found acceptable all through the 5-day preservation period. After Day 2, the control sample was no longer acceptable while seven out of the twelve samples preserved were found to be acceptable up to Day 3. Details of the overall acceptability of the samples are presented in Table 7. In terms of overall acceptability, the most acceptable kunun zaki sample was the sample treated with 3 ml of honey. It was found acceptable all through the 5-day preservation period. After Day 2, the control sample was no longer acceptable while seven out of the twelve samples preserved were found to be acceptable up to Day 3. Details of the overall acceptability of the samples are presented in Table 7. In terms of overall acceptability, the most acceptable kunun zaki sample was the sample treated with 3 ml of honey. It was found acceptable all through the 5-day preservation period. Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 Table 5 shows the mean daily ratings for colour acceptability of the kunun zaki samples. Table 6 Texture acceptability ratings of kunun zaki samples preserved with lemon juice and honey Samples (ml) Day 1 Day 2 Day 3 Day 4 Day 5 LJ(1.5) 4.75±0.50a 3.75±0.50ab 2.75±0.50bc 2.25±0.96c 1.75±0.50c LJ(2.0) 4.5±0.58 a 3.75±0.50ab 2.75±0.50bc 2.50±0.58c 1.75±0.50c LJ(2.5) 4.25±0.50a 3.75±0.50ab 2.75±0.50bc 2.25±0.96c 1.75±0.50c LJ(3.0) 4.00±0.82a 4.00±0.00a 2.50±0.58b 2.50±0.58b 2.00±0.00b H(1.5) 4.75±0.50a 4.25±0.50ab 3.25±0.50bc 2.75±0.50cd 1.75±0.50d H(2.0) 4.50±0.58a 3.75±0.50ab 3.00±0.00abc 2.75±1.26bc 1.75±0.50c H(2.5) 4.50±0.58a 3.75±0.50ab 3.00±0.00bc 2.75±0.50cd 2.00±0.00d H(3.0) 4.00±0.82a 3.75±0.50ab 3.25±0.50ab 3.00±0.82ab 2.25±0.50b H+LJ(1.5) 4.00±0.00a 3.50±1.00ab 2.50±0.58ab 1.25±1.00b 1.00±0.00b H+LJ(2.0) 4.00±0.00a 3.25±0.50ab 2.75±0.50abc 2.50±1.00bc 1.75±0.50c H+LJ(2.5) 4.00±0.82a 3.75±0.50a 2.00±0.58b 1.50±1.00b 1.25±0.00b H+LJ(3.0) 3.75±0.96ab 3.50±0.58a 2.50±0.50ab 1.25±0.96b 1.00±0.00b Control 3.50±0.58a 2.75±0.50a 1.50±0.50b 1.25±0.96b 1.00±0.00b LJ = Lemon Juice; H = Honey. Each value is the mean ± standard deviation of duplicate readings. Values with different letters in the same row are significantly different (p < .05). Control: 100 ml Kunun zaki kept at room temperature without any preservative. able 6 Texture acceptability ratings of kunun zaki samples preserved with lemon juice and honey Table 6 Texture acceptability ratings of kunun zaki samples preserved with lemon juice and hone The texture quality of the control sample was not acceptable after two days while that of five other samples were acceptable up to Day 3. Only the sample preserved with 3 ml of honey remained acceptable up to Day 4. Details of the texture acceptability of kunun zaki samples preserved with lemon juice and honey are presented in Table 6. 5 5 Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 In terms of overall acceptability, the most acceptable kunun zaki sample was the sample treated with 3 ml of honey. It was found acceptable all through the 5-day preservation period. After Day 2, the control sample was no longer acceptable while seven out of the twelve samples preserved were found to be acceptable up to Day 3. Details of the overall acceptability of the samples are presented in Table 7. 4.2. Sensory quality of kunun zaki [10] The sensory quality (aroma, taste, colour, texture and overall acceptability) of the variously treated kunun zaki samples declined with increase in keeping time. This agrees with the findings of Fapohunda and Adeware [13] and Ojimelukwe [10] who reported that the sensory quality of kunun zaki under preservation with natural products declined with time. This decline in sensory quality is suggestive of continuous microbial activities in the product which would result in the production of metabolic wastes such as hydrogen sulphide and other nitrogenous wastes that affect the sensory quality of the product. In spite of the aforesaid, the fact that the observed decline in taste, color, texture and overall acceptability for the treated samples remained statistically non-significant for a period of three days as against two days observed for the control sample suggests that the preservatives used must have had some inhibitory effects on spoilage organisms in the product. Though there were gradual decreases in the taste, aroma, color, texture and overall acceptability qualities of the sample treated with honey (3 ml), the changes were not statistically significant all through the five days of preservation. This retention of sensory quality of kunun zaki sample treated with 3ml of honey points to the preservative quality of honey on the product. The observed decline in aroma quality of the samples could be as result of production of off flavors associated with the production of hydrogen sulphide (H2S) and acetic acid [14]. The decline in taste of the samples could be as a result of continuous accumulation of acids and toxic metabolites as a product of microbial fermentation. Colour deterioration of the samples could have resulted from possible accumulation of microbial metabolic products which may have affected the colour. The decline in texture of the kunun zaki samples 6 Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 could have, among other reasons, resulted from possible increase in microbial load of the product over time. Kunun zaki preserved with 3 ml of honey was found to be the best in terms of overall acceptability. The overall acceptability of the kunun zaki samples under preservation would naturally depend on how acceptable the resultant changes in the other sensory characteristics are to the consumers. could have, among other reasons, resulted from possible increase in microbial load of the product over time. 4.3. Conclusion Appropriate concentrations of lemon juice, honey, and their combinations can be used as preservatives for extension of the shelf life of kunun zaki at room temperature (25±2 oC). Honey, when used at a concentration of 3 ml per 100 ml of kunun zaki, slows down undesirable changes in the physicochemical characteristics of kunun zaki. This concentration of honey brought about the least changes in pH and titratable acidity compared to the other preservative treatments and also exhibited the best potential for use in the preservation of kunun zaki. Honey can be used to extend shelf life of kunun zaki up to a period of five days when preserved at room temperature (25±2 oC). In addition to having the ability to extend the shelf life of kunun zaki, honey has been considered to be a highly nutritious substance due to the various nutrients found in it. Its addition to Kunun zaki will mean more nutrients for the product, thus improving its nutritional benefits for consumers. Though honey is relatively expensive, the extension of shelf life of kunun zaki by 3 days (from 2 to 5 days) probably justifies its use. Acknowledgments Acknowledgments Authors are grateful to Mrs. Mojisola Job of the Department of Science Laboratory Technology of the University of Jos for her technical assistance. 4.2. Sensory quality of kunun zaki [10] Kunun zaki preserved with 3 ml of honey was found to be the best in terms of overall acceptability. The overall acceptability of the kunun zaki samples under preservation would naturally depend on how acceptable the resultant changes in the other sensory characteristics are to the consumers. The acceptability of the samples treated with lemon juice (1.5, 2 and 2.5 ml), honey (1.5 and 3 ml) and their combination (2 and 3 ml) up to a period of four days of keeping suggests that these treatments possess preservative properties. The sample treated with 3 ml of honey, however, exhibited the best preservative property, being found to be acceptable throughout the 5-day preservation period. Honey can therefore be used at this concentration to extend the shelf life of kunun zaki for five days. The samples that contained the combination of lemon juice and honey showed the least preservative effect on the products. This could be as a result of the effect of lemon juice and honey on each other with possible formation of a more complex product. This is in agreement with the findings of [15] who reported that preservative effect of natural products decreased with increase in complexity of the food. Disclosure of conflict of interest Disclosure of conflict of interest The authors, Onyimba, Isaac Amechi and Dishon, Charibu Hurdison declare that there is no conflict of interest that exists. Onyimba IA and Dishon CH. (2019). Effectiveness of lemon juice and honey in shelf life enhancement of kunun zaki. World Journal of Advanced Research and Reviews, 1(1), 01-08. References [1] Sengev IA, Akpapunam MA and Ingbian EK. (2012). Physicochemical and sensory properties of instant kunun zaki flour bends from sorghum and mango mesocarp flour. Nigerian Food Journals, 30, 8-16. [1] Sengev IA, Akpapunam MA and Ingbian EK. (2012). Physicochemical and sensory properties of instant kunun zaki flour bends from sorghum and mango mesocarp flour. Nigerian Food Journals, 30, 8-16. [2] Osuntogun B and Aboabo OO. (2004). Microbial and physico-chemical evaluation of some non-alcoholic beverages. Pakistan Journal of Nutrition, 3(3), 188-192. [2] Osuntogun B and Aboabo OO. (2004). Microbial and physico-chemical evaluation of some non-alcoholic beverages. Pakistan Journal of Nutrition, 3(3), 188-192. [3] Elmahmood AM and Doughari JH. (2007). Microbial quality assessment of kunun zaki beverage sold in Girei town of Adamawa State, Nigeria. African Journal of Food Science, 1, 11-15 [3] Elmahmood AM and Doughari JH. (2007). Microbial quality assessment of kunun zaki beverage sold in Girei town of Adamawa State, Nigeria. African Journal of Food Science, 1, 11-15 [4] Akoma O, Daniel AA, Ajewole AE and Nwodo PG. (2014). Quality characteristics of kunun zaki (a Nigerian fermented cereal beverage) sold within Bida metropolis. Global Advanced Research Journal of Agricultural Science, 3(9), 298-303. [4] Akoma O, Daniel AA, Ajewole AE and Nwodo PG. (2014). Quality characteristics of kunun zaki (a Nigerian fermented cereal beverage) sold within Bida metropolis. Global Advanced Research Journal of Agricultural Science, 3(9), 298-303. [5] Udensi EA, Odom TC and Dike CO. (2012). Comparative studies of ginger (Zingiber officinale) and West African black pepper (piper gumeense) extracts at different concentration on the microbial quality of soyamilk and kunun zaki. Nigerian Food Journal, 30(2), 38-43. [5] Udensi EA, Odom TC and Dike CO. (2012). Comparative studies of ginger (Zingiber officinale) and West African black pepper (piper gumeense) extracts at different concentration on the microbial quality of soyamilk and kunun zaki. Nigerian Food Journal, 30(2), 38-43. [6] Onyimba IA, Itelima JU, Job MO, Ogbonna AI, Ode CO. (2017). Production of sorghum based kunun zaki using selected starter cultures. International Journal of Sciences, 6, 1-7. [6] Onyimba IA, Itelima JU, Job MO, Ogbonna AI, Ode CO. (2017). Production of sorghum based kunun zaki using selected starter cultures. International Journal of Sciences, 6, 1-7. [7] Larmond E. (1977). Laboratory methods for evaluation of food. Department of Agriculture, Ottawa. [7] Larmond E. (1977). Laboratory methods for evaluation of food. Department of Agriculture, Ottawa. References 7 7 Onyimba and Dishon / World Journal of Advanced Research and Reviews, 2019, 01(01), 001–008 [8] Agary OO, Nkama I and Akoma O. (2010). Production of kunun zaki (a Nigerian fermented cereal beverage) using starter cultures. International Research Journal of Microbiology, 1(2), 18-25. [8] [9] Adedokun II, Okorie SU, Nwokeke BC and Onyeneke EN. (2012). Effects of Aframomum danelli and black pepper crude extracts on physico-chemical sensory properties of kunun zaki during storage. Journal of Food Technology, 10(3), 97-102. [10] Ojimelukwe P, Elijah A, Ekong U and Nwokocha K. (2013). Effect of different preservatives on the shelf life of kunun zaki: a traditional fermented cereal based non-alcoholic beverages. Nigerian Journal of Agriculture, Food and Environment, 9(1), 76-79. [11] Krushna NS, Kowsalya A, Radha A and Narayanan RB. (2006). Honey as a natural preservative of milk. Indian Journal of Experimental Biology, 45, 459-464. [12] Hatayama LA. (2006). The effects of honey on longevity of fruits and vegetables. California State Science 2006 Project Summary, Project Number J0407. [13] Fapohunda SO and Adeware A. (2012). Microbial load and keeping quality of kunu under preservative regimes. Nutrition and Food Sciences, 2(4). [14] Seema R. (2015). Food spoilage: Microorganisms and their prevention. Asian Journal of Plant Sciences and Research, 5(4), 47-56. [15] Sultanbawa, Y. (2011). Plant antimicrobials in food applications Miniview. Ed. A. Me’ndez-vilas, Science against Microbial Pathogens: Communicating Current Research and Technological Advances, 1084-1093. How to cite this article 8 8

W2899024773.txt

https://www.degruyter.com/document/doi/10.1515/9783110618945-004/pdf

de

Der Superbulle und der Gangster

De Gruyter eBooks

cc-by

Hanno Berger Der Superbulle und der Gangster ich spreng’ euch alle in die luft! – inspektor blomfields fall nr. 1 a.k.a der superbulle Am 16. April 1968 erfolgte die Uraufführung von Rudolf Zehetgrubers ich spreng’ euch alle in die luft! – inspektor blomfields fall nr. 1 mit Götz George. Freigegeben wurde der Film ab 18 Jahren, zudem wurde er als „nicht feiertagsfrei“ eingestuft.1 Heute kann man den Film unter dem Titel der superbulle auf DVD kaufen. Allein schon, wenn man sich auf diese Eckdaten konzentriert, ergibt sich eine Fülle von Assoziationen: Götz George, der Sohn von Heinrich George, spielt im Jahre 1968 einen Inspektor. Durch den Namen George ist die deutsche Nazi-Vergangenheit präsent; durch „1968“ ist die Auseinandersetzung mit dieser Vergangenheit markiert; die Figur des Inspektors wiederum gehört zur Polizei, die Louis Althusser – ein Theoretiker, dessen Bedeutung für ’68 man nicht überschätzen kann – dem repressiven Staatsapparat zuordnet, der in erster Linie auf der Ebene der Repression (auch der physischen) und in zweiter Linie auf der Ebene der Ideologie arbeitet, um die Reproduktion der bestehenden Verhältnisse zu sichern.2 Und – erlaubt man sich für einen Augenblick eine eigentlich unzulässige nachträgliche Teleologisierung, die allerdings durch den Titel der DVD-Veröffentlichung nahegelegt wird – ein von Götz George gespielter Kommissar verweist natürlich auch auf „Schimanski“ und somit auf einen zentralen Protagonisten der späteren bundesrepublikanischen Unterhaltungskultur. Noch verstärkt werden diese Assoziationen, wenn man sich das – allerdings nicht aus dem Jahre 1968 stammende – DVD-Cover des Filmes anschaut (Abb. 1). Kaum möglich, bei dem „Super“ und dem vor Muskeln nahezu berstenden George nicht an das Konzept des Übermenschen zu denken, bzw. auf die nationalsozialistische Bezugnahme darauf. Auf der anderen Seite der „Bulle“: auch hier wieder die Autorität der Polizei, jedoch, und dies verweist dann wiederum auf die anti-autoritäre Seite von ’68, nicht der Inspektor oder der Polizist, sondern der Bulle. 1 Zu diesen Informationen vgl. https://www.filmportal.de/film/ich-spreng-euch-alle-in-dieluft-inspektor-blomfields-fall-nr-1 (14. 05. 2018). 2 Louis Althusser: Ideologie und ideologische Staatsapparate. In: ders: Marxismus und Ideologie. Westberlin 1973, S. 111–172, hier S. 130. Ein ideologischer Staatsapparat arbeitet in erster Linie auf der Ebene der Ideologie und in zweiter Linie auf der Ebene der Repression. Zu den ideologischen Staatsapparaten vgl. Althusser: Ideologie und ideologische Staatsapparate. https://doi.org/10.1515/9783110618945-004 88 Hanno Berger Abb. 1: Der Superbulle und seine Muskeln. Schaut man sich nach all diesen Assoziationen dann den Film an, so ist man zumindest überrascht und wahrscheinlich auch ein wenig enttäuscht. Der „Superbulle“ entpuppt sich als dann doch sehr zahmer, britischer Kommissar, der sich – wie uns das Voice-over gleich zu Beginn des Filmes wissen lässt – einzig durch seine „ausgeprägte Pedanterie“ von seinen Kollegen unterscheidet. Von einem Schimanski oder einem Über-Polizisten also keine Spur. Und dieser Kommissar eignet sich wohl genauso wenig als Repräsentant eines repressiven Staatsapparates, wie man an ihm eine Auseinandersetzung mit der nationalsozialistischen Vergangenheit Deutschlands festmachen könnte. Auch die anti- Der Superbulle und der Gangster 89 autoritäre Seite von ’68 scheint zunächst einmal nicht wiederzufinden zu sein. Durch die Besetzung des unvergleichlichen Eddi Arent als einen von Blomfields Kollegen unterhält der Film viel eher Beziehungen zum populären westdeutschen Genrekino der 1950er und frühen 1960er Jahre: ist es doch fast unmöglich, bei diesem Schauspieler nicht spontan an seine Rollen in den Edgar-Wallace- und Winnetou-Filmen zu denken. Und vergleicht man den Film mit dem ihn zeitgenössisch umgebenden Kino, so lassen sich wohl kaum Bezüge zu den Autorenfilmen der damaligen Zeit finden, die 1968 auch zu einem filmischen Aufbruch haben werden lassen. Vielmehr könnte man Parallelen zu den unzähligen Erotikfilmen ziehen, die zu dieser Zeit im Kino liefen. Arbeitet doch auch Zehetgrubers Film mit dem – stets als männlich inszenierten – voyeuristischen Blick auf Frauenkörper. Und auch wenn man normative Geschmacksurteile außen vor lassen will, so fällt es schwer, gewisse handwerkliche Ungereimtheiten zu übersehen. Es scheint also, als ob das Urteil des Lexikons des internationalen Films durchaus treffend formuliert ist: „Auf oberflächliche Spannung bedachter, routinierter Serienkrimi mit einigen geschmacklichen Entgleisungen.“ 3 Doch gibt es da noch die Figur des Johnny Smith. Weitaus interessanter als der pedantische „Superbulle“ erscheint die Figur, aus deren Perspektive der erste Teil des ursprünglichen Filmtitels formuliert ist: das ‚lyrische Ich‘ in der Aussage „Ich spreng’ Euch alle in die Luft“ ist Johnny Smith, der Blomfield die Schuld für den Tod seines Bruders gibt und, um diesen zu rächen, Blomfield umbringen möchte. Als er Blomfield nicht in der Polizeistation antrifft, nimmt er die dort anwesenden Personen als Geiseln und bedroht sie nicht nur mit einer Pistole, sondern droht auch, sich und alle anderen mit Hilfe einer Flasche Nitroglyzerins in die Luft zu sprengen. Gespielt wird Johnny Smith dabei von Werner Pochath, und schon ein flüchtiger Blick auf die Filmografie dieses Schauspielers in den Jahren nach seinem Auftritt in ich spreng’ euch alle in die luft! lässt erahnen, in welche Richtung seine Figur und sein Schauspiel gehen könnten. Er spielte vornehmlich in Exploitation-Filmen wie la banda j.s.: cronaca criminale del far west (die rote sonne der rache (Sergio Corbucci, I/BRD/E 1972)), mosquito der schänder (englischer Verleihtitel: bloodlust: the black forest vampire (Marijan Vajda, CH 1977)), il cacciatore di squali (dschungel-django (Enzo G. Castellari, MEX/I/E 1979)), la ragazza del vagone letto (horrorsex im nachtexpress (Ferdinando Baldi, I 1980)) oder maria – nur die nacht war zeuge (Ernst Hofbauer, BRD/I 1980). Doch was nun ist seine Rolle in Zehetgrubers Film aus dem Jahr 1968? Und wie wird er inszeniert? Insbesondere sein erster Auftritt in der Polizeistation verdient eine genauere Betrachtung. 3 https://www.zweitausendeins.de/filmlexikon/?sucheNach=titel&wert=31984 (14. 05. 2018). 90 Hanno Berger Zu einem bedrohlich anschwellenden Streicherton wird, in leicht-schräger Aufsicht auf deren unteres Ende, eine Tür geöffnet, und es treten zwei schwarze Schuhe mit einer Silberverzierung in ihrer Mitte ins Bild. Die Kamera fährt zu einigen wenigen tiefen Gitarrentönen eine eng anliegende Jeans entlang hoch und eröffnet dadurch zunächst den Blick auf einen opulenten weißen Gürtel und dann auf einen orangefarbenen Pullover, über dem eine lässige schwarze Lederjacke liegt und in dessen Mitte eine aus einem kleinen, sichelförmigen Stück Elfenbein bestehende Kette baumelt. Dann, mit dem Finale der kleinen Melodie auf der Tonspur, die mit einem Schlag auf eine Hi-Hat untermalt wird, erscheint Johnny Smiths Gesicht: kurze, an den Spitzen blondierte und leicht zur Seite gekämmte Haare, leichter Kotelettenansatz und ein Dreitagebart, die Augen hinter einer gelblich eingetönten eckigen Brille. Abgeschlossen wird die Inszenierung dieses Eintritts in die Polizeistation durch eine dem bisherigen Spannungsaufbau entsprechende Auflösung: fast schon wie in einer Persiflage auf einen Film-Bösewicht blickt Johnny Smith ostentativ einmal nach rechts und links. Sofort wird er von den anwesenden Personen für sein Aussehen und Auftreten kritisiert: „Junger Mann, bevor Sie in diesem Aufzug auf die Straße gehen, werfen Sie erstmal einen Blick in den Spiegel!“ Und kurz darauf insinuiert eine andere Figur, dass Johnny nicht arbeite. Johnny erwidert: „Der Kollege scheint einer von denen zu sein, die etwas gegen freie Meinungsäußerung zu haben scheinen. Bisschen verkalkt, wie?“, und setzt sich lässig an die Wand gelehnt auf ein Geländer innerhalb der Polizeistation. Später bekommt er zu hören: „Dann gibt es Schwierigkeiten für Dich, Kleiner“, worauf er erwidert: „Ich bin nicht Ihr Kleiner!“ Als Johnny einen Revolver zieht und auf die beiden Polizisten richtet, stellt einer von ihnen fest: „Er schießt wirklich. Er ist bis oben hin voll mit Koks.“ In den folgenden Einstellungen wird Johnny dann von der linken Seite im Halbschatten beleuchtet, was innerdiegetisch durch ein Fenster mit halbgeöffneter Jalousie motiviert wird, und ihn – alleine durch die Lichtsetzung – wie aus einem Film noir gefallen erscheinen lässt. Er wird so deutlich als Film-Figur markiert. Die Szene endet dann auch mit einer Kamerafahrt auf sein Gesicht, sodass dieses am Ende fast den gesamten Bildraum ausfüllt, die Umgebung in einem schwarzen Nichts versinkt und der Fokus, nachdem er die Brille abgenommen hat, nicht nur auf seinem Gesicht, sondern insbesondere auch auf seinen funkelnden blauen Augen liegt. Die Beziehung zu dem ihn umgebenden Raum, der eine Verbindung zum Raum der menschlichen Wahrnehmung zulassen könnte, wird mit und in dieser Kamerafahrt gekappt (Abb.2). Zum Vergleich: Die Einführung des Kommissars Blomfield erfolgt in einer Totalen auf die Außenseite der Polizeistation. In adrettem Hemd, Krawatte, Sakko und Trenchcoat tritt er heraus, die Kamera fährt zurück, er schaut kurz Der Superbulle und der Gangster 91 Abb. 2: Der Gangster als Filmfigur. auf die Uhr, geht über eine Straße und kommt so auf die Kamera zu. Während er gähnt (!), beschreibt ihn ein Voice-over-Kommentar als „kein Einfaltspinsel, aber auch kein Sherlock Holmes“, bevor dann das bereits erwähnte Attribut der ausgeprägten Pedanterie genannt wird. Der Unterschied in der Inszenierung und der durch die Inszenierung erfolgenden affektiven Modulierung könnte also kaum größer sein. Doch zurück zu Johnny: Die fast schon ostentative und an eine Persiflage grenzende Inszenierung als Filmbösewicht, das dandyhafte Outfit, die ihm zugeschriebene Arbeitsmoral eines Gammlers, die politischen Ansichten eines Mitglieds der APO und der Drogenkonsum eines Rockstars: Durch diesen Auftritt erscheint die Figur des Johnny Smith wie eine Amalgamierung aller gegenkulturellen Fantasien. Und man wundert sich nicht, wie sehr sich der Look dieses Film-Bösewichtes wiederfinden lässt in, zum Beispiel, den Bildern, die man von Andreas Baader kennt (Abb. 3). Auch hier: der Kotelettenansatz, der Dreitagebart, die Sonnenbrille, die lässig hängende Jacke. Damit soll keineswegs die These verfolgt werden, die Filmfigur nehme den Mode-Stil der RAF-Terroristen oder gar ihre Taten vorweg;4 auch soll nicht behauptet werden, es ließe sich eine direkte Linie von 4 Durch seine Drohung, mit dem Nitroglyzerin die gesamte Polizeistation und sich selber in die Luft zu sprengen, ließe sich auch eine Verbindung von Johnny Smith zu terroristischen 92 Hanno Berger Abb. 3: Der Terrorist mit Sonnenbrille. (den filmischen Inszenierungen um) 1968 zur RAF ziehen. Es soll keine Kausalität oder gar ein wie auch immer geartetes Spiegelverhältnis postuliert werden. Vielmehr geht es darum, die im einleitenden Essay dieses Sammelbandes angesprochene Heterogenität, die ’68 ausmache, herauszuarbeiten.5 So ist der superbulle eben nicht nur das Biedere und Pedantische sowie das SchlüpfrigVoyeuristische, sondern auch dessen – durchaus doppelbödige – Kehrseite. Lässt man sich den Blick nicht durch (nachträgliche) Pseudoevidenzen verstellen, so wird aus einem scheinbar so belanglosen ‚routinierten Serienkrimi‘ ein Beispiel für die Pluralität und Komplexität, die man im Kino von 1968 – auch abseits der in den Kanon des Autorenfilms aufgenommenen Klassikern – finden kann. Literaturverzeichnis Althusser, Louis: Ideologie und ideologische Staatsapparate. In: ders: Marxismus und Ideologie. Westberlin 1973, S. 111–172. Selbstmordattentätern ziehen. Doch wäre es auch in diesem Falle sicherlich verkürzt, diesbezüglich eine direkte Linie zu ziehen. 5 Vgl. Hermann Kappelhoff: Auf- und Abbrüche – die Internationale der Pop-Kultur, in diesem Band. Der Superbulle und der Gangster Filmografie ich spreng’ euch alle in die luft! – inspektor blomfields fall nr. 1. Reg. Rudolf Zehetgruber. BRD 1968. il cacciatore di squali. Reg. Enzo G. Castellari. MEX/I/S 1979. la banda j.s.: cronaca criminale del far west. Reg. Sergio Corbucci. I/BRD/S 1972. la ragazza del vagone letto. Reg. Ferdinando Baldi. I 1980. maria – nur die nacht war zeuge. Reg. Ernst Hofbauer. BRD/I 1980. mosquito der schänder. Reg. Marijan Vajda. CH 1977. 93 

https://openalex.org/W2883166516

https://hal.science/hal-02976326/document

English

Cryochemically Obtained Nanoforms of Antimicrobial Drug Substance Dioxidine and Their Physico-chemical and Structural Properties

Crystals

cc-by

To cite this version: Tatyana I. Shabatina, Olga I. Vernaya, Vladimir P. Shabatin, Iuliia V. Evseeva, Michail Ya Mel- nikov, et al.. Cryochemically Obtained Nanoforms of Antimicrobial Drug Substance Dioxidine and Their Physico-chemical and Structural Properties. Crystals, 2018, 8 (7), pp.298-1-298-15. ￿10.3390/cryst8070298￿. ￿hal-02976326￿ Received: 1 June 2018; Accepted: 17 July 2018; Published: 19 July 2018 Abstract: Nanoforms of the antimicrobial drug substance 2,3-bis-(hydroxymethyl) quinoxaline-N,N′- dioxide with particles sizes between 50 and 300 nm were obtained by cryochemical modification of the initial pharmaceutical substance using a freeze-drying technique and were characterized by different physicochemical methods (FTIR, UV-Vis, 1H-NMR, DSC, TG and X-ray diffraction) and transmission electron microscopy (TEM). The data obtained from FTIR- and UV–Vis-spectroscopy confirmed the unaltered chemical structure of dioxidine molecules due to the cryochemical modification method. At the same time, X-ray diffraction and thermal analysis data show the change of the crystal structure compared to the parameters of the initial pharmaceutical dioxidine substance. A higher dissolution rate was revealed for cryomodified dioxidine nanoforms. The existence of three polymorphic crystal phases was established for cryomodified dioxidine samples possessed by some thermal activation processes: two anhydrous polymorphic phases, triclinic (T) and monoclinic (M), and one hydrated form (H). Keywords: drug nanoforms; cryochemical synthesis; polymorphism; antimicrobial substances; dioxid crystals crystals www.mdpi.com/journal/crystals Crystals 2018, 8, 298; doi:10.3390/cryst8070298 HAL Id: hal-02976326 https://hal.science/hal-02976326v1 Submitted on 23 Oct 2020 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Cryochemically Obtained Nanoforms of Antimicrobial Drug Substance Dioxidine and Their Physico-chemical and Structural Properties Tatyana I. Shabatina 1,2,*, Olga I. Vernaya 1, Vladimir P. Shabatin 1, Iuliia V. Evseeva 1,2, Michail Ya Melnikov 1, Andrew N. Fitch 3 and Vladimir V. Chernyshev 1,4 ID 1 Department of Chemistry, M.V. Lomonosov Moscow State University, Leninskie Gori build. 1/3, Moscow 119991, Russia; [email protected] (O.I.V.); [email protected] (V.P.S.); [email protected] (I.V.E.); [email protected] (M.Y.M.); [email protected] (V.V.C.) 2 Bauman Moscow State Technical University, 2nd Bauman Str. 5, Moscow 105905, Russia 3 European Synchrotron Radiation Faculty, B.P. 220, 38043 Grenoble CEDEX, France; fi[email protected] 4 Frumkin Institute of Physical Chemistry and Electrochemistry RAS, Leninskii prospect 31, Moscow 119071, Russia d h b d l 1 Department of Chemistry, M.V. Lomonosov Moscow State University, Leninskie Gori build. 1/3, Moscow 119991, Russia; [email protected] (O.I.V.); [email protected] (V.P.S.); [email protected] (I.V.E.); [email protected] (M.Y.M.); [email protected] (V.V.C.) 2 Bauman Moscow State Technical University, 2nd Bauman Str. 5, Moscow 105905, Russia 3 European Synchrotron Radiation Faculty, B.P. 220, 38043 Grenoble CEDEX, France; fi[email protected] 4 Frumkin Institute of Physical Chemistry and Electrochemistry RAS, Leninskii prospect 31, Moscow 119071, Russia C d h b d l * Correspondence: [email protected]; Tel.: +7-495-939-5442 1. Introduction Development of a new commercial medicine is time-consuming and expensive, and the majority of costs are caused by pre-clinical and clinical research. Health safety of a new drug is one of the top priorities, implying the need for investigation of toxicity, optimal and maximal doses, and side effects that would manifest over decades and even generations. In that respect, obtaining new medications from effectiveness-proven components can reduce the research costs compared to investments into the development of novel drugs from scratch. The techniques of such drug renovation include combination of active pharmaceutical ingredients (API) from different medicines, reduction of ingredient particle size [1–6], amorphisation [7–10], co-crystal formation [1,11,12], self-emulsification [13], pH-modification [14] and complexation [15]. Particle diminution (micronization, nanonization or nanocrystallization) is a powerful approach that can increase the dissolution rate and saturation solubility, subsequently improving the bioavailability of drugs. It may also enhance the skin penetration ability of drugs and decrease systemic side effects [2,5]. Drug nanocrystals can be used for carrier-free colloidal drug delivery Crystals 2018, 8, 298; doi:10.3390/cryst8070298 www.mdpi.com/journal/crystals Crystals 2018, 8, 298 2 of 15 systems. According to the definition [2], the drug nanocrystals can be considered as species containing 100% drug substance with particle sizes ranging from 50 to 1000 nm. A stabilizing agent could be located only on their surface. Nanocrystals of drugs can be created by “top-down” or “bottom-up” approaches, or combinations of the two [5,16]. The “top-down” route refers to the breaking down of larger particles by milling or homogenization, while the “bottom-up” processes imply assembling and controlling of precipitations at nanometer scale. Currently, a number of nanosized dosage forms are presented in the pharmaceutical market. Rapamune®(Sirolimus), Emend®(Aprepitant), Tricor®(Fenofibrate), Megace ES®(Megestrol acetate), Avinza®(Morphine sulfate), Focalin®XR (Dexmethyl-phenidate HCl), Ritalin®LA (Methylphenidate HCl), Zanaflex Capsules™(Tizanidine HCl) are being obtained using “top-down” technique—wet milling, whereas Triglide®(Fenofibrate) is being produced by the “bottom-up” high-pressure homogenization method [2,5]. Many API solids can exist in more than one crystalline phase—polymorphs. They may form solvates—crystal forms including solvents. If the solvent is water, such forms are called hydrates. APIs can also be amorphous and not possess a distinguishable lattice. All these forms have the same chemical composition, but different solid-state structure, and therefore they may exhibit different physical properties (compatibility, apparent solubility, dissolution rate), solid state chemistry (reactivity), stability and bioavailability. 2.1. Materials Dioxidine substance produced by Mir-Pharma (Moscow, Russia) (98.9%) was used without further purification. 1. Introduction In order to produce new crystal forms, some special experimental techniques can be used, such as re-crystallization from a single or mixed solvents [8,17,18] using supercritical fluids [19,20], cryochemical sublimation of frozen solutions (freeze drying technique) [21–24], thermal activation of the solid substrates [9], crystallization from the melt [9], condensation from gas phase using inert gas carrier on the cooled supports [25,26]. It was shown recently that cryochemical treatment can be used in order to obtain new nanoforms of some pharmaceutical substances with nanosized particles and modified crystal structure [22,24,25,27–29]. The aim of this work is to develop cryochemical modification methods and to produce new cryochemical forms of drug substance dioxidine, 2,3-bis-(hydroxymethyl)-quinoxaline-N,N′-dioxide, possessing nanosized particles and modified crystal structure and high dissolution rate in comparison to the initial pharmaceutical substance. Dioxidine is well known antimicrobial substance used for therapy treatment of superlative infections due to inhibition of growth of many of gram-positive and gram-negative bacteria [30,31]. The therapeutic effect of this substance is based on the termination of DNA biosynthesis of microbial cells due to the deep disturbance of its nucleotide structure even by presence of sub-inhibiting concentrations. 2.3.1. UV-visible UV-visible absorption double beam spectrophotometer “Jasco V-770” (JASCO Corporation, Tokyo, Japan) in the range of 300–600 nm at room temperature was used to scrutinize the spectra of different forms of dioxidine in aqueous solutions and for spectrophotometric determination of dioxidine concentrations. 2.3.2. 1H-NMR Determination of the chemical shifts of the initial and cryochemically modified substances were held in the saturated solutions in D2O using high resolution NMR-spectrometer of VXR-400 “VARIAN” (VARIAN Inc. Agielent Technologies, Santa Clara, CA, USA). 2.3.6. BET-surface Area Measurements The BET surface area (S) of samples was defined using the surface analyzer assembled on the basis of gas-chromatograph Chrom 5. Average dioxidine particle diameters (d) of the samples were calculated as d = 6/ϱ S, where ϱ is a density of dioxidine substance. 2.3.3. FTIR Spectra FTIR spectra of samples were recorded in KBr discs between 4000 and 400 cm−1 using spectrometer Bruker Tensor II (Bruker GmbH, Mannheim, Germany). 2.3.4. Thermoanalytical Investigations 2.3.4. Thermoanalytical Investigations 2.3.4. Thermoanalytical Investigations Thermoanalytical investigations (thermogravimetry-TG and differential scanning calorimetry- DSC) were performed on Thermoanalyzer STA 449 C Jupiter, NETZSCH (NETZSCH GmbH, Selb Germany), using argon flow and samples heating rate of 10 K/min. 2.3.5. Transmission Electron Microscopy 2.3.5. Transmission Electron Microscopy 2.3.5. Transmission Electron Microscopy Transmission electron microscopy (TEM) images were recorded using microscope LEO 912 AB Omega ZEISS (ZEISS, Oberkochen, Germany) with electrons accelerating voltage 100–200 kV. Transmission electron microscopy (TEM) images were recorded using microscope LEO 912 AB Omega ZEISS (ZEISS, Oberkochen, Germany) with electrons accelerating voltage 100–200 kV. 2.3.6. BET-surface Area Measurements 2.2. Samples Preparation Cryomodified dioxidine: 4% dioxidine aqueous solution was heated to 50 ◦C, injected through a pneumatic nozzle into liquid nitrogen and freeze dried for 24 h. In order to obtain and solve the structure of pure crystal forms we carried out thermal treatment of cryomodified dioxidine substance by three different procedures. y p Thermally activated cryodioxidine samples: y p Thermally activated cryodioxidine samples: Sample 1: (D (50)-40-2 h): Cryomodified dioxidine was heated at 40 ◦C for 2 h, contains mainly H-crystal phase with small addition of M-phase (less than 25%) Sample 2: (D (50)-120-80 s): Cryomodified dioxidine was heated at 120 ◦C for 80 s contains three different crystal forms M (>30%), H and T (>60%). Sample 2: (D (50)-120-80 s): Cryomodified dioxidine was heated at 120 ◦C for 80 s contains three different crystal forms M (>30%), H and T (>60%). 3 of 15 Crystals 2018, 8, 298 Sample 3: (D (50)-120-8 h): Cryomodified dioxidine was heated at 120 ◦C for 8 h, contains only T crystal form. y where T—anhydrous triclinic crystal polymorph phase, M—anhydrous monoclinic crystal polymorph phase, H—hydrated crystal form. 2.3. Characterization Techniques 2.3.7. X-ray Powder Diffraction X-ray powder diffraction measurements were carried out at 250 K at beam line ID22 of the European Synchrotron Radiation Facility (ESRF, Grenoble, France). The instrument is equipped with a cryogenically cooled, double-crystal Si 111 monochromator and Si 111 analyzers. The powder was loaded into a 1-mm-diameter borosilicate thin-walled glass capillary, which was rotated during measurements at a rate of 1200 rpm to improve the powder averaging. Calibration of the instrument and refinement of the X-ray wavelength (= 0.399996(3) Å) were performed via NIST silicon standard 640c (see Table 1 for data collection details). 4 of 15 Crystals 2018, 8, 298 ble 1. Crystallographic data for two anhydrous (T and M) and a hydrated (H) forms of Dioxidine 1. Table 1. Crystallographic data for two anhydrous (T and M) and a hydrated (H) forms of Dioxidine 1. Crystal Polymorph Forms T M H Empirical formula C10H10N2O4 C10H10N2O4 C10H10N2O4·0.33H2O Formula weight 222.20 222.20 228.21 Crystal system triclinic monoclinic orthorhombic Space group P-1 P21/c Pbca a, Å 7.3194 (7) 9.1709 (9) 18.0035 (15) b, Å 8.0774 (8) 15.3486 (14) 31.8369 (19) c, Å 8.9872 (8) 7.0367 (7) 10.2555 (12) θ, deg 71.516 (11) 90 90 β, deg 70.815 (9) 110.316 (12) 90 θ, deg 79.317 (12) 90 90 V, Å3 473.98 (8) 928.87 (16) 5878.2 (9) M20 260 154 144 F30 583 (0.0014, 38) 421 (0.0023, 43) 397 (0.0021, 68) Z 2 4 24 Diffractometer ID22, ESRF ID22, ESRF ID22, ESRF Wavelength, Å 0.399996 (3) 0.399996 (3) 0.399996 (3) ρcalc, g/cm3 1.557 1.589 1.547 µ, mm−1 0.027 0.028 0.027 2θmin–2θmax, increment, deg 2.001–19.902, 0.003 1.200–19.995, 0.003 1.200–20.877, 0.003 Number of parameters/restraints 124/56 118/56 235/177 Rp/Rwp/Rexp 0.0316/0.0438/0.0168 0.0235/0.0317/0.0178 0.0243/0.0325/0.0165 goodness-of-fit 2.608 1.764 1.947 1 For the powder patterns with M and H forms the multi-phase Rietveld refinement was applied. 2.3.9. Structure Determination All crystal structures were solved with the use of simulated annealing technique [37]. For T and H, the low-angle part of powder pattern was used in simulated annealing runs, while for M, the search for the solution was based on the set of 150 low-angle Xobs values [38] extracted from three-phase powder pattern after a Pawley fit with the program MRIA [39]. The model of dioxidine molecule used in a direct space search without H atoms was taken from the literature [40]. 2.3.8. Indexing Unit-cell dimensions were determined using three indexing programs: TREOR90 [32], ITO [33], and AUTOX [34,35]. Indexing of M was possible after removing the peaks of T and H forms from the three-phases pattern of the sample 2. Space groups for all phases were assigned taking into account the systematic extinctions. The unit-cell parameters and space groups were tested further with the use of the Pawley fit [36] and confirmed by crystal structure solution. 2.3.9. Structure Determination 3.1. Physicochemical Characterization of Cryomodified Dioxidine The H-NMR spectra [1] of the initial dioxidine (D2O) δ: 4.93–5.21 (m, 4H, 2 CH2), 7.85–8.05 (m, 2H, H-Ar), 8.38–8.52 (m, 2H, H-Ar), and cryomodified dioxidine (D2O) δ: 4.95–5.25 (m, 4H, 2 CH2), 7.86–8.05 (m, 2H, H-Ar), 8.35–8.50 (m, 2H, H-Ar) are similar, confirming the identity of chemical structure of cryomodified and initial samples. The UV spectra of all forms of dioxidine aqueous solutions were identical and contained a high-intensity absorption band at 250 nm (doublet 241 and 259 nm) related to π→π* electron transition of the aromatic system and a low-intensity band at 375 nm related to n→π* electron transition of n-electrons of the nitrogen related to dioxidine molecule. In FTIR-spectra of the cryomodified dioxidine samples the vibrations of quinoxaline ring appear at 1510 cm−1, the vibrations of C-H of benzene ring appear at 975, 113cm−1, and the band of vibrations of C–O–H is indicated at 1288 cm−1 (Figure S1). While in the case of original dioxidine, the vibrations of the quinoxaline ring appear at 1506 cm−1, and the vibrational bands of the C–H bonds of aromatic ring are manifested by the peak at 971 cm−1 and doublet at 1117 and 1113 cm−1, the vibration band of the C–O–H is also doublet at 1280 and 1288 cm−1 (Figure S1). These data show the differences in dioxidine molecular conformations for the initial and cryomodified forms and combined with the results of UV-vis-spectroscopy confirmed the unchanged chemical structure of dioxidine molecules by cryochemical modification. At the same time, X-ray diffraction data show the difference between the initial pharmaceutical dioxidine substance (d (Å)-I, (%): 8.638–100.0%; 7.508–68.4%; 3.299–24.8%; 2.242–16.8%;) and the cryomodified form (d (Å)-I, (%): 8.740–100.0%; 8.026–94.2%; 3.358–99.3%; 3.304–67.6%;). Thermal analysis curves (TG and DSC) also confirmed the difference of physico-chemical behavior of the initial and cryomodified forms (Figure 1). For the initial dioxidine form the loss of physically adsorbed water is 0.83%, melting occurs at (175.5 ± 0.5 ◦C) with thermal an endothermic effect (99.8 ± 0,4) J/g, and thermal degradation proceeds at (199.2 ± 0.5) ◦C with an exothermic thermal effect (1135 ± 4) J/g and a mass loss 51,48%. 2.3.10. Structure Refinement The solutions were refined with the program MRIA via a multi-phase bond-restrained Rietveld approach in the same way as was reported earlier [27,41]. In the refinements, anisotropic line broadening [42] and preferred orientation approximated with a symmetrized harmonics expansion up to the fourth order [43,44] were taken into account. All non-H atoms were refined isotropically; H atoms were positioned geometrically (C–H 0.93–0.97 Å; O–H 0.82–0.85 Å) and refined as riding. The crystal structure H was solved and initially refined as anhydrous form ignoring the peaks of minor additional phase M (<25%). The preliminary refined structural model of H and final model of T were fixed and used in three-phase refinement of powder pattern 2, containing M in 30%, H- as a minor phase (<10%) and T as a main crystalline phase (>60%). Thus, a final structural model of M was obtained, fixed and then used in the two-phase Rietveld refinement of powder pattern 1, 5 of 15 Crystals 2018, 8, 298 containing H as a main crystalline phase. After several cycles of refinement, the positive residual peak with the height of 1.7 e Å−3 was observed far from the three independent molecules. It was assigned to the oxygen atom from the water molecule and freely refined. 2.4. Determination of Dissolution Rate of Dioxidine Forms Dioxidine samples (0.19 g) were placed into a flask with 20 mL of distilled water located on the activated magnetic stirrer with a stirring speed of 100 rpm at 25 ◦C. Solution samples with a volume of 0.1 mL were removed from the flask at defined intervals of time. Then the samples were diluted with distilled water by 20 times and then used for spectrophotometric determination of dioxidine concentration in the solution. 3. Results and Discussion 3.1. Physicochemical Characterization of Cryomodified Dioxidine 3.1. Physicochemical Characterization of Cryomodified Dioxidine For the cryomodified dioxidine form the loss of adsorbed water is 1.05%, melting occurs at (173.9 ± 0.5) ◦C with an endothermic effect (75.4 ± 0.4) J/g, and thermal degradation proceeds at (193.6 ± 0.5) ◦C with an exothermic effect of (1472 ± 4) J/g and a mass change of 31.74%. 6 of 15 6 of 15 Crystals 2018, 8, 298 Crystals 2018 8 x FO (a) (b) Figure 1. Thermogravimetry (TG) (green) and differential scanning calorimetry (DSC) (blue) curves of initial (a) and cryomodified (b) dioxidine forms. Figure 1. Thermogravimetry (TG) (green) and differential scanning calorimetry (DSC) (blue) curves of initial (a) and cryomodified (b) dioxidine forms. (a) (a) (b) (b) Figure 1. Thermogravimetry (TG) (green) and differential scanning calorimetry (DSC) (blue) curves of initial (a) and cryomodified (b) dioxidine forms. Figure 1. Thermogravimetry (TG) (green) and differential scanning calorimetry (DSC) (blue) curves of initial (a) and cryomodified (b) dioxidine forms. Cryochemical treatment led to the micronisation of the dioxidine substance and the formation of nanosized particles. By low-temperature argon adsorption, the value of specific surface area was determined, and average particle size was calculated for the initial (S = 0.7 m2/g, d = 5700 nm) and cryomodified dioxidine (21.3 m2/g and 190 nm). The similar data on cryomodified and initial dioxidine particles size were obtained by TEM (Figure 2). It was shown from the analysis of several micrographs that the particle size of the cryomodified dioxidine form is equal to 50–300 nm (average particle size 170 nm) and the particle size of the initial dioxidine form is equal to 150–18,000 nm (average particle size d = 5500 nm). Cryochemical treatment led to the micronisation of the dioxidine substance and the formation of nanosized particles. By low-temperature argon adsorption, the value of specific surface area was determined, and average particle size was calculated for the initial (S = 0.7 m2/g, d = 5700 nm) and cryomodified dioxidine (21.3 m2/g and 190 nm). The similar data on cryomodified and initial dioxidine particles size were obtained by TEM (Figure 2). It was shown from the analysis of several micrographs that the particle size of the cryomodified dioxidine form is equal to 50–300 nm (average particle size 170 nm) and the particle size of the initial dioxidine form is equal to 150–18,000 nm (average particle size d = 5500 nm). 3.1. Physicochemical Characterization of Cryomodified Dioxidine 7 of 15 7 f 15 Crystals 2018, 8, 298 C l 2018 8 FOR Figure 2. TEM micrograph of cryomodified dioxidine form. Figure 2. TEM micrograph of cryomodified dioxidine form. E IE Figure 2. TEM micrograph of cryomodified dioxidine form. Figure 2. TEM micrograph of cryomodified dioxidine form. Figure 2. TEM micrograph of cryomodified dioxidine form. Figure 2. TEM micrograph of cryomodified dioxidine form. Dissolution curves of the initial and cryomodified dioxidine forms are presented in Figure 3. While the cryomodified dioxidine completely dissolves in water after 5 s, reaching maximal concentration of 470 mg/mL at 25 °C, it takes more than 50 s for the initial dioxidine form to be fully dissolved. A higher dissolution rate for the cryomodified dioxidine is possibly due to the smaller size of its particles and to the larger value of its specific surface area compared to the original dioxidine. Also, different dissolution rates of two dioxidine forms can be related to differences in their crystal structure. Dissolution curves of the initial and cryomodified dioxidine forms are presented in Figure 3. While the cryomodified dioxidine completely dissolves in water after 5 s, reaching maximal concentration of 470 mg/mL at 25 ◦C, it takes more than 50 s for the initial dioxidine form to be fully dissolved. A higher dissolution rate for the cryomodified dioxidine is possibly due to the smaller size of its particles and to the larger value of its specific surface area compared to the original dioxidine. Also, different dissolution rates of two dioxidine forms can be related to differences in their crystal structure. Dissolution curves of the initial and cryomodified dioxidine forms are presented in Figure 3. While the cryomodified dioxidine completely dissolves in water after 5 s, reaching maximal concentration of 470 mg/mL at 25 °C, it takes more than 50 s for the initial dioxidine form to be fully dissolved. A higher dissolution rate for the cryomodified dioxidine is possibly due to the smaller size of its particles and to the larger value of its specific surface area compared to the original dioxidine. Also, different dissolution rates of two dioxidine forms can be related to differences in their crystal structure. Figure 3. Dissolution curves of initial and cryomodified dioxidine forms. .2. Crystal Structures Figure 3. Dissolution curves of initial and cryomodified dioxidine forms. Figure 3. Dissolution curves of initial and cryomodified dioxidine forms. ures Figure 3. 3.1. Physicochemical Characterization of Cryomodified Dioxidine Dissolution curves of initial and cryomodified dioxidine forms. Figure 3. Dissolution curves of initial and cryomodified dioxidine forms. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures shown in Table 1. Rietveld plots after the final refinements for powder patterns 1–3 are shown in Figures 4–6, respectively. Figure 7 shows powder pattern 2 and difference plot, calculated taking into account the contribution of two phases only—H and T, to demonstrate significant diffraction peaks from the third phase M. Crystal data, data collection, and refinement parameters for crystalline phases T, M and H are shown in Table 1. Rietveld plots after the final refinements for powder patterns 1–3 are shown in Figures 4–6, respectively. Figure 7 shows powder pattern 2 and difference plot, calculated taking into account the contribution of two phases only—H and T, to demonstrate significant diffraction peaks from the third phase M. Crystal data, data collection, and refinement parameters for crystalline phases T, M and H are shown in Table 1. Rietveld plots after the final refinements for powder patterns 1–3 are shown in Figures 4–6, respectively. Figure 7 shows powder pattern 2 and difference plot, calculated taking into account the contribution of two phases only—H and T, to demonstrate significant diffraction peaks from the third phase M. 8 of 15 Crystals 2018, 8, 298 Figure 4. Rietveld plot for the two-phase pattern of 1 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw) and M (bottom raw). Figure 4. Rietveld plot for the two-phase pattern of 1 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw) and M (bottom raw). Crystals 2018, 8, x FOR PEER REVIEW 8 of 15 Figure 4. Rietveld plot for the two-phase pattern of 1 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw) and M (bottom raw). Figure 4. Rietveld plot for the two-phase pattern of 1 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw) and M (bottom raw). Figure 4. Rietveld plot for the two-phase pattern of 1 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw) and M (bottom raw). Figure 4. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures Rietveld plot for the single-phase pattern of 3 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase T. Figure 6. Rietveld plot for the single-phase pattern of 3 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase T. Figure 6. Rietveld plot for the single-phase pattern of 3 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase T. Figure 7. A portion of the three-phase pattern of 2 showing the experimental intensities (black dots), calculated contribution of two phases—H + T (green curve), and difference red curve, corresponding to the contribution of phase M. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase M. The molecule of 2,3-bis(Hydroxymethyl)-quinoxaline-1,4-di-N-oxide (dioxidine) possesses only Figure 7. A portion of the three-phase pattern of 2 showing the experimental intensities (black dots), calculated contribution of two phases—H + T (green curve), and difference red curve, corresponding to the contribution of phase M. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase M. Figure 7. A portion of the three-phase pattern of 2 showing the experimental intensities (black dots), calculated contribution of two phases—H + T (green curve), and difference red curve, corresponding to the contribution of phase M. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase M. 7 A po tio of the th ee phase patte of 2 sho i g the expe i e tal i te sities (black do calculated contribution of two phases—H + T (green curve), and difference red curve, corresponding to the contribution of phase M. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase M. The molecule of 2,3-bis(Hydroxymethyl)-quinoxaline-1,4-di-N-oxide (dioxidine) possesses only Figure 7. A portion of the three-phase pattern of 2 showing the experimental intensities (black dots), calculated contribution of two phases—H + T (green curve), and difference red curve, corresponding to the contribution of phase M. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase M. Figure 7. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures Rietveld plot for the two-phase pattern of 1 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw) and M (bottom raw). Figure 5. Rietveld plot for the three-phase pattern of 2 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw), T (middle raw) and M (bottom raw). Figure 5. Rietveld plot for the three-phase pattern of 2 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw), T (middle raw) and M (bottom raw). Figure 5. Rietveld plot for the three-phase pattern of 2 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw), T (middle raw) and M (bottom raw). difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw), T (middle raw) and M (bottom raw). Figure 5. Rietveld plot for the three-phase pattern of 2 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw), T (middle raw) and M (bottom raw). Figure 5. Rietveld plot for the three-phase pattern of 2 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phases H (top raw), T (middle raw) and M (bottom raw). 9 of 15 Crystals 2018, 8, 298 Figure 6. Rietveld plot for the single-phase pattern of 3 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase T. Figure 6. Rietveld plot for the single-phase pattern of 3 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase T. Crystals 2018, 8, x FOR PEER REVIEW 9 of 15 Figure 6. Rietveld plot for the single-phase pattern of 3 showing the experimental (black) and difference (red) curves. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase T. Figure 6. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures In T, classical intermolecular O–H···O hydrogen bonds (Table 2) link the molecules into chains extended in [1-1-1] (Figure 10, prepared with Mercury [46]), and weak intermolecular C–H···O interactions (Table 2) held these chains together in the crystal. Table 2 Hydrogen-bonding geometry (Å degrees) in T M and H crystalline phases The dioxidine molecule has two hydroxy-groups, which can serve both as donors and acceptors in hydrogen-bonding, and four oxygen atoms can be considered as acceptors. Thus, one can expect various hydrogen-bonding networks, which can consolidate the crystal packing of the dioxidine molecules. In T, classical intermolecular O–H···O hydrogen bonds (Table 2) link the molecules into chains extended in [1-1-1] (Figure 10, prepared with Mercury [46]), and weak intermolecular C–H···O interactions (Table 2) held these chains together in the crystal. The dioxidine molecule has two hydroxy-groups, which can serve both as donors and acceptors in hydrogen-bonding, and four oxygen atoms can be considered as acceptors. Thus, one can expect various hydrogen-bonding networks, which can consolidate the crystal packing of the dioxidine molecules. In T, classical intermolecular O–H···O hydrogen bonds (Table 2) link the molecules into chains extended in [1-1-1] (Figure 10, prepared with Mercury [46]), and weak intermolecular C–H···O interactions (Table 2) held these chains together in the crystal. D–H···A D–H H···A D···A ∠O–H···A T O14−H14···O11i 0.82 1.95 2.721(3) 156 O16−H16···O12ii 0.82 1.92 2.733(3) 171 C13−H13B···O16iii 0.97 2.42 3.205(5) 138 C15−H15A···O11iv 0.97 2.52 3.241(5) 131 C8−H8···O12v 0.93 2.54 3.382(5) 150 M O14−H14···O11vi 0.82 2.21 3.001(6) 162 O16−H16···O14vii 0.82 1.86 2.588(6) 148 C13−H13B···O16viii 0.93 2.28 3.106(8) 147 C15−H15A···O14vii 0.97 2.37 2.922(8) 115 H O14A−H14A···O11C 0.82 2.13 2.741(8) 131 O14B−H14B···O12Cix 0.82 1.93 2.692(8) 154 O14C−H14C···O16Bx 0.82 1.92 2.710(8) 163 Table 2. Hydrogen-bonding geometry (Å, degrees) in T, M and H crystalline phase D–H···A D–H H···A D···A ∠O–H···A T O14−H14···O11i 0.82 1.95 2.721(3) 156 O16−H16···O12ii 0.82 1.92 2.733(3) 171 C13−H13B···O16iii 0.97 2.42 3.205(5) 138 C15−H15A···O11iv 0.97 2.52 3.241(5) 131 C8−H8···O12v 0.93 2.54 3.382(5) 150 M O14−H14···O11vi 0.82 2.21 3.001(6) 162 O16−H16···O14vii 0.82 1.86 2.588(6) 148 C13−H13B···O16viii 0.93 2.28 3.106(8) 147 C15−H15A···O14vii 0.97 2.37 2.922(8) 115 H O14A−H14A···O11C 0.82 2.13 2.741(8) 131 O14B−H14B···O12Cix 0.82 1.93 2.692(8) 154 O14C H14C O16B 0 82 1 92 2 710(8) 163 Table 2. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing (b) (b) (a) 018, 8, x FOR PEER REVIEW (a) (b) (b) (a) (a) Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres. Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres. ( ) ( ) Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres. Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres. Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres. ( ) ( ) Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres. Figure 9. The asymmetric unit of H containing three independent dioxidine molecules—A, B and C, and crystalline water molecule (O1W). Displacement spheres are shown at the 50% probability level. The dioxidine molecule has two hydroxy-groups which can serve both as donors and accep Figure 9. The asymmetric unit of H containing three independent dioxidine molecules—A, B and C, and crystalline water molecule (O1W). Displacement spheres are shown at the 50% probability level. Figure 9. The asymmetric unit of H containing three independent dioxidine molecules—A, B and C, and crystalline water molecule (O1W). Displacement spheres are shown at the 50% probability level. and crystalline water molecule (O1W). Displacement spheres are shown at the 50% probability level. The dioxidine molecule has two hydroxy groups which can serve both as donors and accept Figure 9. The asymmetric unit of H containing three independent dioxidine molecules—A, B and C, and crystalline water molecule (O1W). Displacement spheres are shown at the 50% probability level. Figure 9. The asymmetric unit of H containing three independent dioxidine molecules—A, B and C, and crystalline water molecule (O1W). Displacement spheres are shown at the 50% probability level. in hydrogen-bonding, and four oxygen atoms can be considered as acceptors. Thus, one can expect various hydrogen-bonding networks, which can consolidate the crystal packing of the dioxidine molecules. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures A portion of the three-phase pattern of 2 showing the experimental intensities (black dots), calculated contribution of two phases—H + T (green curve), and difference red curve, corresponding to the contribution of phase M. The vertical blue bars denote the calculated positions of diffraction peaks for crystalline phase M. o i e a eg ee o ee o , o i o o a io i a iou y a i e p a e ay i e i torsion angles N–C–C–O(H) only. In anhydrous polymorphs T and M, the asymmetric unit contains only one independent molecule, and conformations of these molecules are different (Figure 8, prepared with PLATON [45]). The asymmetric unit of the hydrated crystalline phase H contains three independent dioxidine molecules with different conformations and one water molecule (Figure 9). , ( y y y ) q , ( ) p y two internal degrees of freedom, so its conformations in various crystalline phases may differ in torsion angles N–C–C–O(H) only. In anhydrous polymorphs T and M, the asymmetric unit contains only one independent molecule, and conformations of these molecules are different (Figure 8, prepared with PLATON [45]). The asymmetric unit of the hydrated crystalline phase H contains three independent dioxidine molecules with different conformations and one water molecule (Figure 9). The molecule of 2,3-bis(Hydroxymethyl)-quinoxaline-1,4-di-N-oxide (dioxidine) possesses only two internal degrees of freedom, so its conformations in various crystalline phases may differ in torsion angles N–C–C–O(H) only. In anhydrous polymorphs T and M, the asymmetric unit contains only one independent molecule, and conformations of these molecules are different (Figure 8, prepared with PLATON [45]). The asymmetric unit of the hydrated crystalline phase H contains three independent dioxidine molecules with different conformations and one water molecule (Figure 9). 10 of 15 10 of 15 Crystals 2018, 8, 298 (a) 018, 8, x FOR PEER REVIEW (a) (a) (b) Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing the atomic numbering and 50% probability displacement spheres Figure 8. The molecular conformations of independent dioxidine molecule in T (a) and M (b) showing h i b i d 50% b bili di l h stals 2018, 8, x FOR PEER REVIEW 10 o (a) (b) Figure 8. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures Symmetry codes: (i) 2 −x, −y, −z; (ii) 1 −x, 1 −y, 1 −z; (iii) x, y −1, z; (iv) 1 −x, 1 −y, −z; (v) x, 1 + y, z; (vi) −x, −y, −z; (vii) 1 −x, −y, −z; (viii) −x, y + 1/2, 1 2 −z; (ix) 1 + x, 1 2 −y, z −1/2; (x) x −1/2, y, 3/2 −z; (xi) 1 −x, −y, 1 −z; (xii) x −1, y, z; (xiii) x −1/2, y, 1 2 −z; (xiv) 1 2 + x, 1 2 −y, 1 −z; (xv) 1 + x, y, z; (xvi) 1 2 −x, −y, z −1/2; (xvii) x −1, 1 2 −y, 1 2 −z. Symmetry codes: (i) 2 − x, −y, −z; (ii) 1 − x, 1 − y, 1 − z; (iii) x, y − 1, z; (iv) 1 − x, 1 − y, −z; (v) x, 1 + y, z; (vi) −x, −y, −z; (vii) 1 − x, −y, −z; (viii) −x, y + 1/2, ½ − z; (ix) 1 + x, ½ − y, z − 1/2; (x) x − 1/2, y, 3/2 − z; (xi) 1 − x, −y, 1 − z; (xii) x − 1, y,z; (xiii) x − 1/2, y, ½ − z; (xiv) ½ + x, ½ − y, 1 − z; (xv) 1 + x, y, z; (xvi) ½ − x, −y, z − 1/2; (xvii) x − 1, ½ − y, ½ − z. Figure 10. A portion of the crystal packing of T showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [1-1-1]. Figure 10. A portion of the crystal packing of T showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [1-1-1]. Figure 10. A portion of the crystal packing of T showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [1-1-1]. Figure 10. A portion of the crystal packing of T showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [1-1-1]. Classical intermolecular O H O hydrogen bonds in M (Table 2) link the molecules into centrosymetric dimers, which are further linked into chains running along axis a (Figure 11). These hydrogen-bonded chains in M are distinct from those in T. One hydroxy group in T (O14–H14) serves not only as a donor of hydrogen bond but as its acceptor too. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures O16B−H16B···O16Axi 0.82 2 2.769(8) 157 O16C−H16C···O11Bxii 0.82 2.07 2.827(8) 153 O1W−H1A···O16C 0.85 2.02 2.828(8) 157 O1W−H1B···O14Bxiii 0.85 2.07 2.815(8) 147 C5A−H5A···O16B 0.93 2.51 3.287(11) 141 C5B−H5B···O14Axi 0.93 2.57 3.490(9) 171 C6A−H6A···O16Cxiv 0.93 2.56 3.289(9) 136 C7B−H7B···O11Axi 0.93 2.28 3.180(11) 161 C13C−H13E···O11A 0.97 2.5 3.306(11) 140 C15A−H15B···O11Cxvi 0.97 2.37 3.138(11) 135 C15C−H15E···O11Bxvii 0.97 2.36 3.222(10) 148 Symmetry codes: (i) 2 − x, −y, −z; (ii) 1 − x, 1 − y, 1 − z; (iii) x, y − 1, z; (iv) 1 − x, 1 − y, −z; (v) x, 1 + y, z; (vi) −x, −y, −z; (vii) 1 − x, −y, −z; (viii) −x, y + 1/2, ½ − z; (ix) 1 + x, ½ − y, z − 1/2; (x) x − 1/2, y, 3/2 − z; (xi) 1 − x, −y, 1 − z; (xii) x − 1, y,z; (xiii) x − 1/2, y, ½ − z; (xiv) ½ + x, ½ − y, 1 − z; (xv) 1 + x, y, z; (xvi) ½ − x, −y, z − 1/2; (xvii) x − 1, ½ − y, ½ − z. Figure 10. A portion of the crystal packing of T showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [1-1-1]. Classical intermolecular O−H···O hydrogen bonds in M (Table 2) link the molecules int Figure 10. A portion of the crystal packing of T showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [1-1-1]. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures Hydrogen-bonding geometry (Å, degrees) in T, M and H crystalline phase D–H···A D–H H···A D···A ∠O–H···A T O14−H14···O11i 0.82 1.95 2.721(3) 156 O16−H16···O12ii 0.82 1.92 2.733(3) 171 C13−H13B···O16iii 0.97 2.42 3.205(5) 138 C15−H15A···O11iv 0.97 2.52 3.241(5) 131 C8−H8···O12v 0.93 2.54 3.382(5) 150 M O14−H14···O11vi 0.82 2.21 3.001(6) 162 O16−H16···O14vii 0.82 1.86 2.588(6) 148 C13−H13B···O16viii 0.93 2.28 3.106(8) 147 C15−H15A···O14vii 0.97 2.37 2.922(8) 115 H O14A−H14A···O11C 0.82 2.13 2.741(8) 131 O14B−H14B···O12Cix 0.82 1.93 2.692(8) 154 O14C−H14C···O16Bx 0.82 1.92 2.710(8) 163 D–H···A D–H H···A D···A ∠O–H···A T Table 2. Hydrogen-bonding geometry (Å, degrees) in T, M and H crystalline phases. D–H···A D–H H···A D···A ∠O–H···A Table 2. Hydrogen-bonding geometry (Å, degrees) in T, M and H crystalline phases. 11 of 15 Crystals 2018, 8, 298 Table 2. Cont. Table 2. Cont. O16A−H16A···O12A 0.82 2.37 2.884(8) 122 O16A−H16A···O12Axi 0.82 2.03 2.755(8) 147 O16B−H16B···O16Axi 0.82 2 2.769(8) 157 O16C−H16C···O11Bxii 0.82 2.07 2.827(8) 153 O1W−H1A···O16C 0.85 2.02 2.828(8) 157 O1W−H1B···O14Bxiii 0.85 2.07 2.815(8) 147 C5A−H5A···O16B 0.93 2.51 3.287(11) 141 C5B−H5B···O14Axi 0.93 2.57 3.490(9) 171 C6A−H6A···O16Cxiv 0.93 2.56 3.289(9) 136 C7B−H7B···O11Axi 0.93 2.28 3.180(11) 161 C13C−H13E···O11A 0.97 2.5 3.306(11) 140 C15A−H15B···O11Cxvi 0.97 2.37 3.138(11) 135 C15C−H15E···O11Bxvii 0.97 2.36 3.222(10) 148 Symmetry codes: (i) 2 −x, −y, −z; (ii) 1 −x, 1 −y, 1 −z; (iii) x, y −1, z; (iv) 1 −x, 1 −y, −z; (v) x, 1 + y, z; (vi) −x, −y, −z; (vii) 1 −x, −y, −z; (viii) −x, y + 1/2, 1 2 −z; (ix) 1 + x, 1 2 −y, z −1/2; (x) x −1/2, y, 3/2 −z; (xi) 1 −x, −y, 1 −z; (xii) x −1, y, z; (xiii) x −1/2, y, 1 2 −z; (xiv) 1 2 + x, 1 2 −y, 1 −z; (xv) 1 + x, y, z; (xvi) 1 2 −x, −y, z −1/2; (xvii) x −1, 1 2 −y, 1 2 −z. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures O16A−H16A···O12A 0.82 2.37 2.884(8) 122 O16A−H16A···O12Axi 0.82 2.03 2.755(8) 147 O16B−H16B···O16Axi 0.82 2 2.769(8) 157 O16C−H16C···O11Bxii 0.82 2.07 2.827(8) 153 O1W−H1A···O16C 0.85 2.02 2.828(8) 157 O1W−H1B···O14Bxiii 0.85 2.07 2.815(8) 147 C5A−H5A···O16B 0.93 2.51 3.287(11) 141 C5B−H5B···O14Axi 0.93 2.57 3.490(9) 171 C6A−H6A···O16Cxiv 0.93 2.56 3.289(9) 136 C7B−H7B···O11Axi 0.93 2.28 3.180(11) 161 C13C−H13E···O11A 0.97 2.5 3.306(11) 140 C15A−H15B···O11Cxvi 0.97 2.37 3.138(11) 135 C15C−H15E···O11Bxvii 0.97 2.36 3.222(10) 148 Symmetry codes: (i) 2 −x, −y, −z; (ii) 1 −x, 1 −y, 1 −z; (iii) x, y −1, z; (iv) 1 −x, 1 −y, −z; (v) x, 1 + y, z; (vi) −x, −y, −z; (vii) 1 −x, −y, −z; (viii) −x, y + 1/2, 1 2 −z; (ix) 1 + x, 1 2 −y, z −1/2; (x) x −1/2, y, 3/2 −z; (xi) 1 −x, −y, 1 −z; (xii) x −1, y, z; (xiii) x −1/2, y, 1 2 −z; (xiv) 1 2 + x, 1 2 −y, 1 −z; (xv) 1 + x, y, z; (xvi) 1 2 −x, −y, z −1/2; (xvii) x −1, 1 2 −y, 1 2 −z. O16B−H16B···O16Axi 0.82 2 2.769(8) 157 O16C−H16C···O11Bxii 0.82 2.07 2.827(8) 153 O1W−H1A···O16C 0.85 2.02 2.828(8) 157 O1W−H1B···O14Bxiii 0.85 2.07 2.815(8) 147 C5A−H5A···O16B 0.93 2.51 3.287(11) 141 C5B−H5B···O14Axi 0.93 2.57 3.490(9) 171 C6A−H6A···O16Cxiv 0.93 2.56 3.289(9) 136 C7B−H7B···O11Axi 0.93 2.28 3.180(11) 161 C13C−H13E···O11A 0.97 2.5 3.306(11) 140 C15A−H15B···O11Cxvi 0.97 2.37 3.138(11) 135 C15C−H15E···O11Bxvii 0.97 2.36 3.222(10) 148 Symmetry codes: (i) 2 − x, −y, −z; (ii) 1 − x, 1 − y, 1 − z; (iii) x, y − 1, z; (iv) 1 − x, 1 − y, −z; (v) x, 1 + y, z; (vi) −x, −y, −z; (vii) 1 − x, −y, −z; (viii) −x, y + 1/2, ½ − z; (ix) 1 + x, ½ − y, z − 1/2; (x) x − 1/2, y, 3/2 − z; (xi) 1 − x, −y, 1 − z; (xii) x − 1, y,z; (xiii) x − 1/2, y, ½ − z; (xiv) ½ + x, ½ − y, 1 − z; (xv) 1 + x, y, z; (xvi) ½ − x, −y, z − 1/2; (xvii) x − 1, ½ − y, ½ − z. 4. Conclusions Th i 4. Conclusions The improvement of medical therapeutic effects of known medications by obtaining nanoforms possessing nanosized particle and the modification of their crystal structure without changing of the chemical nature is one of the most important problems in modern nanomedicine and physicochemical pharmacy. The developed approaches allowed enhance biomedical efficiency of many drugs along with the reduction of unwanted side effects due to lowering of the required therapeutic doses and decreasing of their toxicity. The method developed by the authors in this article employed the capabilities of cryochemical The improvement of medical therapeutic effects of known medications by obtaining nanoforms possessing nanosized particle and the modification of their crystal structure without changing of the chemical nature is one of the most important problems in modern nanomedicine and physicochemical pharmacy. The developed approaches allowed enhance biomedical efficiency of many drugs along with the reduction of unwanted side effects due to lowering of the required therapeutic doses and decreasing of their toxicity. The method developed by the authors in this article employed the capabilities of cryochemical modification of solid organic compounds, which allowed us to convert the drug substances into nanoforms, and modify their crystal packing without changing their chemical structure under the conditions far from equilibrium. In most cases, amorphous or metastable crystal forms were obtained by this approach, which are kinetically frozen in the solid state and can exist for a long time. The bioavailability of the thus prepared nanoforms of drug substances are usually higher due to the higher dissolution rates and due to the ability of the nanosized particles to be transported by the bloodstream, to overcome the skin barriers and high permeability of cell membranes by nanosized particles. In this work we have succeeded to produce cryochemically modified nanoforms of antimicribial The method developed by the authors in this article employed the capabilities of cryochemical modification of solid organic compounds, which allowed us to convert the drug substances into nanoforms, and modify their crystal packing without changing their chemical structure under the conditions far from equilibrium. In most cases, amorphous or metastable crystal forms were obtained by this approach, which are kinetically frozen in the solid state and can exist for a long time. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures And, one of two N-oxide oxygen atoms (O12) is not involved in classical O−H···O hydrogen-bonding. These considerations led to the conclusion that phase T should be more stable than M, i.e., phase M is metastable one. Weak intermolecular C–H···O interactions in M (Table 2), like in T, consolidate further the crystal packing (Figure 11). Classical intermolecular O−H···O hydrogen bonds in M (Table 2) link the molecules into centrosymetric dimers, which are further linked into chains running along axis a (Figure 11). These hydrogen-bonded chains in M are distinct from those in T. One hydroxy group in T (O14–H14) serves not only as a donor of hydrogen bond but as its acceptor too. And, one of two N-oxide oxygen atoms (O12) is not involved in classical O−H···O hydrogen-bonding. These considerations led to the conclusion that phase T should be more stable than M, i.e., phase M is metastable one. Weak intermolecular C–H···O interactions in M (Table 2), like in T, consolidate further the crystal packing (Figure 11). 12 of 15 12 f 15 Crystals 2018, 8, 298 C l 2018 8 FOR Figure 11. A portion of the crystal packing of M showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [100]. Figure 11. A portion of the crystal packing of M showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [100]. Figure 11. A portion of the crystal packing of M showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [100]. Figure 11. A portion of the crystal packing of M showing the O−H···O hydrogen-bonded (thin purple lines) chain extended in [100]. Classical intermolecular O−H···O hydrogen bonds in H (Table 2) link dioxidine molecules and crystalline water molecules directly into three-dimensional networks, so numerous weak C−H···O interactions (Table 2) in H can be considered forced. Interestingly, some of the independent dioxidine molecules in H have the same peculiarities aforementioned in M, namely, one hydroxy group serves as a donor and an acceptor of hydrogen bonds, and one N-oxide oxygen is not involved in classical O−H···O hydrogen-bonding. The established crystal structures of H, M and T-forms allow us to estimate phase content of Classical intermolecular O−H···O hydrogen bonds in H (Table 2) link dioxidine molecules and crystalline water molecules directly into three-dimensional networks, so numerous weak C−H···O interactions (Table 2) in H can be considered forced. Crystal data, data 3.2. Crystal Structures C t l d t d t 3.2. Crystal Structures Interestingly, some of the independent dioxidine molecules in H have the same peculiarities aforementioned in M, namely, one hydroxy group serves as a donor and an acceptor of hydrogen bonds, and one N-oxide oxygen is not involved in classical O−H···O hydrogen-bonding. The established crystal structures of H, M and T forms allow us to estimate phase content of source samples of dioxidine. So multi-phase Rietveld refinement of powder pattern of commercial sample (see Figure S2) has shown that it contains two crystalline phases H and M in a ratio of 1:5. Powder pattern of the cryomodified sample (Figure S3) indicates that it contains mainly amorphous phase, some amount of H-phase and trace amount of M-phase. The established crystal structures of H, M and T-forms allow us to estimate phase content of source samples of dioxidine. So multi-phase Rietveld refinement of powder pattern of commercial sample (see Figure S2) has shown that it contains two crystalline phases H and M in a ratio of 1:5. Powder pattern of the cryomodified sample (Figure S3) indicates that it contains mainly amorphous phase, some amount of H-phase and trace amount of M-phase. There is patent [29] resulting from the work reported in this manuscript. There is patent [29] resulting from the work reported in this manuscript. Supporting Materials: X-ray Crystallographic file in CIF format. Crystallographic data for H, M and T have been deposited with the Cambridge Crystallographic Data Centre as the supplementary publication nos. CCDC 1533096-1533098. Copies of data can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: (44)1223-336-033; e-mail: [email protected]). Figures of the X-ray powder patterns of the commercial source and just after cryomodification of the samples. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4352/8/7/298/s1. Figure S1. FTIR spectra of the samples were recorded in KBr discs between 4000 cm−1 and 400 cm−1 using spectrometer Bruker Tensor II (Germany). The FTIR spectra of the cryomodified (C) and commercial source (S) samples of dioxidine are presented in Figure S1. Figure S2, X-ray powder patterns of the commercial source (S) sample was measured on a Guinier-Huber camera in a transmission mode, CuK1-radiation. Figure S3, X-ray powder pattern of the just after cryomodification (C) sample was measured on a Guinier-Huber camera in a transmission mode, CuK1-radiation. Author Contributions: Conceptualization, Methodology and Editing—T.I.S. and V.V.C., Supervision. Funding Acquisition and Validation—M.Y.M., Investigation and Writing-Original Draft—O.I.V., I.V.E., V.P.S. Resources, Data Curation and Formal Analysis—A.N.F. Funding: This research was funded by [Russian Science Foundation] grant number [16-13-10365] for financial support. V.V.C. is grateful to the Russian Ministry of Science and Education for the financial support of X-ray part of this study (grant No. RFMEFI61616X0069). Acknowledgments: Authors thank RFBR 18-53-80072, T.I.S., O.I.V., V.P.S., I.V.E. & M.Y.M. thank Russian Science Foundation (grant No. 16-13-10365) for financial support. V.V.C. is grateful to the Russian Ministry of Science and Education for the support of X-ray part of this study (grant No. RFMEFI61616X0069). We also thank ESRF for the access to ID22 station, experiment MA-3313. Conflicts of Interest: The authors declare no conflicts of interest. 4. Conclusions Th i 4. Conclusions The bioavailability of the thus prepared nanoforms of drug substances are usually higher due to the higher dissolution rates and due to the ability of the nanosized particles to be transported by the bloodstream, to overcome the skin barriers and high permeability of cell membranes by nanosized particles. In this work we have succeeded to produce cryochemically modified nanoforms of antimicribial drug substance 2,3-bis-(hydroxymethyl)quinoxaline-N,N'-dioxide (dioxidine) with particles size varied from 50 till 300 nm. The cryomodified samples were characterized by different physicochemical methods (FTIR, UV-Vis, 1H-NMR, DSC, TG, BET-chromatography and X-ray diffraction) and transmission electron microscopy (TEM). We established the identity of the chemical In this work we have succeeded to produce cryochemically modified nanoforms of antimicribial drug substance 2,3-bis-(hydroxymethyl)quinoxaline-N,N′-dioxide (dioxidine) with particles size varied from 50 till 300 nm. The cryomodified samples were characterized by different physicochemical methods (FTIR, UV-Vis, 1H-NMR, DSC, TG, BET-chromatography and X-ray diffraction) and transmission electron 13 of 15 Crystals 2018, 8, 298 13 of 15 microscopy (TEM). We established the identity of the chemical nature of the dioxidine drug substance. We have shown the modification of crystal structure of dioxidine cryoforms. The formation of three polymorph crystal forms was revealed for cryomodified dioxidine samples possessed by some thermal activation processes—two anhydrous polymorphs: triclinic (T), monoclinic (M), and hydrated form (H). The extending of the number of antibiotic drug substances in the experiments on cryochemical modification and detailed study of their biomedical efficiency and bioavailability are in our future plans. References Pharm. Res. 2003, 20, 531–536. [CrossRef] [PubMed] g pp 0. Guranda, D.T.; Gil’deeva, G.N. Preparation of drug polymorphs. Pharm. Chem. J. 2010, 44, 254–260. [CrossRe 11. Blagden, N.; de Matas, M.; Gavan, P.T.; York, P. Crystal engineering of active pharmaceutical ingredients to improve solubility and dissolution rates. Adv. Drug Deliv. Rev. 2007, 59, 617–630. [CrossRef] [PubMed] 12. Bak, A.; Gore, A.; Yanez, E.; Stanton, M.; Tufekcic, S.; Syed, R.; Akrami, A.; Rose, M.; Surapaneni, S.; Bostick, T.; et al. The co-crystal approach to improve the exposure of a water-insoluble compound: amg 517 sorbic acid co-crystal characterization and pharmacokinetics. J. Pharm. Sci. 2008, 97, 3942–3956. [CrossRef] [PubMed] 13. Kang, B.K.; Lee, J.S.; Chon, S.K.; Jeong, S.Y.; Yuk, S.H.; Khang, G.; Lee, H.B.; Cho, S.H. Development of self-microemulsifying drug delivery systems (SMEDDS) for oral bioavailability enhancement of simvastatin in beagle dogs. Int. J. Pharm. 2004, 274, 65–73. [CrossRef] [PubMed] 14. Taniguchi, C.; Kawabata, Y.; Wada, K.; Yamada, S.; Onoue, S. Microenvironmental pH-modification to improve dissolution behavior and oral absorption for drugs with pH-dependent solubility. Expert. Opin. Drug. Deliv. 2014, 11, 505–516. [CrossRef] [PubMed] 5. Loftsson, T.; Brewster, M.E. Pharmaceutical applications of cyclodextrins. 1. Drug solubilization and stabiliza J. Pharm. Sci. 1996, 85, 1017–1025. [CrossRef] [PubMed] 16. Muller, R.H.; Gohla, S.; Keck, C.M. State of the art of nanocrystals—special features, production, nanotoxicology aspects and intracellular delivery. Eur. J. Pharm. Biopharm. 2011, 78, 1–9. [CrossRef] [PubMed] 17. Henwood, S.Q.; Liebenberg, W.; Tiedt, L.R. Characterization of the solubility and dissolution properties of several new rifampicin polymorphs, solvates, and hydrates. Drug Dev. Ind. Pharm. 2001, 27, 1017–1030. [CrossRef] [PubMed] 18. Braun, D.E.; Gelbrich, T.; Kahlenberg, V.; Tessadri, R.; Wieser, J.; Griesser, U.J. Characterization of the solubility and dissolution properties of several new rifampicin polymorphs, solvates, and hydrates. J. Pharm. Sci. 2008, 98, 2010–2026. [CrossRef] [PubMed] 19. Velaga, S.P.; Berger, R.; Carlfors, J. Supercritical fluids crystallization of budesonide and flunisolide. Pharm. Res. 2002, 19, 1564–1571. [CrossRef] [PubMed] 20. Pasquali, I.; Bettini, R.; Giordano, F. Supercritical fluid technologies: an innovative approach for manipulating the solid-state of pharmaceuticals. Adv. Drug Deliv. Rev. 2008, 60, 399–410. [CrossRef] [PubMed] 21. Sergeev, G.B.; Komarov, V.S.; Shabatin, V.P. Method of Obtaining Powders of Drugs. Patent RU 2195264, 25 December 2002. 22. Vernaya, O.I.; Shabatin, V.P.; Semenov, A.M.; Shabatina, T.I. Obtaining ultradispersed dioxidine powder modified via cryochemical synthesis and determining its antibacterial activity. Mosc. Univ. Chem. Bull. 2016, 71, 295–298. [CrossRef] 23. References . Peltonen, L.; Hirvonen, J. Pharmaceutical Nanocrystals by Nanomilling: Critical process parameters, par fracturing and stabilization methods. J. Pharm. Pharmacol. 2010, 62, 1569–1579. [CrossRef] [PubMed] . Sun, B.; Yeo, Y. Nanocrystals for the parenteral delivery of poorly water-soluble drugs. Curr. Opin. Solid S Mater. Sci. 2012, 16, 295–301. [CrossRef] [PubMed] 2. Sun, B.; Yeo, Y. Nanocrystals for the parenteral delivery of poorly water-soluble drugs. Curr. Opin. Solid State Mater. Sci. 2012, 16, 295–301. [CrossRef] [PubMed] 3. Keck, C.M.; Muller, R.H. Drug nanocrystals of poorly soluble drugs produced by high pressure homogenization. Eur. J. Pharm. Biopharm. 2006, 62, 3–16. [CrossRef] [PubMed] 4. Gao, L.; Zhang, D.; Chen, M. Drug nanocrystals for the formulation of poorly soluble drugs and its application as a potential drug delivery system. J. Nanopart. Res. 2008, 10, 845–862. [CrossRef] 5. Junyaprasert, V.B.; Morakul, B. Nanocrystals for enhancement of oral bioavailability of poorly water-soluble drugs. Asian J. Pharm. Sci. 2015, 10, 13–23. [CrossRef] 5. Junyaprasert, V.B.; Morakul, B. Nanocrystals for enhancement of oral bioavailability of poorly water-soluble drugs. Asian J. Pharm. Sci. 2015, 10, 13–23. [CrossRef] 6. Sarnes, A.; Kovalainen, M.; Häkkinen, M.R.; Laaksonen, T.; Laru, J.; Kiesvaara, J.; Ilkka, J.; Oksala, O.; Rönkkö, S.; Järvinen, K.; et al. Nanocrystal-based peroral itraconazole delivery: superior in vitro dissolution enhancement versus sporanox®is not realized in In Vivo drug absorption. J. Control. Release 2014, 180, 109–116. [CrossRef] [PubMed] 7. Sarma, B.; Chen, J.; His, H.-Y.; Myerson, A.S. Solid forms of pharmaceuticals: polymorphs, salts and cocrystals. Korean J. Chem. Eng. 2011, 28, 315–322. [CrossRef] 7. Sarma, B.; Chen, J.; His, H.-Y.; Myerson, A.S. Solid forms of pharmaceuticals: polymorphs, salts and cocrystals. Korean J. Chem. Eng. 2011, 28, 315–322. [CrossRef] 8. Lee, E.H. A practical guide to pharmaceutical polymorph screening & selection. Asian J. Pharm. Sci. 2014, 9, 163–175. 8. Lee, E.H. A practical guide to pharmaceutical polymorph screening & selection. Asian J. Pharm. Sci. 2014, 9, 163–175. 14 of 15 14 of 15 Crystals 2018, 8, 298 9. Yu, L.X.; Furness, M.S.; Raw, A.; Woodland, O.K.P.; Nashed, N.E.; Ramos, E.; Miller, S.P.F.; Adams, R.C.; Fang, F.; Patel, R.M.; et al. Scientific considerations of pharmaceutical solid polymorphism in abbreviated new drug applications. Pharm. Res. 2003, 20, 531–536. [CrossRef] [PubMed] 9. Yu, L.X.; Furness, M.S.; Raw, A.; Woodland, O.K.P.; Nashed, N.E.; Ramos, E.; Miller, S.P.F.; Adams, R.C.; Fang, F.; Patel, R.M.; et al. Scientific considerations of pharmaceutical solid polymorphism in abbreviated new drug applications. References Vernaya, O.I.; Shabatin, V.P.; Semenov, A.M.; Shabatina, T.I. Cryochemical synthesis and antibacterial activity of a hybrid composition based on ag nanoparticles and dioxidine. Mosc. Univ. Chem. Bull. 2017, 72, 6–9. [CrossRef] 24. Chakkittakandy, R.; Corver, J.A.W.M.; Planken, P.C.M. Terahertz Spectroscopy to Identify the Polymorphs in Freeze-Dried Mannitol. J. Pharm. Sci. 2010, 99, 932–940. [CrossRef] [PubMed] 25. Morozov, Y.N.; Utekhina, A.Y.; Shabatin, V.P.; Chernyshev, V.V.; Sergeev, G.B. Cryosynthesis of nan drug substances. Russ. J. Gen. Chem. 2014, 84, 1010–1017. [CrossRef] 26. Morozov, Y.N.; Fedorov, V.V.; Shabatin, V.P.; Vernaya, O.I.; Chernyshev, V.V.; Abel, A.S.; Arhangel’skii, I.V.; Shabatina, T.I.; Sergeev, G.B. cryochemical modification of drugs: nanosized form iii piroxicam and its physical and chemical properties. Mosc. Univ. Chem. Bull. 2016, 71, 287–294. [CrossRef] 27. Chernyshev, V.V.; Morozov, Y.N.; Bushmarinov, I.S.; Makoed, A.A.; Sergeev, G.B. New polymorph of dehydroepiandrosterone (dhea) obtained via cryomodification. Cryst. Growth Des. 2016, 16, 1088–1095. [CrossRef] 28. Vernaya, O.I.; Shabatin, V.P.; Shabatina Khvatov, D.I.; Semenov, A.M.; Yudina, T.P.; Danilov, V.S. Cryochemical modification, activity, and toxicity of dioxidine. Russ. J. Phys. Chem. A 2017, 91, 229–232. [CrossRef] 29. Shabatin, V.P.; Vernaya, O.I.; Semenov, A.M.; Shabatina, T.I. Crystalline β—Modification Of 2,3-Bis- (Hydroxymethyl)-Quinoxaline-N,N′-Dioxide, Method for its Preparation and Pharmaceutical Composition on Its Basis. Patent RU2563256, 11 April 2014. Crystals 2018, 8, 298 15 of 15 30. Padeiskaya, E.N.; Pershin, G.N.; Kostyuchenok, B.M. Dioxidine, a new antibacterial drug for the treatment of suppurative infections. Pharm. Chem. J. 1977, 11, 1148–1154. [CrossRef] 31. Padeiskaya, E.N.; Shipilova, L.D.; Budanova, L.I.; Sizova, T.N.; Ovdeenko, N.I.; Sedov, V.V. Pharmacokinetics of dioxidine: penetration of the drug into organs and tissues after single and repeated introductions. Pharm. Chem. J. 1983, 17, 391–396. [CrossRef] 32. Werner, P.-E.; Eriksson, L.; Westdahl, M. TREOR: A semi-exhaustive trial-and-error powder indexing program for all symmetries. J. Appl. Crystallogr. 1985, 18, 367–370. [CrossRef] 33. Visser, J.W. A fully automatic program for finding the unit cell from powder data. J. Appl. Crystallogr. 1969, 2, 89–95. [CrossRef] 34. Zlokazov, V.B. MRIAAU a program for autoindexing multiphase polycrystals. J. Appl. Crystallogr. 1992, 25, 69–72. [CrossRef] 35. Zlokazov, V.B. AUTOX—A program for autoindexing reflections from multiphase polycrystals. Comput. Phys. Commun. 1995, 85, 415–422. [CrossRef] 36. Pawley, G.S. Unit-Cell Refinement from Powder Diffraction Scans. J. Appl. Crystallogr. 1981, 14, 357–361. [CrossRef] 37. Zhukov, S.G.; Chernyshev, V.V.; Babaev, E.V.; Sonneveld, E.J.; Schenk, H. Application of simulated annealing approach for structure solution of molecular crystals from x-ray laboratory powder data. Z. Kristallogr. References 2001, 216, 5–9. [CrossRef] 38. Chernyshev, V.V.; Schenk, H. A grid search procedure of positioning a known molecule in an unknown crystal structure with the use of powder diffraction data. Z. Kristallogr. 1998, 213, 1–3. [CrossRef] 39. Zlokazov, V.B.; Chernyshev, V.V. mria—A program for a full profile analysis of powder multiphase neutron- diffraction time-of-flight (direct and fourier) spectra. J. Appl. Crystallogr. 1992, 25, 447–451. [CrossRef] 40. Dvoryantseva, G.G.; Lindeman, S.V.; Aleksanyan, S.V.; Struchkov, Y.T.; Teten’chuk, K.P.; Khabarova, L.S.; Elina, A.S. Connection Between the structure and the antibacterial activity of the n-oxides of quinoxalines. molecular structure of dioxidine and quinoxidine. Pharm. Chem. J. 1990, 24, 80–84. [CrossRef] 41. Isaeva, V.I.; Belyaeva, E.V.; Fitch, A.N.; Chernyshev, V.V.; Klyamkin, S.N.; Kustov, L.M. synthesis and structural characterization of a series of novel zn(ii)-based mofs with pyridine-2,5-dicarboxylate linkers. Cryst. Growth Des. 2013, 13, 5305–5315. [CrossRef] y 42. Popa, N.C. The (hkl) dependence of diffraction-line broadening caused by strain and size for all laue groups in rietveld refinement. J. Appl. Crystallogr. 1998, 31, 176–180. [CrossRef] y 42. Popa, N.C. The (hkl) dependence of diffraction-line broadening caused by strain and size for all laue groups in rietveld refinement. J. Appl. Crystallogr. 1998, 31, 176–180. [CrossRef] 43. Ahtee, M.; Nurmela, M.; Suortti, P.; Järvinen, M. correction for preferred orientation in rietveld refinement. J. Appl. Crystallogr. 1989, 22, 261–268. [CrossRef] 43. Ahtee, M.; Nurmela, M.; Suortti, P.; Järvinen, M. correction for preferred orientation in rietveld refinement. J. Appl. Crystallogr. 1989, 22, 261–268. [CrossRef] 44. Järvinen, M. Application of symmetrized harmonics expansion to correction of the preferred orientation effect. J. Appl. Crystallogr. 1993, 26, 525–531. [CrossRef] 44. Järvinen, M. Application of symmetrized harmonics expansion to correction of the preferred orientation effect. J. Appl. Crystallogr. 1993, 26, 525–531. [CrossRef] 45. Spek, A.L. Structure validation in chemical crystallography. Acta Crystallogr. D 2009, D65, 148–155. [CrossRef] [PubMed] 45. Spek, A.L. Structure validation in chemical crystallography. Acta Crystallogr. D 2009, D65, 148–155. [CrossRef] [PubMed] 46. Macrae, C.F.; Bruno, I.J.; Chisholm, J.A.; Edgington, P.R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de Streek, J.; Wood, P.A. Mercury CSD 2.0—new features for the visualization and investigation of crystal structures. J. Appl. Crystallogr. 2008, 41, 466–470. [CrossRef] 46. Macrae, C.F.; Bruno, I.J.; Chisholm, J.A.; Edgington, P.R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de Streek, J.; Wood, P.A. Mercury CSD 2.0—new features for the visualization and investigation of crystal structures. J. Appl. Crystallogr. References 2008, 41, 466–470. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

https://openalex.org/W2483445223

https://scielo.conicyt.cl/pdf/rchog/v81n3/art08.pdf, http://www.scielo.cl/pdf/rchog/v81n3/art08.pdf

es

Barreras en la comunicación percibidas por un grupo de mujeres histerectomizadas: Un desafío para entregar una educación apropiada

Revista chilena de obstetricia y ginecología

cc-by

218 REV CHIL OBSTET GINECOL 2016; 81(3): 218 - 222 Trabajos Originales Barreras en la comunicación percibidas por un grupo de mujeres histerectomizadas: Un desafío para entregar una educación apropiada María Teresa Urrutia. PhD 1a, Paloma Beoriza2b, Alejandra Araya PhD.1a 1 Facultad de Enfermería, Universidad Andrés Bello. 2 Escuela de Enfermería, Facultad de Medicina, Pontificia Universidad Católica de Chile. aEnfermera-Matrona. bAlumna de Enfermería. Financiamiento para publicación: Fondo APA, Escuela de Enfermería Pontificia Universidad Católica de Chile. RESUMEN Antecedentes: La destreza comunicacional del profesional de la salud es reconocida como un aspecto crucial que facilita la educación a los pacientes. Objetivo: Describir las barreras que la mujer histerectomizada percibe en relación a la comunicación durante el proceso quirúrgico. Método: Estudio cualitativo realizado en 15 mujeres histerectomizadas. Para la recolección de los datos se utilizaron entrevistas en profundidad las que fueron analizadas con análisis de contenido. Resultados: Las mujeres identifican diferentes barreras tanto de ellas como de los profesionales que dificultan la comunicación y por ende la educación que ellas reciben en relación al proceso quirúrgico. Las barreras de los profesionales son el lenguaje técnico, falta de tiempo, enojo, contradicciones en las indicaciones y lejanía. Las barreras en las mujeres son la timidez, la vergüenza, la falta de confianza, el miedo y la conformidad. Conclusión: Los profesionales de salud deben enfocarse en desarrollar técnicas de comunicación que ayuden a los pacientes a disminuir las barreras y por ende mejorar los resultados en salud; deben entregar la información de forma clara y simple, verificando que lo comunicado ha sido entendido. PALABRAS CLAVES: Histerectomía, educación comunicación SUMMARY Background: The communication skills of health care professional has been recognized as a crucial aspect that facilitates patient education. Aim: To describe the barriers that women perceive hysterectomy in relation to communication during the surgical procedure. Method: Qualitative study conducted in 15 hysterectomized women. Deep interview and content analysis were used. Results: Women identify different barriers from them and health care professionals, that difficult the communication and therefore the education that the women receive in relation to the surgical process. The professional´s barriers are technical language, lack of time, anger, contradictions in the directions and distance. The barriers for women are shyness, shame, the lack of trust, fear and conformity. Conclusion: Health professionals should focus on developing communication techniques that help patients to reduce barriers and thus improve health outcomes; they must provide information in a clear and simple way, making sure that the statement has been understood. KEY WORDS: Hysterectomy, education, communication BARRERAS EN LA COMUNICACIÓN PERCIBIDOS POR UN GRUPO ... / MARÍA TERESA URRUTIA y cols. INTRODUCCIÓN La histerectomía es una de las cirugías más frecuentes después de la cesárea (1,2) lo que hace que un gran número de mujeres se vean enfrentadas a este proceso quirúrgico (3). La presencia de síntomas de larga data (i.e. sangramiento, dolor abdominal, entre otros), producto de alguna patología benigna (3,4) es una realidad frecuente en las mujeres histerectomizadas previo a esta cirugía (4,5). El hecho que la mayoría de las histerectomías son cirugías programadas (3,4), permite desarrollar una intervención profesional acorde a las necesidades que la mujer presenta y evita, por lo tanto, que acuda a fuentes de información no profesionales para poder satisfacer sus dudas e inquietudes. Si bien la histerectomía puede parecer una cirugía de rutina para el profesional, el significado que se le ha dado culturalmente al útero, hace que no sea considerada de rutina por la mujer. La asociación del órgano con conceptos tan potentes en nuestra cultura como son la maternidad, sexualidad y feminidad entre otros (6-8) sumado a los temores (9,10) y el propio significado que la mujer le da a la cirugía, hacen de este proceso quirúrgico un desafío para los profesionales de salud en lo que respecta a la educación y por ende las vías de comunicación. La destreza comunicacional del profesional de la salud es reconocida como un aspecto crucial que facilita la educación a los pacientes (11,12). El objetivo de este estudio es describir la percepción que un grupo de mujeres histerectomizadas tiene en relación a las barreras comunicacionales, tanto de los profesionales como de ellas, durante su proceso quirúrgico. PACIENTES Y MÉTODO Estudio descriptivo cualitativo, realizado en 15 mujeres histerectomizadas mediante entrevistadas en profundidad. El grupo de estudio corresponde a una muestra por conveniencia de mujeres histerectomizadas por patología benigna durante los 6 meses previos a la recolección de los datos del Complejo Asistencial Dr. Sotero del Río (Santiago, Chile). La recolección de los datos fue realizada entre los meses de mayo a septiembre de 2010. Cada entrevista estuvo direccionada por tres preguntas: 1) ¿Qué ha significado para usted que le hayan sacado el útero? 2) Como fue el apoyo/ educación que recibió? 3) Cual sería el apoyo/educación que le hubiese gustado recibir? El presente artículo analiza una de las dimensiones que emergieron frente a la segunda pregunta. Los resultados de la primera pregunta fueron publicados previamente (13). Las entrevistas fueron realizadas en un lugar privado y grabadas, para posteriormente ser transcritas de manera textual. La duración promedio de 219 cada entrevista fue de 40 minutos. Para el análisis de los datos se utilizó el análisis de contenido según Krippendorf (14). Los escritos fueron analizados por 4 investigadores de manera separada. Cada investigador propuso los significados y las dimensiones encontradas de manera independiente, discutiéndose cada propuesta hasta llegar a un consenso. Para evaluar la credibilidad de los análisis y asegurar la validez descriptiva, se usaron los criterios de Creswell (15). Primero: se utilizó el chequeo de los resultados obtenidos entre las mujeres, después de ser analizadas las entrevistas en profundidad, se realizaron grupos focales para corroborar la validez de los datos obtenidos en las entrevistas. Segundo: se incorporó una descripción detallada de los relatos en cada uno de los significados y dimensiones encontradas, de manera que el lector pudiera determinar su utilidad y aplicabilidad de los resultados de este estudio. Finalmente, el investigador principal examinó los resultados con 3 investigadores expertos en el área, los cuales consideraron válidos el análisis de datos. Esta investigación contó con la aprobación de los Comités de Ética de la Escuela de Enfermería de la Pontificia Universidad Católica de Chile y del Servicio de Salud Metropolitano Sur Oriente. RESULTADOS La edad de las mujeres tiene un rango entre 35 a 73 años. La escolaridad de 3 a 17 años. Del total de mujeres: 11 son casadas, 3 solteras y 1 viuda. De aquellas que tienen pareja, el periodo de convivencia tiene un rango entre 14 y 50 años. La causa más frecuente de HT fue la miomatosis uterina (9 casos). La vía de abordaje de la cirugía fue en 6 casos abdominal, 8 vaginal y en un caso indeterminado, dado que la usuaria afirma que ella aún conserva su útero y que no fue histerectomizada. En 7 casos hubo ooforectomía asociada a la histerectomía. Las mujeres identifican diversas barreras que limitan la comprensión con los profesionales del área de la salud y que se transforman en barreras para lograr una comunicación efectiva entre las mujeres y los profesionales. Al categorizarlas se identifican limitaciones por parte de los profesionales y otras que evidencian la dificultad de las mujeres para relacionarse con el equipo de salud (Tabla I). Barreras por parte del profesional: En relación al profesional de la salud, una de las barreras identificadas es el lenguaje técnico con el cual ellos se comunican, el cual dificulta la comprensión por parte de la mujer. Mujer: “Hablan en términos médicos que uno no entiende y… hablan entre ellos, entonces uno escucha no más y que más hace?, … que mas, que más hago yo? …con términos médicos que nadie entiende en realidad”. 220 REV CHIL OBSTET GINECOL 2016; 81(3) Tabla I BARRERAS EN LA COMUNICACIÓN PERCIBIDAS POR LAS MUJERES DURANTE EL PROCESO QUIRÚRGICO Del profesional de salud: Lenguaje técnico Falta de tiempo Percepción de enojo Indiferencia Contradicciones en la información Lejanía De la mujer: Timidez Vergüenza Falta de confianza Miedo Conformidad La falta de tiempo es otra de las barreras que las mujeres perciben por parte de los profesionales, y que les impide la posibilidad de conversar con ellos. Mujer: “Porque ellos andan muy apurados, andan muy rápido, le miran la ficha a uno, ¿usted es la fulana de tal? sí doctor, ¿cómo se siente? bien, ¿le duele algo? no, ¿le hicieron aseo? si, ¿ha comido bien? si. Y le preguntan todas esas cosas, son muy gentiles, muy agradables, pero andan así… yummm… pasan rápido, uno no puede ponerse, no se puede sentar a conversar con los médicos”. El enojo por parte del personal de salud y la indiferencia son otras barreras percibidas por las mujeres. Mujer: “Llegaba otra persona media cascarrabias y yo estaba con las sondas, necesitaba cosas, que me vieran, entonces había que estar con harta delicadeza, porque andaba como media enojada”. Mujer: “Es bueno que a uno se le informe y se le tome en cuenta lo que uno les dice… porque …es como que uno habla y no se le escucha”. Las contradicciones en las indicaciones del personal de salud es otra barrera que la mujer identifica al momento de ser atendida. Mujer: “Una enfermera da una orden otra enfermera da otra, entonces no sabe a quién hacerle caso y si uno le hace caso a una la otra se molesta. Por ejemplo lo de las medias, yo no sabía… yo llegué del pabellón y una niña me las sacó, después vino otra enfermera se enojó porque me las habían sacado y yo le dije yo acato las órdenes que dicen ustedes”. La lejanía o frialdad en la atención es otra de las barreras que las mujeres señalan marca una distancia entre ellas y el personal de salud. Mujer: “Porque es todo tan frío… uno llega al hospital, lo ve una persona, la opera otra… estás en una sala con 6 personas más que no tienen el mismo problema tuyo, que ni siquiera los puedes conversar… te revisan todos los días físicamente cómo estás, pero nadie te pregunta nada más… Entonces, yo creo que es como muy frío todo… para una persona que sabe … de qué se trata, no va a sentir mayormente ese proceso, pero para una persona que no tiene idea, estar ahí, que te vean la herida y sería todo… no sé, yo lo encuentro demasiado frío” Barreras por parte de la mujer: Dentro de las barreras que las mujeres identifican en ellas misma se encuentran la timidez y la vergüenza. Mujer: “Es que a veces hay personas tímidas, aunque no parezca… uno estando enferma es tímida y necesita que la ayuden, no sé, en ese momento que se siente como débil”. Mujer: “Vergüenza, vergüenza me daba… mucha vergüenza, ¿pero qué iba a hacer?, además que ellos … son médicos y están acostumbrados a ver tanta operación y tanta cosa… que a ellos no les llama la atención me imagino, pero igual… igual yo sentía mucha vergüenza”. La falta de seguridad o “confianza” es otra de las barreras identificadas por las mujeres. Mujer: “Lo atribuyo a no haber conocido al médico, porque cuando uno conoce al médico tiene como esa… el médico te dice estás bien, qué sé yo, entonces está como la como la confianza, aunque uno lo haya visto una vez, pero está la confianza”. El miedo es otra barrera que la mujer refiere la aleja del profesional y le impide resolver sus dudas. Mujer: “Por miedo no pregunto. Porque el doctor va a decir “¿oye ésta a qué vino?, ¿cómo no va a saber a lo que vino?”. Yo pensé que el doctor se podía enojar conmigo por tantas preguntas que uno a veces le hace” La conformidad es otra causa por la cual la mujer no pregunta o resuelve dudas. Mujer: “…no puedo pedir más… del hospital uno no puede pedir demasiado…porque… no sé, es público, es pobre, no puede pedir demasiado uno, entonces tiene que conformarse con lo que hay no más”. DISCUSIÓN Es importante mencionar la escasez de publicaciones respecto a la educación que reciben las mujeres sometidas a una histerectomía, asimismo sobre las barreras o dificultades que pueden presentar tanto ellas como el equipo de salud. Sin embargo, ha sido descrita previamente la entrega de información insuficiente a las pacientes acerca de la cirugía, la anatomía femenina (16) y los efectos post-quirúrgicos (17). No existen publicaciones que refieran las barreras comunicacionales de los profesionales y/o pacientes que se someten a una histerectomía, BARRERAS EN LA COMUNICACIÓN PERCIBIDOS POR UN GRUPO ... / MARÍA TERESA URRUTIA y cols. sin embargo existen estudios que mencionan las dificultades en la comunicación y/o educación en pacientes con otras patologías. En cuanto a las barreras que pueden presentar los equipos de salud se menciona el uso de jerga médica o lenguaje técnico con el que no se encuentran familiarizados los usuarios y el que no es explicado con claridad (18), coincidiendo con los resultados de este estudio. Por otro lado, se describe la entrega de mensajes poco claros e inadecuados como una causa por la que los pacientes no recibían una educación adecuada respecto a diagnósticos, pronósticos o cuidados post hospitalización (18). Junto con lo anterior, se habla del uso de expresiones médicas sin una explicación o comprensión por parte del usuario como un factor influyente en la aparición de angustia y ansiedad por parte del paciente (18). Cabe señalar que una buena educación a las mujeres sometidas a una histerectomía se asocia a una reducción de la ansiedad y experiencias de dolor (19-22). Se hace un especial énfasis en la importancia de entregar explicaciones claras a los pacientes, particularmente en brindarles asistencia para que realicen una elección de tratamiento informada (23). Si la información no es entregada a los pacientes, estos pueden llegar a obtener sus propias conclusiones según la conducta no verbal de los profesionales de la salud (23). Lo mencionado anteriormente podría explicar los sentimientos percibidos por las usuarias en este estudio. Cabe señalar que la frustración que las mujeres perciben por la falta de información, también ha sido documentada en relación a los profesionales. Las preguntas reiteradas por parte de los usuarios, quienes están confundidos o que no han comprendido del todo la información, generan en los profesionales un efecto que provoca resentimiento en el equipo de salud, además de frustración (18). Lo que podría afectar de manera negativa el proceso de aprendizaje de los usuarios al ser atendidos por personas con sentimientos negativos hacia el paciente que no logra comprender la información que se le entrega. Además, podría influir la percepción que el profesional tiene de sus propias habilidades y capacidades para entregar una atención de salud de calidad y centrada en los usuarios. Otra de las barreras que arroja este estudio se relaciona con la contradicción en las indicaciones del personal de salud. La comunicación efectiva entre los profesionales de salud, se asocia con una mejoría de los resultados de los cuidados de los usuarios, de la satisfacción y retención de los profesionales y en algunos casos la disminución de los costos en salud (18). Además, genera una mejoría en cuanto la adherencia a las recomendaciones de tratamiento, bienestar psicológico y emocional y eleva la satisfacción de los pacientes hacia su proveedor de salud (18). Asimismo, se describe que la falta de una comunicación efectiva entre los 221 funcionarios de la salud provoca confusión y una pérdida de confianza por parte de los usuarios hacia los miembros del equipo (18). La lejanía es considerada como otra de las barreras de los profesionales para establecer una comunicación efectiva. La falta de empatía es manifestada en un comportamiento no verbal frío, mecánico, indiferente o negativo por parte de los funcionarios. En muchos casos este comportamiento es percibido como enojo o impaciencia hacia el usuario, lo que probablemente incrementa las barreras de los pacientes para comunicarse con su equipo de atención de salud (18). Las pacientes reconocen como dificultades la timidez y vergüenza, la falta de seguridad o confianza, el miedo y la conformidad en el momento de interactuar con los profesionales. En relación a las barreras por parte de las mujeres, se ha establecido una relación entre la vergüenza y timidez con una baja literacidad de los pacientes, además de la posibilidad de experimentar discriminación (24,25). Por lo anterior es vital proveer un ambiente adecuado a las mujeres para que se sientan capaces de revelar abiertamente su necesidad de asistencia o confusión respecto a la información recibida (18). Cabe señalar que incluso aquellos individuos que poseen literacidad adecuada en salud pueden igualmente encontrar el ambiente médico amenazador y tener problemas para comprender la información que están recibiendo, situación que se acentúa si se encuentran enfermos, con dolor y/o preocupados sobre los efectos de su medicación (18). CONCLUSIÓN Los profesionales de salud deben enfocarse en desarrollar técnicas de comunicación que ayuden a los pacientes a disminuir las barreras y mejorar los resultados en salud; deben entregar la información de forma clara y simple, verificando que lo comunicado ha sido entendido. La comprensión de la información entregada, y la capacidad de autocuidado de las mujeres debe ser considerado un derecho, el cual todo profesional de salud debe proteger como también los proveedores de la atención de salud. REFERENCIAS 1. Garry R, Fountain J, Mason S, Napp V, Brown J, Hawe J, et al. The eVALuate study: two parallel randomised trials, one comparing laparoscopic with abdominal hysterectomy, the other comparing laparoscopic with vaginal hysterectomy. BMJ 2004;328(7432):129-33. 2. Sankaran S, Manyonda IT. Medical management of fibroids. Best Pract Res Clin Obstet Gynaecol 2008;22(4):655-76. 3. Flory N, Bissonnette F, Binik YM. Psychosocial effects of hysterectomy: literature review. J Psychosomatic Res 2005;59(3):117-29. 222 4. Sabry M, Al-Hendy A. Medical treatment of uterine leiomyoma. Reprod Sci 2012;19(4):339-53. 5. Zimmermann A, Bernuit D, Gerlinger C, et al. Prevalence, symptoms and management of uterine fibroids: an international internet-based survey of 21,746 women. BMC Women’s Health 2012;12:6 6. Lonnée-Hoffmann R1, Pinas. Effects of hysterectomy on sexual function. Curr Sex Health Rep 2014;6(4):244-51. 7. Bradford A, Meston C (2007) Sexual outcomes and satisfaction with hysterectomy: influence of patient education. J Sex Med 2007;4:106-14. 8. Reis N, Engin R, Ingec M, Bag B. A qualitative study: beliefs and attitudes of women undergoing abdominal hysterectomy in Turkey. Int J Gynecol Cancer 2008;18:921-8. 9. Gallicchio L, Harvey LA, Kjerulff KH. Fear of cancer among women undergoing hysterectomy for benign conditions. Psychosom Med 2005;67(3):420-4. 10. Wu SM, Chao Yu YM, Yang CF, Che HL. Decisionmaking tree for women considering hysterectomy. J Adv Nurs 2005;51(4):361-8. 11. Deledda G, Moretti F, Rimondini M, Zimmermann C. How patients want their doctor to communicate. A literature review on primary care patients' perspective. Patient Educ Couns 2013;90:297-306. 12. Bensing J, Rimondini M, Visser A. (2014). What patients want. Patient Educ Couns 2013;90(3):287-90. 13. Urrutia MT, Araya A, Flores C, Jara D, Silva S, Lira MJ. Histerectomía: la experiencia de no tener útero para un grupo de mujeres chilenas Rev Chil Obstet Ginecol 2013;78(4):262-8. 14. Krippendorf K. Content Analysis. An introduction to its methodology. Thousands Oaks: Sage Publications; 2004. REV CHIL OBSTET GINECOL 2016; 81(3) 15. Creswell JW. Standards of validation and evaluation. Qualitative inquiry and research design: choosing among five approaches. 2 ed. Thousand Oaks, CA: Sage Publications; 2007. p. 201-21. 16. Wade J, Pletsch P, Morgan S, Menting S. Hysterectomy: what do women need and want to know? J Obstet Gynecol Neonatal Nurs 2000;29(1):33-42. 17. Corney R, Everett H, Howells A, Crowther M. The care of patients undergoing surgery for gynaecological cancer: the need for information, emotional support and counselling. J Adv Nurs 1992;17(6):667-71. 18. Farahani M, Sahragard R, Carroll J, Mohammadi E. Communication barriers to patient education in cardiac inpatient care: A qualitative study of multiple perspectives. Inter J Nurs Pract 2011;17(3):322-8. 19. Cheung L, Callaghan P, Chang A. A controlled trial of psycho-educational interventions in preparing Chinese women for elective hysterectomy. Int J Nurs Stud 2003;402)207-16. 20. Priya P, Roach E. Effect of pre-operative instruction on anxiety among women undergoing abdominal hysterectomy. Nurs J India 2013;104(6):245-8. 21. Callaghan P, Li H. The effect of pre-operative psychological interventions on post-operative outcomes in Chinese women having an elective hysterectomy. Br J Health Psychol 2002;7(Pt2):247-52. 22. Araya A, Urrutia M. El proceso de ser histerectomizada: Aspectos educativos a considerar. Rev Chil Obstet Ginecol 2008;73(5):347-52. 23. Mody R. Overcoming barriers to delivering information to cancer patients. Br J Nurs 2003;12(21):1281-7. 24. Lambert V, Keogh D. Health literacy and its importance for effective communication Part 2. Nurs Child Young People 2014;26(4):32-6. 25. Oates D, Silliman R. Health Literacy: Improving patient understanding. Oncology 2009;23(4)376-9. 

https://openalex.org/W4283765476

https://jurnal.ubs-usg.ac.id/index.php/joeb/article/download/78/218

Indonesian

Pengaruh Kualitas Produk terhadap Pengembangan Usaha Sejiwa Coffee Bandung

Journal of Economics and Business/Journal of Economics & Business

cc-by-sa

Abstrak Penelitian ini mengangkat topik Kualitas produk terhadap pengembangan usaha Sejiwa Coffee Bandung. Tujuan dari penelitian ini untuk mengetahui apakah ada pengaruh antara kulitas produk terhadap pengembangan usaha Sejiwa Coffee Bandung. Metode yang digunakan pada penelitian ini adalah metode Kuantitatif dengan jenis penelitian deskriptif, penelitian ini memfokuskan pada pengembangan usaha, yang mencakup kualitas produk dan loyalitas konsumen sebagai variabel independen dan pengembangan usaha sebagai variabel dependen. Skala pengukuran yang digunakan dalam penelitian ini yaitu skala ordinal dengan menggunakan skala likert. Penelitian ini menggunakan teknik non probabilitas dengan jenis convenience sampling dengan jumlah sampel sebanyak 150 responden yang pernah mengunjungi Sejiwa Coffee di Bandung. Kata kunci: Kualitas Produk; Pengembangan Usaha; Loyalitas Konsumen; Sejiwa Coffee. Abstract This research raises the topic of product quality on the business development of Sejiwa Coffee Bandung. The purpose of this study is to determine whether there is an influence between product quality on the business development of Sejiwa Coffee Bandung. The method used in this research is quantitative method with descriptive research type, this research focuses on business development, which includes product quality and consumer loyalty as independent variables and business development as the dependent variable. The measurement scale used in this study is the ordinal scale using a Likert scale. This study uses a non-probability technique with the type of convenience sampling with a sample size of 150 respondents who have visited Sejiwa Coffee in Bandung. PENGARUH KUALITAS PRODUK TERHADAP PENGEMBANGAN USAHA SEJIWA COFFEE BANDUNG Maria Sugiat, Wafda Qinthara Konjala Univeristas Telkom, Bandung, Indonesia [email protected], wafdaqinthara@ Maria Sugiat, Wafda Qinthara Konjala Univeristas Telkom, Bandung, Indonesia Email: [email protected], [email protected] e-ISSN: 2774-7042 p-ISSN: 2302-8025 e-ISSN: 2774-7042 p-ISSN: 2302-8025 Journal of Economics and Business UBS Vol. 11 No. 1 Januari-Juni 2021 Pendahuluan Saat ini coffee shop dikota Bandung sedang digemari oleh masyarakat khususnya anak muda, maka tak mengherankan saat ini coffee shop sangat menjamur di Kota Bandung dengan berbagai menu dan konsep yang ditawarkan. Salah satu nya Sejiwa coffee yang menawarkan menu kopi dan pastry yang enak dan juga menawarkan segi konsep yang menarik atau biasa disebut dengan istilah instagramable. 43 Maria Sugiat, Wafda Qinthara Konjala Gambar 1. Sumber: pergidulu.com Gambar 1. Sumber: pergidulu.com Sejiwa coffee merupakan coffee shop yang berada dibawah manajemen TSVC Establisment dan terletak di pusat Kota Bandung tepatnya di jalan Progo no. 15 yang buka pada 23 Desember 2016, namanya kian melejit setelah disambangi oleh Presiden Jokowi pada 4 Desember 2017 lalu. Setelah berdiri kurang lebih 3 tahun, akhirnya pada bulan Juli 2020 Sejiwa Coffee membuka rumah kedua nya yang berada di jalan Kiputih no. 1A dengan konsep outdoor yang lebih luas (Rahadian, 2020). Sejiwa coffee memiliki konsep yang unik, dapat dilihat dari ruangan lantai satu yang memiliki konsep rumah kaca dipadukan dengan pop art dan memiliki konsep open kitchen bar dimana pengunjung dapat melihat para barista saat sedang meramu minuman yang ditawarkan (Ravenika, 2018). selain itu hal unik dari Sejiwa Coffee adalah seluruh barista dan staf memakai seragam yang mirip dengan jas laboratorium dengan tulisan Sejiwa “staff uniform” dibagian belakang (Akbar, 2018). Menu sejiwa coffee yang paling terkenal dan menjadi best seller adalah es kopi jiwa, saat sedang menikmati secangkir kopi sebagian konsumen mungkin akan terasa hampa jika tidak ditemani dengan pastry yang manis. Beberapa yang menjadi favorit adalah red velvet, oreo mille crepe, dan nastar cheese cake. Es kopi jiwa dan beberapa pastry menjadi produk yang di favoritkan konsumen saat ini. Hal ini lah yang membuat konsumen menjadi loyal terhadap produk yang ditawarkan oleh sejiwa coffee, sehingga membuat konsumen mau untuk datang kembali ke sejiwa coffee (Gemilang, 2019). Coffee shop mulai tumbuh di Indonesia dan menarik banyak pecinta kopi, fenomena tumbuhnya coffee shop juga terjadi di Kota Bandung (Wardhana et al., 2018). Perkembangan coffee shop di Kota Bandung terus tumbuh dan berkembang dengan hadirnya berbagai coffee shop di Kota Bandung, hal ini disebabkan oleh berkembangnya tren minum kopi di coffee shop (Sijabat, 2019). Banyaknya pengusaha yang melihat pasar penikmat kopi yang cukup tinggi maka tak heran banyak cabang 44 Pengaruh Kualitas Produk terhadap Pengembangan Usaha Sejiwa Coffee Bandung baru atau bahkan banyak pengusaha yang membuka coffee shop baru di Bandung (Nilawati, 2019). Pendahuluan Menurut Dinas Kebudayaan dan Pariwisata Kota Bandung persebaran coffee shop dibagi kedalam dua kategori yaitu kelas atas dan kelas menengah. Kelas atas yaitu coffee shop yang menawarkan produk dengan modal awal yang tinggi sedangkan kelas menengah merupakan coffee shop yang menawarkan produk dengan harga yang terjangkau dan modal awal yang tidak terlalu tinggi. Berdasarkan hal tersebut maka dibuatkan tabel sebagai berikut: Tabel 1. Kategori Coffee Shop Coffee Shop Kelas Atas Coffee Shop Kelas Menengah Starbucks Coffee Sejiwa Coffee Maxx Coffee Didago Coffee Excelso Coffee Blue Door The Coffee Bean Lacamera Coffee Djournal Coffee Dua Coffee Fore Coffee Yumaju Coffee Kopi Kenangan Two Cent Janji Jiwa Mimiti Coffee & Space Two Hand Full Marka Coffee Lalune Coffee & Lucheonette Morning Glory Sydwic Sumber: data yang telah diolah 2020 Tabel 1. Kategori Coffee Shop Tabel 1. Kategori Coffee Shop Berdasarkan table tersebut dapat disimpulkan bahwa persebaran coffee shop di Kota Bandung berdasarkan kepemilikan nya lebih di dominasi oleh coffee shop kelas menengah, hal ini dapat terjadi karena banyak pelaku bisnis baru yang membuka usaha coffee shop dengan tempat dan fasilitas yang tidak kalah dengan coffee shop kelas atas. Melihat dari hal tersebut, tak dapat dipungkiri bahwa bisnis coffee shop merupakan bisnis yang memiliki prospek baik kedepannya. Para pelaku bisnis hanya perlu menginovasikan konsep coffee shop yang unik agar menarik minat konsumen sehingga mampu bersaing dengan kompetitor (Sijabat, 2019). Persaingan bisnis coffee shop saat ini sangat ketat, karena saat ini hampir di setiap sudut Kota Bandung terdapat coffee shop, dengan menawarkan produk yang berkualitas dan unik serta tempat dengan konsep yang menarik dan nyaman hal tersebut akan membuat bisnis yang dijalankan dapat bersaing dan mampu mempertahankan konsumen serta dapat membuat konsumen menjadi loyal terhadap bisnis yang dijalankan (BisnisUKM, 2019). 45 45 Maria Sugiat, Wafda Qinthara Konjala Metode Penelitian ini akan menggunakan metode kuantitatif. Sujarweni (2015:39) menyatakan bahwa metode kuantitatif merupakan jenis penelitian yang menghasilkan suatu penemuan yang dapat diperoleh dengan menggunakan prosedur statistik atau cara lain dari suatu pengukuran. Dalam penelitian kuantitatif mempunyai karakteristik tertentu yang dinamakan sebagai variabel, hubungan antar variabel biasa dianalisi dengan menggunakan teori yang objektif. Sedangkan menurut (Silalahi & Atif, 2015) yang menyatakan bahwa metode kuantitatif merupakan suatu penelitian yang menjelaskan suatu fenomena melalui pengumpulan data yang diolah atau dianalisis dengan menggunakan statistik tertentu, hal ini digunakan guna melihat masalah-masalah yang nyata. Jadi dapat diketahui bahwa elemen penting dari metode kuantitatif yaitu menjelaskan fenomena, mengumpulkan data, dan diolah menggunakan statistik guna mendapatkan hasil yang nyata. Jenis penelitian yang digunakan oleh penulis dalam melakukan penelitian ini adalah jenis penelitian deskriptif. Menurut (Bahri, 2018) penelitian deskriptif adalah penelitian yang bertujuan untuk mendeskripsikan suatu keadaan atau objek yang menjadi perhatian dalam kegiatan penelitian secara sistematis. Sedangkan menurut (Silalahi & Atif, 2015) penelitian deskriptif menyajikan satu gambar yang rinci tentang satu situasi khusus, setting sosial, atau hubungan. Pada penelitian ini tipe penyelidikan yang digunakan yaitu penelitian kausal. Penelitian kausal menurut (Silalahi & Atif, 2015) yaitu mengenai hubungan pengaruh atau hubungan sebab akibat antara satu atau lebih variabel dengan variabel lainnya. Strategi penelitian yang digunakan yaitu dengan menyebarkan kuisioner, menurut (Silalahi & Atif, 2015) kuisioner merupakan satu set tulisan tentang pertanyaan yang diformulasikan guna mendapatkan jawaban dari responden. Pada penelitian ini penulis tidak melakukan intervensi data karena menggunakan data yang sudah ada dan tidak melalukan perubahan. Serta pada penelitian ini unit analisis bersifat individu dan peneliti menggunakan penelitian cross-sectional berdasarkan waktu pelaksanaanya. Menurut (Silalahi & Atif, 2015) waktu pelaksanaan cross-sectional merupakan penelitian yang dilakukan dalam satu periode tertentu untuk mendapatkan data yang dibutuhkan. Karakteristik penelitian ini dijelaskan secara ringkas seperti pada Tabel 2. berikut ini: Tabel 2. Karakteristik Penelitian No Karakteristi Penelitian Jenis 1 Metode yang digunakan Kuantitatif 2 Tujuan Penelitian Deskriptif 3 Tipe penyelidikan penelitian Kausal 4 Strategi penelitian yang digunakan Kuisioner 5 Keterlibatan peneliti dalam penelitian Tidak terjadi intervensi data oleh 46 Pengaruh Kualitas Produk terhadap Pengembangan Usaha Sejiwa Coffee Bandung peneliti 6 Unit analisis individu Individu 7 Waktu pelaksanaan Cross-sectional Sumber: Hasil olah penulis (2020) Pengaruh Kualitas Produk terhadap Pengembangan Usaha Sejiwa Coffee Bandung peneliti 6 Unit analisis individu Individu 7 Waktu pelaksanaan Cross-sectional Sumber: Hasil olah penulis (2020) Penelitian ini termasuk dalam penelitian deskriptif. Metode Penelitian deskriptif adalah penelitian yang dilakukan untuk memastikan dan mampu menggambarkan karakteristik dari variabel yang menarik. Penelitian ini termasuk dalam penelitian deskriptif karena peneliti hanya menggambarkan faktor atau variabel tersebut, tanpa melihat atau melakukan tes hubungan atau pengaruh antar faktor atau variabel (Indrawati, 2015). Pengumpulan data merupakan proses mendapatkan data dari responden atau objek yang telah ditentukan, alat pengumpulan data yang digunakan yaitu kuisioner dan observasi. Pada bagian ini pun akan dibahas mengenai jenis variabel yang digunakan oleh penulis, operasional variabel sebagai rancangan dalam membuat kuisioner serta dibahas pula mengenai skala pengukuran yang akan digunakan pada penelitian ini. Skala pengukuran yang digunakan pada penelitian ini adalah skala ordinal dengan menggunakan metode skala Likert. Menurut (Silalahi & Atif, 2015) menyatakan bahwa skala ordinal digunakan untuk mengidentifikasi adanya perbedaan yang hasilnya dapat dilakukan suatu pengukuran, sedangkan skala Likert menurut (Silalahi & Atif, 2015) adalah teknik pengskalaan yang digunakan untuk mengukur persepsi seseorang tentang dirinya atau kelompok yang berhubungan dengan suatu hal. Sumber: Olahan Peneliti Output IBM SPSS Versi 24 (2021) Sumber: Olahan Peneliti Output IBM SPSS Versi 24 (2021) Berdasarkan tabel diatas diperoleh hasil P1: 0,255X dengan nilai signifikan 0,003 < 0,5 yang berarti bahwa kualitas produk memiliki korelasi terhadap pengembangan usaha dengan pengaruh sebesar 0,255. Dengan terus meningkatkan kualitas produk yang ditawarkan maka akan menarik banyak konsumen dan akan berdampak pada pengembangan usaha dari Sejiwa Coffee itu sendiri karena sudah mengetahui apa yang pasar butuhkan dengan terus meningkatkan kualitas produk yang ditawarkan. Hasil analisis yang didapatkan menyatakan bahwa analisis pengaruh variabel kualitas produk terhadap variabel pengembangan usaha diperoleh nilai korelasi sebesar 0,255. Sehingga dari nilai tersebut dapat dikatakan bahwa secara langsung terdapat pengaruh signifikan variabel kualitas produk terhadap variabel pengembangan usaha. Hasil penelitian ini diperoleh dari hasil kuesioner yang disebar kepada 150 responden dengan kriteria: pria dan wanita, berdomisili di kota Bandung dan sekitarnya, rentang umur 17 - >40 tahun, pernah mengunjungi Sejiwa Coffee, dan pernah melakukan pembelian terhadap produk yang ditawarkan oleh Sejiwa Coffee. Berdasarkan hasil analisis deskriptif dari kedua variabel yaitu kualitas produk dan pengembangan usaha masing – masing mendapatkan skor sebesar 85,3 % untuk kualitas produk dan 85,5 % untuk pengembangan usaha. Berdasarkan garis kontium kedua variabel yang digunakan dalam penelitian ini berada pada kategori sangat setuju. Berdasarkan hasil analisis yang dilakukan bahwa variabel kualitas produk (X) terhadap pengembangan usaha (Y) memiliki pengaruh sebesar 0,255. Hasil dan Pembahasan Berdasarkan hasil pengolah mengenai kualitas produk terhadap pengembangan usaha maka didapatkan hasil sebagai berikut: Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 8.165 1.779 4.590 .000 X1 .245 .081 .255 3.012 .003 X2 .573 .083 .584 6.907 .000 47 47 Maria Sugiat, Wafda Qinthara Konjala a. Dependent Variable: Y Sumber: Olahan Peneliti Output IBM SPSS Versi 24 (2021) Kesimpulan Variabel kualitas produk memiliki persentase skor sebesar 85,3 % dan masuk kedalam kategori sangat setuju, dapat diartikan bahwa kualitas produk dari Sejiwa Coffee dinilai sangat penting bagi konsumen. Dengan indikator paling tinggi berada pada indikator nomer 1 dengan persentase 87,7% yang berarti bahwa konsumen setuju jika Sejiwa Coffee memiliki ciri khas dan harus dipertahan kan, sedangkan indikator terendah berada pada indikator nomer 5 dengan persentase 82,7% yang berarti bahwa tidak ada pengaruh terhadap harga yang ditawarkan oleh Sejiwa Coffee. Variabel pengembangan usaha memiliki persentase skor sebesar 85,5 % dan masuk kedalam kategori sangat setuju, dapat diartikan bahwa skor dari kualitas produk dan loyalitas konsumen yang tinggi dapat berpengaruh terhadap pengembangan usaha dari Sejiwa Coffee itu sendiri. Dengan indikator tertinggi berada pada indikator nomer 1 dengan persentase 90,7% yang berarti konsumen setuju bahwa Sejiwa Coffee menawarkan tempat yang nyaman, sedangkan indikator terendah berapa pada indikator nomer 4 dengan persentase 80,3% yang berarti tidak semua konsumen mengetahu Sejiwa Coffee dari sosial media atau aplikasi review. 48 Pengaruh Kualitas Produk terhadap Pengembangan Usaha Sejiwa Coffee Bandung DAFTAR PUSTAKA Abdullah, T., & Tantri, F. (2012). Manajemen Pemasaran. Depok: Raja Grafindo Persada. Adam, M. (2015). Manajemen Pemasaran Jasa: teori dan aplikasi. Bandung: Alfabeta. Akbar, M. F. (2018). Berkunjung Ke Cafe Sejiwa Coffee, Sejiwa Dalam Rasa. Keluyuran.Com. https://keluyuran.com/sejiwa-coffee/ Bahri, S. (2018). Metodologi Penelitian Bisnis Lengkap Dengan Teknik Pengolahan Data SPSS. Yogyakarta: ANDI. BisnisUKM. (2019). Strategi Meningkatkan Kepuasan Pelanggan. Bisnisukm.Com. https://bisnisukm.com/strategi-meningkatkan-kepuasan-pelanggan.html Indrawati, P. D. (2015). Metode Penelitian Manajemen dan Bisnis Konvergensi Teknologi Komunikasi dan Informasi. Bandung: Refika Aditama. Kotler, P. H., Armstrong, G., Harris, L., & Piercy, N. (2017). Principles of Marketing. European, Harlow. United Kingdom: Pearson Education Limited. Nilawati, P. P. (2019). Bisnis Kopi Masih Diminati, Terminal Coffee Buka Cabang Kelima di Kota Bandung. Jabar.Tribunnews.Com. https://jabar.tribunnews.com/2019/08/06/bisnis-kopi-masih-diminati-terminal- coffee-buka-cabang-kelima-di-kota-bandung g g Rahadian, B. (2020). Sejiwa Coffee Bandung. Bintangrahadian.Com http://www.bintangrahadian.com/2020/07/15/sejiwa-coffee-bandung/ Ravenika, S. A. (2018). Sejiwa Coffee: Kedai Kopi Instagramable di Bandung Decode.Uai.Ac.Id. https://decode.uai.ac.id/?p=3815 Sijabat, M. (2019). Kedai Kopi: Bisnis Masa Kini, Menjanjikan Di Masa Depan. Hariannusa. https://hariannusa.com/2019/10/25/kedai-kopi-bisnis-masa- kini- menjanjikan-di-masa-depan/ Silalahi, U., & Atif, N. F. (2015). Metode penelitian sosial kuantitatif. Bandung: Refika Aditama. Sunyoto, D., & Admojo, T. (2014). Konsep Dasar Riset Pemasaran dan Perilaku Konsumen. Utami, C. W. (2010). Manajemen Ritel Strategi dan Implementasi Operasional Bisnis Ritel Modern Di Indonesia. Jakarta: Salemba Empat. Wardhana, A., Kartawinata, B. R., & Syahputra, S. (2018). Analisis Positioningtop Brand Coffee Shop berdasarkan Persepsi Pelanggandi Kota Bandung. Prosiding SNaPP: Sosial, Ekonomi Dan Humaniora, 4(1), 303–310. 49 49

https://openalex.org/W2750321578

https://www.scielo.br/j/rbem/a/f3kqKJjVQJxB4985fDMVb8b/?lang=pt&format=pdf

Portuguese

Inteligência Artificial e Medicina

Revista Brasileira de Educação Médica

cc-by

RESUMO Ao mesmo tempo em que se discutem problemas na relação médico-paciente e a deficiência do exame clínico na atenção médica, que torna o diagnóstico clínico mais dependente de exames complemen- tares, enfatiza-se cada vez mais a importância do computador em medicina e na saúde pública. Isto se dá seja pela adoção de sistemas de apoio à decisão clínica, seja pelo uso integrado de novas tecno- logias, incluindo as tecnologias vestíveis/corporais (wearable devices), seja pelo armazenamento de grandes volumes de dados de saúde de pacientes e da população. A capacidade de armazenamento e processamento de dados aumentou exponencialmente ao longo dos recentes anos, criando o conceito de big data. A Inteligência Artificial processa esses dados por meio de algoritmos, que tendem a se aperfeiçoar pelo seu próprio funcionamento (self learning) e a propor hipóteses diagnósticas cada vez mais precisas. Sistemas computadorizados de apoio à decisão clínica, processando dados de pacientes, têm indicado diagnósticos com elevado nível de acurácia. O supercomputador da IBM, denominado Watson, armazenou um volume extraordinário de informações em saúde, criando redes neurais de processamento de dados em vários campos, como a oncologia e a genética. Watson assimilou dezenas de livros-textos em medicina, toda a informação do PubMed e Medline, e milhares de prontuários de pacientes do Sloan Kettering Memorial Cancer Hospital. Sua rede de oncologia é hoje consultada por especialistas de um grande número de hospitais em todo o mundo. O supercomputador inglês Deep Mind, da Google, registrou informações de 1,6 milhão de pacientes atendidos no National Health Service (NHS), permitindo desenvolver novos sistemas de apoio à decisão clínica, analisando dados desses pacientes, permitindo gerar alertas sobre a sua evolução, evitando medicações contraindicadas ou conflitantes e informando tempestivamente os profissionais de saúde sobre seus pacientes. O Deep Mind, ao avaliar um conjunto de imagens dermatológicas na pesquisa de melanoma, mostrou um desempenho melhor do que o de especialistas (76% versus 70,5%), com uma especificidade de 62% versus 59% e uma sensibilidade de 82%. Mas se o computador fornece o know-what, caberá ao mé- dico discutir o problema de saúde e suas possíveis soluções com o paciente, indicando o know-why do seu caso. RESUMO Isto requer uma contínua preocupação com a qualidade da educação médica, enfatizando o conhecimento da fisiopatologia dos processos orgânicos e o desenvolvimento das habilidades de ouvir, examinar e orientar um paciente e, consequentemente, propor um diagnóstico e um tratamento de seu problema de saúde, acompanhando sua evolução. I Professor Honoris Causa da UnB, Professor aposentado da UFRJ e Consultor Sênior da Unasus-Fiocruz. Inteligência Artificial e Medicina Artificial Intelligence and Medicine Luiz Carlos LoboI Revista Brasileira de Educação Médica Revista Brasileira de Educação Médica 41 (2) : 185-193; 2017 185 ABSTRACT ABSTRACT While discussions develop regarding problems in the doctor-patient relationship and the deficiency of the clinical examination in medical practice, which leaves diagnoses more dependent of complementa- ry tests, the importance of the computer in medicine and public health is highlighted. This is happe- ning, either through the adoption of clinical decision support systems, the use of new technologies, such as wearable devices, or the storage and processing of large volumes of patient and population data. Data storage and processing capacity has increased exponentially over recent years, creating the concept of “big data”. Artificial Intelligence processes such data using algorithms that continually improve through intrinsic self-learning, thus proposing increasingly precise diagnostic hypotheses. Computerized clinical decision support systems, analyzing patient data, have achieved a high degree of accuracy in their diagnoses. IBM’s supercomputer, named “Watson”, has stored an extraordinary volume of health information, creating a neural network of data processing in several fields, such as oncology and genetics. Watson has assimilated dozens of medical textbooks, all the information from PubMed and Medline, and thousands of medical records from the Sloan Kettering Cancer Memorial Hospital. Its oncology network is now consulted by numerous specialists from all over the world. The English supercomputer Deep-Mind, by Google, has stored data from 1.6 million National Health Service patients, enabling the development of new clinical decision support systems, analysis of these patient data and generating alerts on their evolution in order to avoid contraindicated or conflicting medications, whilst also sending timely updates to the physicians about the health of their patients. Analyzing a set of dermatological images in a melanoma study, Deep-Mind showed a higher level of performance than that of specialists (76% versus 70.5%), with a specificity of 62% versus 59% and a sensitivity of 82%. Nevertheless, whereas the computer provides the know-what, it is the physi- cian that will discuss the medical problem and the possible solutions with the patient, indicating the know-why of his or her case. This area requires continuous focus on the quality of medical training, emphasizing knowledge of the physiopathology of the organic processes and the development of the abilities to listen to, examine and advise a patient and, consequently, propose a diagnosis and treat- ment, accompanying his or her evolution. KEYWORDS –– Doctor-Patient Relationship. –– Clinical Examination. –– Decision Support Systems. –– Artificial Intelligence. –– Wearable Devices. –– Medical Education. –– Doctor-Patient Relationship. –– Clinical Examination. –– Decision Support Systems. –– Decision Support Systems. –– Artificial Intelligence. –– Wearable Devices. –– Medical Education. –– Medical Education. Recebido em: 08/06/2017 Aprovado em: 12/06/2017 Recebido em: 08/06/2017 Aprovado em: 12/06/2017 Recebido em: 08/06/2017 Aprovado em: 12/06/2017 Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 186 PALAVRAS-CHAVE –– Relação Médico-Paciente. –– Exame Clínico. –– Sistemas de Apoio à Decisão. –– Inteligência Artificial. –– Dispositivos Corporais. –– Educação Médica. –– Relação Médico-Paciente. –– Exame Clínico. –– Sistemas de Apoio à Decisão. –– Inteligência Artificial. –– Dispositivos Corporais. –– Educação Médica. –– Relação Médico-Paciente. –– Exame Clínico. –– Dispositivos Corporais. –– Educação Médica. DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  INTELIGÊNCIA ARTIFICIAL E MEDICINA Inteligência Artificial em medicina é o uso de computadores que, analisando um grande volume de dados e seguindo algo- ritmos definidos por especialistas na matéria, são capazes de propor soluções para problemas médicos. Berner e col.5 discutiram a relevância de se ter uma lista de hipóteses diagnósticas, seja para o clínico, seja para estudantes de Medicina. Uma lista de possíveis diagnósticos, aceitos de for- ma acrítica, poderia resultar no pedido de muitos exames para esclarecer o caso, tornando mais caro o tratamento do paciente. Computadores podem armazenar e recuperar dados so- bre imagens, como lesões dermatológicas ou exames radioló- gicos, de ultrassom, de ressonância magnética, de tomografia por emissão de pósitrons (PET), de ecocardiogramas, de ele- troencefalogramas, eletrocardiogramas, dados de dispositivos vestíveis/corporais (wearable devices) e gerar probabilidades de diagnóstico baseadas em algoritmos de decisão estabeleci- dos e que podem se automodificar em decorrência de resulta- dos obtidos (self improvement). Recente trabalho6 analisa a adoção de sistemas de apoio à decisão clínica para melhorar a acurácia da medicina, conside- rando o aumento da carga de trabalho de médicos em emergên- cias e clínicas de família. A Associação Americana de Escolas de Medicina (AAMC) estima uma demanda de 130.600 médicos até 2025, o que justificaria a adoção de programas de apoio à de- cisão como forma de diminuir a possibilidade de erros médicos. Dados de pacientes podem ser coletados seja diretamente de prontuários médicos eletrônicos, seja por meio da digitação de informações de anamnese, de exame clínico do paciente, exames complementares, evolução da enfermidade e medica- mentos prescritos e usando algoritmos definidos e que podem ser atualizados com a análise desses dados e propor diagnós- ticos diferenciais de enfermidades, com as respectivas proba- bilidades de ocorrência. Em 2009, verificou-se que 32% dos erros médicos nos EUA resultavam da diminuição do tempo de interação do médico com os pacientes, produzindo diagnósticos equivocados, não reconhecimento da urgência ou piora da evolução do pacien- te que demandariam prescrever ou realizar ações pertinentes. Mesmo em hospitais que disponham de prontuários médicos eletrônicos7, com a possibilidade de melhor coleta de dados, admite-se que 78,9% dos erros médicos estariam relacionados a problemas na relação médico-paciente, exame clínico defi- ciente, falha de avaliação dos dados do paciente ou falta de exames que comprovassem a hipótese diagnóstica. INTRODUÇÃO mote do mundo atual, caracterizando uma sociedade em mu- dança rápida e constante. Ao longo de sua evolução, o homem adquiriu uma série de extensões. Com o desenvolvimento da oposição do polegar pôde segurar um porrete e aumentar o tamanho de seu braço ou atirar pedras para se defender de agressores. Com o uso de ondas de rádio pôde atingir populações distantes e com o mi- croscópio ver dimensões muito pequenas, como as bactérias. A criação de redes que interligam computadores, ense- jando a comunicação rápida entre pessoas, de certo modo fragmentou e segmentou a sociedade em grupos de interesses semelhantes. Conversa-se mais através de smartphones do que presen- cialmente. A oferta contínua de vídeos e áudios em streaming reforça esse processo de individualização e isolamento. Talvez isso explique o fato de ser a depressão um dos grandes proble- mas médicos da atualidade. Se o telefone permitiu a conversa entre pessoas que resi- dem em qualquer lugar do planeta, estendendo a capacidade de ouvir, a televisão permitiu ver fatos distantes e no momen- to em que se produzem, aumentando a capacidade de ver. Ao mesmo tempo, houve um grande avanço nas tecnolo- gias médicas voltadas ao diagnóstico pela imagem: o ecocar- diograma, que, alguns acham, tornará obsoleto o estetoscópio, o ultrassom, que substitui, dizem, a palpação e a percussão do abdome, a ressonância, que substitui os exames de radiologia O advento do computador trouxe uma enorme amplia- ção da dimensão do homem, aumentando sua capacidade de calcular e armazenar grandes volumes de informação, e isso em nanossegundos. Ampliou sua memória e o tempo de recu- peração de uma informação. O “aqui e agora” passou a ser o Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 186 DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  Vários sistemas3 foram desenvolvidos com o objetivo de oferecer uma lista de possíveis diagnósticos para um proble- ma de saúde, com as probabilidades de acerto, usando dados da gênese e evolução desse problema, avaliando sinais e sinto- mas de pacientes, analisando resultados de exames realizados e propondo possibilidades diagnósticas. contrastada, e o PET, que permite verificar as áreas envolvi- das em várias atividades cerebrais, vieram substituir o exame clínico dos pacientes e a relação médico-paciente. Hoje, faz-se diagnóstico pelos exames e não pelo raciocínio e julgamento clínico. INTRODUÇÃO Para que avaliar a expansão torácica, auscultar, sentir o frêmito toracovocal ou fazer uma percussão para diagnosti- car uma pneumonite, se um raio X simples de tórax pode fazer o diagnóstico? Recente trabalho4 selecionou quatro sistemas, após eles te- rem analisado dados de 20 casos clínicos apresentados no New England Medical Journal e no Medical Knowledge Self Assessment Program (MKSAP) do American College of Physicians, que visa avaliar os conhecimentos de medicina interna de estu- dantes do terceiro ano de Medicina de escolas dos EUA. Dois desses sistemas de diagnósticos diferenciais – DDX (DXPlain, de Harvard, e Isabel, privado) – indicaram acertos significa- tivos, ainda que não tenham acertado o diagnóstico de dois casos do MKSAP. Trabalhos recentes1 enfatizam “the fading art of the clinical examination” e ressaltam a necessidade de olhar, ouvir e exa- minar o paciente e só então pedir exames para comprovar a hipótese diagnóstica. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 187 CRÍTICAS AO USO DE SISTEMA DE APOIO À DECISÃO CLÍNICA os alunos: hipotético-dedutivo e indutivo. Mostrou predo- minância de alunos (69%) induzindo um diagnóstico. Esses processos ressaltam dois tipos de pensar: o teórico-dedutivo (que vai do geral ao particular), mais comum entre latinos, e o empírico-indutivo (que vai do particular ao geral), que carac- teriza, em geral, os anglo-saxões 11. A tomada de decisão em medicina depende, essencialmente, da proposta de hipóteses diagnósticas sugeridas pelo médico após colher e avaliar dados sobre os problemas de saúde de um paciente. Essas hipóteses seriam avaliadas e indicariam a necessidade de ter novas informações sobre o caso ou a reali- zação de exames complementares que permitam eleger a me- lhor opção na solução do problema. Sistemas de suporte à decisão clínica podem assistir o profissional nesse processo de tomada de decisão. Esses sis- temas podem sugerir hipóteses, com suas probabilidades de ocorrência, mas não explicam o porquê dessas hipóteses. In- formam o know-what, mas não o know-why. A experiência obtida após anos de trabalho clínico permi- tiria ao profissional avaliar melhor essas hipóteses diagnós- ticas fazendo perguntas ao paciente ou sugerindo exames a serem realizados. Por outro lado, o sistema de tomada de decisão, depen- dendo da capacidade de perceber o que ocorre com o paciente, pode levar a erros na conduta proposta. Um trabalho clássico de Elstein e cols.8 indicava, após lon- ga pesquisa registrando o processo de solução de problemas em várias situações controladas, o seguinte: Uma menina de 14 anos procurou o Hospital de Sobra- dinho em Brasília com amenorreia primária. Queixava-se de “ainda não ter ficado moça”. Mas o residente que a atendeu observou a icterícia da paciente e, após exames, fez um diag- nóstico de anemia falciforme. Tratou a paciente e deu-lhe alta, mas a paciente, chorando, dizia ao médico que não a tinham tratado porque “ainda não tinha ficado moça”. 1. Hipóteses de solução de problemas, com frequência, são geradas precocemente ao se entrevistar um pacien- te, seja porque o médico se lembra de casos semelhan- tes, seja porque leu um artigo que discutia situações com as mesmas características; 2. CRÍTICAS AO USO DE SISTEMA DE APOIO À DECISÃO CLÍNICA Hipóteses são, de regra, apresentadas em número limi- tado, mesmo em casos complexos; não obstante, hipó- teses gerais, como dor abdominal, podem ser propos- tas junto com outras; Para obviar essa distorção na percepção de um caso, Weed12 propôs em 1964 a adoção do “Prontuário Orientado para Problemas”, em que, após ouvir e examinar o paciente, o médico deveria listar os problemas apresentados seja de or- dem somática, seja de ordem psicológica e mesmo de ordem social, e propor um plano de ação para cada problema apre- sentado13. 3. Como consequência da proposição precoce de hipó- teses, os médicos podem buscar novos dados para comprovar suas hipóteses e não para reavaliá-las, ne- cessitando de muito mais informação ou resultados de exames para rever essas hipóteses; O “Prontuário Médico Orientado para Problemas” foi adotado em poucos hospitais, talvez porque, reconhecendo os problemas do paciente e propondo um plano de diagnóstico e tratamento para cada problema, a competência médica fique sujeita a uma clara avaliação por seus pares. 4. A solução de um caso está, de regra, relacionada à competência do profissional em relação ao problema apresentado pelo paciente. Como dizia Claude Ber- nard, “quem não sabe o que procura não entende o que encontra”. INTELIGÊNCIA ARTIFICIAL E MEDICINA Atualmente, o uso de wearable devices tem sido introduzi- do na prática médica, obtendo informações contínuas sobre glicemia, ECG e movimento, por exemplo, que podem gerar ações automatizadas, como injetar insulina, dar uma descar- ga elétrica de um desfibrilador subcutâneo ou variar a dose de um medicamento em pacientes com doença de Parkinson. Informações desses gadjets são capturadas pelo celular do pa- ciente e podem ser transmitidas ao seu médico. O uso de sistemas que cruzam remédios prescritos e da- dos do paciente, evitando interações ou doses inapropriadas, está disponível e se reflete numa melhor e mais segura pres- crição medicamentosa. Médicos e farmacêuticos se beneficiam muito com esses sistemas que confrontam dados e prescri- ções, evitando casos de medicações inadequadas a mulheres grávidas, doentes com insuficiência renal, alérgicos à droga prescrita e interações entre medicações feitas por mais de um profissional. Desde algum tempo, busca-se desenvolver sistemas computadorizados de apoio ao diagnóstico clínico. Howard Bleish2, há mais de 50 anos, já oferecia um sistema que, ava- liando dados de um paciente, sugeria ações para restabelecer seu equilíbrio hidroeletrolítico. DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 188 USO DE SUPERCOMPUTADORES EM INTELIGÊNCIA ARTIFICIAL E MEDICINA Os primórdios da IA datam de 1950, quando Alan Turing publicou seu artigo “Computing Machinery and Intelligence”14 e propôs o teste (hoje nomeado TT em sua homenagem) que compara a performance de um computador e de uma pessoa na solução de um problema. Registrando um grande número de casos com seu esque- ma de diagnóstico, tratamentos prescritos e resultados obti- dos, esses sistemas permitem uma expansão do conhecimento médico e a sugestão de condutas a serem seguidas, nesse caso com embasamento probabilístico. O termo Inteligência Artificial foi cunhado numa confe- rência no Dartmouth College em 195615 por McCarthy e cols., e a sua aplicação em medicina iniciou-se com o artigo de Shor- tlife em 196316. Pesquisadores da IBM, usando redes neurais, consegui- ram obter uma acurácia de 86% no diagnóstico de retinopatia diabética feito em 35 mil imagens de retina acessadas através da tecnologia EyePACs da IBM de identificação de lesões e outros sinais observados em vasos sanguíneos. Szlovits, em publicação de 200917, admite que atualmente a IA em medicina está se tornando não apenas uma parte, mas um componente essencial da informática médica e um recurso importante na solução de problemas em atenção à saúde. As pesquisas iniciais visavam capturar o conhecimento de especialistas no desenvolvimento de sistemas de apoio ou proposta de soluções clínicas. Ao contrário do Watson, que processa mais demandas es- pecíficas, o Deep Mind, criado em 2010 e comprado pela Goo- gle em 2014, também usando uma rede neural, pretende não ser programado com fins predefinidos, “aprendendo” com a experiência na solução de problemas. Atualmente, o problema é processar um grande volume de informações, seja por meio de prontuários eletrônicos com da- dos dos pacientes, resultados de seus exames, diagnóstico pro- posto, prescrição e resultados dessas medicações, seja por di- gitação, considerando que dados podem não estar disponíveis ou podem estar incompletos. Deve-se considerar também que informações ainda podem ter que ser digitadas em decorrência da incompatibilidade de sistemas onde elas estejam registra- das. Há que se considerar, no entanto, que essa digitação pode- rá eventualmente introduzir um componente de erro humano. O Deep Mind tem sido usado em medicina na avaliação de scans visuais, buscando causas de cegueira. USO DE SUPERCOMPUTADORES EM INTELIGÊNCIA ARTIFICIAL E MEDICINA Estudos de Rimoldi9 mostraram que o número de dados de anamnese, exame físico e exames complementares requi- sitados por internos, residentes e especialistas para resolver um caso varia significativamente, indicando a importância da experiência na proposição de possíveis diagnósticos. O reconhecimento de padrões (pattern recognition) e de crité- rios de diagnóstico (combinação de sintomas, sinais e resul- tados de exames) é usado na determinação de um diagnós- tico correto. Inteligência Artificial (IA) é um ramo da ciência da compu- tação que se propõe a desenvolver sistemas que simulem a capacidade humana de percepção de um problema, identifi- cando seus componentes para, com isso, resolver problemas e propor/tomar decisões. Outra definição de Inteligência Artificial indica que seria a criação de sistemas inteligentes de computação capazes de realizar tarefas sem receber instruções diretas de humanos (os “robôs” são exemplos disso). Um trabalho da Universidade Estadual de Londrina so- bre o raciocínio clínico dos estudantes de Medicina 10 indicou dois processos de elaboração de hipóteses diagnósticas entre Usando diferentes algoritmos e estratégias de tomada de decisões e um grande volume de dados, sistemas de IA são capazes de propor ações, quando solicitados. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 188 DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  Com a disseminação dos smartphones, sistemas como o Apple Siri, Microsoft Cortana e Google Now permitem inte- ração em linguagem aberta, funcionando como um assistente pessoal, dando informações ao usuário, lembrando ações a serem tomadas, fazendo postagens no Facebook ou Twitter e conversando com o usuário como um amigo virtual. A Inteligência Artificial envolve várias etapas ou compe- tências, como reconhecer padrões e imagens, entender lingua- gem aberta escrita e falada, perceber relações e nexos, seguir algoritmos de decisão propostos por especialistas, ser capaz de entender conceitos e não apenas processar dados, adquirir “raciocínios” pela capacidade de integrar novas experiências e, assim, se autoaperfeiçoar, resolvendo problemas ou reali- zando tarefas. Duas das mais importantes experiências no uso de IA em vários campos, inclusive medicina, são a plataforma Watson Health, da IBM, e o Deep Mind, da Inglaterra, que processam informações armazenadas na “nuvem” de oncologia e de ava- liação de risco e evolução de pacientes. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 189 INTEGRAÇÃO DE BASE DE DADOS – BIG DATA A capacidade de armazenar dados aumentou de forma ex- ponencial nas últimas décadas. De armazenamento em fitas e discos móveis, passou-se a armazenar via internet em sis- temas de grande capacidade (warehouses) e, atualmente, em redes de computadores ditos “na nuvem”. O armazenamento de dados passou a ser gigantesco, le- vando ao conceito de big data. Estima-se em 2,5 exabytes (um exabyte corresponde a 10 à potência 18; para comparação, um gigabyte é 10 à potência 9), com um crescimento anual previs- to de 57% no período de 2014-201920, chegando a 24,3 exabytes em 2019, admitindo-se ainda que em 2019 vídeos deverão cor- responder a 50% do tráfego na internet. Outra experiência relatada por Mukherjee foi desenvolvi- da na Universidade de Columbia: uma radiologista discutia a importância de diagnosticar precocemente um AVC numa tomografia computadorizada (CT scan), permitindo a destrui- ção oportuna de um coágulo no cérebro. É fácil diagnosticar um AVC quando o cérebro já está morto e cinzento, dizia ela. O desafio é diagnosticar e intervir precocemente. O reconhe- cimento de imagens feito por IA poderia obviar esse desafio pelo reconhecimento de pequenos detalhes indicando áreas suspeitas em cortes de CT que poderiam passar despercebi- das. Big data está sendo gradualmente introduzido no sistema de atenção à saúde. Dados de prevalência, incidência e evolu- ção de enfermidades permitiriam gerar dados estatísticos, an- tecipar surtos epidemiológicos e prescrever ações preventivas. Dados de pacientes, como idade, sexo, etnia, local de re- sidência, antecedentes pessoais e familiares, sintomas e sinais apresentados, exames realizados ou obtidos por meios eletrô- nicos (wearable devices), diagnósticos feitos, tratamento e evo- lução coletados, permitiriam estabelecer uma base de dados e aprimorar condutas estabelecidas. Uma pesquisa realizada na USP19 mostrou que a geração de hipóteses e diagnósticos diferenciais, em geral, é feita pelo reconhecimento visual imediato de determinadas lesões. A experiência foi realizada com 25 radiologistas que tentaram reconhecer imagens do pulmão apresentadas a eles dentro de um tubo de ressonância magnético funcional. O tempo mé- dio para propor um diagnóstico foi de 1,33 segundo, ativando sempre a mesma área do cérebro. O importante parece ter sido o conhecimento prévio da forma e características da lesão, o que poderia ser feito por Inteligência Artificial. Esses dados deveriam ser disponibilizados ao paciente, que poderia participar de decisões que envolvessem ações médicas, já que as consequências dessas ações seriam sofridas por ele. USO DE SUPERCOMPUTADORES EM INTELIGÊNCIA ARTIFICIAL E MEDICINA Em 2016, esta- beleceu um projeto com o Royal Free London NHS Founda- tion Trust e com o Imperial College Health Care NHS Trust para acessar os prontuários eletrônicos de pacientes e avaliar seus dados, o que tem gerado grande controvérsia no país. O problema da confidencialidade dos dados tem sido questiona- do, sobretudo pelo fato de eles serem processados livremente por uma empresa privada (Google). Sistemas que funcionam em background podem ser uti- lizados para verificar dados de pacientes, como interação e incompatibilidade de medicamentos, dados de exames com- plementares discrepantes, exames solicitados e/ou a serem realizados. No Brasil, em 2016, a Fiocruz criou o Centro de Integra- ção de Dados e Conhecimentos para Saúde (Cidacs), buscan- do integrar dados de saúde e políticas sociais de mais de cem milhões de brasileiros contemplados no Programa Bolsa Fa- mília e outros programas de proteção social numa única base de dados, mas preservando a confidencialidade desses dados. O problema da perda de confidencialidade dos dados ar- mazenados tem sido ressaltado e deve ser sempre discutido, mas sistemas de saúde, como o NHS da Inglaterra, indicam que o benefício da troca de experiências excede problemas eventuais decorrentes da quebra de confidencialidade. A troca de informa- ções já é feita entre pacientes com websites, como PatientsLike- Me, Acor (association of cancer online), CrowdMed, SmartPatients. Também foram criadas plataformas de vigilância em saú- de, estudos genéticos (Epigen), de incorporação de tecnolo- gias e inovações em sistemas de informação para apoiar o SUS e de Estudos de Equidade e Sustentabilidade Urbana e seus efeitos em saúde. Essa iniciativa deverá ser o começo de um programa de integração de dados sobre as condições de saúde de indivíduos e da população do País. Várias empresas estão muito engajadas em projetos de IA, como IBM, Google, Apple, Microsoft e Amazon. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 189 DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  EXPERIÊNCIAS EM INTELIGÊNCIA ARTIFICIAL EM SAÚDE ponibilidade de recursos e a possibilidade de discutir o caso in loco são escassas ou inexistentes. A disponibilização contínua de sistemas computadorizados de apoio à decisão clínica seria outra alternativa a ser postulada. Em trabalho recente, Mukherjee18 relata a experiência de Se- bastian Thrun, da Universidade de Stanford, que armazena, numa rede neural de computação, 130 mil imagens de lesões da pele classificadas por dermatologistas. O sistema usa algo- ritmos que reconhecem imagens e suas características (pattern recognition). Em junho de 2015, Thrun e equipe começaram a validar o sistema usando um conjunto de 14 mil imagens que haviam sido diagnosticadas por dermatologistas, solicitan- do que o sistema reconhecesse três tipos de lesão: benignas, malignas e crescimentos não cancerosos. O sistema acertou 72% das vezes, comparado com um acerto de 66% obtido por dermatologistas qualificados. A experiência de Thrun foi am- pliada para incluir 25 dermatologistas e uma amostra de 2 mil casos biopsiados. A máquina continuou sendo mais acurada. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 190 INTEGRAÇÃO DE BASE DE DADOS – BIG DATA Eric Topol fala em empoderar o paciente21 na discussão de seu problema de saúde, sobretudo se se considerar que outros sistemas como o Google já oferecem dados sobre problemas médicos, mas sem uma análise criteriosa dessa informação. O médico poderá discutir o caso agregando seu know-why. O problema passou a ser dispor de processos de integra- ção e pesquisa em diferentes bases de dados. Sistemas integra- dos e universais de atenção à saúde, como o National Health Service, da Inglaterra, facilitam a pesquisa por constituírem uma base de dados única. Discute-se a abertura desses dados a toda a comunidade médica, com a certeza de que o benefício de trocar e compartilhar experiências é muito maior que a pre- servação de sua confidencialidade. O dermatologista Lindsay Bordone, da Universidade de Columbia18, tinha 49 pacientes agendados no dia. Examinados em sequência, todos saíam felizes. É que o computador pode acertar o know-what, mas o médico, conversando com seu pa- ciente, explica o know-why. Isto indica por que os médicos nun- ca serão substituídos por uma máquina. Elas não explicam o porquê nem aliviam a angústia do paciente. O apoio à decisão clínica disponível por meio de consul- tores on-line, como oferecido no Brasil pelo Telessaúde/Tele- Medicina, favorece a prática em lugares remotos onde a dis- Um sistema único e padronizado de dados em saúde po- deria ser o trunfo essencial a garantir a qualidade da atenção prestada em saúde no Brasil. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 190 DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  propor drogas mais bem ajustadas e observar a evolução do paciente. Integrar diferentes sistemas de registro eletrônico de da- dos em medicina e saúde é o grande desafio. A adoção de diferentes sistemas informatizados nos estados e municípios brasileiros e a baixa integração com os sistemas privados de saúde dificultam a criação de uma base de dados única e na- cional e a possibilidade de estabelecer condutas e diretrizes para os principais problemas de saúde do País22. O programa Precision Medicine Initiative prevê estabelecer uma base de dados genéticos de um milhão de pessoas para avaliar a eficácia de drogas em condições específicas. Um volume muito grande de informações médicas (80% segundo a IBM) já está armazenado nos bancos de dados do Watson23, supercomputador da empresa. INTEGRAÇÃO DE BASE DE DADOS – BIG DATA O sistema Watson de oncologia é hoje usado em um grande número de instituições de saúde dos EUA24. A obtenção de dados colhidos no Datasus com o fim de pagar contas hospitalares e que seriam processados, eventual- mente, no sistema de computação do Cidacs-Fiocruz poderia iniciar um programa de racionalização e melhoria da atenção médica no País, seja pelo conhecimento da frequência de ações adotadas para resolver casos clínicos e custos envolvidos, seja pela possibilidade de verificar os resultados dos procedimen- tos realizados e tratamentos prescritos, evitando recorrências e sugerindo ações preventivas. No Brasil, o Fleury Medicina e Saúde é o primeiro par- ceiro da unidade IBM Watson em Saúde na América Latina25, iniciando, sobretudo, estudos com o Watson Genomics para au- xiliar médicos a identificar medicamentos e ensaios clínicos relevantes com base em alterações genômicas de um indiví- duo e dados extraídos da literatura médica. O Watson na área de saúde fornece a oncologistas conhecimentos advindos dos mais importantes centros de pesquisa e tratamento de câncer no mundo. Integrar dados dos vários níveis de atenção à saúde, anali- sando-os por região e complexidade, incentivar a troca on-line de experiências, discutir o uso de novas tecnologias na prática médica, não só em grandes centros, mas, sobretudo, em áreas remotas, são imperativos a serem discutidos. Esperamos que outras instituições brasileiras sigam a ex- periência do Fleury e que passem a intercambiar informações sobre problemas de atenção à saúde em nível individual e co- letivo. Dados de notificações de doenças processados tempesti- vamente pela vigilância em saúde permitem estudar não só o diagnóstico desses casos, como os desvios em sua evolução, prever a possibilidade de surtos e epidemias, e propor as me- didas necessárias de prevenção e proteção da saúde da popu- lação. Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 191 CONCLUSÃO A deficiência na interação com o paciente, a falta de seu exame clínico e a dependência de exames complementares no diag- nóstico médico irão aumentar cada vez mais o uso do compu- tador em medicina. Afortunadamente, a informação científica tende a ser pa- dronizada. A Organização Mundial da Saúde e o Instituto de Saúde dos EUA (NIH) desenvolveram padrões, como a Classi- ficação Internacional de Doenças (CID) e a International Health Terminology Standards Development Organization (Snomed). A in- tegração de dados genômicos é feita por meio de sistemas de nomenclatura como o Gene Ontology. Sem uma padronização de dados fica difícil a integração e a adoção de sistemas de apoio à decisão clínica e em sistemas de saúde. A disponibilidade de sistemas de apoio à decisão clínica com grande acurácia, o uso de dispositivos vestíveis/corpo- rais (wearable devices), o aumento exponencial da capacidade de armazenar e processar dados de pacientes e da população (big data) são fatos que já se integraram à realidade em muitos países. A Inteligência Artificial processa esses dados por meio de algoritmos que tendem a se aperfeiçoar pelo seu próprio funcionamento (self learning) e a propor hipóteses diagnósticas cada vez mais precisas. Big data que agregam informações sobre saúde, determi- nantes genéticos de doenças, estudos de órgãos, células, mo- léculas, e até átomos, de mecanismos de transcrição e repres- são do DNA e como modificá-los, de proteínas e metabólitos e a interação de indivíduos com o ecossistema ensejam um conhecimento de possíveis agravos à saúde do indivíduo e da população e como controlá-los e resolvê-los. O sistema de computação Deep Mind inglês, recentemen- te adquirido pela Google, processa atualmente 1,6 milhão de prontuários de pacientes atendidos nos hospitais do Serviço Nacional de Saúde da Inglaterra (NHS), buscando desenvol- ver uma nova geração de sistemas de apoio à decisão clínica, analisando dados desses pacientes e gerando alertas sobre a sua evolução, evitando medicações contraindicadas ou con- flitantes e informando tempestivamente os profissionais de saúde sobre seus pacientes. Nos EUA, o programa de pesquisa em câncer estabele- ceu o projeto NCI-Match, que busca parear tipos de tumor e terapias prescritas. O projeto envolveu mil pacientes que apre- sentavam um tumor que não havia respondido a tratamentos padrões para o caso. CONCLUSÃO Usando marcadores genéticos, buscou-se Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 191 DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  A IBM, por sua vez, criou um supercomputador – o Wat- son – com capacidade de armazenar dados médicos num vo- lume extraordinário. O Watson assimilou dezenas de livros- -textos em medicina, toda a informação do PubMed e Medli- ne, e milhares de prontuários de pacientes do Sloan Kettering Memorial Cancer Hospital. Segundo a revista Forbes, o Wat- son analisou 25 mil casos clínicos com a assistência de 14.770 médicos para buscar melhorar sua acurácia diagnóstica e está ficando mais inteligente a cada ano. Sua rede de oncologia é hoje consultada por especialistas de um grande número de hospitais. 6. Wikipedia contributors. Clinical decision support system. Wikipedia, The Free Encyclopedia [capturado 6 mai. 2017]. Disponível em: https://en.wikipedia.org/wiki/Clinical_ decision_support_system 7. Castaneda C, Nafley K, Mannion C, Bhattacharyya P, Blake P, Pecora A, Goy A, Suh KS. Clinical Decision Su- pport Systems for Improving Diagnostic Accuracy and Achieving Precision Medicine. JClin Bioinforma. 2015:5(4). Disponível em: www.ncbi.nlm.nih.gov/pmc/articles/ PMC4381462/. 8. Elstein AS, Shulman LS. Sprafka SA. Medical Problem Solving: an analysis of clinical reasoning. Cambridge (MA):Harvard University Press; 1978. O programa Precision Medicine Initiative prevê estabelecer uma base de dados genéticos de um milhão de pessoas para avaliar a eficácia de drogas em condições específicas. 9. Rimoldi HJA. Diagnosing the Diagnostic Process. Medical Education.1988;22(4):270-278. O processamento de um grande volume de informações em saúde permitirá melhorar a compreensão da gênese, diag- nóstico e tratamento de problemas de saúde não só do indiví- duo, como da população. Permitirá, em consequência, propor novas ações voltadas à promoção, prevenção e recuperação da saúde, o que incluiria a necessidade de eventual reestrutura- ção dos sistemas voltados a desenvolver essas ações. 10. Fornaziero CC, Gordan PA, Garanhani ML. O raciocínio clínico dos estudantes de medicina em uma universidade pública do Brasil. Rev Bras Educ Med. 2012;36(4):463-469. 11. Conant JB. Two Modes of Thought. New York: Trident Press; 1964. 12. Weed LL. Medical Records, patient care and medical edu- cation. Irish J Med Sci.1964;6:271-282. Mas se o computador fornece o know-what, caberá ao mé- dico discutir o problema de saúde e suas possíveis soluções com o paciente, indicando o know-why do seu caso. CONCLUSÃO Isto requer uma contínua preocupação com a qualidade da formação médica e o entendimento de que o médico talvez seja o mais importante agente terapêutico, pela orientação que dá a seu paciente e, consequentemente, pelo alívio de suas tensões e necessidades. 13. Lobo LCG. Prontuário Médico Orientado para Problemas. Rev Hosp Clin Fac Med S Paulo.1976;31:67-72. 14. Turing AM. Computing Machinery and Intelligence. Mind.1950;49:433-460. 15. McCarthy J, Minsky ML, Rochester N, Shanon CE. A Pro- posal of the Dartmouth Summer Research Project on Ar- tificial Intelligence. Disponível em:http://en.wikipedia. org/wiki/History_of_artificial_intelligence CONFLITO DE INTERESSES O autor negou haver qualquer conflito de interesses. O autor negou haver qualquer conflito de interesses. O autor negou haver qualquer conflito de interesses. REFERÊNCIAS 16. Shortlife EH. The Adolescence of AI in medicine: will the field come of age in the ‘90s. Artif Intell Med. 1993;5(2):93-166. 1. Knox R. The Fading Art of the Physical Exam. National Public Radio [internet]. 2010 Sep 20;Science. Disponí- vel em: http://www.npr.org/templates/story/story. php?storyId=129931999 1. Knox R. The Fading Art of the Physical Exam. National Public Radio [internet]. 2010 Sep 20;Science. Disponí- vel em: http://www.npr.org/templates/story/story. php?storyId=129931999 17. Patel VL, Shortlife EH, Stefanelli M, Szolovits P, Berthold MR, Bellazi R, Abu-Hanna A. The Coming Age of Artificial Intelligence in Medicine. Artif Intell Med 2009;46(1):5-17. 2. Bleish H. Computer Evaluation of Acid-Base Disorders, JClinInv. 1972;53:285-291. 2. Bleish H. Computer Evaluation of Acid-Base Disorders, JClinInv. 1972;53:285-291. 18. Mukherjee S. A.I. Versus M.D: what happens when diag- nosis is automated? The New Yorker [on line] 2010 april 3. [capturado 3 mai. 2017] Disponível em: <http://www. newyorker.com/magazine/2017/04/03/ai-versus-md>. 3. Wikipedia contributors. Clinical decision support sys- tem. Wikipedia, The Free Encyclopedia [capturado 5 mai. 2017]. Disponível em: https://en.wikipedia. org/w/index.php?title=Clinical_decision_support_ system&oldid=778842100. 19. MeloM, ScarpinDJ, Amaro JR,EPassos, RBD, SatoJR, Fris- ton KJ,Price CJ. How Doctors Generete Diagnostic Hypo- theses: Study of Radiological Diagnosis with Functional Magnetic Ressonance Imaging. Disponível em: http:// journal.plos.org/plosone/article?id=10.1371/journal. pone.0028752 4. Bond WF, Schartz LM, Weaver KR, Levick D, Giuliano M, Graber ML. Differential Diagnosis Generators: an evalua- tion of currently available computer programs. JGenIntern Med. 2011;27(2):213-319. 20. CISCO VNI Mobile,2015, Cyber Academy. Global Mobile Traffic ForecastCAGR = Computed Annual Growth Rate. Edinburgh. 5. Berner ES. Clinical Decision Support Systems: theory and practice.New York:Springer; 1999. 5. Berner ES. Clinical Decision Support Systems: theory and practice.New York:Springer; 1999. Revista Brasileira de Educação Médica 2 Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 2 192 DOI: http://dx.doi.org/10.1590/1981-52712015v41n2esp Luiz Carlos Lobo  25. Merker J. Fleury Testa Watson para Saúde. Baguete [cap- turado 7 mai. 2017] Disponível em: https://www.baguete. com.br/noticias/04/10/2016/fleury-testa-watson-para- -saude 21. Topol E. The Patient Will See You Now:the future of medi- cine is in your hands. New York:Basic Books,2015. 22. LoboL CG. Sistemas de Informação e Planejamento de Saúde. Brasília: Ed.Univ. Brasília,1993. 23. IBM. IBM’s Watson Supercomputer May Soon be the Best Doctor in the World, Busines Insider [capturado 8 mai. 2017] Disponível em: Wikipedia contributors. Clinical de- cision support system. Wikipedia, The Free Encyclopedia [capturado 5 mai. 2017]. Disponível em: http://www. businessinsider.com/ibms-watson-may-soon-be-the-best- -doctor-in-the-world-2014-4 Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 193 ENDEREÇO PARA CORRESPONDÊNCIA Luiz Carlos Lobo Rua Raul Pompéia, 94 – apto 1007 Copacabana – Rio de Janeiro CEP 22080-002 – RJ E il l i l l b @ il Luiz Carlos Lobo Rua Raul Pompéia, 94 – apto 1007 Copacabana – Rio de Janeiro CEP 22080-002 – RJ E-mail: [email protected] 24. IBM. Como o Watson, a plataforma de computação cogni- tiva da IBM, está ajudando a revolucionar o perfil de cinco verticais econômicas, IDGNOW [capturado 6 Mai.2017] Disponível em: http://idgnow.com.br/ti-corporati- va/2016/06/20/como-o-watson-a-plataforma-de-compu- tacao-cognitiva-da-ibm-esta-ajudando-a-revolucionar-o- -perfil-de-cinco-verticais-economicas Revista Brasileira de Educação Médica 41 (2) : 185 – 193 ; 2017 193

https://openalex.org/W2947266462

https://www.frontiersin.org/articles/10.3389/fendo.2019.00318/pdf

English

Caregiver Emotional Burden in Testicular Cancer Patients: From Patient to Caregiver Support

Frontiers in endocrinology

cc-by

Caregiver Emotional Burden in Testicular Cancer Patients: From Patient to Caregiver Support Silvia De Padova 1, Chiara Casadei 2, Alejandra Berardi 1, Tatiana Bertelli 1, Alessia Filograna 2, Maria Concetta Cursano 3, Cecilia Menna 2, Salvatore Luca Burgio 2, Amelia Altavilla 2, Giuseppe Schepisi 2, Sabrina Prati 2, Sandra Montalti 2, Michal Chovanec 3,4, Giuseppe Luigi Banna 5, Luigi Grassi 6, Michal Mego 3,4 and Ugo De Giorgi 2* Silvia De Padova 1, Chiara Casadei 2, Alejandra Berardi 1, Tatiana Bertelli 1, Alessia Filograna 2, Maria Concetta Cursano 3, Cecilia Menna 2, Salvatore Luca Burgio 2, Amelia Altavilla 2, Giuseppe Schepisi 2, Sabrina Prati 2, Sandra Montalti 2, Michal Chovanec 3,4, Giuseppe Luigi Banna 5, Luigi Grassi 6, Michal Mego 3,4 and Ugo De Giorgi 2* 1 Psycho-Oncology Unit, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST), IRCCS, Meldola, Italy, 2 Medical Oncology Department, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST), IRCCS, Meldola, Italy, 3 Medical Oncology Department, Campus Bio-Medico University, Rome, Italy, 4 2nd Department of Oncology, Faculty of Medicine, Comenius University and National Cancer Institute, Bratislava, Slovakia, 5 Division of Medical Oncology, Cannizzaro Hospital, Catania, Italy, 6 University Hospital Psychiatry Unit, Integrated Department of Mental Health and Addictive Behavior, St. Anna University Hospital and NHS Community Health Trusts, Ferrara, Italy Edited by: Edited by: Gabriella Castoria, Second University of Naples, Italy Reviewed by: Silvio Naviglio, Università degli Studi della Campania Luigi Vanvitelli Caserta, Italy Erika Di Zazzo, Università degli Studi della Campania Luigi Vanvitelli Caserta, Italy *Correspondence: Ugo De Giorgi [email protected] Specialty section: This article was submitted to Cancer Endocrinology, a section of the journal Frontiers in Endocrinology Received: 26 February 2019 Accepted: 01 May 2019 Published: 28 May 2019 REVIEW published: 28 May 2019 doi: 10.3389/fendo.2019.00318 REVIEW published: 28 May 2019 doi: 10.3389/fendo.2019.00318 INTRODUCTION Testicular cancer (TC) is the most frequent solid tumor in young adult men aged between 15 and 35 years, and is a highly curable cancer with survival rates close to 99% for stage I disease cases (localized tumor) and 80–90% for cases with metastatic disease treated with cisplatin-based chemotherapy and surgery on residual masses, when indicated (1). In addition, the 10–20% of metastatic patients who are not cured with first-line cisplatin-based chemotherapy, increase their chances of long-term remission in nearly 50% of cases treated with second-line treatments, such as high-dose chemotherapy (HDCT) or standard-dose chemotherapeutic regimens, and in nearly 15–30% of cases treated in the following lines with other salvage regimens (2–5). A young age at diagnosis and excellent prognosis, physical, psychological and social well-being represent a significant indicator for follow-up and survivorship of these people. In fact, despite the excellent prognosis, cured patients may experience long-term somatic sequelae and psychosocial distress Keywords: caregiver, testicular, cancer, patients, long-term survivors Testicular cancer is the most common tumor in young males aged 15–40 years. The overall cure rate for men with testicular cancer is >90%, so a huge number of these patients will become testicular cancer survivors. These people may feel some stress in the experience of diagnosis, treatment, and consequences that affects the quality of life, and during follow-up, especially when new issues and emotional distresses appear over time, such as late side-effects of treatments and emotional challenges including fear of tumor relapse, fertility and sexuality concerns, and social and workplace issues. The cancer experience has an impact not only on patients, but also on their relatives (e.g., spouses, parents, or siblings), who often have to assume a caregiving role for the duration of and following treatment for cancer. Moreover, the caregiver plays an important role in supporting a man with a testicular cancer, providing physical and emotional care. This review presents a summary of existing knowledge regarding the impact and the burden of testicular cancer on caregivers. Citation: De Padova S, Casadei C, Berardi A, De Padova S, Casadei C, Berardi A, Bertelli T, Filograna A, Cursano MC, Menna C, Burgio SL, Altavilla A, Schepisi G, Prati S, Montalti S, Chovanec M, Banna GL, Grassi L, Mego M and De Giorgi U (2019) Caregiver Emotional Burden in Testicular Cancer Patients: From Patient to Caregiver Support. Front. Endocrinol. 10:318. doi: 10.3389/fendo.2019.00318 May 2019 | Volume 10 | Article 318 Frontiers in Endocrinology | www.frontiersin.org 1 De Padova et al. Caregiver in Testicular Cancer Patients GRAPHICAL ABSTRACT | Factors influencing Caregiver emotional burden in TC patients. GRAPHICAL ABSTRACT | Factors influencing Caregiver emotional burden in TC patients. Cancer is a family experience that exerts a change in family’s system, balance and identity, redefines the rules, changes the lives of all its members, brings an immense amount of stress, and presents many challenging situations. Cancer and the approaches used to treat it can introduce a complex array of lifestyle changes and emotional responses, which can be difficult for family members to handle. The diagnosis of cancer, its treatment and symptoms both of the illness and of the chemotherapy have an influence on how patients and their caregivers experience distress. A review on psychological impact of cancer on patients’ partners and other relatives affirmed that an important minority of carers become highly distressed, clinically depressed and anxious: in particular, prevalence of clinically significant distress among caregivers was reported to be 20–30% in studies using self-report questionnaires, whereas in studies that used diagnostic interviews rates are approximately 10% (15). However, data concerning caregivers’ distress are sparse. This review analyzes TC-related distress and burden of caregivers. according to the tumor and treatment burden (6, 7). However, type and duration of each treatment depends on initial stage of the disease (Table 1). As a consequence, different physical and psychological loads correlate with different treatment loads. Both tumor diagnosis and tumor treatment are usually stressful events affecting not only patients but the whole family system (Graphical Abstract) (8). Frontiers in Endocrinology | www.frontiersin.org Role of Caregiver Caregivers are individuals, usually family members or relatives (e.g., partner, parents, but also close friends), who have a significant relationship with the patient suffering from a life- threatening illness and provide assistance (9, 10). Along the whole process of the disease and its treatments, they are engaged in the practical help and psychological support for coping with the situation, including the emotions of uncertainty and fear (11). This role requires many abilities that may be physically, emotionally and financially demanding. The burden of caregiving has been defined a “multidimensional biopsychosocial reaction resulting from an imbalance of care demands relative to caregivers’ personal time, social roles, physical and emotional states, financial resources, and formal care resources given the other multiple roles they fulfill” (12). 1https://www.nccn.org/professionals/physician_gls/pdf/testicular.pdf CAREGIVER BURDEN Histology Stage Primary treatment Most common adverse events Detrimental effect on QoL Stage Secondary treatment Most common adverse events Detrimental effect on QoL Seminoma I Single agent Carboplatin (AUC7 for 1 cycle) or RT (20 or 25 Gy) Myelotoxicity Fatigue Low risk Relapsed/Refractory TC VeIP (4 cycles) or TIP (4 cycles) or HDCT ± RT Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Cardiovascular toxicity Pulmonary toxicity Solid secondary tumors Leukemia High risk IIA RT (30 Gy) Fibrosis Low risk IIB BEP (3 cycles) or EP (4 cycles) or RT in selected non-bulky cases Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Pulmonary toxicity Solid secondary tumors Leukemia Intermediate risk IIC, III BEP (3 or 4 cycles) Non-seminoma I Surveillance or BEP (1 cycle) or RPLND If BEP: Myelotoxicity Fatigue Vomiting Alopecia Intermediate risk II, IIIA BEP (3 cycles) or EP (4 cycles) or Nerve-sparing RPLND Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Pulmonary toxicity Solid secondary tumors Leukemia Intermediate risk IIIB, IIIC BEP (4 cycles) or VIP (4 cycles) RT, radiotherapy; BEP, Bleomycin, Etoposide, and Cisplatin; EP, Etoposide and Cisplatin; RPLND, retroperitoneal lymph-node dissection; VIP, Etoposide, Ifosfamide, and Cisplatin; VeIP, VeIP Vinblastine, Ifosfamide, Cisplatin; TIP, Paclitaxel, Ifosfamide, and Cisplatin; HDCT, high dose chemotherapy. Stage and primary treatment according to NCCN guidelines version 1.20191. tumors Leukemia IIA RT (30 Gy) Fibrosis Low risk IIB BEP (3 cycles) or EP (4 cycles) or RT in selected non-bulky cases Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Pulmonary toxicity Solid secondary tumors Leukemia Intermediate risk IIC, III BEP (3 or 4 cycles) Non-seminoma I Surveillance or BEP (1 cycle) or RPLND If BEP: Myelotoxicity Fatigue Vomiting Alopecia Intermediate risk II, IIIA BEP (3 cycles) or EP (4 cycles) or Nerve-sparing RPLND Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Pulmonary toxicity Solid secondary tumors Leukemia Intermediate risk IIIB, IIIC BEP (4 cycles) or VIP (4 cycles) RT, radiotherapy; BEP, Bleomycin, Etoposide, and Cisplatin; EP, Etoposide and Cisplatin; RPLND, retroperitoneal lymph-node dissection; VIP, Etoposide, Ifosfamide, an VeIP, VeIP Vinblastine, Ifosfamide, Cisplatin; TIP, Paclitaxel, Ifosfamide, and Cisplatin; HDCT, high dose chemotherapy. Stage and primary treatment according to NCCN version 1.20191. CAREGIVER BURDEN The experience of illness perceived by the caregiver depends on some specific aspects of the disease such as the type of cancer and the stage of life in which it is diagnosed. Cancer could determine major effects both on caregivers and patients, with literature clearly indicating that cancer affects the emotional, social, physical, and spiritual well-being of patients and their family members (13). TC has a profound effect on body image and on the personal image of oneself and, and often occurs in adolescence and young adulthood, times characterized by significant life changes and psychosocial challenges. These men are in the prime of their lives, when health is often taken for granted, while interpersonal relationships and the desire to start a family may be major interests (16–19). Most studies in family caregivers of cancer patients reported diverse problems as a consequence of their role, ranging from diminished physical health and psychological distress to an adverse impact on their work (14). The most prevalent physical problems included sleep disturbance, fatigue, pain, loss of physical strength, loss of appetite, and weight loss. May 2019 | Volume 10 | Article 318 2 Caregiver in Testicular Cancer Patients De Padova et al. TABLE 1 | Therapeutic strategies, clinical complications, and correlation with QoL. CAREGIVER BURDEN Main psychological problems in TC caregivers Need of information Anxiety Depression Inadequacy Lack of practical and emotional support Risk of infertility Sexual difficulties TABLE 3 | Differences in emotional burden between TC patients and their caregivers. Caregivers (%) Patients (%) Anxiety 30–50 (24) 19–34 (25) Depression 12–59 (24) 10–25 (25) TABLE 2 | Main psychological issues in TC caregivers. Main psychological problems in TC caregivers Need of information Anxiety Depression Inadequacy Lack of practical and emotional support Risk of infertility Sexual difficulties TABLE 2 | Main psychological issues in TC caregivers. Main psychological problems in TC caregivers Need of information Anxiety Depression Inadequacy Lack of practical and emotional support Risk of infertility Sexual difficulties TABLE 3 | Differences in emotional burden between TC patients and their caregivers. TABLE 2 | Main psychological issues in TC caregivers. Patients, caregivers and care providers had different expectations about TC survivorship: psychological distress was considered as highly relevant by 35% of patients and caregivers and 93% of care providers; the couple’s relationship was quite or very difficult for 12% of patients and caregivers in comparison to 64% in the perception of care providers (34). A different perception of the illness experience could affect the recognition and ability to respond to the needs of patients and caregivers. In another study, close relatives of men suffering from TC highlight four themes: the disease and its course, normalization, the long-term consequences, and the social network (35). The results showed that relatives suffer from social isolation (35). (15, 23). In this contest, TC survivors’ caregivers are a high risk population. Several studies reported rates of anxiety of between 30 and 50% (24) among family caregivers in comparison to rates of anxiety of between 19 and 34% (25) in patient samples. Likewise, rates of depression are reported to be between 10 and 25% (25) in patients samples compared to between 12 and 59% (24) among family caregivers. This underlines that, in many cases, the psychological burden on caregivers is even greater than in patients (Table 3). Caregivers receive less practical and emotional support from friends and professionals than patients (26). However, their high self-efficacy can improve their own mental state and also the mental well-being of patients. CAREGIVER BURDEN Another source of caregiver burden could be the fertility issue: the paternity rates among men who attempted to conceive a child after treatment were 71% at 15 years and 76% at 20 years after orchiectomy, but this rate ranged from 48% in the HDCT group to 92% in the surveillance group (36). Sandén and Söderhamn (37) reported a conversational interview to a young woman whose partner had TC using a semistructured guide with open- ended questions. Caring became primary for female partner, and she focused less on her own needs in order to support the patient; everyday life changed, as more time was spent at the hospital, the home, and the parent’s home. The third keypoint was the shortness of time: from the discovery of the disease and the start of chemotherapy, time was reported as passing very quickly, and the felt like they spent a lot of time with their physicians. In their caregiver role, partners of TC patients have a social task: commonly they give information to family members and friends while her husband is in the hospital (27). Another considerable aspect of these women’s burden is emotional experiences connected to the illness and to the period of life in which it occurs. Often the diagnosis of TC happens in an important period of life, characterized by major life changes and specific developmental tasks, when forming intimate and long- term emotional and sexual relationships, and starting families may be major concerns (28). Each member of the couple, both the patient and the partner, is faced with the possibility of treatment-related infertility and sexual difficulties in a period of life where partners are often focused on starting a family (27). However, only a minority of the couples experienced more serious and long lasting testicular cancer-induced disturbances in sexual and marital relationships. In general, couples felt their relationship became more tightly bonded and stronger following the confrontation with TC (29, 30). In literature, little data exists about the role of mothers as caregiver of TC patients and the dramatic changes in their lives. Unlike their healthy peers, young TC patients often face greater challenges in life: they may experience delays in developmental milestones, difficulties in employment and interpersonal relationships, and medical and institutional problems (e.g., economy, education, transport). CAREGIVER BURDEN These challenges can hinder their transition to independence, which is not favored by mothers who continue to take care even when their sons progress to adulthood (38). Both psychosocial and QoL consequences occur years after the experience of the tumor and the end of treatments. Tuinman et al. (31) showed that spouses who experienced the diagnosis and treatment process had better physical QoL than the average woman. Their stress response levels were low and were related to the stress response level of TC survivors and to the duration of treatments received. However, these women, even years after the completion of treatment, were experiencing more stress response symptoms than the TC survivors. Frontiers in Endocrinology | www.frontiersin.org CAREGIVER BURDEN IIA RT (30 Gy) Fibrosis Low risk IIB BEP (3 cycles) or EP (4 cycles) or RT in selected non-bulky cases Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Pulmonary toxicity Solid secondary tumors Leukemia Intermediate risk IIC, III BEP (3 or 4 cycles) Non-seminoma I Surveillance or BEP (1 cycle) or RPLND If BEP: Myelotoxicity Fatigue Vomiting Alopecia Intermediate risk II, IIIA BEP (3 cycles) or EP (4 cycles) or Nerve-sparing RPLND Myelotoxycity Fatigue Alopecia Vomiting Neurotoxicity Infertility Pulmonary toxicity Solid secondary tumors Leukemia Intermediate risk IIIB, IIIC BEP (4 cycles) or VIP (4 cycles) RT, radiotherapy; BEP, Bleomycin, Etoposide, and Cisplatin; EP, Etoposide and Cisplatin; RPLND, retrope VeIP, VeIP Vinblastine, Ifosfamide, Cisplatin; TIP, Paclitaxel, Ifosfamide, and Cisplatin; HDCT, high dose ch version 1.20191. IIIB, IIIC BEP (4 cycles) or VIP (4 cycles) RT, radiotherapy; BEP, Bleomycin, Etoposide, and Cisplatin; EP, Etoposide and Cisplatin; RPLND, retroperitoneal lymph-node dissection; VIP, Etoposide, Ifosfamide, and Cisplatin; VeIP, VeIP Vinblastine, Ifosfamide, Cisplatin; TIP, Paclitaxel, Ifosfamide, and Cisplatin; HDCT, high dose chemotherapy. Stage and primary treatment according to NCCN guidelines version 1.20191. Diagnosis of TC causes a sort of emotional earthquake which involves relatives, partners and close friends. Cancer patients during treatment and also during the follow-up period experience many needs, and caregivers are often unprepared to respond to this important burden (20). Main psychological problems experienced by TC caregivers are summarized in Table 2. Of note is that the physical and mental health of patients and their caregivers are often related. In this contest, improving knowledge and social support to caregivers could help to ameliorate patients’ global health. Confirming this, depression symptoms are less frequent in patients living in couples (21). Taking care of a patient with cancer is described as a full- time job (22) and caregivers themselves often are in great need of psychological support. Caregivers of TC patients usually are parents or partners, typically young females and it is recognized that the highest predisposition to symptoms of distress have been showed by female caregivers of young age and lower social status (19). Therefore, for female partners of cancer patients, there is a risk of developing psychological and psychiatric morbidity and lower quality-of-life (QoL) than women in healthy couples May 2019 | Volume 10 | Article 318 Frontiers in Endocrinology | www.frontiersin.org 3 Caregiver in Testicular Cancer Patients De Padova et al. TABLE 2 | Main psychological issues in TC caregivers. CONCLUSIONS Cancer treatment is not an individual experience, but induces deep effects on patients’ families. Couples who achieve the survivorship phase often have to change life plans, deal with TC and treatment-related effects, and manage worries about future health (52). Patients partnered at diagnosis experience a better emotional and physical adaptation to disease (53, 54) and the majority of follow-up studies reported that the rate of divorce or broken relationship was 5% to 10% (55). The impact of disease on caregivers depends also on patients’ story. TC patients who undergo, after orchiectomy, one cycle of chemotherapy probably have a lower burden of distress compared with patients who complete four standard cycles of chemotherapy, due to the reduced treatment load. Moreover, some patients are not cured with first-line chemotherapy and have to be treated with further intensified chemotherapeutic regimens, including standard dose chemotherapy supported by granulocyte-colony stimulating factor (G-CSF) (3) and/or HDCT with support of autologous peripheral-blood stem-cell (56, 57). HDCT is a stressful program both for patients and their family, due to long hospitalization periods and a high risk of treatment-related toxicity that requires the adoption of specific precautionary measures. These therapeutic options are able to lead to long-term remission of disease, but leave a stressful emotional burden on the patient and his caregivers. There is some data available about stressful burden of caregivers of elderly persons with physical dependence (58), but there are little evidence about young patients’ caregivers, though often they are young people at risk of psychological distress which could have an impact on long-term effects (59, 60). According to the different stage of the disease at diagnosis, several treatment strategies are recommended (Table 1). These modalities are associated with different complications and late toxicities and a negative impact on QoL. TC survivors have a high risk of leukemia; the relative risk, associated with the previous use of etoposide, ranges between 3.5 and 4.5 and appears often within 10 years following the end of treatment (44). Younger age at radiotherapy and/or chemotherapy increases risk for solid secondary tumors and remains elevated for at least 35 years (45). In long-term setting pulmonary toxicity, infections and cardiovascular events are higher compared with the general population (46, 47). For TC survivors and their caregivers, preserved fertility is a fundamental subject which has an important impact on their QoL (48). Frontiers in Endocrinology | www.frontiersin.org CONCLUSIONS The prospect of paternity improves with the decreasing number of cycles of chemotherapy, therefore the correct management of TC requires a careful balance between the intensity of treatment and burden of disease, in order to limit short and long term adverse events (49). In sight of this, the correct management of late toxicities is essential in order to preserve the higher QoL of patients and their caregivers. A recent multidisciplinary consensus conference by the Italian Germ cell cancer Group (IGG) and the Associazione Italiana di Oncologia Medica (AIOM) has provided recommendations for surveillance and follow-up appointments of men with TC, suggesting a visit with caregiver at the beginning of follow-up and, eventually, a psychological consultation (50). Unfortunately, there are little data available in literature concerning the role of caregiver in TC patients, probably because of the lower number of persons involved compared with breast cancer, for example. However, in our opinion, according to the young age of patients and the very good prognosis, it is important to consider a more integrated system of the patient and his social support, with the purpose of improving QoL not only during active treatment, but also in the follow-up period, and to encourage a less traumatic return to the everyday life. To date there are no evidence about the importance of the role of nurse in supporting caregivers of TC-patients. In our opinion, it is important to develop educational programs with the aim of creating a cancer clinical nurse specialist role, who could support patients and their families during the course of disease. This program could guarantee personalized nursing assistance and aid (both psychological and practical), leaving from the psychosocial contest of the patients. The role of nurse is expressed before, during and after the identification and CAREGIVER THERAPY Caregivers need a large volume of information, including: diagnosis-related information, prognosis-related information, treatment-related information, information on homecare, and information about impact on the family or on relationship with partner. Therefore, psychoeducational interventions have been conceived to increase their knowledge. Bultz et al. (39) reported that this sort of intervention had a positive impact on caregivers’ ability to provide care and also improve marital satisfaction Relevant components of the caregiver’s burden consist of the support, assistance and information needs that, if not unmet, leads to reduced QoL and high levels of distress (32). Kim et al. (33) demonstrated that caregivers of cancer patients frequently have a variety of unmet needs and that unmet needs strongly predicted their QoL. May 2019 | Volume 10 | Article 318 Frontiers in Endocrinology | www.frontiersin.org 4 Caregiver in Testicular Cancer Patients De Padova et al. of patients. Pelusi et al. (40) revealed that caregivers sharing their cancer experience with others in storytelling is essential to offer educational information and emotional support to those who hear it, but also care for self is an important component of managing the course of these events. A Chinese study explored the relationship between family resilience and the post- traumatic growth, and the quality of life of survivors of breast cancer, demonstrating that family resilience decreased caregivers’ and patients’ burden (41). One intervention used telephone interpersonal counseling, which was delivered to patients and their caregivers separately to improve cancer education and resulted in significant decreases in depression and anxiety levels in the caregivers group (42). Kozachik et al. (43) conducted a quasi-experimental study to describe the use of complementary therapy (such as reflexology, guided imagery, and reminiscence therapy) to cancer patients undergoing chemotherapy and their family caregivers. The authors were unable to draw conclusions regarding the impact of complementary therapy on caregiver burden, however they suggest that one complementary therapy may be incorporated into patients’ and caregivers’ courses of cancer treatment (43). the monitoring of signs and symptoms of the disease and the treatment, in order to promote the well-being of the patients and their caregivers. TC patients and their families have to be included; they should have the opportunity to be involved in the planning of the assistance and in the decision making process through individualized services which could adapt to the changing psychophysical status of the patients (51). REFERENCES Traeger L, Greer JA, Fernandez-Robles C, Temel JS, Pirl WF. Evidence-based treatment of anxiety in patients with cancer. J Clin Oncol. (2012) 30:1197–205. doi: 10.1200/JCO.2011.39.5632 26. Gröpper S, van der Meer E, Landes T, Bucher H, Stickel A, Goerling U. Assessing cancer-related distress in cancer patients and caregivers receiving outpatient psycho-oncological counseling. Support Care Cancer. (2016) 24:2351–7. doi: 10.1007/s00520-015-3042-9 6. Rossen PB, Pedersen AF, Zachariae R, Von Der Maase H. Health-related quality of life in long-term survivors of testicular cancer. J Clin Oncol. (2009) 27:5993–9. doi: 10.1200/JCO.2008.19.6931 27. Tuinman MA, Hoekstra HJ, Sleijfer DT, Fleer J, Vidrine DJ, Gritz ER, et al. Testicular cancer: a longitudinal pilot study on stress response symptoms and quality of life in couples before and after chemotherapy. Support Care Cancer. (2007), 15:279–86. doi: 10.1007/s00520-006-0119-5 7. Shinn EH, Swartz RJ, Thornton BB, Spiess PE, Pisters LL, Basen-Engquist KM. Testis cancer survivors’ health behaviors: comparison with age-matched relative and demographically matched population controls. J Clin Oncol. (2010) 28:2274–9. doi: 10.1200/JCO.2009.23.9608 28. Warner EL, Kent EE, Trevino KM, Parsons HM, Zebrack BJ, Kirchhoff AC. Social well-being among adolescents and young adults with cancer: a systematic review. Cancer. (2016) 122:1029–37. doi: 10.1002/cncr.29866 8. Rolland JS. Cancer and the family: an integrative model. Cancer. (2005) 104 (Suppl. 11):2584–95. doi: 10.1002/cncr.21489 9. Hudson P, Payne S. Family caregivers and palliative care: current status and agenda for the future. J Palliat Med. (2011) 14:864–9. doi: 10.1089/jpm.2010.0413 29. Gritz ER, Wellisch DK, Siau J, Wang HJ. Long-term effects of testicular cancer on sexual functioning in married couples. Cancer. (1989) 64: 1560–7. 10. Hudson P, Payne S. The future of family caregiving: research, social policy and clinical practice. In: Hudson P, Payne S, editors. Family Carers in Palliative Care: A Guide for Health and Social Care Professionals. Oxford: Oxford University Press (2009). p. 277–303. 30. Hannah MT, Gritz ER, Wellisch DK, Fobair P, Hoppe RT, Bloom JR, et al. Changes in marital and sexual functioning in long-term survivors and their spouses: testicular cancer versus Hodgkin’s disease. Psychooncology. (1992) 1:89–103. 31. Tuinman MA, Fleer J, Hoekstra HJ, Sleijfer DT, Hoekstra-Weebers JE. Quality of life and stress response symptoms in long-term and recent spouses of testicular cancer survivors. Eur J Cancer. (2004) 40:1696–703. doi: 10.1016/j.ejca.2004.03.020 11. Romito F, Goldzweig G, Cormio C, Hagedoorn M, Andersen BL. Informal caregiving for cancer patients. Cancer. (2013) 119 (Suppl. 11):2160–9. doi: 10.1002/cncr.28057 12. REFERENCES Given CW, Stommel M, Given BA, Osuch J, Kurtz ME, Kurtz JC. The influence of cancer patients’ symptoms and functional status on patients’ depression and family caregivers’ reaction and depression. Health Psychol. (1993) 12:277–85. 32. Printz C. Cancer caregivers still have many unmet needs. Cancer. (2011) 117:1331. doi: 10.1002/cncr.26075 33. Kim H, Yi M. Unmet needs and quality of life of family caregivers of cancer patients in south korea. Asia Pac J Oncol Nurs. (2015) 2:152–9. doi: 10.4103/2347-5625.158019 13. Northouse LL. Helping families of patients with cancer. Oncol Nurs Forum. (2005) 32 743–50. doi: 10.1188/04.ONF.743-750 34. De Padova S, Rosti G, Scarpi E, Salvioni R, Amadori D, De Giorgi U. Expectations of survivors, caregivers and healthcare providers for testicular cancer survivorship and quality of life. Tumori. (2011) 97:367–73. doi: 10.1700/912.10036 14. Stenberg U, Ruland CM, Miaskowski C. Review of the literature on the effects of caring for a patient with cancer. Psychooncology. (2010) 19:1013–25. doi: 10.1002/pon.1670 15. Pitceathly C, Maguire P. The psychological impact of cancer on patients’ partners and other key relatives: a review. Eur J Cancer. (2003) 39:1517–24. doi: 10.1016/s0959-8049(03)00309-5 35. Sanden I, Hyden LC. How everyday life is affected: an interview study of relatives of men suffering from testicular cancer. J Psychosoc Oncol. (2002) 20:27–44. doi: 10.1300/J077v20n02_02 16. Fleer J, Hoekstra HJ, Sleijfer DT, Tuinman MA, Klip EC, Hoekstra-Weebers JE. Quality of life of testicular cancer survivors and the relationship with sociodemographics, cancer-related variables, and life events. Support Care Cancer. (2006) 14:251–9. doi: 10.1007/s00520-005-0879-3 36. Brydøy M, Fosså SD, Klepp O, Bremnes RM, Wist EA, Wentzel-Larsen T, et al. Paternity following treatment for testicular cancer. J Natl Cancer Inst. (2005) 97:1580–8. doi: 10.1093/jnci/dji339 37. Sandén I, Söderhamn O. Experience of living in a disrupted situation as partner to a men with testicular cancer. Am J Mens Health. (2009) 3:126–33. doi: 10.1177/1557988307311289 17. Arai Y, Kawakita M, Hida S, Terachi T, Okada Y, Yoshida O. Psychosocial aspects in long-term survivors of testicular cancer. J Urol. (1996) 155:574–8. doi: 10.1016/S0022-5347(01)66452-8 18. Jones GY, Payne S. Searching for safety signals: the experience of medical surveillance amongst men with testicular teratomas. Psychooncology. (2000) 9:385–94. doi: 10.1002/1099-1611(200009/10)9:5<385::AID-PON467>3.0. CO;2-B 38. Deatrick JA, Hobbie W, Ogle S, et al. Competence in caregivers of adolescent and young adult childhood brain tumor survivors. Health Psychol. (2013) 33:1103–12. doi: 10.1037/a0033756 39. Bultz BD, Speca M, Brasher PM, Geggie PH, Page SA. REFERENCES 20. Kent EE, Rowland JH, Northouse L, Litzelman K, Chou WYS, Shelburne N, et al.. Caring for caregivers and patients: research and clinical priorities for informal cancer caregiving. Cancer. (2016) 122:1987–95. doi: 10.1002/cncr.29939 1. Fizazi K, Oldenburg J, Dunant A, Chen I, Salvioni R, Hartmann JT, et al. Assessing prognosis and optimizing treatment in patients with postchemotherapy viable nonseminomatous germ-cell tumors (NSGCT): results of the sCR2 international study. Ann Oncol. (2008) 19:259–64. doi: 10.1093/annonc/mdm472 1. Fizazi K, Oldenburg J, Dunant A, Chen I, Salvioni R, Hartmann JT, et al. Assessing prognosis and optimizing treatment in patients with postchemotherapy viable nonseminomatous germ-cell tumors (NSGCT): results of the sCR2 international study. Ann Oncol. (2008) 19:259–64. doi: 10.1093/annonc/mdm472 21. Gil T, Sideris S, Aoun F, van Velthoven R, Sirtaine N, Paesmans M, et al. Testicular germ cell tumor: short and long-term side effects of treatment among survivors. Mol Clin Oncol. (2016) 5:258–64. doi: 10.3892/mco.2016.960 2. De Giorgi U, Demirer T, Wandt H, Taverna C, Siegert W, Bornhauser M, et al. Second-line high-dose chemotherapy in patients with mediastinal and retroperitoneal primary non-seminomatous germ cell tumors: the EBMT experience. Ann Oncol. (2005) 16:146–51. doi: 10.1093/annonc/ mdi017 22. Rabow MW, Hauser JM, Adams J. Supporting family caregivers at the end of life: “they don’t know what they don’t know.” J Am Med Assoc. (2004) 291:483–91. doi: 10.1001/jama.291.4.483 23. Hagedoorn M, Buunk B, Kuijer R, Wobbes T, Sanderman R. Couples dealing with cancer: role and gender differences regarding psychological distress and quality of life. Psychooncology. (2000) 9:232–42. doi: 10.1002/1099-1611(200005/06)9:3<232::aid-pon458>3.0.co;2-j 3. Kondagunta GV, Bacik J, Donadio A, Bajorin D, Marion S, Sheinfeld J, et al. Combination of paclitaxel, ifosfamide, and cisplatin is an effective second-line therapy for patients with relapsed testicular germ cell tumors. J Clin Oncol. (2005) 23:6549–55. doi: 10.1200/JCO.2005.19.638 24. Grunfeld E, Coyle D, Whelan T, Clinch J, Reyno L, Earle CC, et al. Family caregiver burden: results of a longitudinal study of breast cancer patients and their principal caregivers. CMAJ. (2004) 170:1795–801. doi: 10.1503/cmaj.1031205 4. Oing C, Seidel C, Bokemeyer C. Therapeutic approaches for refractory germ cell cancer. Expert Rev Anticancer Ther. (2018) 18:389–97. doi: 10.1080/14737140.2018.1450630 5. De Giorgi U, Rosti G, Aieta M, Testore F, Burattini L, Fornarini G, et al. Phase II study of oxaliplatin and gemcitabine salvage chemotherapy in patients with cisplatin-refractory nonseminomatous germ cell tumor. Eur Urol. (2006) 50:1032–8. doi: 10.1016/j.eururo.2006.05.011 25. AUTHOR CONTRIBUTIONS SD, CC, and UD collaborated in the conception, in the data retrieval, and in the drafting of the text. CC, GS, and MC collaborated in the revision of the text and in the completion of the bibliographic research. AB, TB, AF, CM, SB, AA, SP, SM, MC, GB, LG, and MM revised the manuscript. May 2019 | Volume 10 | Article 318 5 Caregiver in Testicular Cancer Patients De Padova et al. REFERENCES A randomized controlled trial of a brief psychoeducational support group for partners of early stage breast cancer patients. Psychooncology. (2000) 9:303–13. doi: 10.1002/1099-1611(200007/08)9:4<303::AID-PON462>3.0.CO;2-M 19. Chovanec M, Vasilkova L, Setteyova L, Obertova J, Palacka P, Rejlekova K, et al. Long-term cognitive functioning in testicular germ-cell tumor survivors. Oncologist. (2018) 23:617–23. doi: 10.1634/theoncologist.2017-0457 May 2019 | Volume 10 | Article 318 Frontiers in Endocrinology | www.frontiersin.org 6 Caregiver in Testicular Cancer Patients De Padova et al. survivors: a review of the literature. J Adolesc Health. (2010) 47:115–25. doi: 10.1016/j.jadohealth.2010.04.005 survivors: a review of the literature. J Adolesc Health. (2010) 47:115–25. doi: 10.1016/j.jadohealth.2010.04.005 40. Pelusi J, Krebs LU, Castro L. Understanding cancer—understanding the stories of life and living. J Cancer Edu. (2005) 20 (Suppl. 1):12–6. doi: 10.1207/s15430154jce2001s_04 53. Brodsky MS. Testicular Cancer Survivors’ Impressions of the Impact of the Disease on their Lives. Qual Health Res. (1995) 5:78–96. doi: 10.1177/104973239500500106 j 41. Liu Y, Li Y, Chen L, Li Y, Qi W, Yu L. Relationships between family resilience and posttraumatic growth in breast cancer survivors and caregiver burden. Psychooncology. (2018) 27:1284–90. doi: 10.1002/pon.4668 54. Sheppard C, Wylie KR. An assessment of sexual difficulties in men after treatment for testicular cancer. Sex Relatsh Ther. (2001) 16:47–58. doi: 10.1080/14681990124325 42. Badger T, Segrin C, Meek P, Lopez AM, Bonham E, Sieger A. Telephone interpersonal counseling with women with breast cancer: symptom management and quality of life. Oncol Nurs Forum. (2005) 32:273–9. doi: 10.1188/05.ONF.273-279 55. Fosså SD, Travis LB, Dahl AA. Medical and psychosocial issues in testicular cancer survivors. In: Oncology: An Evidence-Based Approach. New York, NY: Springer (2006). p. 1825–37. 43. Kozachik SL, Wyatt G, Given CW, Given BA. Patterns of use of complementary therapies among cancer patients and their family caregivers. Cancer Nurs. (2006) 29:84–94. doi: 10.1097/00002820-200603000-00002 56. Adra N, Abonour R, Althouse SK, Albany C, Hanna NH, Einhorn LH. High-dose chemotherapy and autologous peripheral-blood stem- cell transplantation for relapsed metastatic germ cell tumors: the Indiana university experience. J Clin Oncol. (2017) 35:1096–102. doi: 10.1200/JCO.2016.69.5395 44. Richiardi L, Scélo G, Boffetta P, Hemminki K, Pukkala E, Olsen JH, et al. Second malignancies among survivors of germ-cell testicular cancer: a pooled analysis between 13 cancer registries. Int J Cancer. (2007) 120:623–31. doi: 10.1002/ijc.22345 57. Pedrazzoli P, Ferrante P, Kulekci A, Schiavo R, De Giorgi U, Carminati O, et al. REFERENCES Autologous hematopoietic stem cell transplantation for breast cancer in Europe: critical evaluation of data from the European Group for Blood and Marrow Transplantation (EBMT) Registry 1990–1999. Bone Marrow Transplant. (2003) 32:489–94. doi: 10.1038/sj.bmt.1704153 45. Ng AK, Kenney LB, Gilbert ES, Travis LB. Secondary malignancies across the age spectrum. Semin Radiat Oncol. (2010) 20:67–78. doi: 10.1016/j.semradonc.2009.09.002 46. Travis LB, Beard C, Allan JM, Dahl AA, Feldman DR, Oldenburg J, et al. Testicular cancer survivorship: research strategies and recommendations. J Natl Cancer Inst. (2010) 102:1114–30. doi: 10.1093/jnci/djq216 58. Navarro-Sandoval C, Uriostegui-Espìritu LC, Delgado-Quinones EG, Sahagùn-Cuevas MN. Depression and burden on primary caregivers of elderly persons with physical dependence of the UMF 171. Rev Med Inst Mex Seguro Soc. (2017) 55:25–31. 47. Huddart RA, Norman A, Shahidi M, Horwich A, Coward D, Nicholls J, et al. Cardiovascular disease as a long-term complication of treatment for testicular cancer. J Clin Oncol. (2003) 21:1513–23. doi: 10.1200/JCO.2003. 04.173 59. Kenney LB, Antal Z, Ginsberg JP, Hoppe BS, Bober SL, Yu RN, et al. Improving male reproductive health after childhood, adolescent, and young adult cancer: progress and future directions for survivorship research. J Clin Oncol. (2018) 36:2160–8. doi: 10.1200/JCO.2017.76.3839 48. Stoehr B, Schachtner L, Pichler R, Holzner B, Giesinger J, Oberguggenberger A, et al. Influence of achieved paternity on quality of life in testicular cancer survivors. BJU Int. (2013) 111 (4 Pt B):E207–12. doi: 10.1111/j.1464-410X.2012.11579.x 60. De Giorgi U, Rosti G, Slavin S, Yaniv I, Harousseau JL, Ladenstein R, et al. Salvage high-dose chemotherapy for children with extragonadal germ-cell tumours. Br J Cancer. (2005) 93:412–7. doi: 10.1038/sj.bjc.6602724 49. Brydøy M, Fosså SD, Klepp O, Bremnes RM, Wist EA, Wentzel-Larsen T, et al. Paternity and testicular function among testicular cancer survivors treated with two to four cycles of cisplatin-based chemotherapy. Eur Urol. (2010) 58:134–40. doi: 10.1016/j.eururo.2010.03.041 Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. 50. Banna GL, Nicolai N, Palmieri G, Ottaviano M, Balzarini L, Barone D, et al. Recommendations for surveillance and follow-up of men with testicular germ cell tumors: a multidisciplinary consensus conference by the Italian Germ cell cancer Group and the Associazione Italiana di Oncologia Medica. Crit Rev Oncol Hematol. (2019) 137:154–64. Frontiers in Endocrinology | www.frontiersin.org REFERENCES doi: 10.1016/j.critrevonc.2019.03.006 Copyright © 2019 De Padova, Casadei, Berardi, Bertelli, Filograna, Cursano, Menna, Burgio, Altavilla, Schepisi, Prati, Montalti, Chovanec, Banna, Grassi, Mego and De Giorgi. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Copyright © 2019 De Padova, Casadei, Berardi, Bertelli, Filograna, Cursano, Menna, Burgio, Altavilla, Schepisi, Prati, Montalti, Chovanec, Banna, Grassi, Mego and De Giorgi. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 51. Dreyer B, Macfarlane K, Hendry D. The testicular cancer nurse specialist: a pivotal role in patient care. Br J Nurs. (2018) 27:S26–7. doi: 10.12968/bjon.2018.27.18.S26 52. Carpentier MY, Fortenberry JD. Romantic and sexual relationships, body image, and fertility in adolescent and young adult testicular cancer 52. Carpentier MY, Fortenberry JD. Romantic and sexual relationships, body image, and fertility in adolescent and young adult testicular cancer May 2019 | Volume 10 | Article 318 Frontiers in Endocrinology | www.frontiersin.org 7

https://openalex.org/W4386818934

https://zenodo.org/records/8354360/files/CRS_Primer_Dynamic_reporting.pdf

English

Primer: Dynamic Reporting

Zenodo (CERN European Organization for Nuclear Research)

cc-by

What is dynamic reporting? Dynamic reporting goes back to a paradigm called literate programming which was developed by Knuth (1984) for computer programming. At the time, software engineers were faced with the prob- lem that their source code was not easily understandable for other people such as collaborators or clients. Literate programming solves this problem by combining the machine-readable source code and its human-readable documentation into one document. As science became more quantitative and computer-dependent, researchers quickly recognized the importance of literate programming for writing dynamic research reports. Modern dynamic reporting systems allow researchers to perform data analysis and to write the corresponding documentation in a single document, making each step easier to understand, automating boring and error-prone tasks, and improving computational reproducibility. Why should I use dynamic reporting? In many empirical research disciplines, writing scientific analyses starts with importing a data set into a software tool and executing specific commands in order to get the desired numerical results, tables and figures, as illustrated in Figure 1 (top). This output is subsequently copied and pasted manually into a typesetting program, the results are discussed and the scientific report is finalized (Xie, 2015). This two-step procedure has various disadvantages: • Use of two separate software tools for analysing and reporting is awkward and tedious • Use of two separate software tools for analysing and reporting is awkward and tedious • Many manual steps make the approach prone to errors and not directly reproducible • Updates in the original data set require the whole manual process to be repeated • Sharing the full analysis and reporting pipeline with other researchers is difficult • Reusing the analysis procedure on similar problems is cumbersome Dynamic reporting addresses these issues through a combination of analysis code and text in one document. As such the analysis and the reporting are integrated in one software environment as il- lustrated in Figure 1 (bottom). Primer: Dynamic Reporting Primer: Dynamic Reporting Felix Hofmann, Samuel Pawel, Monika Hebeisen, Leonhard Held doi: 10.5281/zenodo.7565735 How do I use dynamic reporting? There are various implementations of dynamic reporting, the most popular ones are R Markdown (Xie et al., 2018) and Jupyter Notebooks (Kluyveretal., 2016). Both ofthemcombinethe simpleMarkdown syntax to specify text and layout with commonly used programming languages (e.g., R, Python, and Julia), and are therefore easy to learn for beginners. Recently, a more modern implementation named Quarto was introduced by the developers of R Markdown. Quarto works with R, Python and Julia, and is quickly gaining popularity. Other implementations (e.g., knitr, Sweave, pweave, and weave.jl) use LaTeX instead of Markdown, a more powerful but also more difficult typesetting system. These implementations are therefore intended for more advanced users. Some commercial programs (e.g., Mathematica or MATLAB) also offer implementations of dynamic reporting but are not very popular because they are inaccessible to people without a license. There are also online platforms for sharing 1 .dat .csv .xlsx Data Data analysis .py .sps .sav .m .R ## clean data dat <- read.csv("data".csv) ………………………………………………………… ………………………………………………………… ## Table table(x, y, z) ………………………………………………………… ………………………………………………………… ## Figure plot(x, y) ………………………………………………………… ………………………………………………………… Tables Figures .dat .csv .xlsx .sav .png .tiff .pdf .jpeg Report .tex .odt .docx Introduction …………………………………………… …………………………………………… Methods …………………………………………… …………………………………………… Results …………………………………………… Table 1 Figure 1 x y z …… …… …… …… …… …… Manual reporting Code editing Copy paste Text editing Copy paste .dat .csv .xlsx Data .sav Report .tex .odt .html x y z …… …… …… …… …… …… Script .ipynb .Rnw .qmd .Rmd ## Introduction ……………………………………………………………… ……………………………………………………………… ## Methods ……………………………………………………………… ……………………………………………………………… ## Results ……………………………………………………………… ```{r "data-cleaning"} ## load data dat <- read.csv("data".csv) ``` ```{r "table1"} ## create table table(x, y, z) ``` ```{r "figure1"} plot(x, y) ``` Text editing Code editing .docx .pdf Introduction …………………………………………… …………………………………………… Methods …………………………………………… …………………………………………… Results …………………………………………… Table 1 Figure 1 x y z …… …… …… …… …… …… Dynamic reporting Automatic Figure 1: Schematic illustration of manual reporting (top) and dynamic reporting (bottom) workflows and collaboratively working on dynamic reports in the cloud, for instance, Binder (Jupyter et al., 2018 and Google Colab. We recommend R Markdown (for R) and Jupyter Notebooks (for Python and Julia) to beginners wh want to learn dynamic reporting. Both are powerful, free and open-source, and easy to use. How do I use dynamic reporting? In th .dat .csv .xlsx Data Data analysis .py .sps .sav .m .R ## clean data dat <- read.csv("data".csv) ………………………………………………………… ………………………………………………………… ## Table table(x, y, z) ………………………………………………………… ………………………………………………………… ## Figure plot(x, y) ………………………………………………………… ………………………………………………………… Tables Figures .dat .csv .xlsx .sav .png .tiff .pdf .jpeg Report .tex .odt .docx Introduction …………………………………………… …………………………………………… Methods …………………………………………… …………………………………………… Results …………………………………………… Table 1 Figure 1 x y z …… …… …… …… …… …… Manual reporting Code editing Copy paste Text editing Copy paste Data analysis Tables Figures Data .dat .csv .xlsx Data .sav Report .tex .odt .html x y z …… …… …… …… …… …… Script .ipynb .Rnw .qmd .Rmd ## Introduction ……………………………………………………………… ……………………………………………………………… ## Methods ……………………………………………………………… ……………………………………………………………… ## Results ……………………………………………………………… ```{r "data-cleaning"} ## load data dat <- read.csv("data".csv) ``` ```{r "table1"} ## create table table(x, y, z) ``` ```{r "figure1"} plot(x, y) ``` Text editing Code editing .docx .pdf Introduction …………………………………………… …………………………………………… Methods …………………………………………… …………………………………………… Results …………………………………………… Table 1 Figure 1 x y z …… …… …… …… …… …… Dynamic reporting Automatic Figure 1: Schematic illustration of manual reporting (top) and dynamic reporting (bottom) workflows .dat .csv .xlsx Data .sav Report .tex .odt .html x y z …… …… …… …… …… …… Script .ipynb .Rnw .qmd .Rmd ## Introduction ……………………………………………………………… ……………………………………………………………… ## Methods ……………………………………………………………… ……………………………………………………………… ## Results ……………………………………………………………… ```{r "data-cleaning"} ## load data dat <- read.csv("data".csv) ``` ```{r "table1"} ## create table table(x, y, z) ``` ```{r "figure1"} plot(x, y) ``` Text editing Code editing .docx .pdf Introduction …………………………………………… …………………………………………… Methods …………………………………………… …………………………………………… Results …………………………………………… Table 1 Figure 1 x y z …… …… …… …… …… …… Dynamic reporting Automatic Figure 1: Schematic illustration of manual reporting (top) and dynamic reporting (bottom) workflows. Dynamic reporting Data Figure 1: Schematic illustration of manual reporting (top) and dynamic reporting (bottom) workflows. and collaboratively working on dynamic reports in the cloud, for instance, Binder (Jupyter et al., 2018) and Google Colab. and collaboratively working on dynamic reports in the cloud, for instance, Binder (Jupyter et al., 2018) and Google Colab. We recommend R Markdown (for R) and Jupyter Notebooks (for Python and Julia) to beginners who want to learn dynamic reporting. Both are powerful, free and open-source, and easy to use. In the following, we will briefly explain the basics of R Markdown and Juypter Notebook. R Markdown Dynamic reporting with R Markdown is done via R Markdown files which have the file extension .Rmd. These files can be edited in any text editor. Figure 2 shows an R Markdown file opened in the RStudio editor. RStudioisnotrequiredforusingRMarkdownbutwerecommenditforbeginnerssinceitmakes using R Markdown much easier. 2 Markdown file opened in RStudio (left) and its html output (right). Markdown file opened in RStudio (left) and its html output (right). ht). pdf output ut (r ut (r outpu html htm arkd arkd yter Notebook opened in JupyterLab (left) and its pdf output (right). Jupyter Notebooks Dynamic reporting with Jupyter Notebooks is based on iPython notebook ipynb files. These files are typically edited in the Jupyter Notebook or the JupyterLab browser interface. JupyterLab is more modern and has more features but both essentially link a notebook editor with an interactive session of a specific programming language like Python, R, or Julia. An advantage of this design is that the browser serves as a file editor which provides simple access to all of the built-in functionality. In gen- eral, working with Jupyter Notebooks offers much more of a point-and-click experience than editing R Markdown files which allows beginners to be productive from the beginning. This can be seen in Figure 3 (left), which shows an example notebook opened in JupyterLab. However, due to the simple user interface, Jupyter Notebooks are also a bit limited when it comes to the layout of the output doc- ument. In terms of supported output formats, Jupyter Notebooks offer a wide variety of options to export a Notebook as html, tex, md, pdf, and other file formats. Figure 3 (right) shows the notebook exported to a pdf document. and eft) b (l b ( b ( erL er No Every R Markdown file starts with a header code block enclosed by three hyphens (---) where meta data and the output is specified.1 In our example, we choose html as output format, however, many other types are possible (for example pdf or docx). After the file header, the main part of the R Mark- down file starts. Text can now be written in the usual Markdown syntax. For example, *text* becomes text, **text** becomes text, or #section and ##subsection become headings (the official syntax is doc- umented here http://daringfireball.net/projects/markdown/syntax, see also Ovadia, 2014). Between passages of text, it is possible to embed R and/or Python code in so-called code chunks that start with ‘‘‘{r} or ‘‘‘{python}, respectively, and end with ‘‘‘.2 Figures are automatically inserted in the output. Function, such as kable from the knitr package or xtable from the xtable package (Dahl et al., 2019) can be used for dynamically generating tables. R code can also be placed inline, i.e. em- beddedintextsentences. Thisisusefulforreportinganalysisresults, suchaseffectsizesorconfidence intervals, automatically in the text. An inline R code chunk starts with ‘r and ends with ‘. By clicking on the “knit” button in RStudio the Rmd source file is converted to the specified output format.3 For an R Markdown file to compile successfully, it is required that the code in the Rmd file runs successfully without errors. This can be seen as a further advantage of dynamic reporting with R Markdown as it ensures a minimum degree of error-free code. By clicking on the “knit” button in RStudio the Rmd source file is converted to the specified output format.3 For an R Markdown file to compile successfully, it is required that the code in the Rmd file runs successfully without errors. This can be seen as a further advantage of dynamic reporting with R Markdown as it ensures a minimum degree of error-free code. 1The header code block is also called the YAML header because the YAML configuration language is used within. 2 Fore more details on the Python integration in R Markdown see the reticulate cheat sheet (RStudio, 2019). 3Internally, RStudio runs the function rmarkdown::render, so if RStudio is not used this functions needs to be called instead. 1The header code block is also called the YAML header because the YAML configuration language is used within. 2Fore more details on the Python integration in R Markdown see the reticulate cheat sheet (RStudio 2019) more details on the Python integration in R Markdown see the reticulate cheat sheet (RStudio, 2019). nally, RStudio runs the function rmarkdown::render, so if RStudio is not used this functions needs to be called Want to learn more? The code from the examples above is available at https://github.com/crsuzh/dynamicReporting. The repository also includes knitr and Quarto versions of the same examples. More about R Markdown can be learned in the online book by Xie et al. (2018). There is also an excellent book on knitr (Xie, 2015) with comprehensive online documentation (https://yihui.org/knitr/). The Turing Way hand- book to reproducible, ethical and collaborative data science (Turing Way Community et al., 2019) provides further information on dynamic reporting with Jupyter notebooks and other tools for re- producible research. Peikert and Brandmaier (2021) explain how dynamic reporting can be com- binedwithothercomputationaltoolssuchascontainerization,versioncontrol,anddependencyman- agement. Hands-on tutorials for dynamic reporting are given in the online courses “Reproducible ResearchusingJupyterNotebooks”(DataCarpentry,2016)and“RforReproducibleScientificAnalysis” (Zimmermanetal.,2019)bythesoftware/datacarpentry, andtheRMarkdowntutorial(RStudio,2020) from RStudio. 5 Acknowledgments WethankUlrikeHeld, EvaFurrer, andRachelHeyardforhelpfulcommentsandaccesstotheirteaching materials. Version from September 18, 2023. This work is published under a Creative Commons Attribution 4.0 International license (CC-BY 4.0). Edited by Felix Hofmann (0000-0002-3891-6239), Samuel Pawel (0000-0003-2779-320X), and Leonhard Held (0000-0002-8686-5325). Version from September 18, 2023. This work is published under a Creative Commons Attribution 4.0 International license (CC-BY 4.0). Edited by Felix Hofmann (0000-0002-3891-6239), Samuel Pawel (0000-0003-2779-320X), and Leonha References Dahl, D. B., Scott, D., Roosen, C., Magnusson, A., and Swinton, J. (2019). xtable: Export Tables to LaTeX or HTML. URL https://CRAN.R-project.org/package=xtable. R package version 1.8-4. Data Carpentry (2016). Reproducible Research using Jupyter Notebooks. URL https://reproducible-science-curriculum.github.io/workshop-RR-Jupyter/. URL Jupyter, P., Bussonnier, M., Forde, J., Freeman, J., Granger, B., Head, T., Holdgraf, C., Kelley, K., Nalvarte, G., Osheroff, A., Pacer, M., Panda, Y., Perez, F., Ragan-Kelley, B., and Willing, C. (2018). Binder 2.0 – reproducible, interactive, sharable environments for science at scale. doi:10.25080/majora-4af1f417-011. Kluyver, T., Ragan-Kelley, B., Pérez, F., Granger, B. E., Bussonnier, M., Frederic, J., Kelley, K., Hamrick, J. B., Grout, J., Corlay, S., et al. (2016). Jupyter Notebooks – a publishing format for reproducible computational workflows, volume 2016. doi:10.3233/978-1-61499-649-1-87. Knuth, D. E. (1984). Literate Programming. The Computer Journal, 27(2):97–111. doi:10.1093/comjnl/27.2.97. Ovadia, S. (2014). Markdown for librarians and academics. Behavioral & Social Sciences Librarian, 33(2):120–124. doi:10.1080/01639269.2014.904696. Peikert, A. and Brandmaier, A. M. (2021). A reproducible data analysis workflow. Quantitative and Computational Methods in Behavioral Sciences, 1. doi:10.5964/qcmb.3763. RStudio (2019). Use Python with R with reticulate. URL https://ugoproto.github.io/ugo_r_doc/pdf/reticulate.pdf. RStudio (2020). R Markdown from RStudio: Introduction. URL https://rmarkdown.rstudio.com/lesson-1.html. URL Turing Way Community, Arnold, B., Bowler, L., Gibson, S., Herterich, P., Higman, R., Krystalli, A., Mor- ley, A., O’Reilly, M., and Whitaker, K. (2019). The Turing Way: A Handbook for Reproducible Data Science. Zenodo. doi:10.5281/zenodo.3233986. Xie, Y. (2015). Dynamic Documents with R and knitr. CRC Press, Boca Raton, FL. Xie, Y. (2015). Dynamic Documents with R and knitr. CRC Press, Boca Raton, FL. Xie, Y., Allaire, J. J., and Grolemund, G. (2018). R markdown: The definitive guide. Chapman and Hall/CRC. URL https://bookdown.org/yihui/rmarkdown/. Xie, Y., Allaire, J. J., and Grolemund, G. (2018). R markdown: The definitive guide. Chapman and Hall/CRC. URL https://bookdown.org/yihui/rmarkdown/. Zimmerman, N., Wilson, G., Silva, R., Ritchie, S., Michonneau, F., Oliver, J., Dashnow, H., Boughton, A., Teucher, A., et al. (2019). swcarpentry/r-novice-gapminder: Software Carpentry: R for Reproducible Scientific Analysis, June 2019. doi:10.5281/zenodo.3265164. 6

https://openalex.org/W2151182907

https://www.scielo.br/j/hcsm/a/pQZCMdFXKNvWLZVcJCr6fHM/?lang=pt&format=pdf

Portuguese

Poesia e ciência: quase crônica

História, ciências, saúde-Manguinhos

cc-by

Marco Lucchesi Professor da Pós-Graduação em Letras Neolatinas Faculdade de Letras – Universidade Federal do Rio de Janeiro – UFRJ Av. Brigadeiro Trompowski, s/n, Cidade Universitária 21941-590 Rio de Janeiro – RJ – Brasil [email protected] POESIA E CIÊNCIA – QUASE CRÔNICA E N S A I O POESIA E CIÊNCIA – QUASE CRÔNICA E N S A I O NCIA – QUASE CRÔNICA E N S A I O Poesia e ciência – quase crônica Poetry and science – almost a short-short story Poetry and science – almost a short-short story LUCCHESI, M.: Poesia e ciência – quase crônica. História, Ciências, Saúde – Manguinhos, v. 13 (suplemento), p. 257-67, outubro 2006. q História, Ciências, Saúde – Manguinhos, Relação entre ciência e poesia. O diálogo possível entre o verso e o universo (ou multiverso). Testemunhos do conhecimento pessoal do autor do texto com Ilya Prigogine e Carlo Rubbia, no Rio de Janeiro. A poesia descobre o horizonte quântico. E suas implicações com relação à idéia de uma física das partículas do simples ao complexo. O paradigma antigo e o novo paradigma. O rosto inesperado e belo da complexidade. A beleza e a inteligência, como interfaces de um mesmo processo. PALAVRAS-CHAVE: poesia; ciência; crônica. PALAVRAS-CHAVE: poesia; ciência; crônica. LUCCHESI, M.: Poetry and science – almost a short-short story. História, Ciências, Saúde – Manguinhos, v. 13 (supplement), p. 257-67, October 2006. Relation between science and poetry. The possible dialogue between verse and universe (or multiverse). Testimony from Ilya Prigogine and Carlo Rubbia, in Rio de Janeiro , personal acquaintances of the author. Poetry discovers the quantum horizon. And its implications vis-à-vis the notion of particle physics, from simple to complex. The old paradigm and the new paradigm. The beautiful, unexpected face of complexity. Beauty and intelligence, as interfaces of one same process. KEYWORDS: poetry; science; short-short story. 257 v. 13 (suplemento), p. 257-67, outubro 2006 v. 13 (suplemento), p. 257-67, outubro 2006 MARCO LUCCHESI MARCO LUCCHESI Uma grande verdade é aquela cujo contrário Também é uma grande verdade Uma grande verdade é aquela cujo contrário Também é uma grande verdade Niels Bohr C C om uma pequena luneta e um atlas celeste eu tentava, aos treze anos, decifrar, desengonçado e boquiaberto, os compli- cados arabescos com que os homens mapearam o céu. Dezembro terminava, e a história de minha solidão adquiria um estranho componente noturno. O Boieiro e os Cães de Caça da palavra seguem a Ursa eternamente e mais se atrevem na distância de outras órbitas nos fúlgidos anéis de Berenice em que se perdem os contrafortes da linguagem A palavra e seu destino seguem como a Via-Látea para a constelação de Hércules ao sempre suspirado cabo Não [“Sphera”, de Marco Lucchesi, p. 63] [“Sphera”, de Marco Lucchesi, p. 63] E tremia com Pascal diante do Silêncio. E mergulhava com Leopardi no Infinito. De um lado, o firmamento pontilhado de es- trelas. De outro, minhas divagações trespassadas de incerteza. Poetry and science – almost a short-short story Que mais poderia esperar ver senão o zodíaco da vida? Espelhos que me refletissem e demarcassem o rosto de minha transitoriedade. O resto de uma estrela contemplando a própria origem. Sonhando o retorno. Meus observatórios eram a praia de Piratininga e a igreja de São Francisco, onde padre Dante apreciava, com o seu grande telescópio, as estrelas da constelação de Sagitário. p g Minha primeira emoção foram as crateras da Lua. Depois de Ariosto e Júlio Verne, eis-me cosmonauta virtual entre cosmonautas literários. Ou talvez a encantadora nebulosa de Órion, de cuja incomparável beleza não é possível dissertar sem assombro. Ou talvez o impacto de saber que a luz atual de Andrômeda se originou há dois milhões 258 História, Ciências, Saúde – Manguinhos, Rio de Janeiro POESIA E CIÊNCIA – QUASE CRÔNICA de anos. Ou ainda a descoberta desconcertante de que o número de galáxias varia de cem a duzentos bilhões! Tudo isso era espantoso! “As Plêiades” São mais de mil demônios que povoam, estrelas solitárias!, o vórtice da noite... Órion volta para as Plêiades seu arco luminoso e a flecha pontiaguda torna mais fria nossa dor e mais espessa... Súbita flecha fere e arrebata os mais de mil demônios que povoam, no vórtice do tempo, a noite fria [“Poemas Reunidos”, Marco Lucchesi, p. 74-5] Ia tateando, como um cego – já no segundo científico –, as leis de anos. Ou ainda a descoberta desconcertante de que o número de galáxias varia de cem a duzentos bilhões! Tudo isso era espantoso! “As Plêiades” São mais de mil demônios que povoam, estrelas solitárias!, o vórtice da noite... Órion volta para as Plêiades seu arco luminoso e a flecha pontiaguda torna mais fria nossa dor e mais espessa... Súbita flecha fere e arrebata os mais de mil demônios que povoam, no vórtice do tempo, a noite fria [“Poemas Reunidos”, Marco Lucchesi, p. 74-5] [“Poemas Reunidos”, Marco Lucchesi, p. 74-5] Ia tateando, como um cego – já no segundo científico –, as leis da mecânica celeste e quântica. Ocupado com idéias gerais, sentia- me assombrado, raptado pela grandeza do Universo, tomado pela noção do mistério. Quase como Sono de Cipião. v. 13 (suplemento), p. 257-67, outubro 2006 História, Ciências, Saúde – Manguinhos, Rio de Janeiro Poetry and science – almost a short-short story As certezas das aulas de física e química do Salesiano começavam a ser demolidas pelos golpes de Fritjof Capra, Mário Schenberg, Abdus Salam, John Brockman, Stephen Hawking, Harald Fritzsch, Steven Weinberg, Michael Talbot, Werner Heisenberg, além de Ilya Prigogine, que conheci no Rio, em 1996. Uma inteligência impressionadora, a de Prigogine. 259 v. 13 (suplemento), p. 257-67, outubro 2006 MARCO LUCCHESI Após a demolição das correntes fixistas, tais leituras me acerca- vam do insólito edifício da astrofísica, ao redor do qual era possível identificar diversos andaimes e mecanismos, que se destinavam a completar o que faltava, a unificação das forças fundamentais, para cuja tarefa não poucos operários trabalhavam usando supercordas. j p p p A origem do Universo me intrigava. E bastava apenas a idéia de origem para me intrigar. Pensar o Universo... g p g Pairava, no Princípio, a totalidade sem tempo, a supersimetria absoluta e o vazio quântico imponderável. Tudo constituía uma unidade primordial. Não havia nome. Não havia coisa. Não havia tempo. Não havia sequer o verbo haver. O silêncio da unidade parecia consumir a força precária da dualidade. Faltava o brilho das estrelas e o calor do Sol. A Terra, de onde o brilho das estrelas e o calor do Sol seriam apreciados quinze bilhões de anos mais tarde, não passava de um sonho (se existisse o sonhador). Ó misté- rio insondável! Ó solidão infinita! Meus antepassados e meus descendentes dormiam informes no seio desta assombrosa unidade. Ainda não éramos filhos do carbono e do amoníaco. Não havia car- bono ou amoníaco. Tudo era tudo em Tudo. Senti-me trespassado por uma forte emoção metafísica. Essa monstruosa unidade só engendrou vida a partir da quebra de sua própria simetria, quando se romperam os grilhões da tota- lidade. A matéria começava a superar a antimatéria. O número de elétrons era maior do que o de pósitrons. Os prótons superavam os antiprótons e os neutrinos ultrapassavam os antineutrinos. Assim, graças ao desequilíbrio da balança cósmica, originou-se a vida: os átomos e as células, as estrelas e as galáxias, os rios e os oceanos, a consciência e a autoconsciência, esta página precária e seu generoso leitor. A noosfera teilhardiana. “De rerum natura” Alheios ao destino dos mortais além das nuvens claras e sombrias vivem os deuses raros nas alturas... v. 13 (suplemento), p. 257-67, outubro 2006 Poetry and science – almost a short-short story livres de enganos dores nostalgias de morte vil que aos poucos nos invade; da chuva de átomos em que se evade indefinidamente a natureza 260 POESIA E CIÊNCIA – QUASE CRÔNICA em sua eterna mas avara empresa de reunir os átomos-enxame, seguindo a força rude do cliname, para formar compostos provisórios, que se desfazem noutros repertórios: estrelas, águas, nuvens, tempestades, cristais, abelhas, glórias ou cidades e flores, pedras, corpos, consciência – figuram como pálida aparência... e acima desse mundo sempre em guerra acima da miragem dessa terra repousam esquecidos nos meatos mais livres os celestes, mais beatos [“Poemas Reunidos”, Marco Lucchesi, p. 59-60] [“Poemas Reunidos”, Marco Lucchesi, p. 59-60] [“Poemas Reunidos”, Marco Lucchesi, p. 59-60] O átomo primordial fazia-me pensar nos pré-socráticos e no imenso Lucrécio, especialmente na idéia do rompimento. Não fosse um inesperado desvio, a infinita chuva de átomos lucreciana – semina rerum, corpuscola minima – pareceria marcada por uma universal monotonia. Não fosse aquele desvio produtor do entrechoque dos átomos em todas as direções, quando principiaram a ricochetear no espaço vazio, nada do que existe jamais teria existido. Essa inclinação voluntária e autônoma explicava o agregado de átomos formadores dos corpos visíveis e invisíveis do cosmos. O desvio da chuva lucreciana discrepava do redemoinho democrítico. O De rerum natura abandonava o dínos e assumia o clinamen. Mas o resultado era o mesmo e dependia da colisão das partículas. Átomos infinitos, combinações infinitas, espaços infinitos. Pelo vigor da chuva ou do redemoinho, abscissão do infinito ou permanência do infinito. O casamento com a metafísica estava comprometido. O tristíssimo Lucrécio consolava o jovem estudante, que estava prestes a pedir o divórcio. Mors universalis. 261 v. 13 (suplemento), p. 257-67, outubro 2006 MARCO LUCCHESI A alma, um composto de átomos lisos e redondos. A morte, implausível – sendo tão-somente desagregação dos átomos do corpo e da alma, agora decompostos e ulteriormente recompostos em novas combinações. O corpo e a alma, a matéria e o mundo se agregavam e se desagregavam e voltavam a se agregar formando uma torrente que parecia não ter fim. Apesar disso, o Universo havia de se decompor. As palavras com as quais Lucrécio define esse momento guardam um sopro de poesia triste e superior. Pálidas analogias poderiam associar o Big Bang e o Big Crunch à decomposição do De rerum natura. História, Ciências, Saúde – Manguinhos, Rio de Janeiro Poetry and science – almost a short-short story Da mesma maneira pela qual se formou o nosso mundo, assim também nasceram outros iguais ou diferentes deste. Formaram-se espaços vazios, metacósmicos, cognominados intermundia, onde moravam os deuses. A supernova que brilha pouco acima do horizonte e o café que se resfria sobre a mesa: assim opera em todos os quadrantes a lei terrível da entropia [“Sphera”, de Marco Lucchesi, p. 99] opera em todos os quadrantes [“Sphera”, de Marco Lucchesi, p. 99] Quase a idéia dos Universos-bolha. Para Gell Mann, devíamos chamar o uni-verso de multi-verso. Uma bolha-mãe gerando uma bolha-filho. Diversos mundos gerados uns pelos outros e ligados por túneis de tempo e de espaço (as famosas pontes de Einstein- Rosen, conhecidas como buracos de minhoca), indicam aspectos da simetria quebrada de todas as bolhas. Como não lembrar de Giordano Bruno e de grande parte da ficção científica de um Cyrano ou de um Wikórski? Ou de O túnel do tempo, seriado televisivo de minha adolescência? Analogias insipientes, pois nada mais distante de Epicuro, Demócrito e Lucrécio do que a teoria quântica dos campos, onde entra em jogo – além das diferenças primárias – a compreensão da antimatéria. Para cada partícula corresponderia uma antipartícula, sendo que a antipartícula da antipartícula seria a própria partícula. Tomemos um exemplo. O hidrogênio é composto de um próton e de um elétron, enquanto o anti-hidrogênio é composto de um antipróton e de um pósitron. Donde Richard Feynman explicar a diferença traçada pelo ponteiro do relógio: a matéria seguia a dire- ção horária e a antimatéria seguia a direção anti-horária. E aqui começavam meus devaneios. Voltar no tempo. Passar pelo Aqueronte. Abraçar os mortos. Adiar o futuro. 262 História, Ciências, Saúde – Manguinhos, Rio de Janeiro POESIA E CIÊNCIA – QUASE CRÔNICA POESIA E CIÊNCIA – QUASE CRÔNICA POESIA E CIÊNCIA – QUASE CRÔNICA Tudo isso fazia girar a cabeça. Basta lembrar o princípio da incerteza, de Heisenberg, segundo o qual podemos determinar a posição exata de uma partícula ou a sua velocidade, mas não as duas coisas ao mesmo tempo; ou a passagem instantânea de um elétron da órbita infe-rior para a órbita superior, sem atravessar o espaço intermediá- rio, quando um fóton atinge o átomo, a que se dá o nome de salto quântico; ou o gato de Schrödinger, que demonstrou a impossibilidade de separação e de interação entre o ato de observar e a coisa observada. 263 v. Poetry and science – almost a short-short story 13 (suplemento), p. 257-67, outubro 2006 MARCO LUCCHESI Quem não se espantou com a teoria quântica não a compreen- deu! Foi o que comentou Carlo Rubbia numa conferência em 1993, na Praia Vermelha. Eu queria conhecer o homem que demonstrara indiretamente o reverso da medalha atômica. Rubbia construíra um acelerador próton-antipróton e seus resultados foram coroados de êxito. Seguindo as idéias de Van der Meer, a tarefa quase impossível de produzir e armazenar antimatéria dependia do resfriamento estocástico dos antiprótons. Algo extre- mamente árduo conseguido através de uma férrea disciplina e uma ciclópica determinação para aperfeiçoar os aceleradores de partículas. O resultado é que, de um bilhão de interações, obtiveram-se apenas cinco que encerravam provas cabais da subpartícula W. As expe- riências davam conta da simetria das forças fundamentais. No caso de Rubbia, em especial, a força fraca e o eletromagnetismo, como teoriza Abdus Salam. O Nobel da Física seria conquistado em 1982. q Aquele era o homem que comprovara a idéia extraordinária de que habitávamos uma parte do multiverso. Rubbia mostrou uma grande afabilidade e conversava com total desembaraço, num inglês impecável. Um ser vigoroso. Uma das questões que guardei daquele encontro foi a de Luís Calife, acerca da complexidade do modelo padrão da física quântica, que mais se parecia com um tratado de botânica. Rubbia não se fez de rogado e teceu considerações favoráveis diante da complexidade, ao contrário do pensamento grego, em que a razão mais simples devia presidir à compreensão do não-simples. Tratava-se de um esforço para se alcançar a arché. Rubbia defendeu a complexidade como o alfabeto básico da natureza, diante de cuja leitura não devíamos recuar. Mas eram mais de dois milênios sob a influência do paradigma da simplicidade. Não seria fácil modificar a idéia incomparavelmente cristalina da unidade para aceitar de pronto que a complexidade abrigava os raios de uma impressentida beleza. A hiperfísica não havia superado a metafísica. p De todo modo, Rubbia defendia a complexidade com argumentos fortes que decorriam de sua prática de físico experimental. Fiz-lhe uma só pergunta: recordei-lhe as palavras de Dirac, Nobel de Física, para quem a especulação cosmológica repousava na beleza matemática. E que os físicos teóricos aceitavam a neces- sidade da beleza matemática como um artigo de fé. Dirac afirmava que a teoria da relatividade gozou de uma imediata aceitação uni- versal em virtude de sua beleza matemática. Poetry and science – almost a short-short story Desejava saber se Carlo Rubbia concordava com esta opinião. Mesmo porque, Dirac sofria a nostalgia do simples. A resposta veio como um raio: a beleza era uma das faces do Universo e a matemática era a forma de traduzir aquela beleza. 264 História, Ciências, Saúde – Manguinhos, Rio de Janeiro POESIA E CIÊNCIA – QUASE CRÔNICA POESIA E CIÊNCIA – QUASE CRÔNICA Rubbia insistia na beleza do complexo e se orgulhava de ter parti- cipado da construção da ponte que unia a força fraca ao eletromag- netismo, e que significava uma parte da beleza virtual da Teoria da Grande Unificação. Alguém na sala lembrou a história cósmica que o nosso corpo registra e o abismo de tempo que podemos alcançar, partindo de qualquer uma de nossas células. Comecei então a recordar as leituras que proporcionavam a célebre viagem ao microscópico, da baixa para a alta energia, do presente ao passado, da célula ao quark. Vemos dentro da célula – e aqui tomo a liberdade de parafrasear T. Ferris – uma confusão de tortuosos ribossomos e ondulantes mitocôndrias, lisossomos esféricos e estrelados centríolos, conjun- tos de complexos aparelhos garantidores das funções respiratória, sanitária e produtora de energia. Conquanto esta célula tenha ape- nas alguns anos (no meu caso trinta e três), a sua estrutura re- monta a mais de um bilhão de anos, quando se formaram as células eucariontes. Indo mais fundo, começamos a descortinar dentro do núcleo os singulares contornos das macromoléculas do ADN. Cada qual en- cerrando, em sua estrutura de duplas hélices e em suas pontes de hidrogênio, uma grande quantidade de informação genética, acu- mulada ao longo de quatro bilhões de anos de evolução. Armaze- nada num alfabeto nucleotídeo de quatro letras – feito de açúcar e moléculas de fosfato – que determinam a singularidade do indiví- duo e a relação deste com a espécie. Prosseguindo a viagem pelo túnel do tempo, deparamo-nos com as moléculas do ADN, formadas por átomos, cujas camadas ele- trônicas aparecem engrinaldadas numa espantosa variedade de formas. Alguns desses elétrons acabaram de chegar, arrancados como foram dos átomos adjacentes. Outros se juntaram aos seus núcleos atômicos, há mais de cinco bilhões de anos, na nebulosa de que a Terra foi formada. q Se aumentarmos mais cem mil vezes um átomo de carbono, teremos apenas o núcleo diante de nosso campo visual. v. 13 (suplemento), p. 257-67, outubro 2006 Poetry and science – almost a short-short story Os núcleos foram reunidos dentro de uma estrela que explodiu antes do apare- cimento do Sol. Finalmente, examinando mais de perto, podemos perceber os quarks que formam cada próton e cada nêutron no núcleo. Os quarks foram reunidos quando o Universo não passava de uns poucos segundos de idade. Assim, as estruturas menores e mais fundamentais encontram-se ligadas por níveis altíssimos de ener- gia, porque foram forjadas no calor altíssimo do instante zero, da Grande Explosão. 265 v. 13 (suplemento), p. 257-67, outubro 2006 MARCO LUCCHESI [“Poemas Reunidos”, Marco Lucchesi, p. 69-70] 266 História, Ciências, Saúde – Manguinhos, Rio de Janeiro POESIA E CIÊNCIA – QUASE CRÔNICA POESIA E CIÊNCIA – QUASE CRÔNICA Deixei a conferência esmagado pelo sentimento do tempo. Silêncios sobre-humanos e espaços interminados. Pensamento fabri- cando imagens insondáveis. Fui passear na Urca, e a cabeça cheia de células, neurônios e estrelas. Meus hábitos metafísicos decadentes empurravam-me para uma altitude maior. Não me perguntem a origem do empurrão. Talvez estivesse emergindo de volta, às coisas da superfície. E ia passeando na beleza do Pão de Açúcar. Do outro lado. Pelo caminho do Bem-te-vi. Era isso! Eu buscava o outro lado. Ouso confessar que Deus e a altitude coincidiam em minha angústia avassaladora? A resposta mais candente emergia do Cântico cósmico, de Ernesto Cardenal: Pregunto por el mundo mas allá de los cuantos. Ao plano vivo e denso, ao intermundo, que pode estar além dos quarks e revelar uma sombra, um grito, um rosto. v. 13 (suplemento), p. 257-67, outubro 2006 Recebido para publicação em janeiro de 2006. Aprovado para publicação em abril de 2006. O Outro lado da noite? A natureza, em seu amor ardente, no círculo da própria negação, em ouro, pedra e sal ambivalente, trabalha na perene transição. Dissolve e coagula eternamente a vida, que renasce, em floração, da morte, como a lua refulgente, surgindo na profunda escuridão. Na síntese do velho Magofonte, a vívida matéria se desfaz em águas claras, na secreta fonte: até que inesperada se refaz, envolta, como a Uroburos insonte, num círculo sutil que não se esfaz. [“Sphera”, de Marco Lucchesi, p. 95] A natureza, em seu amor ardente, no círculo da própria negação, em ouro, pedra e sal ambivalente, trabalha na perene transição. Dissolve e coagula eternamente a vida, que renasce, em floração, da morte, como a lua refulgente, surgindo na profunda escuridão. Na síntese do velho Magofonte, a vívida matéria se desfaz em águas claras, na secreta fonte: até que inesperada se refaz, envolta, como a Uroburos insonte, num círculo sutil que não se esfaz. A natureza, em seu amor ardente, no círculo da própria negação, em ouro, pedra e sal ambivalente, trabalha na perene transição. Dissolve e coagula eternamente a vida, que renasce, em floração, da morte, como a lua refulgente, surgindo na profunda escuridão. Na síntese do velho Magofonte, a vívida matéria se desfaz em águas claras, na secreta fonte: até que inesperada se refaz, envolta, como a Uroburos insonte, num círculo sutil que não se esfaz. Recebido para publicação em janeiro de 2006. Aprovado para publicação em abril de 2006. 267 v. 13 (suplemento), p. 257-67, outubro 2006 Detalhe de ‘‘Fetiches’’, padronagem em tecido criado por Fayga Ostrower Detalhe de ‘‘Fetiches’’, padronagem em tecido criado por Fayga Ostrower Detalhe de ‘‘Fetiches’’, padronagem em tecido criado por Fayga Ostrower História, Ciências, Saúde – Mang História, Ciências, Saúde – Mangu

https://openalex.org/W2809480381

https://zenodo.org/records/1293760/files/113.pdf

English

IDENTIFICATION OF GROUND WATER POTENTIAL ZONE USING REMOTE SENSING AND GIS TECHNIQUES.

Zenodo (CERN European Organization for Nuclear Research)

cc-by

IDENTIFICATION OF GROUND WATER POTENTIAL ZONE USING REMOTE SENSING AND GIS TECHNIQUES. Abstract ……………………………………………………………… The ground water is one of the important resources of natural water; ground water recharge potential zone to protect the water quality and to manage the ground water system. Using Geographic Information System (GIS) technology parameters like slope, geology, geomorphology, drainage density, land and land cover are obtained using top sheets. Vertical Electrical Soundings (VES) were conducted at 15 different locations at Channapatna Taluk. The fracture zones were available at 50, 90 and 100 feet depth in VES-7 at Kudluru. The depth at 50, 60 and 70 feet, fracture zones were also available and hence this area is considered as excellent zone in VES-10 at Sulleri. And below 200 feet depth, fracture zones were not observed which indicates the failure for the bore well in VES-15 at Sankalagere village. Abstract ……………………………………………………………… The ground water is one of the important resources of natural water; ground water recharge potential zone to protect the water quality and to manage the ground water system. Using Geographic Information System (GIS) technology parameters like slope, geology, geomorphology, drainage density, land and land cover are obtained using top sheets. Vertical Electrical Soundings (VES) were conducted at 15 different locations at Channapatna Taluk. The fracture zones were available at 50, 90 and 100 feet depth in VES-7 at Kudluru. The depth at 50, 60 and 70 feet, fracture zones were also available and hence this area is considered as excellent zone in VES-10 at Sulleri. And below 200 feet depth, fracture zones were not observed which indicates the failure for the bore well in VES-15 at Sankalagere village. Introduction:- The resistivity method is aimed at measuring the potential differences on the surface due to the current flow within the ground. Since the mechanisms that control the fluid flow and electric current and conduction are generally governed by the same physical parameters and litho logical attributes, the hydraulic and electrical conductivities are dependent on each other (George et al., 2015). Ground water is an important resource contributing significantly in total annual supply. Assessing the potential zone of ground water recharge, a ground water developing program requires a large volume of data from various sources. Hence, identification and quantization of these features are important for generating a ground water potential model of a study area. The different hydro geological themes could be used to identify the ground water potential zone of the present area. Journal Homepage: -www.journalijar.com Article DOI:10.21474/IJAR01/7116 DOI URL: http://dx.doi.org/10.21474/IJAR01/7116 DOI URL: http://dx.doi.org/10.21474/IJAR01/7116 RESEARCH ARTICLE IDENTIFICATION OF GROUND WATER POTENTIAL ZONE USING REMOTE SENSING AND GIS TECHNIQUES. Dr. Jagadeesha. M. Kattimani1, Ananda. B. Murthy2, Ranjan. M. S3, Srujan. D. Kumar4 and Vaibhavi. K. R5. 1. Assistant Professor, Department of Civil Engineering, JSS Academy of Technical Education, Bangalore, Karnataka. 2. Under Graduate Students, Department of Civil Engineering, JSS Academy of Technical Education, Bangalore, Karnataka. …………………………………………………………………………………………………….... Manuscript Info Abstract ……………………. ……………………………………………………………… Manuscript History Received: 16 March 2018 Final Accepted: 18 April 2018 Published: May 2018 Keywords:- Vertical Electrical Soundings (VES), Ground water potential zones, Resistivity, Fracture zones. The ground water is one of the important resources of natural water; ground water recharge potential zone to protect the water quality and to manage the ground water system. Using Geographic Information System (GIS) technology parameters like slope, geology, geomorphology, drainage density, land and land cover are obtained using top sheets. Vertical Electrical Soundings (VES) were conducted at 15 different locations at Channapatna Taluk. The fracture zones were available at 50, 90 and 100 feet depth in VES-7 at Kudluru. The depth at 50, 60 and 70 feet, fracture zones were also available and hence this area is considered as excellent zone in VES-10 at Sulleri. And below 200 feet depth, fracture zones were not observed which indicates the failure for the bore well in VES-15 at Sankalagere village. Copy Right, IJAR, 2018,. All rights reserved. …………………………………………………………………………………………………….... Introduction:- The resistivity method is aimed at measuring the potential differences on the surface due to the current flow within the ground. Since the mechanisms that control the fluid flow and electric current and conduction are generally governed by the same physical parameters and litho logical attributes, the hydraulic and electrical conductivities are dependent on each other (George et al., 2015). Ground water is an important resource contributing significantly in total annual supply. Assessing the potential zone of ground water recharge, a ground water developing program requires a large volume of data from various sources. Hence, identification and quantization of these features are important for generating a ground water potential model of a study area. The different hydro geological themes could be used to identify the ground water potential zone of the present area. ISSN: 2320-5407 ISSN: 2320-5407 Int. J. Adv. Res. 6(5), 948-953 Journal Homepage: -www.journalijar.com Article DOI:10.21474/IJAR01/7116 DOI URL: http://dx.doi.org/10.21474/IJAR01/7116 Location of the study area:- The present study area is located in Channapatna, District Ramanagara. Channapatna is located at 12.65°N and 77.22°E. It has an average elevation of 739 meters (2424 ft.). The total selected area of the study is 54 sq.km. There are 3 sub-watersheds (i.e. Harokoppa, Malurpatna and Santemagenahalli). 948 Corresponding Author:-Jagadeesha. M. Kattimani. Address:-Assistant Professor, Department of Civil Engineering, JSS Academy of Technical Education, Bangalore, Karnataka. 948 Corresponding Author:-Jagadeesha. M. Kattimani. Address:-Assistant Professor, Department of Civil Engineering, JSS Academy of Technical Education, Bangalore, Karnataka. 948 Int. J. Adv. Res. 6(5), 948-953 ISSN: 2320-5407 Fig.1:-Location map of the study area Materials and Methods:- The most widely used geophysical method for ground water exploration is electrical resistivity method as it is efficient in detecting water-bearing layers, being simple and inexpensive to carry out field investigations (Zohdy, 1974). Vertical Electrical Sounding (VES) is a geophysical method for investigation of a geological medium. The Vertical Electrical Sounding (VES) technique adopting a schlumberger array was used. The half current electrode spacing (AB/2) was varied from 1 meter to a maximum of 300 meters. There were totally 15 VES were conducted in the study area. Fig.1:-Location map of the study area Fig.1:-Location map of the study area Fig.1:-Location map of the study area Materials and Methods:- Materials and Methods:- The most widely used geophysical method for ground water exploration is electrical resistivity method as it is efficient in detecting water-bearing layers, being simple and inexpensive to carry out field investigations (Zohdy, 1974). Vertical Electrical Sounding (VES) is a geophysical method for investigation of a geological medium. The Vertical Electrical Sounding (VES) technique adopting a schlumberger array was used. The half current electrode spacing (AB/2) was varied from 1 meter to a maximum of 300 meters. There were totally 15 VES were conducted in the study area. 949 Int. J. Adv. Res. 6(5), 948-953 ISSN: 2320-5407 Fig.2:-Drainage map of the study area Fig.3:-Slope map of the study area Fig 4:-Geomorphology map of the study area Fig.5:-Land used map of the study area Fig.3:-Slope map of the study area Fig.2:-Drainage map of the study area F Fig.3:-Slope map of the study area Fig.2:-Drainage map of the study area Fig Fi F Fig 4:-Geomorphology map of the study are Fig.5:-Land used map of the study area Fig 4:-Geomorphology map of the study area Fig.5:-Land used map of the study area 950 Int. J. Adv. Res. Location of the study area:- 6(5), 948-953 ISSN: 2320-5407 Fig.6:-VES Location Map and Resistivity Data at 150 meters depth Fig: 7:-Ground water potential zone of study area Fig.6:-VES Location Map and Resistivity Data at 150 meters depth Fig.6:-VES Location Map and Resistivity Data at 150 meters depth Fig.6:-VES Location Map and Resistivity Data at 150 meters depth Fig: 7:-Ground water potential zone of study area Fig: 7:-Ground water potential zone of study area Fig: 7:-Ground water potential zone of study area 951 Int. J. Adv. Res. 6(5), 948-953 ISSN: 2320-5407 Table 1.1:-Ground water potential zone of study area S.No Thematic layers Map weight (wt) Individual features Feature score (wi) Total 1 Geomorphology 25 Valley fill 80 2000 Pediment valley 70 1750 Pediment inselberge 60 1500 Moderately weathered 50 1250 Shallow weathered 40 1000 Residual hill 30 750 Water body mask 0 0 2 Soil 20 Sandy skeletal 100 2000 Fine loamy 90 1800 Fine 80 1600 Clayey soil 70 1400 Water body mask 0 0 3 LULC 15 Agricultural land 70 1050 Forest 60 900 Built up land 40 600 Wasteland 30 450 Water body mask 0 0 4 Slope 15 Moderate slope 90 1350 Nearly level 80 1200 Gentle slope 70 1050 Moderately slope 60 900 Very gentle slope 50 750 Strong slope 40 600 5 Litho logy 10 Granodiorite 60 600 6 Ground water 10 Very good to good 50 500 Moderate 40 400 Moderate to poor 30 300 Poor 20 200 Poor to nil 10 100 Water body mask 0 0 7 Forest 5 Forest land 40 200 S.No Ranges Class Area (Sq.kms) Percentage 1 1800-3550 Poor 3.031 5.55 2 3550-4900 Good 19.132 35.01 3 4900-5350 Very good 11.297 20.67 4 5350-6450 Excellent 21.186 38.77 Total 54.646 100 Acknowledgement:- g Authors are thankful to Malurpatna Formers and JSS Academic Technical Education (JSSATE) Bengalore management for providing the apportunity to do the research activity in the study area. Results:- Drainages are dominantly spread in the western part of the study area around the Malurpatna shows less drainage pattern. 0 to 1% to 3% of slope observed in the western part of the study area, valleys are dominantly observed in the eastern part of the study area. Moderately weathered and shallow weathered pediplain and agricultural lands are also observed in the study area. Apparently resistivity at 150 meters (fig 6) shows those areas such as Sulleri, Akkuru, Malagala, Kudluru and surrounding areas possess low resistivity ranging between 0.00-172.80Ω⋅m. The yellow zone consists of Chakluru and Sankalagere where the resistivity ranges between 172.80-370.18Ω⋅m. Tubinkere, A V Halli and Chakkere comes under Red Zone which is the high resistivity ranging between 370.18- 673.07Ω⋅m (fig 6). Using Arc GIS 10.3 through integration method identified ground water potential zone in the study area classified poor to excellent in very good to good 55%, excellent 39% and poor is 5% (Table 1.1). Drainages are dominantly spread in the western part of the study area around the Malurpatna shows less drainage pattern. 0 to 1% to 3% of slope observed in the western part of the study area, valleys are dominantly observed in the eastern part of the study area. Moderately weathered and shallow weathered pediplain and agricultural lands are also observed in the study area. Apparently resistivity at 150 meters (fig 6) shows those areas such as Sulleri, Akkuru, Malagala, Kudluru and surrounding areas possess low resistivity ranging between 0.00-172.80Ω⋅m. The yellow zone consists of Chakluru and Sankalagere where the resistivity ranges between 172.80-370.18Ω⋅m. Tubinkere, A V Halli and Chakkere comes under Red Zone which is the high resistivity ranging between 370.18- 673.07Ω⋅m (fig 6). Using Arc GIS 10.3 through integration method identified ground water potential zone in the study area classified poor to excellent in very good to good 55%, excellent 39% and poor is 5% (Table 1.1). 952 Int. J. Adv. Res. 6(5), 948-953 ISSN: 2320-5407 4. Selvarani, A. et al., (2014): "Evaluation of ground water potential zones using GIS and remote sensing in Noyyal Basin, Tamil Nadu, India." International Journal of Environmental Technology and Management 17.5 (2014): 377-392.. 3. Lokesha, N. et al., (2005): "Delineation of ground water potential zones in a hard rock terrain of Mysore district, Karnataka using IRS data and GIS techniques." Journal of the Indian Society of Remote Sensing 33.3 (2005): 405. References:- 1. Chandra et al., (2006): "Integrated studies for characterization of lineaments used to locate groundwater potential zones in a hard rock region of Karnataka, India." Hydrogeology Journal 14.6 (2006): 1042-1051. 1. Chandra et al., (2006): "Integrated studies for characterization of lineaments used to locate groundwater potential zones in a hard rock region of Karnataka, India." Hydrogeology Journal 14.6 (2006): 1042-1051. 2. Jagadeesha, M.K & T,J.Renuka Prasad. (2016): “Vertical Electrical Soundings (VES) Geophysical Exploration analysis of Part of South-East Agro climatic Zone of Karnataka, Kolar District, Karnataka, India Using Geographical Information System (GIS) Techniques.” International Journal of Advanced Research (2016), Volume 4, Issue 7, 2019-2023. 3. Lokesha, N. et al., (2005): "Delineation of ground water potential zones in a hard rock terrain of Mysore district, Karnataka using IRS data and GIS techniques." Journal of the Indian Society of Remote Sensing 33.3 (2005): 405. 3. Lokesha, N. et al., (2005): "Delineation of ground water potential zones in a hard rock terrain of Mysore district, Karnataka using IRS data and GIS techniques." Journal of the Indian Society of Remote Sensing 33.3 (2005): 405. 4. Selvarani, A. et al., (2014): "Evaluation of ground water potential zones using GIS and remote sensing in Noyyal Basin, Tamil Nadu, India." International Journal of Environmental Technology and Management 17.5 (2014): 377-392.. 4. Selvarani, A. et al., (2014): "Evaluation of ground water potential zones using GIS and remote sensing in Noyyal Basin, Tamil Nadu, India." International Journal of Environmental Technology and Management 17.5 (2014): 377-392.. 953

https://openalex.org/W2138438730

https://dspace.mit.edu/bitstream/1721.1/92475/1/Doyle-2014-Chromatin%20Loops%20as%20A.pdf

English

Chromatin Loops as Allosteric Modulators of Enhancer-Promoter Interactions

PLOS computational biology/PLoS computational biology

cc-by

MIT Open Access Articles The MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters. Terms of use: Creative Commons Attribution Abstract The classic model of eukaryotic gene expression requires direct spatial contact between a distal enhancer and a proximal promoter. Recent Chromosome Conformation Capture (3C) studies show that enhancers and promoters are embedded in a complex network of looping interactions. Here we use a polymer model of chromatin fiber to investigate whether, and to what extent, looping interactions between elements in the vicinity of an enhancer-promoter pair can influence their contact frequency. Our equilibrium polymer simulations show that a chromatin loop, formed by elements flanking either an enhancer or a promoter, suppresses enhancer-promoter interactions, working as an insulator. A loop formed by elements located in the region between an enhancer and a promoter, on the contrary, facilitates their interactions. We find that different mechanisms underlie insulation and facilitation; insulation occurs due to steric exclusion by the loop, and is a global effect, while facilitation occurs due to an effective shortening of the enhancer-promoter genomic distance, and is a local effect. Consistently, we find that these effects manifest quite differently for in silico 3C and microscopy. Our results show that looping interactions that do not directly involve an enhancer-promoter pair can nevertheless significantly modulate their interactions. This phenomenon is analogous to allosteric regulation in proteins, where a conformational change triggered by binding of a regulatory molecule to one site affects the state of another site. Citation: Doyle B, Fudenberg G, Imakaev M, Mirny LA (2014) Chromatin Loops as Allosteric Modulators of Enhancer-Promoter Interactions. PLoS Comput Biol 10(10): e1003867. doi:10.1371/journal.pcbi.1003867 Editor: Marc A. Marti-Renom, Centre Nacional d’Ana`lisi Geno`mica and Centre de Regulacio´ Geno`mica, Spain Editor: Marc A. Marti-Renom, Centre Nacional d’Ana`lisi Geno`mica and Centre de Regulacio´ Geno`mica, Spai Received April 22, 2014; Accepted August 20, 2014; Published October 23, 2014 Received April 22, 2014; Accepted August 20, 2014; Published October 23, 2014 Copyright:  2014 Doyle et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright:  2014 Doyle et al. This is an open-access article distributed under th unrestricted use, distribution, and reproduction in any medium, provided the original Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. Polymer models were simulated with OpenMM, a high-performance GPU-assisted molecular dynamics software available at https://simtk.org/home/openmm. Abstract We used an in-house openmm-polymer library to efficiently set up polymer simulations with OpenMM, and to analyze simulation results. openmm-polymer is publicly available on the Bitbucket online repository: http://bitbucket.org/mirnylab/openmm-polymer. Funding: The research of BD has been supported by NSF-funded PRIMES high school outreach program at MIT and MIT UROP programs. The work was supported by the National Cancer Institute-funded Center for Physical Sciences in Oncology at MIT U54-CA143874-04. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exi * E il l id@ i d Competing Interests: The authors have declared that no competing interests exist. * Email: [email protected] Competing Interests: The authors have declared that no competing interests exist. * Email: [email protected] * Email: [email protected] specific looping interactions are formed [6,11,12]. These observa- tions raise an important question; namely, how can E-P contacts be affected by looping interactions between other regulatory elements in their genomic neighborhood? Boryana Doyle1,2,3, Geoffrey Fudenberg4, Maxim Imakaev3, Leonid A. Mirny3,4,5* 1 Program for Research in Mathematics, Engineering and Science for High School Students, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America, 2 Undergraduate Research Opportunities Program, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America, 3 Department of Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America, 4 Graduate Program in Biophysics, Harvard University, Cambridge, Massachusetts, United States of America, 5 Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America PLOS Computational Biology | www.ploscompbiol.org October 2014 | Volume 10 | Issue 10 | e1003867 Model and analysis of simulations Using equilibrium simulations of a confined polymer chain, we study how chromatin loops affect E-P contact frequency in their vicinity. We model chromatin as a semi-flexible polymer fiber with excluded volume; the fiber consists of 15 nm diameter monomers, each representing three nucleosomes or 500 bp, with a persistence length of 3 monomers (Figure 1G, Methods) [32]. Unless otherwise noted, we allow occasional chromatin fiber crossing by setting a finite energy cost (using a truncated repulsive potential) for two monomers to occupy the same volume, which accounts for topoisomerase II (topo-II) activity (see Methods). Thus, two regions of the chain can spontaneously cross through each other with a probability controlled by the energy penalty of co- occupancy. To account for the dense arrangement of chromatin within the nucleus, we confine the chromatin fiber to impose a 2% volume density. We later vary volume density from 1% to 20% (see below), which is consistent with current estimates of chromatin volume density in the interphase nuclei of higher eukaryotes [33]. Since the flexibility of the chromatin fiber in vivo is incompletely characterized, we varied flexibility in our simulations and found quantitatively similar results (see below). It remains unclear whether, and to what extent, the looping interactions between other regulatory elements can mediate E-P contacts. When these looping interactions do not directly involve the E-P pair, their effect is reminiscent of allosteric regulation in proteins [26,27], where binding of a regulator molecule to one site changes the state of another site or the whole protein. A classic example is the binding of allolactose to the lac repressor at the regulatory domain. While allosteric interactions in proteins are mediated by the protein structure, we propose that interactions between genomic sites could be mediated by local changes in the conformational ensemble of the chromatin fiber. Polymer simulations provide an ideal testing ground to investigate the allosteric effects of a loop on E-P contacts; many loci can be probed simultaneously at high resolution, and more complicated looping arrangements can be systematically charac- terized. Previously, Mukhopadhyay et al. [28] used polymer simulations to demonstrate that the topological model of insulation applies to an unconfined system of two fused chromatin rings; namely, two loci within the same ring interact more frequently than loci in different rings. We extend this line of inquiry by asking whether forming loops may affect interactions at scales exceeding the loop size, e.g. Author Summary The regulatory effects of loops may also be relevant at larger genomic distances; in mice, a regulatory element with multiple larger (25 kb and 55 kb) loops was suggested to control multiple E-P contacts at the H19 locus [22]. Analogously, loops between insulating elements were suggested to modulate the activity of silencing elements [23]. and orientation. The authors explain these observations by gypsy- gypsy looping interactions. We note that while this gypsy element consisted of twelve repeated copies of the Su(Hw) binding site, elements with fewer sites are sufficient for insulation [24,25]. The regulatory effects of loops may also be relevant at larger genomic distances; in mice, a regulatory element with multiple larger (25 kb and 55 kb) loops was suggested to control multiple E-P contacts at the H19 locus [22]. Analogously, loops between insulating elements were suggested to modulate the activity of silencing elements [23]. Chromatin Loops as Modulators of Enhancer-Promoter Interactions interactions between specific elements, as described below. Synthesizing results from the literature, we primarily focused our simulation analysis on two important arrangements of the loop- forming elements relative to an E-P pair: (1) an enhancer is flanked by loop-forming elements, while a promoter is beyond the loop (Figure 1E); and (2) both loop-forming elements are located in the genomic region between an enhancer and a promoter (Figure 1F). Author Summary Author Summary In eukaryotes, enhancers directly contact promoters over large genomic distances to regulate gene expression. Characterizing the principles underlying these long-range enhancer-promoter contacts is crucial for a full under- standing of gene expression. Recent experimental map- ping of chromosomal interactions by the Hi-C method shows an intricate network of local looping interactions surrounding enhancers and promoters. We model a region of chromatin fiber as a long polymer and study how the formation of loops between certain regulatory elements can insulate or facilitate enhancer-promoter interactions. We find 2–5 fold insulation or facilitation, depending on the location of looping elements relative to an enhancer- promoter pair. These effects originate from the polymer nature of chromatin, without requiring additional mecha- nisms beyond the formation of a chromatin loop. Our findings suggest that loop-mediated gene regulation by elements in the vicinity of an enhancer-promoter pair can be understood as an allosteric effect. This highlights the complex effects that local chromatin organization can have on gene regulation. We find that loops can significantly insulate or facilitate the frequency of E-P interactions, depending on the loop location relative to the E-P pair. We consider a variety of situations and parameters, including: E-P genomic distance, stiffness of the chromatin fiber, size of the loop, topological constraints on the chromatin fiber (i.e. topoisomerase II activity), chromatin density, the number of looping elements, and excluded volume interac- tions. We find that different mechanisms underlie insulation and facilitation; insulation occurs due to steric exclusion by the loop, while facilitation occurs due to an effective shortening of the E-P genomic distance. We additionally consider how insulation and facilitation would be observed in microscopy studies and find substantial differences from how they would manifest in 3C-based studies. Taken together, our results suggest that due to its polymer nature, chromatin allows for interactions to be mediated in an allosteric manner, i.e. formation of a contact between two sites can insulate or facilitate interactions between other loci in the vicinity. and orientation. The authors explain these observations by gypsy- gypsy looping interactions. We note that while this gypsy element consisted of twelve repeated copies of the Su(Hw) binding site, elements with fewer sites are sufficient for insulation [24,25]. Introduction Distal enhancer elements in higher eukaryotes are essential for regulating gene expression [1–4]. In conjunction with transcrip- tion factor binding and nucleosome modifications, the classic model of enhancer function requires the direct spatial contact between enhancers and their target promoters (Figure 1A) [1–4]. Recent studies have started to reveal the complexity of the enhancer-promoter (E-P) interaction network, where each en- hancer can influence multiple promoters, and each promoter may be influenced by multiple enhancers [5–8]. In addition, gene expression and E-P interactions occur within higher-order three- dimensional chromatin organization, which is characterized by an intricate network of interactions at multiple scales. For example, below 1 Mb, chromatin is organized into continuous 500–900 kb regions of enriched contact frequency called topologically associ- ated domains (TADs) [9,10]. TADs were found to be about 90% cell-type independent (2763/3000 conserved boundaries between two assayed cell types [9]). Within TADs, additional cell-type Two models for how proximal looping interactions can modulate E-P contacts have been proposed: the decoy model and the topological model (experiments [13–15], reviewed in [16– 19]). The decoy model suggests that insulating elements directly interact with the enhancer, sequestering it from the promoter, and thereby directly hinder E-P interactions. The topological model proposes that two regulatory elements in the vicinity of the enhancer and promoter can interact with each other to form a chromatin loop; this, in turn, affects E-P contacts. Evidence supporting the topological model includes experi- ments in multiple organisms (Figure 1B–D) [20–23]. For example, Kyrchanova et al. [21] recently observed that a single Drosophila gypsy element placed between an enhancer and a promoter did not change their interactions; however, introducing two gypsy elements changed E-P interactions depending on gypsy position 1 October 2014 | Volume 10 | Issue 10 | e1003867 October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org Chromatin Loops as Modulators of Enhancer-Promoter Interactions PLOS Computational Biology | www.ploscompbiol.org Model and analysis of simulations interactions of a loop with the rest of the chromosome or between loci in the vicinity of the loop. For each set of conditions and looping interactions, we performed Langevin dynamics simulations using OpenMM [34] (see Methods and Video S1) and sampled conformations from the resulting equilibrium ensemble; these conformations were subse- quently analyzed to compute contact frequencies (see below, and Methods). To investigate the effects of a chromatin loop on a larger region of chromatin, we model a loop by forming an irreversible bond between a pair of monomers and allowing the whole polymer to equilibrate (Figure 1E, 1F, see Methods). We considered loops of sizes L = 15 kb, 30 kb, and 60 kb, and a 2.5 kb loop with a more flexible chromatin fiber (see Discussion), in a proportionally sized genomic region of length 33*L, i.e. 1 Mb for a 30 kb loop (Figure S1). Our polymer model contains no additional sequence-specific details, and thus generally addresses how E-P interactions are altered in the vicinity of a loop. The model remains agnostic to the chromatin organization at larger genomic Here we use polymer models to study how 15–60 kb chromatin loops can influence E-P contacts. We note that 3C-based studies have only begun to provide unbiased data at sufficient resolution to build polymer models of a particular locus [29], and the fine structure of the chromatin fiber in vivo remains largely unknown [30,31]. Thus, for generality, we model chromatin as a long homogeneous flexible fiber with only a few additional looping October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org 2 Chromatin Loops as Modulators of Enhancer-Promoter Interactions Figure 1. Enhancer-promoter pairs in the context of other interactions. Experimental Studies, (A) Illustration of an enhancer (in yellow) spatially interacting with a promoter (blue) along a chromatin fiber. This coloring convention continues throughout the paper. (B) A recent study in Drosophila suggested a 7 kb chromatin loop formed between Su(Hw) insulators (orange) could decrease E-P interactions (red ‘‘X’’) [20]. (C) Conversely, a 3 kb chromatin loop in the region between enhancer and promoter was proposed to increase E-P interactions. (D) Five arrangements for proposed looping interactions from three studies, left to right, [21], [22], and [23]. Chromatin loops can insulate or facilitate enhancer- promoter interactions For each set of parameter values and loops, we generate an equilibrium ensemble of conformations and compute the contact frequency between loci (monomers) in this ensemble (Figure 2A, Table S1 for parameter values). We display pairwise contact frequencies using heatmaps (Figure 2B), as typical for Hi-C and 5C experiments. Our simulated heatmaps are characterized by two features: (i) a decay of contact frequency as a function of increasing genomic distance, and (ii) an off-diagonal interaction between the loop bases. The first feature follows from the polymer connectivity of the simulated chromatin fiber. The second feature alters the typical decline in the contact frequency and is of primary interest in this study. We used the simulated heatmaps of pairwise contact frequency to investigate the two important arrangements of the E-P pair and the loop from the literature (Figure 1). The first arrangement involves a chromatin loop formed by elements flanking an enhancer, such that the enhancer is located within the chromatin loop and the promoter is located outside of the loop (Figure 2C). Since the enhancer and promoter are equivalent in our polymer model, this scenario also describes a promoter flanked by a pair of loop-forming elements and an enhancer located outside of the loop. Simulations show that for 50 kb E-P spacing, formation of such a 30 kb loop leads to a ,35% reduction in E-P contacts, serving as an insulator (contact frequency ratio of 0.64, Figure 2D). Below we refer to this arrangement as insulation. For a given position of the enhancer and the promoter, we can compute the contact frequency ratio as the contact frequency in the model with a loop, divided by the contact frequency for an otherwise equivalent model without a loop. Contact frequency ratios below 1 indicate insulation, whereas ratios above 1 correspond to facilitation. Unless noted otherwise, we report contact frequency ratios for a 30 kb loop and a 50 kb E-P genomic distance. We note that each simulation contains information regarding every possible position of the enhancer and the promoter. From this, we can compute contact frequency ratios as a function of E-P distance and location. Below we examine how the loop length and the E-P spacing affect observed phenomena. The second arrangement constitutes a chromatin loop located in the genomic region between the enhancer and promoter, i.e. both loop-forming elements are located between the enhancer and promoter (Figure 2C). Model and analysis of simulations (left) a single Drosophila gypsy element between an enhancer and a promoter did not change their interactions (top), however an additional gypsy element upstream of the enhancer decreased E-P interactions (bottom) [21]. (center) at the mouse H19 locus, a regulatory element with multiple larger loops (55 kb and 25 kb) was suggested to control multiple E- P contacts; the enhancer can regulate the promoter before the loop, but cannot regulate the promoter within the loop [22]. (right) chromatin loops may also modulate spatial interactions between silencing elements (e.g. PRE, black triangles) and their target promoters [23]. The promoter within the loop is not silenced (top), whereas the promoter beyond the loop is silenced (bottom). Polymer Simulations, (E) Arrangement 1: polymer conformation where an enhancer is within a chromatin loop and a promoter is beyond the loop. (F) Arrangement 2: polymer conformation where an enhancer is before the loop and a promoter is after the loop. (G) (left) zoom-in on our polymer model of chromatin. The three large circles represent one monomer each; each monomer consists of three nucleosomes (small circles) or 500 bp. (right) full view of a sample polymer conformation showing a 30 kb chromatin loop (black) with highlighted loop-bases (orange) within a 1 Mb region. doi:10.1371/journal.pcbi.1003867.g001 Figure 1. Enhancer-promoter pairs in the context of other interactions. Experimental Studies, (A) Illustration of an enhancer (in yellow) spatially interacting with a promoter (blue) along a chromatin fiber. This coloring convention continues throughout the paper. (B) A recent study in Drosophila suggested a 7 kb chromatin loop formed between Su(Hw) insulators (orange) could decrease E-P interactions (red ‘‘X’’) [20]. (C) Conversely, a 3 kb chromatin loop in the region between enhancer and promoter was proposed to increase E-P interactions. (D) Five arrangements for proposed looping interactions from three studies, left to right, [21], [22], and [23]. (left) a single Drosophila gypsy element between an enhancer and a promoter did not change their interactions (top), however an additional gypsy element upstream of the enhancer decreased E-P interactions (bottom) [21]. (center) at the mouse H19 locus, a regulatory element with multiple larger loops (55 kb and 25 kb) was suggested to control multiple E- P contacts; the enhancer can regulate the promoter before the loop, but cannot regulate the promoter within the loop [22]. (right) chromatin loops may also modulate spatial interactions between silencing elements (e.g. Model and analysis of simulations PRE, black triangles) and their target promoters [23]. The promoter within the loop is not silenced (top), whereas the promoter beyond the loop is silenced (bottom). Polymer Simulations, (E) Arrangement 1: polymer conformation where an enhancer is within a chromatin loop and a promoter is beyond the loop. (F) Arrangement 2: polymer conformation where an enhancer is before the loop and a promoter is after the loop. (G) (left) zoom-in on our polymer model of chromatin. The three large circles represent one monomer each; each monomer consists of three nucleosomes (small circles) or 500 bp. (right) full view of a sample polymer conformation showing a 30 kb chromatin loop (black) with highlighted loop-bases (orange) within a 1 Mb region. doi:10.1371/journal.pcbi.1003867.g001 scales, assuming that the simulated region is contained within a single TAD [35]. October 2014 | Volume 10 | Issue 10 | e1003867 Chromatin loops can insulate or facilitate enhancer- promoter interactions (D) Contact frequency ratios (Methods) for insulation and facilitation arrangements with a 30 kb loop and 50 kb E-P genomic distance. Here and below, error bars indicate one standard deviation about the mean. doi:10.1371/journal.pcbi.1003867.g002 (Figure 3A). These results reveal an important difference between loop-induced facilitation and insulation: facilitation is a local phenomenon, and insulation is a global effect. First, we investigated the importance of chromatin fiber flexibility by simulating chromatin fibers with different persistence lengths. We found that fiber flexibility does not significantly affect insulation or facilitation (Figure S3). This is consistent with the fact that both phenomena are observed at distances much larger than the persistence length, and thus in our simulations do not emerge solely due to fiber stiffness. As such, cartoons with rigid, stiff loops should in many cases be understood as schematics [36]; renderings of three-dimensional chromatin loops from our models are shown in Figure 1E, 1F, and 2A. We also note that for simulations with larger and smaller loop sizes, the main qualitative features of the heatmap remain the same (Figure S1). To better understand insulation, we varied the position of the enhancer within the loop. We found that insulation is weaker when the enhancer is placed in the middle of the loop (0.75 contact frequency ratio), and strengthens as the enhancer approaches the base of the loop (0.49 contact frequency ratio, Figure 3B). We note that an extreme case of topological-model insulation is in fact similar to decoy-model insulation, which occurs when the enhancer is placed at the base of the loop. In this scenario, we observe stronger insulation because the enhancer is permanently interacting with the other loop base, sterically hindering interactions between the enhancer and all other loci. This can be seen as dark stripes at the positions of the loop base monomers on the heatmap; the profile of interactions of the loop base with the rest of the fiber is detailed in Figure S2. Below we identify and discuss mechanisms underlying insulation and facilitation. Next, we studied the effect of topological constraints, as they have been suggested to play an important role in chromosome organization [33,37,38]. To investigate this, we performed simulations both with and without allowing two regions of the chromatin fiber to cross, which may respectively correspond to cells with active and inactive topo-II. We found that insulation and facilitation are observed irrespective of the topological constraints (Figure S4A). Chromatin loops can insulate or facilitate enhancer- promoter interactions Formation of such a loop facilitates E-P interactions by increasing their contact frequency by more than 4- fold (contact frequency ratio of 4.15, Figure 2D). Next we examined how E-P spacing affects the magnitude of loop-induced insulation or facilitation. Interestingly, the two effects behave differently; while facilitation diminishes with E-P genomic distance, insulation appears to be independent of distance October 2014 | Volume 10 | Issue 10 | e1003867 October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org 3 Chromatin Loops as Modulators of Enhancer-Promoter Interactions Figure 2. A chromatin loop alters the frequency of enhancer-promoter interactions. (A) Five sample conformations from polymer simulations with a 30 kb permanent loop (black) formed between two loop bases (orange) in a 1 Mb region of fiber. (B) Average heatmap (300 kb by 300 kb) for polymer simulations of the permanent, one-loop system, with a 30 kb loop (aggregated over 800,000 simulated conformations). Top and left edges show positions of the enhancer (yellow), promoter (blue), and loop bases (orange) for insulation and facilitation arrangements. (C) Schematics of E-P arrangements. (top) chromatin fiber without a fixed loop and with E-P genomic distance of 50 kb, as used to calculate expected (no-loop) contact frequencies (Methods). (middle) arrangement where insulation is observed, represented by the red ‘‘X’’. (bottom) arrangement where facilitation is observed. (D) Contact frequency ratios (Methods) for insulation and facilitation arrangements with a 30 kb loop and 50 kb E-P genomic distance. Here and below, error bars indicate one standard deviation about the mean. doi:10.1371/journal.pcbi.1003867.g002 Figure 2. A chromatin loop alters the frequency of enhancer-promoter interactions. (A) Five sample conformations from polymer simulations with a 30 kb permanent loop (black) formed between two loop bases (orange) in a 1 Mb region of fiber. (B) Average heatmap (300 kb by 300 kb) for polymer simulations of the permanent, one-loop system, with a 30 kb loop (aggregated over 800,000 simulated conformations). Top and left edges show positions of the enhancer (yellow), promoter (blue), and loop bases (orange) for insulation and facilitation arrangements. (C) Schematics of E-P arrangements. (top) chromatin fiber without a fixed loop and with E-P genomic distance of 50 kb, as used to calculate expected (no-loop) contact frequencies (Methods). (middle) arrangement where insulation is observed, represented by the red ‘‘X’’. (bottom) arrangement where facilitation is observed. Chromatin loops can insulate or facilitate enhancer- promoter interactions We note that the terms topological model, topologically-associated domains (TADs), and topological con- straints all refer to distinct, and likely unrelated, concepts. In particular, our results demonstrate that the topological model of insulation is independent from topological constraints on the Chromatin fiber flexibility, topological constraints, and overall density do not underlie insulation or facilitation Chromatin fiber flexibility, topological constraints, and overall density do not underlie insulation or facilitation To test the generality of insulation and facilitation, we varied several biologically relevant and physical characteristics of our model, many of which have not been fully characterized in vivo. y To test the generality of insulation and facilitation, we varied several biologically relevant and physical characteristics of our model, many of which have not been fully characterized in vivo. October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org 4 Chromatin Loops as Modulators of Enhancer-Promoter Interactions Figure 3. Insulation and facilitation strength depends on enhancer-promoter positions. (A) Insulation (left) and facilitation (right) as a function of E-P genomic distance. For insulation, enhancer position remains fixed. For facilitation, an E-P pair is positioned symmetrically around the loop at each genomic distance. (B) Insulation for different positions of the enhancer within the loop with a constant genomic distance of 50 kb. doi:10.1371/journal.pcbi.1003867.g003 Figure 3. Insulation and facilitation strength depends on enhancer-promoter positions. (A) Insulation (left) and facilitation (right) as a function of E-P genomic distance. For insulation, enhancer position remains fixed. For facilitation, an E-P pair is positioned symmetrically around the loop at each genomic distance. (B) Insulation for different positions of the enhancer within the loop with a constant genomic distance of 50 kb. doi:10.1371/journal.pcbi.1003867.g003 chromatin fiber. Additionally, topological constraints are distinct from other topological effects such as supercoiling of the chromatin fiber [39]; supercoiling can lead to significant insulation and facilitation [39], but may be more relevant for bacterial chromosome organization [40]. of the fiber is depleted near the loop, i.e. the loop sterically excludes interactions with the rest of the polymer (Figure 4A). Interestingly, regions immediately outside the loop are also sterically excluded by the loop; we find 20–50% insulation for regions up to 6 kb away from the loop (Figure S6A). Third, we assessed the influence of chromatin density on insulation and facilitation. In particular, active and inactive chromatin environments are known to have respectively lower and higher densities [33,41]. We performed simulations at densities ranging from low (1%) to high (20%) volume density (Figure S4B). We found that while both insulation and facilitation remain qualitatively present at all densities, they are roughly twice as strong at 1% vs. 20% density. Chromatin fiber flexibility, topological constraints, and overall density do not underlie insulation or facilitation This finding indicates that low density, as found in active chromatin, is important for the magnitude of both possible regulatory effects. Facilitation does not depend on excluded volume interactions, but depends on E-P distance. Therefore we considered how facilitation might arise from an effectively shortened E-P distance imposed by the intervening loop; in particular, we compared contact frequency ratios for the facilitation arrange- ment and for a simulation without a loop but at a 30 kb smaller genomic distance (i.e. shortened by the loop size). Indeed, we see that these are in almost complete agreement, demonstrating that facilitation results from the effectively shortened genomic distance (Figure 4B). To get further insight into the mechanisms of insulation and facilitation, we performed in silico fluorescence in situ hybrid- ization (FISH) by calculating the distribution of E-P spatial distances across many conformations (Figure 4C). To consis- tently compare insulation and facilitation, we considered them at an E-P distance of 90 kb, where both effects have approximately the same fold change (contact frequency ratios of 0.75 and 1.3). For insulation, we observe only a small shift in the overall distribution of E-P spatial distances (mean increased by 3%). This confirms that insulation occurs not because the E- P pair is much further away on average, but due to steric exclusion of the promoter by the loop engulfing the enhancer. For facilitation, however, the distribution of E-P spatial distances shifted more strongly (mean decreased by 9%). These results highlight that differences in contact frequency are not always proportionally reflected in differences in mean spatial distances. Moreover, our results show that both effects could be hard to detect by microscopy, but facilitation would be more evident than insulation. Together, these variations in our model suggest that insulation and facilitation exist across a range of biologically relevant parameter values. However, they indicate that insulation and facilitation do not mechanistically follow from the fiber stiffness, topological constraints, or overall density. PLOS Computational Biology | www.ploscompbiol.org October 2014 | Volume 10 | Issue 10 | e1003867 Fundamental properties of polymers underlie insulation and facilitation To understand the mechanisms of insulation and facilitation, we performed simulations of a phantom polymer chain, which lacks excluded volume interactions (Figure S5). Remarkably, elimina- tion of excluded volume interactions completely abolishes the insulation effect. In contrast, the degree of facilitation remains largely unaffected by the elimination of excluded volume (reduced from 4.15 to 3.20). We note that phantom chain simulations do not adequately describe chromosomes, but nevertheless can provide useful insights into polymer behavior; here, they demonstrate how steric exclusion by a loop can give rise to insulation. Together, these results provide evidence for the mechanisms underlying insulation and facilitation. For insulation, regions within the loop are sterically excluded from making contacts with the rest of the polymer fiber. For facilitation, the E-P pair has an effectively shorter genomic distance. The loss of insulation in simulations without excluded volume interactions led us to investigate the spatial relation- ship between the loop and the rest of the polymer fiber. We found that the spatial density of monomers from other regions October 2014 | Volume 10 | Issue 10 | e1003867 October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org PLOS Computational Biology | www.ploscompbiol.org 5 Chromatin Loops as Modulators of Enhancer-Promoter Interactions Figure 4. Mechanisms of insulation and facilitation. (A) (top) Illustration of the insulation mechanism: strong dynamic steric exclusion by a chromatin loop is shown by a superposition of loops in multiple conformations (grey, with enhancer and promoter) and their sterically excluded region (dashed lines), surrounded by other distal regions of chromatin (grey). (bottom) Density of distal monomers (i.e. outside the loop and.10 kb from the loop base) as a function of radial distance from the center of mass of the loop. The loop-free control exactly repeats this procedure for an equivalent region without a loop. Both are normalized using respective radial-position dependent spatial density (Methods). (B) (top) Illustration of facilitation mechanism: an E-P pair flanking a loop has an effectively shorter genomic distance; here an E-P pair with 50 kb separation and a 30 kb loop behaves similarly to an E-P pair separated by 20 kb in a region without a loop. (bottom) Comparison of contact frequency ratios for the above situations, as a function of E-P distance. (C) Simulated cumulative distribution of spatial distances (in silico FISH) for an E-P pair with a genomic distance of 90 kb. doi:10.1371/journal.pcbi.1003867.g004 Figure 4. Intra-loop interactions and two-loop models The analyses above focused on understanding how a single loop affects E-P contact frequency at genomic distances exceeding the loop size. For E-P genomic distances less than the loop size, both elements can be positioned within the loop. In this case, we found that interactions are facilitated, consistent with previous results [28]. However, the degree of facilitation depends on the relative position of the elements in the loop (Figure S6B). Placing the enhancer at one loop base and the promoter at the other can greatly facilitate their interaction frequency. On the contrary, with the enhancer at the loop base and the promoter in the enhancer- proximal portion of the loop, the facilitation effect may disappear, likely due to the superposition of intra-loop facilitation with the insulating properties of the loop bases (Figure S2B). Fundamental properties of polymers underlie insulation and facilitation Mechanisms of insulation and facilitation. (A) (top) Illustration of the insulation mechanism: strong dynamic steric exclusion by a chromatin loop is shown by a superposition of loops in multiple conformations (grey, with enhancer and promoter) and their sterically excluded region (dashed lines), surrounded by other distal regions of chromatin (grey). (bottom) Density of distal monomers (i.e. outside the loop and.10 kb from the loop base) as a function of radial distance from the center of mass of the loop. The loop-free control exactly repeats this procedure for an equivalent region without a loop. Both are normalized using respective radial-position dependent spatial density (Methods). (B) (top) Illustration of facilitation mechanism: an E-P pair flanking a loop has an effectively shorter genomic distance; here an E-P pair with 50 kb separation and a 30 kb loop behaves similarly to an E-P pair separated by 20 kb in a region without a loop. (bottom) Comparison of contact frequency ratios for the above situations, as a function of E-P distance. (C) Simulated cumulative distribution of spatial distances (in silico FISH) for an E-P pair with a genomic distance of 90 kb. doi:10 1371/journal pcbi 1003867 g004 doi:10.1371/journal.pcbi.1003867.g004 sterically excludes the other, as well as the rest of the chromatin fiber. PLOS Computational Biology | www.ploscompbiol.org October 2014 | Volume 10 | Issue 10 | e1003867 Polymer model The polymer mechanisms underlying insulation and facilitation arise due to two different effects. Facilitation results from the effectively shortened E-P genomic distance due to the loop and is likely robust to the molecular details of the chromatin fiber. Insulation arises due to excluded volume interactions and steric exclusion by the loop, and thus depends on the chromatin- chromatin affinity. In polymer physics terms, insulation would require good solvent conditions, which are likely satisfied in active decondensed chromatin. We further note that altering chromatin- chromatin affinity may allow further modulation of insulation strength, for example through chromatin modifications that change the net charge of nucleosome tails. Model overview. We modeled chromatin as a fiber of monomers connected by harmonic bonds. Unless noted, each spherical monomer had a diameter of 15 nm and represented 500 bp, or approximately three nucleosomes. A permanent loop was formed by connecting two monomers with a harmonic bond of the same strength as the bonds between all consecutive monomers. This permanently brings the loop bases into contact. Two such loops were formed in the two-loop simulations. A three- point interaction force was used to impose a bending energy and account for the rigidity of the fiber. To model volume interactions, monomers interacted via a shifted Lennard-Jones potential, which is a computationally efficient purely-repulsive potential. Unless noted otherwise, the Lennard-Jones potential was truncated at U = 3kT as specified below to allow occasional fiber crossing. Simulated polymers were confined to a sphere at a given density and initialized from an unentangled polymer conformation. In specific biological systems, the detailed structure and flexibility of the chromatin fiber may become relevant. When the chromatin loop size approaches several persistence lengths, the loop could become very rigid. Consequentially, its effects may depend on the molecular details of the loop-forming elements, including their orientation as observed in a recent study [21]. However, the insulation and facilitation we observe may still manifest with similar strength at smaller genomic distances for more flexible or loosely packed chromatin fiber; changing these parameters causes small loops to behave similarly to larger loops (Figure S1D). We note that many processes may locally increase chromatin flexibility, either uniformly or through the formation of kinks [38], including the loss/unwrapping of nucleosomes [48]. Chromatin Loops as Modulators of Enhancer-Promoter Interactions Chromatin Loops as Modulators of Enhancer-Promoter Interactions while they are present. Roughly speaking, the effect on insulation or facilitation for a given loop is proportional to its frequency of occurrence in a cell. interactions. Specific mechanisms include: sliding along DNA [43], lamina attachments [19,44], and inter-nucleosome interac- tions [17]. Our simulations show that the formation of the loop itself can insulate or facilitate E-P interactions, due to the polymer nature of chromatin, independent of specific molecular mecha- nisms. Given the complexity of the local looping network, it is likely that there are multiple dynamic loops in the vicinity of the enhancer and promoter. While we studied the permanent single and double loop systems, our results provide intuition even to these more complicated systems. For instance, the global nature of insulation implies it can hinder interactions between enhancers and any number of promoters. Conversely, facilitation is local and thus specific to the regions that directly flank the loop. Together, our results highlight the complex and non-local grammar of regulatory elements surrounding enhancers and promoters. In conjunction with emerging biological data, future simulations will provide additional insight into the consequences of chromatin’s polymer nature for allosteric modulation E-P interactions. These effects can be best understood in terms of allosteric regulation. In particular, interactions that are responsible for the formation of the loop do not necessarily directly prevent or form E-P contacts. Instead, they steer the conformational ensemble of the chromatin fiber toward or away from conformations where an enhancer and a promoter are in contact. This mechanism of action is analogous to classical allosteric regulation in proteins [26], and particularly to disordered proteins, where binding of an allosteric substrate changes the protein conformation, which in turn alters the structure of a distant active site [45]. We note that the concept of allostery has also been useful for understanding other systems, including nucleosome-mediated transcription factor binding cooperativity [46,47]. Polymer model Finally, the fine details of loop formation may be very important when the E-P pair is within the loop or near the loop bases, as we observed large variations in facilitation from subtle differences in E-P position in these cases. Polymer models were simulated with OpenMM, a high- performance GPU-assisted molecular dynamics software (https://simtk.org/home/openmm). We used an in-house openmm-polymer library to efficiently set up polymer simulations with OpenMM, and to analyze simulation results. openmm- polymer is publicly available on the Bitbucket online repository: http://bitbucket.org/mirnylab/openmm-polymer. Scripts used to perform simulations, build contact maps, and calculate insulation/ facilitation are available in the ‘‘examples’’ folder of the openmm- polymer library; those scripts can be modified to incorporate any arrangement of loops and calculate facilitation or insulation for any parameter values. Simulations were characterized by 4 parameters: loop size, number of loops, fiber stiffness, and system density. Total polymer length was always chosen to be approximately 33 * loop size. The initial conformation for all simulations was an unentangled polymer ring. Simulations for a phantom chain were performed by switching off inter-monomer Lennard-Jones interactions. Choice of parameters for various models is summarized in Table S1. Reconciling views of chromosome organization from 3C-based and microscopy studies remains an important challenge [49]. In our simulations, we found that changes in contact frequency are not always accompanied by equal changes in the mean spatial distance between two loci. In particular, insulation changes the distribution of spatial distances at small values, while having little effect on the mean. Changes in the spatial distribution at small distances could be difficult to detect experimentally and would require many cells to be assayed. Our results also suggest that integrating 3C-based and microscopy data can provide mecha- nistic insights. PLOS Computational Biology | www.ploscompbiol.org Discussion Using a polymer model of chromatin, we found that a single loop in the vicinity of an E-P pair can either insulate or facilitate their interactions. These effects have a considerable magnitude, with about 2-fold insulation and 3–5 fold facilitation of E-P contact frequency, which is comparable to generally observed changes in gene expression [42]. Collectively, experiments have observed that different local arrangements of regulatory elements can lead to complex patterns of gene expression. For example, one insulating element between an E-P pair can decrease gene expression, yet two elements between the same E-P pair do not [15,16,18,19]; it was hypothesized that the two elements cancel each other out by forming a loop. Our model shows that this loop would in fact facilitate E-P interactions. Additionally, our model predicts that if the second element were placed outside the E-P pair, the resulting loop would indeed insulate E-P interactions. Note that since the exact quantitative relationship between E-P contact frequency and gene expression or phenotype remains largely unknown, we focus on qualitative comparisons between our model and these experimental studies. Many enhancers, promoters, and loop-forming elements can be present in a given genomic region, opening the possibility for more complicated scenarios of insulation and facilitation. Towards this end, we performed simulations where two consecutive loops were formed. We observed qualitatively similar insulation and facilita- tion in the two-loop case, for the two arrangements similar to those we initially focused on (Figure S7). Within this two-loop element, the average contact frequency between loci within one loop is higher than the average contact frequency between loci from different loops. In this sense, the two loops are insulated from each other as well as from the rest of the fiber. This is consistent with simulations of an isolated system of two fused rings [28]. Moreover, these results show that the concept of steric exclusion, which underlies insulation for a single loop, applies to the two-loop case as well. In particular, each loop in the two-loop model The indirect modulation of E-P contacts by chromatin loops is often referred to as the topological model [16,17,19], a term used rather vaguely. Studies that consider the topological model often assume a particular mechanism whereby the loop alters E-P October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org 6 Calculating and analyzing heatmaps Neighboring monomers interacted via a shifted Lennard-Jones (LJ) repulsive potential U = 4 * (1/r1221/r6)+1, for r,21/6; U = 0 for r.21/6 (for details see [32]). To account for the activity of type- II topoisomerase, we allowed fiber crossing by truncating the shifted LJ potential at an energy of Ecutoff = 3 kT. For energy U more than 0.5 * Ecutoff, the LJ potential was modified as: Usoftened = 0.5 * Ecutoff * (1+tanh(2*U/Ecutoff21)). To avoid numerical instabilities, the interaction radius r was truncated at r = 0.3 via: rtruncated = (r10+0.310)1/10, which introduced a negligi- ble shift in the final softened potential. We note that our simulations were performed at thermodynamic equilibrium, and thus the rate at which the fibers were allowed to pass through each other does not influence the equilibrium properties of the system; the only relevant factor is whether the system is allowed to change its topological state or is ‘‘locked’’ to the unknotted topological state. We explored both scenarios in our simulations. Spherical confinement was realized as a potential linearly increasing at a rate of k = 5kT/mon when the radius was larger than the confinement radius. To obtain heatmaps, we first found all contacts within each polymer conformation. A contact was defined as two monomers being at a distance less than 2 monomer diameters. Contacts for all pairs of monomers were then put on a heatmap (i.e. a 2000- monomer polymer produced a 200062000 heatmap). When calculating contact frequency ratios for insulation and facilitation, averaging was performed over small regions of the heatmap to reduce sampling noise. Unless noted, we report the average value for insulation over a region of the heatmap, by averaging over monomers in the promoter-proximal third of the loop and over a +/23 monomer E-P separation. The range of insulation values for different positions in the promoter-proximal half of the loop is shown in Figure 3B. For facilitation, we average over a region of the heatmap defined by a +/23 monomer E-P distance and a +/2 ((E-P genomic distance – loop length)/6) monomer offset from a symmetric placement of an E-P pair around the loop bases (e.g. +/26 monomers in Figure 2B). For the case when the E-P pair was within the loop, no averaging was used, since this occurs at short E-P distances where many contacts occur. Forces and Langevin Dynamics simulations Adjacent monomers were connected by harmonic bonds with a potential U = 25*(r21)2 (here and below, energy is in units of kT). The stiffness of the fiber was modeled by a three point interaction term, with the potential U = k*(1-cos(a)), where a is an angle between neighboring bonds, and k is a parameter controlling stiffness. A value of k = 3 was used for most simulations; k = 2 and Another important aspect of in vivo networks of local looping interactions is that they may be both dynamic over the course of the cell cycle [32] and different between cells [50]. Our results for insulation and facilitation by fixed loops, where the bases of the loop are always connected, remain relevant for dynamic loops October 2014 | Volume 10 | Issue 10 | e1003867 October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org October 2014 | Volume 10 | Issue 10 | e1003867 7 Chromatin Loops as Modulators of Enhancer-Promoter Interactions bond and a kink, this process preserves the original topology and creates an un-entangled polymer. k = 4 were used for simulations with lower and higher stiffness; k = 2 was used for simulations with the smallest (10 monomer) loop. Initialization and starting conformations Since our simulations were performed at thermodynamic equilibrium, the starting conformation does not affect properties of the resulting heatmap; for simulations with fixed topology (i.e. no fiber crossing), only the topological state of the starting conformation is relevant. For simulations with or without fiber crossings, we initialized our simulations from an un-entangled polymer state created as described below. We started with a 4- monomer ring on a cubic lattice. We then chose one bond at random, and tried to extend the polymer at this location by two monomers, by making a bond into a kink. To do this, we considered another bond, obtained by shifting this bond by one in a random direction perpendicular to the bond (choosing one out of 4 possible directions). If both locations of the shifted bond were free, the polymer was extended to incorporate this bond. For example, if a chosen bond was going in +z direction: …R(0,0,0)R(0,0,1)R…, and we attempted to grow it in the 2y direction (chosen randomly out of +x, 2x, +y, 2y), we would check positions (0,21,0) and (0,21,1). If both of them were free, the polymer sequence would be changed to …R(0,0,0)R(0, 21,0)R(0,21,1)R(0,0,1)R…. If at least one of the positions of the shifted bond was occupied, selection of the random bond was repeated. The process was repeated until the polymer grew to the desired length. Since no polymer fibers can pass between the old 2. We then account for the fact that at larger spatial distances, a greater portion of the spatial shell exceeds the confining boundary. To this end, we took the position of the loop center of mass in a given conformation, and performed step 1 for a COM from a conformation in a different run (i.e., for conformation X in run Y, we took COM of the loop from the conformation X in the run (Y+1) modulus 10). 3. We then divided spatial densities from 1 by spatial densities from 2 to obtain normalized spatial densities. To create a no-loop control, we repeated steps 1–3 for the monomers exactly opposite from the center of the loop, exactly repeating the same procedure (i.e. assuming 60-monomer loop began at monomer 1970, recalling that the polymer is closed into a length-2000 ring). Calculating and analyzing heatmaps To calculate the contact frequency ratio, we used simulations without a loop to calculate the expected frequency; for all parameter values, two simulations without a loop and ten simulations with a loop were performed, with a newly generated starting conformation for each simulation. We simulated our model using Langevin Dynamics, performing 80,000 blocks of 3000 time steps (240,000,000 time steps total). For the fiber lengths considered here, polymer simulations reached equilibrium in less than 1000 blocks; this was confirmed by observing that monomer displacement saturates at about 500 blocks. Polymer conformations starting with block 1000 were used for our analysis. We note that this study focuses on equilibrium aspects of chromatin loops and that simulated time is not specifically matched to the time-scale of E-P interactions in vivo; chromatin loop dynamics are beyond the scope of this study. Calculating spatial density around a loop To calculate normalized spatial density around the loop, we analyzed the model with default parameters (i.e. as in Figure 2), where a loop connects monomers 970 and 1030 of a 2000- monomer ring. An Andersen thermostat was used to keep the kinetic energy of the system from diverging. The time step was then chosen to ensure conservation of kinetic energy and lack of fiber crossing with the non-truncated Lennard-Jones potential. The absence of fiber crossing in this case was confirmed by the conservation of Alexander’s polynomial for a 50000-monomer ring simulated at a high density of 0.85 for 1,000,000,000 time steps. 1. For each conformation, we found the center of mass (COM) of the 60 loop monomers. We then defined ‘‘distal’’ monomers as monomers 0–950 and 1050–2000 (outside of the loop, plus 20 monomers, or 10 kb, away from the loop base). We then counted how many distal monomers were at each spatial distance from the loop COM, averaging over all conformations and using bins starting at 0 with a step size of.5 monomers. Supporting Information Figure S1 Effects of loop size. Schematics show insulation and facilitation arrangements including the enhancer (yellow), the promoter (blue), and the loop bases (orange) for the heatmaps below. In all cases the main qualitative features remain the same. (A) A 300 kb by 300 kb heatmap for a 30 kb loop, as shown in Figure 2B. (B) A 300 kb by 300 kb heatmap for a smaller loop of length 15 kb. (C) A 300 kb by 300 kb heatmap for a larger loop of length 60 kb. (D) A 50 kb by 50 kb heatmap for a very small loop of length 2.5 kb. In this simulation only, each monomer represents 250 bp rather than 500 bp of a more flexible fiber (k = 2, see Methods), representing a loosely arranged chromatin fiber. This heatmap indicates that insulation and facilitation may still manifest at smaller genomic distances for a more flexible or loosely packed chromatin fiber, as these changes cause small loops to behave similarly to larger loops. Note the color of the map differs due to the smaller dynamic range in total number of interactions for this shorter chromatin fiber, but the same qualitative features are present. (PDF) Figure S5 Effects of phantom polymer chain. (A) Heat- map for phantom polymer chain with a 30 kb loop, where insulation and facilitation arrangements are shown as in Figure 2B. The vertical and horizontal stripes of depleted interactions are almost non-existent, indicating dramatically reduced insulation. (B) Bar plot displays insulation and facilitation for the regular scenario (Figure 2B) on the left and the phantom chain on the right. Facilitation is slightly diminished, whereas insulation completely disappears. (PDF) Figure S5 Effects of phantom polymer chain. (A) Heat- map for phantom polymer chain with a 30 kb loop, where insulation and facilitation arrangements are shown as in Figure 2B. The vertical and horizontal stripes of depleted interactions are almost non-existent, indicating dramatically reduced insulation. (B) Bar plot displays insulation and facilitation for the regular scenario (Figure 2B) on the left and the phantom chain on the right. Facilitation is slightly diminished, whereas insulation completely disappears. (PDF) Figure S6 Loop shadowing and intra-loop facilitation (A) Regions immediately outside the loop are also sterically excluded by the loop; in other words the loop’s steric ‘‘shadow’’ can cause insulation when the E-P pair is near, but outside, of the loop. Supporting Information Note the black line has been slightly offset, so that error bars are visible. (B) (left) Intra-loop facilitation when the E-P pair is positioned symmetrically within the loop. When E-P distances are much less than the loop size, the loop has a negligible influence on their contact frequency, and the contact frequency ratio is ,1. However the magnitude of facilitation increases very quickly as E-P distance approaches the loop size because the loop bases are always in contact (corresponding strong peak in Figure S2 at 30 kb = loop size). Note truncated y-axis (at contact frequency ratios of 10.0). (right) Asymmetric E-P placement with increasing E-P distance, where the enhancer stays in the middle of the loop, while the promoter moves towards the loop base. Intra-loop facilitation drops off approaching 15 kb (half of the loop size), due to a superposition with the insulating properties of the loop bases. (PDF) Figure S2 Loop-base profile. Contact frequency ratio of the loop base vs. all other loci (i.e. a 4C-like profile); an enhancer placed at one loop base (0 kb) displays a complex pattern of insulation and facilitation, which we summarize in terms of five regions (A–E). The x-axis shows the upstream or downstream distance to the loop base where this enhancer is placed; note the position of the other loop base is at 30 kb. The y-axis is truncated at contact frequency ratios of 3.0, as when both the enhancer and promoter are positioned at loop bases (i.e. x = 30 kb), the magnitude of facilitation is very large since the loop bases are always in contact. (A) Insulation of the loop base from upstream regions of chromatin. (B) Intra-loop insulation when E-P distance is less than half the loop size. (C) Intra-loop facilitation when E-P distance exceeds half the loop size. (D) Facilitation when the E-P distance slightly exceeds the loop size. (E) Insulation of the loop base from distal downstream regions of chromatin. (PDF) Figure S7 Effects of two consecutive loops. (A) Four sample polymer conformations from simulations of the two-loop system with loops (black) and loop bases (orange) highlighted. (B) Heatmap which shows log (total # of contacts) for the two-loop system. Each loop is 30 kb in the 300 kb by 300 kb region shown. The four red dots closer to the diagonal are the direct interaction of the loop bases from the formation of two loops. Supporting Information The two, weaker, red dots further from the diagonal are the interaction between the base at the start of the first loop and at the end of the second loop. The horizontal and vertical stripes of darker blue are indicative of strong insulation. Annotations show two loops formed from three bases (orange) along with the insulation and facilitation E-P placements. (C) Schematics of E-P arrangement for the two- loop system (top) insulation, indicated the red ‘‘X’’, (bottom) facilitation. ( ) PDF Figure S3 Effects of chromatin fiber flexibility. (A) Heatmap on left displays log (total # of contacts) for simulations with a more flexible polymer and standard parameters: 30 kb chromatin loop, 2% density, fiber crossing (topoisomerase activity). On the right is a heatmap for the less flexible polymer. In both cases, the loop features observed in Figure 2B are still present. (B) Bar plot shows insulation and facilitation: at the stiffness presented in the main figures, for a more flexible polymer, and for a less flexible polymer. (C) (top) shows a 20 monomer or 10 kb stretch from a conformation of a more flexible polymer. (bottom) shows a 500 monomer or 250 kb region from a conformation of a more flexible polymer. (D) Same as (C), but for a less flexible polymer. Note the smoother appearance of the less flexible chromatin fiber conformation. (PDF) Figure S3 Effects of chromatin fiber flexibility. (A) Heatmap on left displays log (total # of contacts) for simulations with a more flexible polymer and standard parameters: 30 kb chromatin loop, 2% density, fiber crossing (topoisomerase activity). On the right is a heatmap for the less flexible polymer. In both cases, the loop features observed in Figure 2B are still present. (B) Bar plot shows insulation and facilitation: at the stiffness presented in the main figures, for a more flexible polymer, and for a less flexible polymer. (C) (top) shows a 20 monomer or 10 kb stretch from a conformation of a more flexible polymer. (bottom) shows a 500 monomer or 250 kb region from a conformation of a more flexible polymer. (D) Same as (C), but for a less flexible polymer. Note the smoother appearance of the less flexible chromatin fiber conformation. (PDF) Simulated FISH distributions To calculate simulated FISH distributions, we considered an E- P distance at which the magnitude of insulation and facilitation are comparable (90 kb) and analyzed the model with default parameters. We then iterated over conformations and calculated the spatial distance between the E-P pair for both arrangements as October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org October 2014 | Volume 10 | Issue 10 | e1003867 8 Chromatin Loops as Modulators of Enhancer-Promoter Interactions tion with chromatin threaded through the loop. Without topo-II (with topological constraints), chromatin fibers cannot cross and the two conformations cannot interconvert. Bar plot shows the contact frequency ratio for an E-P genomic distance of 50 kb. (B) Effect of density on insulation and facilitation; bar plots show results for 50 kb E-P genomic distance. (PDF) well as for a control arrangement without a loop (i.e. monomers exactly opposite from the center of the loop, as for the spatial density calculation). The spatial distances were binned starting at 0 with a step size of 0.5 monomers. References Proc Natl Acad Sci U S A 100: 5223–5228. 16. Raab JR, Kamakaka RT (2010) Insulators and promoters: closer than we think. Nat Rev Genet 11: 439–446. p 40. Le TB, Imakaev MV, Mirny LA, Laub MT (2013) High-resolution mapping of the spatial organization of a bacterial chromosome. Science 342: 731–734. patial organization of a bacterial chromosome. Science 342: 731–734 17. Kulaeva OI, Nizovtseva EV, Polikanov YS, Ulianov SV, Studitsky VM (2012) Distant activation of transcription: mechanisms of enhancer action. Mol Cell Biol 32: 4892–4897. 41. Littau VGA VC., Frenster JH, Mirsky AE (1964) Active and Inactive Regions of Nuclear Chromatin as Revealed by Electron Microscope Autoradiography. Proc Natl Acad Sci U S A 52: 93–100. 18. Gerasimova TI, Corces VG (2001) Chromatin Insulators and Boundaries: Effects on Transcription and Nuclear Organization. Annual Review Genetics 35: 193–208. 42. Loven J, Orlando DA, Sigova AA, Lin CY, Rahl PB, et al. (2012) Revisiting global gene expression analysis. Cell 151: 476–482. 43. Zhu X, Ling J, Zhang L, Pi W, Wu M, et al. (2007) A facilitated tracking and transcription mechanism of long-range enhancer function. Nucleic Acids Res 35: 5532–5544. 19. Geyer PK, Clark I (2002) Protecting against promiscuity: the regulatory role of insulators. Cellular and Molecular Life Sciences 59: 2112–2127. 20. Savitskaya E, Melnikova L, Kostuchenko M, Kravchenko E, Pomerantseva E, et al. (2006) Study of long-distance functional interactions between Su(Hw) insulators that can regulate enhancer-promoter communication in Drosophila melanogaster. Mol Cell Biol 26: 754–761. 44. Harmston N, Lenhard B (2013) Chromatin and epigenetic features of long-range gene regulation. Nucleic Acids Res 41: 7185–7199. 45. Vendruscolo M (2011) Protein regulation: the statistical theory of allostery. Nat Chem Biol 7: 411–412. 21. Kyrchanova O, Maksimenko O, Stakhov V, Ivlieva T, Parshikov A, et al. (2013) Effective blocking of the white enhancer requires cooperation between two main mechanisms suggested for the insulator function. PLoS Genet 9: e1003606. 46. Mirny LA (2010) Nucleosome-mediated cooperativity between transcription factors. Proc Natl Acad Sci U S A 107: 22534–22539. mechanisms suggested for the insulator function. PLoS Genet 9: e1 . Miller JA, Widom J (2003) Collaborative Competition Mechanism 47. Miller JA, Widom J (2003) Collaborative Competition Mechanism for Gene Activation In Vivo. Molecular and Cellular Biology 23: 1623–1632. 22. Kurukuti S, Tiwari VK, Tavoosidana G, Pugacheva E, Murrell A, et al. References Trends Biochem Sci 36: 1–6. 8. Montavon T, Soshnikova N, Mascrez B, Joye E, Thevenet L, et al. (2011) A regulatory archipelago controls Hox genes transcription in digits. Cell 147: 1132–1145. 32. Naumova N, Imakaev M, Fudenberg G, Zhan Y, Lajoie BR, et al. (2013) Organization of the mitotic chromosome. Science 342: 948–953. 9. Dixon JR, Selvaraj S, Yue F, Kim A, Li Y, et al. (2012) Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature 485: 376–380. 33. Halverson JD, Smrek J, Kremer K, Grosberg AY (2014) From a melt of rings to chromosome territories: the role of topological constraints in genome folding. Rep Prog Phys 77: 022601. 10. Nora EP, Lajoie BR, Schulz EG, Giorgetti L, Okamoto I, et al. (2012) Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature 485: 381–385. p g y 34. Eastman P, Friedrichs MS, Chodera JD, Radmer RJ, Bruns CM, et al. (2013) OpenMM 4: A Reusable, Extensible, Hardware Independent Library for High Performance Molecular Simulation. J Chem Theory Comput 9: 461–469. 11. Sanyal A, Lajoie BR, Jain G, Dekker J (2012) The long-range interaction landscape of gene promoters. Nature 489: 109–113. 35. Dekker J, Marti-Renom MA, Mirny LA (2013) Exploring the three-dimensional organization of genomes: interpreting chromatin interaction data. Nat Rev Genet 14: 390–403. 12. Phillips-Cremins JE, Sauria ME, Sanyal A, Gerasimova TI, Lajoie BR, et al. (2013) Architectural protein subclasses shape 3D organization of genomes during lineage commitment. Cell 153: 1281–1295. 36. Fudenberg G, Mirny LA (2012) Higher-order chromatin structure: bridging physics and biology. Curr Opin Genet Dev 22: 115–124. 13. Bondarenko VA, Jiang YI, Studitsky VM (2003) Rationally designed insulator- like elements can block enhancer action in vitro. EMBO Journal 22: 4728–4737. 37. Dorier J, Stasiak A (2009) Topological origins of chromosomal territories. Nucleic Acids Res 37: 6316–6322. 38. Rosa A, Everaers R (2008) Structure and dynamics of interphase chromosomes. PLoS Comput Biol 4: e1000153. 14. Gohl D, Aoki T, Blanton J, Shanower G, Kappes G, et al. (2011) Mechanism of chromosomal boundary action: roadblock, sink, or loop? Genetics 187: 731–748. 39. Benedetti F, Dorier J, Burnier Y, Stasiak A (2013) Models that include supercoiling of topological domains reproduce several known features of interphase chromosomes. Nucleic Acids Res 42: 2848–2855. 15. Majumder P, Cai HN (2003) The functional analysis of insulator interactions in the Drosophila embryo. We thank Mirny lab members for useful discussions. We thank Mirny lab members for useful discussions. We thank Mirny lab members for useful discussions. (XLSX) Video S1 Langevin Dynamics of a 30 kb permanent loop formed in a 1 Mb region of chromatin fiber, as in Figure 2. The polymer is colored according to the facilitation arrangement, where the loop (black) occurs in the region between the enhancer (yellow) and the promoter (blue). The movie is presented at a rate of 1000 simulated time-steps per one second of real time; every seventh frame of the movie corresponds to a Video S1 Langevin Dynamics of a 30 kb permanent loop formed in a 1 Mb region of chromatin fiber, as in Figure 2. The polymer is colored according to the facilitation arrangement, where the loop (black) occurs in the region between the enhancer (yellow) and the promoter (blue). The movie is presented at a rate of 1000 simulated time-steps per one second of real time; every seventh frame of the movie corresponds to a Figure S4 Effect of topoisomerase and chromatin density on local loop-mediated interactions. (A) Effect of topological constraints on insulation and facilitation. With topo-II, there are no topological constraints and a conformation without chromatin threaded through the loop can convert to a conforma- October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org 9 Chromatin Loops as Modulators of Enhancer-Promoter Interactions Chromatin Loops as Modulators of Enhancer-Promoter Interactions References 1. Bulger M, Groudine M (2011) Functional and mechanistic diversity of distal transcription enhancers. Cell 144: 327–339. 24. Chetverina D, Aoki T, Erokhin M, Georgiev P, Schedl P (2014) Making connections: insulators organize eukaryotic chromosomes into independent cis- regulatory networks. Bioessays 36: 163–172. 2. Levine M (2010) Transcriptional enhancers in animal development and evolution. Curr Biol 20: R754–763. 25. Scott KS, Taubman A.D., Geyer P.K. (1999) Enhancer Blocking by the Drosophila gypsy Insulator Depends Upon Insulator Anatomy and Enhancer Strength. Genetics 153: 787–798. 3. Pennacchio LA, Bickmore W, Dean A, Nobrega MA, Bejerano G (2013) Enhancers: five essential questions. Nat Rev Genet 14: 288–295. g 26. Monod J, Changeux JP, Jacob F (1963) Allosteric Proteins and Cellular Control Systems. Journal of Molecular Biology 6: 306–329. 4. Tolhius B, Palstra RJ, Splinter E, Grosveld F, de Laat W (2002) Looping and Interaction between Hypersensitive Sites in the Active B-globin Locus. Molecular Cell 10: 1453–1465. J g J J ( ) Systems. Journal of Molecular Biology 6: 306–329 27. Wyman J (1964) Linked functions and reciprocal effects in hemoglobin: a second look. Adv Protein Chem 19: 91. 5. Consortium EP, Bernstein BE, Birney E, Dunham I, Green ED, et al. (2012) An 5. Consortium EP, Bernstein BE, Birney E, Dunham I, Green ED, et al. (2012) An integrated encyclopedia of DNA elements in the human genome. Nature 489: 57–74. 28. Mukhopadhyay S, Schedl P, Studitsky VM, Sengupta AM (2011) Theoretical analysis of the role of chromatin interactions in long-range action of enhancers and insulators. Proc Natl Acad Sci U S A 108: 19919–19924. integrated encyclopedia of DNA elements in the human genome. Nature 489: 57–74. 6. Jin F, Li Y, Dixon JR, Selvaraj S, Ye Z, et al. (2013) A high-resolution map of the three-dimensional chromatin interactome in human cells. Nature 503: 290– 294. 29. Giorgetti L, Galupa R, Nora EP, Piolot T, Lam F, et al. (2014) Predictive polymer modeling reveals coupled fluctuations in chromosome conformation and transcription. Cell 157: 950–963. 7. Andrey G, Montavon T, Mascrez B, Gonzalez F, Noordermeer D, et al. (2013) A Switch Between Topological Domains Underlies HoxD Genes Collinearity in Mouse Limbs. Science 340: 1195–1205. p 30. Joti Y, Hikima T, Nishino Y, Kamada F, Hihara S, et al. (2012) Chromosomes without a 30-nm chromatin fiber. Nucleus 3: 404–410. 31. Fussner E, Ching RW, Bazett-Jones DP (2011) Living without 30 nm chromatin fibers. Author Contributions Conceived and designed the experiments: BD MI GF LAM. Performed the experiments: BD. Analyzed the data: BD. Contributed reagents/materials/ analysis tools: BD MI GF LAM. Wrote the paper: BD MI GF LAM. Conceived and designed the experiments: BD MI GF LAM. Performed the experiments: BD. Analyzed the data: BD. Contributed reagents/materials/ analysis tools: BD MI GF LAM. Wrote the paper: BD MI GF LAM. October 2014 | Volume 10 | Issue 10 | e1003867 Acknowledgments computationally obtained conformation, with quadratic interpo- lation performed between subsequent conformations (done to avoid high-frequency fluctuations between neighboring frames). The total simulation time for each run of each parameter set was approximately 5,000 times longer than the part of simulation displayed in this movie. References (2006) CTCF binding at the H19 imprinting control region mediates maternally inherited higher-order chromatin conformation to restrict enhancer access to Igf2. Proc Natl Acad Sci U S A 103: 10684–10689. Activation In Vivo. Molecular and Cellular Biology 23: 1623–16 gy 48. Li G, Levitus M, Bustamante C, Widom J (2005) Rapid spontaneous accessibility of nucleosomal DNA. Nat Struct Mol Biol 12: 46–53. 49. Belmont AS (2014) Large-scale chromatin organization: the good, the surprising, and the still perplexing. Curr Opin Cell Biol 26: 69–78. 49. Belmont AS (2014) Large-scale chromatin organization: 23. Comet I, Schuettengruber B, Sexton T, Cavalli G (2010) A chromatin insulator driving three-dimensional Polycomb response element (PRE) contacts and Polycomb association with the chromatin fiber. Proc Natl Acad Sci U S A 108: 2294–2299. 50. Nagano T, Lubling Y, Stevens TJ, Schoenfelder S, Yaffe E, et al. (2013) Single- cell Hi-C reveals cell-to-cell variability in chromosome structure. Nature 502: 59–64. October 2014 | Volume 10 | Issue 10 | e1003867 PLOS Computational Biology | www.ploscompbiol.org 10

https://openalex.org/W4214933660

https://www.scielo.br/j/riem/a/NCyT6vTvxBJy93jRKy3pV9j/?lang=en&format=pdf

Latin

A method for considering the influence of distinct casting stages in the flexural design of prestressed concrete cross sections

Revista IBRACON de Estruturas e Materiais

cc-by

A method for considering the influence of distinct casting stages in the flexural design of prestressed concrete cross sections Um método para consideração da influência de moldagens distintas no dimensionamento à flexão de seções transversais de concreto protendido Eduardo Vicente Wolf Trentinia  Guilherme Aris Parsekiana  Túlio Nogueira Bittencourtb  Eduardo Vicente Wolf Trentinia  Guilherme Aris Parsekiana  Túlio Nogueira Bittencourtb  aUniversidade Federal de São Carlos – UFSCar, Programa de Pós-graduação em Engenharia Civil, São Carlos, SP, Brasil bUniversidade de São Paulo – USP, Escola Politécnica, São Paulo, SP, Brasil Abstract: Composite elements are structures of concrete, or other materials, constructed in different casting stages that act jointly under external loads. These elements are used when it is intended to combine the constructive advantages of precast structures with the monolithic behavior of cast-in-place structures. In regular civil engineering applications, such as the construction of a bridge or viaduct, the precast section is used as shoring before casting the slab in place. This process leads to imposed deformations prior to the ultimate limit state and a discontinuity in the specific strain of the composite cross section. This work proposes a methodology to design composite cross sections, built in two casting stages, evaluating the specific strain provided the construction process that can be easily implemented in precise computational routines. From applying the methodology on study-case numerical example, it is observed that the beam casted in two stages presents a factored moment resistance smaller than an identical beam casted in a single stage. However, further investigations should be conducted to assess the extent of this difference. Received 09 September 2021 Accepted 24 January 2022 Keywords: composite sections, construction process, precast, bending, structural design. Resumo: Elementos compostos são estruturas de concreto, ou outros materiais, executados em moldagens distintas que atuam de maneira conjunta sob ações externas. Esses elementos são utilizados quando se pretende aliar as vantagens construtivas de estruturas pré-moldadas ao comportamento monolítico de estruturas moldadas no local. Em situações comuns na engenharia civil, como a construção de uma ponte ou viaduto, a seção pré-moldada é utilizada como cimbramento para a moldagem no local da laje. Esse processo proporciona deformações prévias ao estado-limite último e descontínuas na seção transversal composta. Este trabalho propõem uma metodologia para dimensionamento de seções compostas, em duas etapas de concretagem, avaliando as deformações proporcionadas pelo processo construtivo que pode ser facilmente implementada em rotinas computacionais precisas. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 ORIGINAL ARTICLE A method for considering the influence of distinct casting stages in the flexural design of prestressed concrete cross sections Um método para consideração da influência de moldagens distintas no dimensionamento à flexão de seções transversais de concreto protendido Eduardo Vicente Wolf Trentinia  Guilherme Aris Parsekiana  Túlio Nogueira Bittencourtb  aUniversidade Federal de São Carlos – UFSCar, Programa de Pós-graduação em Engenharia Civil, São Carlos, SP, Brasil bUniversidade de São Paulo – USP, Escola Politécnica, São Paulo, SP, Brasil ORIGINAL ARTICLE A method for considering the influence of distinct casting stages in the flexural design of prestressed concrete cross sections Um método para consideração da influência de moldagens distintas no dimensionamento à flexão de seções transversais de concreto protendido Eduardo Vicente Wolf Trentinia  Guilherme Aris Parsekiana  Túlio Nogueira Bittencourtb  aUniversidade Federal de São Carlos – UFSCar, Programa de Pós-graduação em Engenharia Civil, São Carlos, SP, Brasil bUniversidade de São Paulo – USP, Escola Politécnica, São Paulo, SP, Brasil 1 INTRODUCTION The demand for quality, speed of execution and rationalization of building materials has increased the choice for precast concrete structures. The use of precast elements introduces a repetitive character into the construction process reducing waste and directly reflecting a better productivity of the workforce [1]. To connect the precast elements to the rest of the structure, it is common to cast in place part of the element, creating a monolithic connection. In civil construction these elements are known as pre-beams and pre-slabs and NBR 9062 [2] defines them as composite elements, being: “structures of concrete, or other materials, constructed in different casting stages that act jointly under external loads”. The use of concrete composite elements combines the advantages of precast concrete structures, such as the use of more complex cross sections, reuse of formwork, possible prestress on plant, excellent quality control, reduction of shoring, the monolithic behavior induced by cast in place concrete [3], [4]. The use of composite sections in concrete introduces peculiar situations the structural an influence of the factors below must be checked: osite sections in concrete introduces peculiar situations the structural analysis process. Each rs below must be checked: tions in the precast concrete section prior to the curing of the second stage cast; deformations in the precast concrete section prior to the curing of the second stage cast; different strain resulting from concrete shrinkage and creep, on each stage, due to the difference between materials and casting age, and; different strain resulting from concrete shrinkage and creep, on each stage, due to the difference between materials and casting age, and; the existence of slip between the interface of the two concrete casts [2]. Regarding occurrence of slip verification between the contact surfaces, many experimental studies were carried out and proposed equations to evaluate the load capacity of the interface to the horizontal shear are available in [5]–[12]. Design codes, such as ACI [13], AASHTO [14] and NBR 9062 [2], include equations to evaluate this capacity. The NBR 9062 [2] prescribes that if the horizontal shear strength is greater than the shear stress load, it can be considered that the composite element presents monolithic behavior, as also observed by [15]–[17]. Different age, rheology, and stress level in each casting stage, causes different volume changes due to shrinkage and creep of the concrete. A method for considering the influence of distinct casting stages in the flexural design of prestressed concrete cross sections Aplicando a metodologia em um exemplo numérico, é observado que a seção concretada em duas etapas apresenta um momento resistente menor que uma seção idêntica concretada em etapa única. Porém são necessárias mais investigações para avaliar a amplitude desta divergência em outros elementos. Palavras-chave: seções compostas, processo construtivo, pré-moldado, flexão, dimensionamento. How to cite: E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt, “A method for considering the influence of distinct casting stages in the flexural design of prestressed concrete cross sections,” Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022, https://doi.org/10.1590/S1983- 41952022000400010 Corresponding author: Eduardo Vicente Wolf Trentini. E-mail: [email protected] Financial support: None. Conflict of interest: Nothing to declare. Data Availability: The data that support the findings of this study are available from the corresponding author, EVWT, upon reasona This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permit and reproduction in any medium, provided the original work is properly cited. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022| https://doi.org/10.1590/S1983-41952022000400010 Corresponding author: Eduardo Vicente Wolf Trentini. E-mail: [email protected] Financial support: None. Conflict of interest: Nothing to declare. g vailability: The data that support the findings of this study are available from the corresponding author, EVWT, upon reasonable request. Data Availability: The data that support the findings of this study are available from the corresponding author, EVWT, upon rea This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted and reproduction in any medium, provided the original work is properly cited. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022| https://doi.org/10.1590/S1983-41952022000400010 1/21 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 1 INTRODUCTION The strain discontinuity along the cross section makes the use of conventional design methods unfeasible, since usually continuous strain distribution along the height is assumed. Few works have been developed considering the influence of strain in the precast section when used as shoring for the second casting stage. The strain discontinuity along the cross section makes the use of conventional design methods unfeasible, since usually continuous strain distribution along the height is assumed. Few works have been developed considering the influence of strain in the precast section when used as shoring for the second casting stage. In 1955 [27], experimental studies were carried out in which previous prestressing and self-weight strain in the precast section are considered to estimate the stresses, in elastic regime, of composite cast-in-place slab and precast concrete beams, built in two steps. Authors conclude that if the rough interface provides enough friction, the composite cross section can be monolithic considered. Taha, in 1978 [28], developed a software for the design of composite sections of steel beams and concrete slabs. The permanent strain in the steel section due to the slab casting load were considered, but still in an elastic regime. The design was carried out using the allowable stress method. Dritsos et al. in 1995 [29] reports the efficiency of reinforcing concrete structures by adding a new casting step. The authors considered the slip at the interface between the two casting steps which, in turn, promotes a similar strain discontinuity situation as the construction of non-shored composite cross sections. The equilibrium was evaluated with non-linear constitutive laws for steel and concrete. Hwang et al. in 2015 [30] proposed a method to evaluate the factored moment resistance and the deformation of precast post-tensioned beams, composed with cast-in-place slab and shored. The authors proposed a complex analysis where, in addition to the non-linear behavior of the materials, the existence of slippage at the interface is also considered with a non-linear response. As this method is intended for shored while casting structures, the analysis developed only evaluates the strain discontinuity due to the prestressing of the precast beam before the slab curing, and due to the interface slippage of the two concretes. This work proposes a method for flexural design, of precast post-tensioned composite concrete sections with cast- in-place slab without additional shoring. 1 INTRODUCTION Since there is a relative strain restriction between the two casting stages, the differential volume variation results in the development of stresses in the section [18]. Models for evaluating the resulting stresses due to differential volume variation are extensively addressed in the literature [19]–[24]. As shown by [25], the creep effect ends up decreasing the tension difference between the precast section and the cast in place section, thus making the stress distribution in the composite cross section more similar to that developed in a single-step casted section. The shrinkage effect is more intense in the first days. Therefore, after the in place casting, the concrete of the second cast shrinks more than the first cast, developing tensile stresses and even cracks that may reduce durability [26]. In bridge construction, the girders usually are joined to the slab in a later cast in place step. This connection occurs in the upper region of the cross section subject to compression in the case of beams demanded by positive bending moment. In this case, the tensile stress mentioned above is beneficial for the analysis of the section in Ultimate Limit State (ULS), therefore, the evaluation of these both effects are not included in the scope of this article assuming that the ULS occurs when the effects of shrinkage and creep have not yet fully developed. During the construction process, the precast girder is lifted and placed on the supports, and then used as shoring for the cast in place slab. The self-weight load deforms only the girders, once only after the curing of the second casting stage can the additional loads be considered acting on the monolithic composite section. This constructive process results in a cross section that presents a strain, 𝜀𝜀, discontinuity along of its z-height due to previous strain imposed to the precast member prior to the slab curing. Figure 1 illustrates the strain and normal stresses of part of a beam, resulting from this construction process. To show the contrast between the developed stresses and strains, in Figure 1 the behavior of beam built within this construction process is compared to that of a beam built in a single casting. 2/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Figure 1. Comparison between construction processes. Figure 1. Comparison between construction processes. 1 INTRODUCTION This method evaluates the strains in the precast beam, which precede the curing of the concrete of the second casting stage due to prestressing and self-weight, considering the monolithic behavior after curing, neglecting the effects of the differential volume variation between the concrete of the two casting stages. To simplify and limit the problem, the following hypotheses are adopted: 2 HYPOTHESES OF THE MODEL Stress-strain relationship of the materials. Figure 2. Stress-strain relationship of the materials. Where 𝜎𝜎𝑐𝑐 is 𝜎𝑓 𝜎𝜎𝑐𝑐= 0,85 ∙𝑓𝑓𝑐𝑐𝑐𝑐ቂ1 −ቀ1 − 𝜀𝜀𝑐𝑐 𝜀𝜀𝑐𝑐2ቁ 𝑛𝑛 ቃ 𝑓𝑐𝑐𝜀𝑐𝜀𝑐𝑢𝑛 (1) where 𝑓𝑓𝑐𝑐𝑐𝑐, 𝜀𝜀𝑐𝑐2, 𝜀𝜀𝑐𝑐𝑢𝑢 and 𝑛𝑛 are defined in [31] as a function of the characteristic strength of concrete 𝑓𝑓𝑐𝑐𝑐𝑐. where 𝑓𝑓𝑐𝑐𝑐𝑐, 𝜀𝜀𝑐𝑐2, 𝜀𝜀𝑐𝑐𝑢𝑢 and 𝑛𝑛 are defined in [31] as a function of the characteristic strength of concrete 𝑓𝑓𝑐𝑐𝑐𝑐. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 2 HYPOTHESES OF THE MODEL To simplify and limit the problem, the following hypotheses are adopted: 3/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt 1. The beams are long one-dimensional structural elements; thus, the cross sections remain plane after deformation – Bernoulli's theorem. 1. The beams are long one-dimensional structural elements; thus, the cross sections remain plane after deformation – Bernoulli's theorem. 2. The connection between passive and prestressing reinforcement with concrete is assumed to be perfect. There is no slippage between the elements and the strain of the reinforcement is the same as the concrete in its vicinity. 2. The connection between passive and prestressing reinforcement with concrete is assumed to be perfect. There is no slippage between the elements and the strain of the reinforcement is the same as the concrete in its vicinity. 3. The interaction between the precast beam and the cast-in-place slab is full, that is, there is no slippage at the interface, thus assuming a monolithic behavior. Any variation in curvature or axial deformation that occurs after cure of the second step is the same for the entire cross section. This hypothesis is valid, according to NBR 9062 [2], if the horizontal shear load is less than the interface strength. 4. Prestressing is applied before the cast of the second step. 5. The cast of the second step is carried out without additional shoring, that is, all self-weight inc of the slab deforms only the precast section. y p 6. Between the concrete of the two castings, the differential effects of temperature, shrinkage and creep are neglected. 6. Between the concrete of the two castings, the differential effects of temperature, shrinkage The constitutive law of the materials involved in this analysis are described in NBR 6118:2014 [31] and presented in Figure 2. The constitutive law for compressed concrete has two parts, the first a polynomial curve and the second linear. The stress of the tensioned concrete is neglected. The steel used for passive reinforcement is CA-50 which has an elastic modulus 𝐸𝐸𝑠𝑠 and a factored yield stress 𝑓𝑓𝑦𝑦𝑦𝑦 defined in [31]. The steel of the prestressing reinforcement is CP- 190 RB with modulus of elasticity 𝐸𝐸𝑝𝑝, factored conventional yield strength 𝑓𝑓𝑝𝑝𝑝𝑝𝑝𝑝 and factored tensile strength 𝑓𝑓𝑝𝑝𝑝𝑝𝑝𝑝 defined in [32]. Both steels have symmetrical behavior Figure 2. Stress-strain relationship of the materials. 𝜎 Figure 2. 3 EQUILIBRIUM AND COMPATIBILITY EQUATIONS Since 𝑑𝑑𝑑𝑑 is function of the concrete stress 𝑓𝑓𝑐𝑐, where 𝑑𝑑𝑑𝑑= 𝑓𝑓𝑐𝑐∙𝑑𝑑𝐴𝐴, the 𝑅𝑅𝑐𝑐𝑐𝑐 and 𝑀𝑀𝑐𝑐𝑐𝑐 are then written as: 𝑅𝐹𝑅𝑓𝐴 𝑅𝑅𝑐𝑐𝑐𝑐= ∫d𝐹𝐹 𝐴𝐴 ∴ 𝑅𝑅𝑐𝑐𝑐𝑐= ∫𝑓𝑓𝑐𝑐∙d𝐴𝐴 𝐴𝐴 , (4) 𝑀𝑀𝑐𝑐𝑐𝑐= ∫𝑧𝑧∙d𝐹𝐹 𝐴𝐴 ∴ 𝑀𝑀𝑐𝑐𝑐𝑐= ∫𝑧𝑧∙𝑓𝑓𝑐𝑐∙d𝐴𝐴 𝐴𝐴 . (5) 𝑅𝑅𝑐𝑐𝑐𝑐= ∫d𝐹𝐹 𝐴𝐴 ∴ 𝑅𝑅𝑐𝑐𝑐𝑐= ∫𝑓𝑓𝑐𝑐∙d𝐴𝐴 𝐴𝐴 , 𝑀𝑀𝑐𝑐𝑐𝑐= ∫𝑧𝑧∙d𝐹𝐹 𝐴𝐴 ∴ 𝑀𝑀𝑐𝑐𝑐𝑐= ∫𝑧𝑧∙𝑓𝑓𝑐𝑐∙d𝐴𝐴 𝐴𝐴 . 𝑅𝑅𝑐𝑐𝑐𝑐= ∫d𝐹𝐹 𝐴𝐴 ∴ 𝑅𝑅𝑐𝑐𝑐𝑐= ∫𝑓𝑓𝑐𝑐∙d𝐴𝐴 𝐴𝐴 , 𝑀𝑧𝐹𝑀𝑧𝑓𝐴 (4) 𝑀𝑀𝑐𝑐𝑐𝑐= ∫𝑧𝑧∙d𝐹𝐹 𝐴𝐴 ∴ 𝑀𝑀𝑐𝑐𝑐𝑐= ∫𝑧𝑧∙𝑓𝑓𝑐𝑐∙d𝐴𝐴 𝐴𝐴 . (5) The integrals of the Equations 4 and 5 can be solved by one of two processes: analytical integration or numerical integration by discretizing the area into small elements. Numerical integration is versatile in terms of allowing to consider different constitutive laws. On the other hand, the quality of the response is related to the number of elements used, and this alternative has a higher computational cost. The computational cost is an inconvenience that can make the use of this alternative unfeasible when incorporated in iterative processes. p Analytical integration presents a negligible computational cost when compared to numerical integration, and its answer is precise within the approximations of the mathematical model. Thus, the integrals of Equations 4 and 5 are here analytically evaluated. 3 EQUILIBRIUM AND COMPATIBILITY EQUATIONS The internal forces must be in equilibrium with the external forces. In the case of the current structure, the equilibrium must be verified in two situations. The first verification is performed during the casting of the addition on site, which here is called the initial step. Figure 3 illustrates the internal forces and strains of the cross section, when it is submitted to the initial step. This figure also highlights the position of layers A, B, C, and S, which are respectively: surface level between the two castings, finished level of the cast in place addition, level of the beam base and level of the passive reinforcement outermost to the beam. 𝜀𝜀 Checking the equilibrium in the initial step is necessary to evaluate the permanent deformation prior to the slab casting. The strain at layer A, 𝜀𝜀𝐴𝐴,𝑖𝑖, and the strain in the layer S, 𝜀𝜀𝑆𝑆,𝑖𝑖, are the variables chosen to define the curvature and the axial deformation of the beam during the initial step. The strain between the layers A and B are null. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 4/21 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Figure 3. Internal stresses and strain diagram in initial step. Figure 3. Internal stresses and strain diagram in initial step. The second equilibrium is developed with the composite section in ULS. This includes the strains on the initial step plus then addition of curvature and axial deformation after the slab casting. This phase is called total step. Figure 4 illustrates internal forces and the strains on the composed cross section, in the total step. Figure 4. Internal stresses and strain diagram in total step. 𝑥 Figure 4. Internal stresses and strain diagram in total step. The equilibrium equations that apply to the case of symmetrical bending are: (2) Sum of forcer in the 𝑥𝑥 direction, 𝑆𝑆𝑆𝑆; (3) Sum of moments in the 𝑦𝑦 direction, 𝑆𝑆𝑆𝑆. These equations are defined at the point CG, which represents the centroid of the cross section. 𝐹𝑅𝑅𝑅𝑁𝑆𝑆 ∑𝐹𝐹𝑥𝑥= 0 ∴ 𝑅𝑅𝑠𝑠𝑠𝑠+ 𝑅𝑅𝑝𝑝𝑝𝑝+ 𝑅𝑅𝑐𝑐𝑐𝑐−𝑁𝑁𝑑𝑑= 0 → 𝑆𝑆𝑆𝑆= 0 (2) Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 5/21 E. V. W. Trentini, G. A. Parsekian, and T. N. 3 EQUILIBRIUM AND COMPATIBILITY EQUATIONS Bittencourt 𝑀𝑀𝑀𝑀𝑀 (3) 𝑅 ∑𝑀𝑀𝑦𝑦= 0 ∴ 𝑀𝑀𝑠𝑠𝑠𝑠+ 𝑀𝑀𝑝𝑝𝑝𝑝+ 𝑀𝑀𝑐𝑐𝑐𝑐+ 𝑀𝑀𝑑𝑑= 0 → 𝑆𝑆𝑆𝑆= 0 Where: 𝑅𝑅𝑠𝑠𝑠𝑠 normal reaction of the passive reinforcement, 𝑅𝑅𝑝𝑝𝑝𝑝 normal reaction of the prestressing reinforcement, 𝑅𝑅𝑐𝑐𝑐𝑐 normal reaction of the concrete, 𝑁𝑁𝑑𝑑 factored normal force of the member, 𝑀𝑀𝑠𝑠𝑑𝑑 moment of the reaction of the passive reinforcement, 𝑀𝑀𝑝𝑝𝑝𝑝 moment of the reaction of the active reinforcement, 𝑀𝑀𝑐𝑐𝑐𝑐 moment of the reaction of the concrete, 𝑀𝑀𝑑𝑑 factored bending moment of the member. All these moments are defined around the 𝑦𝑦-axis. 𝜀 The reactions and moments of the reinforcement are obtained by assuming the constitutive law shown in Figure 2. Assuming perfect bond, the strain of the passive reinforcement is equal to that of the concrete in its vicinity. The strain of the prestressing reinforcement is the sum of the pre-strain of the steel, 𝜀𝜀𝑝𝑝𝑝𝑝𝑝𝑝, with the strain of the adjacent concrete. The contribution of concrete in the internal equilibrium is analyzed by highlighting the infinitesimal element of area 𝑑𝑑𝑑𝑑, considering that an infinitesimal force 𝑑𝑑𝑑𝑑 is exerted on it. Since 𝑑𝑑𝑑𝑑 is function of the concrete stress 𝑓𝑓𝑐𝑐, where 𝑑𝑑𝑑𝑑= 𝑓𝑓𝑐𝑐∙𝑑𝑑𝐴𝐴, the 𝑅𝑅𝑐𝑐𝑐𝑐 and 𝑀𝑀𝑐𝑐𝑐𝑐 are then written as: 𝑅𝐹𝑅𝑓𝐴 The reactions and moments of the reinforcement are obtained by assuming the constitutive law shown in Figure 2. Assuming perfect bond, the strain of the passive reinforcement is equal to that of the concrete in its vicinity. The strain of the prestressing reinforcement is the sum of the pre-strain of the steel, 𝜀𝜀𝑝𝑝𝑝𝑝𝑝𝑝, with the strain of the adjacent concrete. The contribution of concrete in the internal equilibrium is analyzed by highlighting the infinitesimal element of area 𝑑𝑑𝑑 id i h i fi i i l f 𝑑𝑑𝑑 i d i Si 𝑑𝑑𝑑 i f i f h 𝑓𝑓 h𝑑𝑑𝑓𝑑𝐴𝑅𝑀 The reactions and moments of the reinforcement are obtained by assuming the constitutive law shown in Figure 2. Assuming perfect bond, the strain of the passive reinforcement is equal to that of the concrete in its vicinity. The strain of the prestressing reinforcement is the sum of the pre-strain of the steel, 𝜀𝜀𝑝𝑝𝑝𝑝𝑝𝑝, with the strain of the adjacent concrete. 𝑑𝑑𝑑𝑑𝑑𝑑𝑓 The contribution of concrete in the internal equilibrium is analyzed by highlighting the infinitesimal element of area 𝑑𝑑𝑑𝑑, considering that an infinitesimal force 𝑑𝑑𝑑𝑑 is exerted on it. 3.1 Analytical integral of concrete stresses 𝑅𝑓𝑐𝑧𝑒 𝑅𝑅𝑐𝑐𝑐𝑐= ∑ ቆ0,85 ∙𝑓𝑓𝑐𝑐𝑐𝑐∙ቀ𝑐𝑐1 ∙𝑧𝑧+ 𝑐𝑐2∙𝑧𝑧2 2 ቁቇ ቤ 𝑏𝑏 𝑒𝑒 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 (13) and 𝑀𝑀𝑐𝑐𝑐𝑐= ∑ ቆ0,85 ∙𝑓𝑓𝑐𝑐𝑐𝑐∙ቀ 𝑐𝑐1∙𝑧𝑧2 2 + 𝑐𝑐2∙𝑧𝑧3 3 ቁቇቤ 𝑏𝑏 𝑒𝑒 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 (14) 𝑅𝑅𝑐𝑐𝑐𝑐= ∑ ቆ0,85 ∙𝑓𝑓𝑐𝑐𝑐𝑐∙ቀ𝑐𝑐1 ∙𝑧𝑧+ 𝑐𝑐2∙𝑧𝑧2 2 ቁቇ ቤ 𝑏𝑏 𝑒𝑒 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 (13) and 𝑀𝑀𝑐𝑐𝑐𝑐= ∑ ቆ0,85 ∙𝑓𝑓𝑐𝑐𝑐𝑐∙ቀ 𝑐𝑐1∙𝑧𝑧2 2 + 𝑐𝑐2∙𝑧𝑧3 3 ቁቇቤ 𝑏𝑏 𝑒𝑒 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 (14) In the analyses, the reinforcement area is not subtracted from the concrete cross section area. In the analyses, the reinforcement area is not subtracted from the concrete cross section area. 3.1 Analytical integral of concrete stresses The methodology for evaluating the axial reaction 𝑅𝑅𝑐𝑐𝑐𝑐 and the moment 𝑀𝑀𝑐𝑐𝑐𝑐 by analytical integration, used here, was developed by Silva and Carvalho in 2019 [33], for the NBR 6118:2014 constitutive law of the concrete. 𝜀𝜀 The method is applied to cross sections described in polygonal form, which have nodes at the vertices and at transition points of the constitutive relation. The transition points of the constitutive relation occur where the section deformation is null, or equal to 𝜀𝜀𝑐𝑐2 or equal to 𝜀𝜀𝑐𝑐𝑐𝑐. 𝐴𝑙𝑧 With the polygonal defined, the cross section is subdivided into trapezoids, 𝐴𝐴𝑙𝑙, contained between the 𝑧𝑧-axis and the lines that define the perimeter of the cross section. Figure 5 illustrates one of these trapezoids contained between the line 𝑙𝑙 and the 𝑧𝑧-axis. Figure 5. Subdivision of the cross section into trapezoids. Figure 5. Subdivision of the cross section into trapezoids. 6/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt The contribution of concrete in the equilibrium is evaluated with the Equations 6, 7, 13 and 14 as shown in [33]. 𝜎𝑐𝜀𝑐 The contribution of concrete in the equilibrium is evaluated with the Equations 6, 7, 13 and 14 as shown in [33]. For the polynomial part of the constitutive relation, 𝜎𝜎𝑐𝑐 of Equation 1, that is, strains smaller than zero and larger than 𝜀𝜀𝑐𝑐2. 𝑒 For the polynomial part of the constitutive relation, 𝜎𝜎𝑐𝑐 of Equation 1, that is, strains smaller than zero and larger than 𝜀𝜀𝑐𝑐2. 3.1 Analytical integral of concrete stresses 𝑒 𝑅𝑅𝑐𝑐𝑐𝑐= ∑ ൮−0,85 ∙𝑓𝑓𝑐𝑐𝑐𝑐∙ቌ− ℎ𝑝𝑝∙൬ 𝑧𝑧𝑁𝑁𝑁𝑁+ℎ𝑝𝑝−𝑧𝑧 ℎ𝑝𝑝 ൰ 𝑛𝑛1 ∙ቀ𝑐𝑐1∙𝑛𝑛2+𝑐𝑐2∙൫ℎ𝑝𝑝+𝑧𝑧𝑁𝑁𝑁𝑁+𝑛𝑛∙𝑧𝑧+𝑧𝑧൯ቁ 𝑛𝑛1∙𝑛𝑛2 −𝑐𝑐1 ∙𝑧𝑧− 𝑐𝑐2∙𝑧𝑧2 2 ቍ൲ ተ 𝑏𝑏 𝑒𝑒 𝑃𝑃𝑃𝑃𝑃𝑃 (6) (6) and 𝑀𝑀𝑐𝑐𝑐𝑐= ∑ ቌ 0,85∙𝑓𝑓𝑐𝑐𝑐𝑐 6∙𝑛𝑛1∙𝑛𝑛2∙𝑛𝑛3 ∙൭3 ∙𝑐𝑐1 ∙ቆ𝑛𝑛1 ∙𝑛𝑛2 ∙𝑛𝑛3 ∙𝑧𝑧2 + 2 ∙ℎ𝑝𝑝∙൬ 𝑧𝑧𝑁𝑁𝑁𝑁+ℎ𝑝𝑝−𝑧𝑧 ℎ𝑝𝑝 ൰ 𝑛𝑛1 ∙ቀ൫ℎ𝑝𝑝+ 𝑧𝑧𝑁𝑁𝑁𝑁൯∙𝑛𝑛3 + 𝑛𝑛1 ∙𝑛𝑛3 ∙𝑧𝑧ቁቇ+ 2 ∙𝑐𝑐2 ∙ 𝑃𝑃𝑃𝑃𝑃𝑃 ቆ𝑛𝑛1 ∙𝑛𝑛2 ∙𝑛𝑛3 ∙𝑧𝑧3 + 3 ∙ℎ𝑝𝑝∙൬ 𝑧𝑧𝑁𝑁𝑁𝑁+ℎ𝑝𝑝−𝑧𝑧 ℎ𝑝𝑝 ൰ 𝑛𝑛1 ∙ቀ2 ∙𝑧𝑧𝑁𝑁𝑁𝑁 2 + 2 ∙ℎ𝑝𝑝 2 + 2 ∙ℎ𝑝𝑝∙𝑛𝑛1 ∙𝑧𝑧+ 𝑛𝑛1 ∙𝑛𝑛2 ∙𝑧𝑧2 + 2 ∙𝑧𝑧𝑁𝑁𝑁𝑁∙ ൫2 ∙ℎ𝑝𝑝+ 𝑛𝑛1 ∙𝑧𝑧൯ቁቇ൱ቍቮ 𝑏𝑏 𝑒𝑒 (7) (7) where 𝑧𝑧 is the coordinate of the polygonal node, sometimes of node 𝑏𝑏, sometimes of node 𝑒𝑒; 𝑧𝑧𝐿𝐿𝐿𝐿 the coordinate of the neutral line; ℎ𝑝𝑝 the height of the polynomial part of the constitutive relationship, highlighted in Figure 5; 𝑐𝑐1 and 𝑐𝑐2 are the constants of the equation of the line 𝑙𝑙 of 𝑦𝑦 as a function of 𝑧𝑧; and 𝑛𝑛1, 𝑛𝑛2 e 𝑛𝑛3 are terms depending on the degree of the polynomial of the constitutive relation. The constants of line 𝑙𝑙 and the terms as a function of the exponent of the constitutive relationship are determined with 𝑐𝑐 𝑐𝑐1 = 𝑧𝑧𝑏𝑏∙𝑦𝑦𝑒𝑒−𝑧𝑧𝑒𝑒∙𝑦𝑦𝑏𝑏 𝑧𝑧𝑏𝑏−𝑧𝑧𝑒𝑒 and 𝑐𝑐2 = 𝑦𝑦𝑏𝑏−𝑦𝑦𝑒𝑒 𝑧𝑧𝑏𝑏−𝑧𝑧𝑒𝑒 (8) and (9) 𝑛𝑛1 = 𝑛𝑛+ 1, 𝑛𝑛2 = 𝑛𝑛+ 2 and 𝑛𝑛3 = 𝑛𝑛+ 3 (10), (11) and (12) 𝑛𝑓 𝑐𝑐1 = 𝑧𝑧𝑏𝑏∙𝑦𝑦𝑒𝑒−𝑧𝑧𝑒𝑒∙𝑦𝑦𝑏𝑏 𝑧𝑧𝑏𝑏−𝑧𝑧𝑒𝑒 and 𝑐𝑐2 = 𝑦𝑦𝑏𝑏−𝑦𝑦𝑒𝑒 𝑧𝑧𝑏𝑏−𝑧𝑧𝑒𝑒 𝑛𝑛1 = 𝑛𝑛+ 1, 𝑛𝑛2 = 𝑛𝑛+ 2 and 𝑛𝑛3 = 𝑛𝑛+ 3 𝑛𝑓 𝑐𝑐1 = 𝑧𝑧𝑏𝑏∙𝑦𝑦𝑒𝑒−𝑧𝑧𝑒𝑒∙𝑦𝑦𝑏𝑏 𝑧𝑧𝑏𝑏−𝑧𝑧𝑒𝑒 and 𝑐𝑐2 = 𝑦𝑦𝑏𝑏−𝑦𝑦𝑒𝑒 𝑧𝑧𝑏𝑏−𝑧𝑧𝑒𝑒 𝑛𝑛𝑛𝑛𝑛𝑛 (8) and (9) (10), (11) and (12) 𝑛𝑛1 = 𝑛𝑛+ 1, 𝑛𝑛2 = 𝑛𝑛+ 2 and 𝑛𝑛3 = 𝑛𝑛+ 3 where 𝑛𝑛 is defined in [31] as function of 𝑓𝑓𝑐𝑐𝑐𝑐. For the part of the constitutive relation where the stress is constant, that is, specific strains smaller than 𝜀𝜀𝑐𝑐2 and larger 𝜀𝑐𝑐 where 𝑛𝑛 is defined in [31] as function of 𝑓𝑓𝑐𝑐𝑐𝑐. h f h i i l i h h i h i ifi i ll h𝜀 d l𝜀 where 𝑛𝑛 is defined in [31] as function of 𝑓𝑓𝑐𝑐𝑐𝑐. where 𝑛𝑛 is defined in [31] as function of 𝑓𝑓𝑐𝑐𝑐𝑐. For the part of the constitutive relation where the stress is constant that is specific strains smaller than 𝜀𝜀𝑐 and larger𝜀𝑐𝑐 𝑛 [ ] 𝑓𝑓𝑐𝑐𝑐𝑐 For the part of the constitutive relation where the stress is constant, that is, specific strains smaller than 𝜀𝜀𝑐𝑐2 and larger than 𝜀𝜀𝑐𝑐𝑐𝑐. 3.2 Strain in the cross section during the construction steps and in ULS The equilibrium equations are a function of the cross-sectional strains. As the member is in equilibrium, these equations can be used to determine the deformations in the cross section. 7/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Assuming that the cross sections remain flat after the loading action, the deformations then vary linearly in relation to height. Thus, in a conventional cross section (consisting of a single casting step), the cross-sectional strains are completely determined by knowing the strain of a pair of points with known height. In the case of a cross section of the type analyzed in this work, the deformation of the initial step is defined by the deformations in the following regions: • 𝜀𝜀𝐴𝐴,𝑖𝑖 – strain at the layer A, in initial step; 𝜀𝑆𝑖 • 𝜀𝜀𝐴𝐴,𝑖𝑖 – strain at the layer A, in initial step; 𝜀𝑆𝑖 • 𝜀𝜀𝐴𝐴,𝑖𝑖 – strain at the layer A, in initial step; 𝜀𝑆𝑖 𝜀𝜀𝐴𝐴,𝑖𝑖 strain at the layer A, in initial step; • 𝜀𝜀𝑆𝑆,𝑖𝑖 – strain at the layer S, in initial step; • 𝜀𝜀𝑆𝑆,𝑖𝑖 – strain at the layer S, in initial step; • with the strains between layers A and B being null. • with the strains between layers A and B being null. 𝜀 The strains in the total step are defined by the strain in the following regions: 𝜀𝐴𝑡𝜀 • 𝜀𝜀𝐴𝐴,𝑡𝑡 – strain at the layer A, in total step; 𝜀𝑆𝑡𝜀 • 𝜀𝜀𝑆𝑆,𝑡𝑡 – strain at the layer S, in total step; 𝜀𝐵𝑎𝜀 • 𝜀𝜀𝐵𝐵,𝑎𝑎 – strain at the layer B in additional and total step; • 𝜀𝜀𝐴𝐴,𝑎𝑎 – strain at the layer A in additional and total step. These strains are illustrated in Figures 3 and 4. The evolution of strains in the cross section throughout the steps of the construction process is shown in Figure 6. With the composed section, admitting full interaction, the section behaves in a monolithic manner and the strain present in the total step are equal to the strain of the initial step plus strain developed in the additional step. The additional step is a virtual step that represents the addition of curvature and axial deformation, which occurs in the composed beam, between the initial step and the total step. 3.2 Strain in the cross section during the construction steps and in ULS For the newly formed composite beam, Bernoulli's hypothesis is also true, thus the addition of strain, as a function of height, is linear, and is valid for the entire cross section. Adding the strain of the additional step to the already developed strain in the initial step, the strain of the total step of the cross section is obtained. 𝜀 The strains in the additional step are defined by the strain in the following regions: 𝜀𝐵𝑎𝜀 • 𝜀𝜀𝐵𝐵,𝑎𝑎 – strain at the layer B, in additional step; 𝜀𝑆𝑎 • 𝜀𝜀𝐵𝐵,𝑎𝑎 – strain at the layer B, in additional step; • 𝜀𝜀𝑆𝑆,𝑎𝑎 – strain at the layer S, in additional step; Figure 6. Evolution of strain in defined steps. Figure 6. Evolution of strain in defined steps. Having defined the three steps, the following relations between the strains are now written (15) 𝜀𝜀𝐴𝐴,𝑡𝑡= 𝜀𝜀𝐴𝐴,𝑖𝑖+ 𝜀𝜀𝐴𝐴,𝑎𝑎 (16) 8/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt 𝜀 where 𝜀𝜀𝐴𝐴,𝑎𝑎 is the strain at the layer A in additional step given by: 𝜀𝜀𝜅𝑧𝑧 where 𝜀𝜀𝐴𝐴,𝑎𝑎 is the strain at the layer A in additional step given by: 𝜀𝜀𝜅𝑧𝑧 (17) 𝜀𝜀𝐴𝐴,𝑎𝑎= 𝜀𝜀𝐵𝐵,𝑎𝑎+ 𝜅𝜅𝑎𝑎∙(𝑧𝑧𝐵𝐵−𝑧𝑧𝐴𝐴), where 𝜅𝜅𝑎𝑎 is the curvature of the additional step calculated by the equation: 𝜅𝜀𝑆𝑎𝜀𝐵𝑎 where 𝜅𝜅𝑎𝑎 is the curvature of the additional step calculated by the equation (18) 𝜅𝜅𝑎𝑎= 𝜀𝜀𝑆𝑆,𝑎𝑎−𝜀𝜀𝐵𝐵,𝑎𝑎 𝑧𝑧𝐵𝐵−𝑧𝑧𝑆𝑆, 𝑧𝑧 where 𝑧𝑧𝑗𝑗 is the 𝑧𝑧 coordinate of the layer 𝑗𝑗. 𝜀𝜀𝜀𝜀 where 𝑧𝑧𝑗𝑗 is the 𝑧𝑧 coordinate of the layer 𝑗𝑗. 𝜀𝜀𝜀𝜀 Note that now, using relations (15), (16), (17) and (18), all the strains of the problem are determined if the deformations 𝜀𝜀𝐴𝐴,𝑖𝑖, 𝜀𝜀𝑆𝑆,𝑖𝑖, 𝜀𝜀𝐵𝐵,𝑎𝑎 and 𝜀𝜀𝑆𝑆,𝑎𝑎 are known. 𝜀𝐴𝑖𝜀𝑆𝑖 Note that now, using relations (15), (16), (17) and (18), all the strains of the problem are determined if the deformations 𝜀𝜀𝐴𝐴,𝑖𝑖, 𝜀𝜀𝑆𝑆,𝑖𝑖, 𝜀𝜀𝐵𝐵,𝑎𝑎 and 𝜀𝜀𝑆𝑆,𝑎𝑎 are known. 𝜀𝐴𝑖𝜀𝑆𝑖 Writing a system of equations with (2) and (3), for the initial step, it is possible to determine the strains 𝜀𝜀𝐴𝐴,𝑖𝑖 e 𝜀𝜀𝑆𝑆,𝑖𝑖. This process is called determination of the strains in initial step and is described in item 4 of this work. Once these strains are known, the strains domains of the additional step that promote ULS in the composite section are then defined, a process described in item 5. 3.2 Strain in the cross section during the construction steps and in ULS Once the strains domains are known, the process for evaluate the factored moment resistance for the composite cross section is described in item 6 and the design method is presented in item 7. 4 DETERMINATION OF THE STRAINS IN INITIAL STEP The initial step is defined as the instant immediately after the cast in place addition. In this situation, the addition concrete is fresh, therefore, it accommodates the strains and does not offer participation in the 𝑅𝑅𝑐𝑐𝑐𝑐,𝑖𝑖 e 𝑀𝑀𝑐𝑐𝑐𝑐,𝑖𝑖. The pre- strain, necessary to determine the prestressing force, must be calculated with the prestressing losses of the current instant. Furthermore, the factored internal forces 𝑁𝑁𝑑𝑑,𝑖𝑖 and 𝑀𝑀𝑑𝑑,𝑖𝑖, are determined with the loads at this moment: self- weight of the precast section and self-weight of the cast in place addition. 𝜀𝐴𝑖𝜀𝑆𝑖 To evaluate the moment resistance of the cross section, it is first necessary to evaluate the strains 𝜀𝜀𝐴𝐴,𝑖𝑖 and 𝜀𝜀𝑆𝑆,𝑖𝑖 that occur in the initial step. The NBR 6118:2014 allows to assume that the stress-strain relationship of concrete is linear if the stress is less than 50% of the compressive strength of concrete. To overcome this limitation, the non-linear constitutive relationship for the concrete in bending shown in Figure 2 will be assumed [31]. 𝜀𝜀 Usually at least the strain at one point in the cross section is known. This does not apply to the current situation since the strains 𝜀𝜀𝐴𝐴,𝑖𝑖 e 𝜀𝜀𝑆𝑆,𝑖𝑖 can assume any values within their limits. The problem requires the evaluation of strains in the initial step for arbitrary bending moment. Thus, the problem consists of the solution of the nonlinear system, written by applying Equations 2 and 3 in the initial step, with two variables to be evaluated, namely 𝜀𝜀𝐴𝐴,𝑖𝑖 e 𝜀𝜀𝑆𝑆,𝑖𝑖. The evaluation of Equations 2 and 3 presupposes a defined problem, that is, the concrete cross section as well as the position, area, and pre-strain, 𝜀𝜀𝑝𝑝𝑝𝑝é,𝑖𝑖, of the reinforcements must be known. 𝑓𝑥𝑓𝑥 To solve the system, the use of the damped Newton-Raphson method is proposed. This method consists of an iterative process to search for the root of a function 𝑓𝑓(𝑥𝑥𝑛𝑛), nonlinear, where the tangent 𝑓𝑓′(𝑥𝑥𝑛𝑛) of the current iteration 𝑛𝑛 is used to estimate the next candidate solution 𝑥𝑥𝑛𝑛+1. 𝑥𝑥𝛼 𝑥𝑥𝑛𝑛+1 = 𝑥𝑥𝑛𝑛−𝛼𝛼∙ 𝑓𝑓(𝑥𝑥𝑛𝑛) 𝑓𝑓′(𝑥𝑥𝑛𝑛) (19) Since the function of the problem is a vector function, the damped Newton-Raphson method is written as 𝒙𝑛𝒙𝑛𝛼𝑱𝒇𝒙𝑛𝒇𝒙𝑛 Since the function of the problem is a vector function, the damped Newton-Raphson method is written as 𝒙𝑛𝒙𝑛𝛼𝑱𝒇𝒙𝑛𝒇𝒙𝑛 𝒙𝒙𝑛𝑛+1 = 𝒙𝒙𝑛𝑛−𝛼𝛼∙𝑱𝑱𝒇𝒇(𝒙𝒙𝑛𝑛) −1 ∙𝒇𝒇(𝒙𝒙𝑛𝑛), 𝑱 (20) 𝒙𝒙𝑛𝑛+1 = 𝒙𝒙𝑛𝑛−𝛼𝛼∙𝑱𝑱𝒇𝒇(𝒙𝒙𝑛𝑛) −1 ∙𝒇𝒇(𝒙𝒙𝑛𝑛), 𝑱𝒇𝒙 where 𝑱𝑱𝒇𝒇(𝒙𝒙𝒏𝒏) is the Jacobian matrix defined on the vector function 𝒇𝒇(𝒙𝒙𝒏𝒏). 𝒇𝒇(𝒙𝒏𝒏)𝒇𝒙𝒏 or this problem, the Equation 20 is rewritten as: 4 DETERMINATION OF THE STRAINS IN INITIAL STEP For this problem, the Equation 20 is rewritten as: where 𝑱𝑱𝒇𝒇(𝒙𝒙𝒏𝒏) is the Jacobian matrix defined on the vector function 𝒇𝒇(𝒙𝒙𝒏𝒏). where 𝑱𝑱𝒇𝒇(𝒙𝒙𝒏𝒏) is the Jacobian matrix defined on the vector function 𝒇𝒇(𝒙𝒙𝒏𝒏). where 𝑱𝑱𝒇𝒇(𝒙𝒙𝒏𝒏) is the Jacobian matrix defined on the vector function 𝒇𝒇(𝒙𝒙𝒏𝒏). where 𝑱𝑱𝒇𝒇(𝒙𝒙𝒏𝒏) is the Jacobian matrix defined on the v For this problem, the Equation 20 is rewritten as: 𝒇𝒇(𝒙𝒏𝒏)𝒇𝒙𝒏 For this problem, the Equation 20 is rewritten as: For this problem, the Equation 20 is rewritten as: 9/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt ቄ𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑖𝑖ቅ 𝑛𝑛+1 = ቄ𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑖𝑖ቅ 𝑛𝑛 −𝛼𝛼∙൦ 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝑆𝑆,𝑖𝑖 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝑆𝑆,𝑖𝑖 ൪ 𝑛𝑛 −1 ∙ቊ𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖, 𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖, 𝜀𝜀𝑆𝑆,𝑖𝑖൯ቋ 𝑛𝑛 (21) ቄ𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑖𝑖ቅ 𝑛𝑛+1 = ቄ𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑖𝑖ቅ 𝑛𝑛 −𝛼𝛼∙൦ 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝑆𝑆,𝑖𝑖 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝑆𝑆,𝑖𝑖 ൪ 𝑛𝑛 −1 ∙ቊ𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖, 𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖, 𝜀𝜀𝑆𝑆,𝑖𝑖൯ቋ 𝑛𝑛 𝛼 (21) (21) The term 𝛼𝛼 is the damping factor determined as follows: 𝛼𝑆𝑆𝑆𝑆𝑛𝑝𝑆𝑆𝑆𝑆𝑛𝑝 The term 𝛼𝛼 is the damping factor determined as follows: 𝛼𝑆𝑆𝑆𝑆𝑛𝑝𝑆𝑆𝑆𝑆𝑛𝑝 1. For 𝛼𝛼= 1, evaluate ฯቄ𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆ቅ 𝑛𝑛 ฯ 𝑝𝑝 and ฯቄ𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆ቅ 𝑛𝑛+1 ฯ 𝑝𝑝 ; 2. Check if ฯቄ𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆ቅ 𝑛𝑛+1 ฯ 𝑝𝑝 < ฯቄ𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆ቅ 𝑛𝑛 ฯ 𝑝𝑝 ; 𝜀𝜀𝐴𝑖𝜀𝑆𝑖𝑛𝛼 3. if 2 is true, ቄ𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑖𝑖ቅ 𝑛𝑛+1 is accepted as a new iteration; 𝛼𝛼 3. if 2 is true, ቄ𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑖𝑖ቅ 𝑛𝑛+1 is accepted as a new iteration; 𝛼𝛼 𝑛𝑛+1 4. if 2 is false, 𝛼𝛼← 𝛼𝛼 2 and the process is repeated from 2. 4. if 2 is false, 𝛼𝛼← 𝛼𝛼 2 and the process is repeated from 2. 4. if 2 is false, 𝛼𝛼← 𝛼𝛼 2 and the process is repeated from 2. Figure 7 graphically shows the problem-solving strategy, which iteratively, the deformations will be approaching the roots of equations. Figure 7 graphically shows the problem-solving strategy, which iteratively, the deformations will be approaching the roots of equations. Figure 7. Problem of determination the strains in the initial step. 𝑆𝑆𝑆𝑆 Figure 7. Problem of determination the strains in the initial step. 𝑆𝑆𝑆𝑆 The Jacobian matrix requires the evaluation of partial derivatives of the 𝑆𝑆𝑆𝑆 and 𝑆𝑆𝑆𝑆 functions. These derivatives are approximated using central finite differences resulting in Equations 22 to 25. 5 DERMINATION OF THE ULS REGIONS OF COMPOSITE CROSS SECTIONS ULS strain domains can be defined. In this work, the concept of domain is replaced by Santos [34] proposal. Santos proposes the grouping of one or more ULS domains in regions that present the same rupture mechanism. 𝜀𝜀𝜀 In composite sections, the ULS can be characterized by the individual or combined state of three situations: • conventional failure due to excessive plastic strain of the passive reinforcement, 𝜀𝜀𝑆𝑆,𝑡𝑡= 𝜀𝜀𝑠𝑠𝑢𝑢, where 𝜀𝜀𝑠𝑠𝑠𝑠= 10 ‰ [31], region 3; 𝜀𝐵𝑡𝜀𝑐𝑐𝜀𝜀 In composite sections, the ULS can be characterized by the individual or combined state of three situations: • conventional failure due to excessive plastic strain of the passive reinforcement, 𝜀𝜀𝑆𝑆,𝑡𝑡= 𝜀𝜀𝑠𝑠𝑢𝑢, where 𝜀𝜀𝑠𝑠𝑠𝑠= 10 ‰ [31], region 3; 𝜀𝐵𝑡𝜀𝑐𝑐𝜀𝜀 [ ], g ; • conventional failure by limit-shortening of concrete in the cast in place addition, 𝜀𝜀𝐵𝐵,𝑡𝑡= [ ], g ; conventional failure by limit-shortening of concrete in the cast in place addition, 𝜀𝜀𝐵𝐵,𝑡𝑡= 𝜀𝜀𝑐𝑐𝑐𝑐, regi conventional failure by limit shortening of concrete in the cast in place addition, 𝜀𝜀𝐵𝐵,𝑡𝑡 𝜀𝜀𝑐𝑐𝑐𝑐, region 2B; • conventional failure by limit-shortening of concrete of the precast section, 𝜀𝜀𝐴𝐴,𝑡𝑡= 𝜀𝜀𝑐𝑐𝑐𝑐, region 2A. y g p𝜀 𝐵𝐵,𝑡𝑡𝜀 𝑐𝑐𝑐𝑐 • conventional failure by limit-shortening of concrete of the precast section, 𝜀𝜀𝐴𝐴,𝑡𝑡= 𝜀𝜀𝑐𝑐𝑐𝑐, region 2 ,𝑡𝜀𝑐𝑐 onventional failure by limit-shortening of concrete of the precast section, 𝜀𝜀𝐴𝐴,𝑡𝑡= 𝜀𝜀𝑐𝑐𝑐𝑐, region 2A. Figure 8 illustrates the deformation of the composite cross section, in ULS, exemplifying the characteristic rupture of each strain region. Figure 8 illustrates the deformation of the composite cross section, in ULS, exemplifying the characteristic rupture of each strain region. Figure 8. Deformation of the composite cross section in ULS in the total step. Figure 8. Deformation of the composite cross section in ULS in the total step. A cross section, depending on its resulting strains in the initial step, can present three possible trajectories of ULS regions. A trajectory is understood as the sequence in which the regions are presented, analyzing the deformations in ULS, when the normal force 𝑁𝑁𝑑𝑑 assumes values in decreasing order from the maximum to the minimum allowed. Thus, the trajectory of ULS regions is determined by evaluating the level of strain value 𝜀𝜀𝐴𝐴,𝑖𝑖 in relation to the limits 𝜀𝜀𝐴𝐴𝑖𝑖23𝐵𝐵 and 𝜀𝜀𝐴𝐴𝑖𝑖12𝐵𝐵, thus classifying the precast cross section in one of three categories: 𝜀𝐴𝑖𝜀𝐴𝑖𝐵 A cross section, depending on its resulting strains in the initial step, can present three possible trajectories of ULS regions. 4 DETERMINATION OF THE STRAINS IN INITIAL STEP This technique approximates the tangent slope of the function, at the point of interest, by the slope of a line formed by two points on the function, distant ℎ from each other. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 10/21 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 ≅ 𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖+ℎ,𝜀𝜀𝑆𝑆,𝑖𝑖൯−𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖−ℎ,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 2∙ℎ (22) 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝑆𝑆,𝑖𝑖 ≅ 𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖+ℎ൯−𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖−ℎ൯ 2∙ℎ (23) 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 ≅ 𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖+ℎ,𝜀𝜀𝑆𝑆,𝑖𝑖൯−𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖−ℎ,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 2∙ℎ (24) (22) (23) 𝜕𝜕𝜕𝜕𝜕𝜕൫𝜀𝜀𝐴𝐴,𝑖𝑖,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 𝜕𝜕𝜀𝜀𝐴𝐴,𝑖𝑖 ≅ 𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖+ℎ,𝜀𝜀𝑆𝑆,𝑖𝑖൯−𝑆𝑆𝑆𝑆൫𝜀𝜀𝐴𝐴,𝑖𝑖−ℎ,𝜀𝜀𝑆𝑆,𝑖𝑖൯ 2∙ℎ (24) Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 10/21 (25) (26) In the examples in this article, it is admitted 𝜉𝜉1 = 10−2 when evaluating 𝑆𝑆𝑁𝑁 and 𝑆𝑆𝑆𝑆 in kN and in kN ∙m, respectively. The distance ℎ for evaluating the partial derivatives was adopted equal to 10−3 for the first iteration and 10−9 for the other iterations, using double-precision. The first iteration starts from the strains 𝜀𝜀𝐴𝐴,𝑖𝑖= 0 and 𝜀𝜀𝑆𝑆,𝑖𝑖= 0. In the examples in this article, it is admitted 𝜉𝜉1 = 10−2 when evaluating 𝑆𝑆𝑁𝑁 and 𝑆𝑆𝑆𝑆 in kN and in kN ∙m, respectively. The distance ℎ for evaluating the partial derivatives was adopted equal to 10−3 for the first iteration and 10−9 for the other iterations, using double-precision. The first iteration starts from the strains 𝜀𝜀𝐴𝐴,𝑖𝑖= 0 and 𝜀𝜀𝑆𝑆,𝑖𝑖= 0. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 5 DERMINATION OF THE ULS REGIONS OF COMPOSITE CROSS SECTIONS A trajectory is understood as the sequence in which the regions are presented, analyzing the deformations in ULS, when the normal force 𝑁𝑁𝑑𝑑 assumes values in decreasing order from the maximum to the minimum allowed. 𝜀𝐴𝑖𝜀𝜀 Thus, the trajectory of ULS regions is determined by evaluating the level of strain value 𝜀𝜀𝐴𝐴,𝑖𝑖 in relation to the limits 𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵 and 𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵, thus classifying the precast cross section in one of three categories: 𝜀𝐴𝑖𝜀𝐴𝑖𝐵 • lightly compressed precast section, when 𝜀𝜀𝐴𝐴,𝑖𝑖> 𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵, with ULS region trajectory from 3 to 2B; • lightly compressed precast section, when 𝜀𝜀𝐴𝐴,𝑖𝑖> 𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵, with ULS region trajectory from 11/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt • moderately compressed precast section, when 𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵< 𝜀𝜀𝐴𝐴,𝑖𝑖≤𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵, with ULS region trajectory from 3 to 2B to 2A; 𝜀𝐴𝑖𝜀𝐴𝑖𝐵𝜀𝜀 • moderately compressed precast section, when 𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵< 𝜀𝜀𝐴𝐴,𝑖𝑖≤𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵, with ULS region trajectory from 3 to 2B to 2A; 𝜀𝐴𝑖𝜀𝐴𝑖𝐵𝜀𝜀 2A; • heavily compressed precast section, when 𝜀𝜀𝐴𝐴,𝑖𝑖≤𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵, with ULS region trajectory from 3 to 2A. 𝜀𝐴𝑖𝐵𝜀𝐴𝑖𝐵 ,𝑖𝜀𝐴 ,𝑖,𝐵 The limits 𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵 and 𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵 are determined by Equations 27 and 28 as a function of the strains obtained in the initial step. 𝜀𝜀𝜀𝜀 𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵= ൫𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚+ 𝜀𝜀𝐶𝐶,𝑖𝑖൯∙ 𝑧𝑧𝐵𝐵−𝑧𝑧𝐴𝐴 𝑧𝑧𝐵𝐵−𝑧𝑧𝐶𝐶 (27) 𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵= ൫𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝑦𝑦𝑦𝑦+ 𝜀𝜀𝑆𝑆,𝑖𝑖 ൯∙ 𝑧𝑧𝐵𝐵−𝑧𝑧𝐴𝐴 𝑧𝑧𝐵𝐵−𝑧𝑧𝑆𝑆 (28) 𝜀𝜀 𝜀𝜀𝐴𝐴,𝑖𝑖,12𝐵𝐵= ൫𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚+ 𝜀𝜀𝐶𝐶,𝑖𝑖൯∙ 𝑧𝑧𝐵𝐵−𝑧𝑧𝐴𝐴 𝑧𝑧𝐵𝐵−𝑧𝑧𝐶𝐶 𝜀𝜀𝐴𝐴,𝑖𝑖,23𝐵𝐵= ൫𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝑦𝑦𝑦𝑦+ 𝜀𝜀𝑆𝑆,𝑖𝑖 ൯∙ 𝑧𝑧𝐵𝐵−𝑧𝑧𝐴𝐴 𝑧𝑧𝐵𝐵−𝑧𝑧𝑆𝑆 𝜀 (27) (28) where 𝜀𝜀𝑠𝑠𝑠𝑠 is the maximum strain allowed in the passive reinforcement, 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚 the strain at layer C of the boundary between regions 1 and 2, being 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚= 0, and 𝜀𝜀𝐶𝐶,𝑖𝑖 the strain at layer C of the initial step given by: 𝜀𝜀𝜅𝑧𝑧 where 𝜀𝜀𝑠𝑠𝑠𝑠 is the maximum strain allowed in the passive reinforcement, 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚 the strain at layer C of the boundary between regions 1 and 2, being 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚= 0, and 𝜀𝜀𝐶𝐶,𝑖𝑖 the strain at layer C of the initial step given by: 𝜀𝜀𝜅𝑧𝑧 (29) 𝜀𝜀𝐶𝐶,𝑖𝑖= 𝜀𝜀𝐴𝐴,𝑖𝑖+ 𝜅𝜅𝑖𝑖∙(𝑧𝑧𝐴𝐴−𝑧𝑧𝐶𝐶), 𝜅 where 𝜅𝜅𝑖𝑖 is the curvature of the initial step evaluated by the equation: 𝜅𝜀𝜀 where 𝜅𝜅𝑖𝑖 is the curvature of the initial step evaluated by the equation: 𝜅𝜀𝜀 where 𝜅𝜅𝑖𝑖 is the curvature of the initial step evaluated by the equation: 𝜅𝜀𝜀 𝜅𝜅𝑖𝑖= 𝜀𝜀𝑆𝑆,𝑖𝑖−𝜀𝜀𝐴𝐴,𝑖𝑖 𝑧𝑧𝐴𝐴−𝑧𝑧𝑆𝑆. (30) Region 1 is not of interest to this work as it exceeds the limits of 𝑥𝑥𝑑𝑑 ⁄ for beams of item 14.6.4.3 of NBR 6118 [31]. 5 DERMINATION OF THE ULS REGIONS OF COMPOSITE CROSS SECTIONS The strain regions in ULS for the composite section are defined in relation to the strains of the additional step Region 1 is not of interest to this work as it exceeds the limits of 𝑥𝑥𝑑𝑑 ⁄ for beams of item 14.6.4.3 of NBR 6118 [31]. Region 1 is not of interest to this work as it exceeds the limits of 𝑥𝑥𝑑𝑑 ⁄ for beams of item 14.6.4.3 of NBR 6118 [31]. The strain regions in ULS for the composite section are defined in relation to the strains of the additional step. Figure 9 represents the strain regions in ULS for composite beams with lightly, moderately and heavily compressed precast section. The strain regions in ULS for the composite section are defined in relation to the strains of the additional step. Figure 9 represents the strain regions in ULS for composite beams with lightly, moderately and heavily compressed precast section. Figure 9. Strain regions in ULS in the additional step. Figure 9. Strain regions in ULS in the additional step. Each strain region in ULS is defined by a strain group of that have the same value at a certain level of the cross section. Each region has a limited curvature as a function of the compression level of the precast section. 𝜀𝐴𝑎𝑚𝑚𝑚𝜀𝑆𝑎𝑚𝑚𝑚𝜀𝐶𝑎𝑚𝑚𝑚𝜀𝜀𝜀 In Figure 9, 𝜀𝜀𝐴𝐴,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚, 𝜀𝜀𝑆𝑆,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚 and 𝜀𝜀𝐶𝐶,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚 are the strains in the additional step, which provides that the strains in the total step are respectively equal 𝜀𝜀𝑐𝑐𝑐𝑐, 𝜀𝜀𝑠𝑠𝑠𝑠 and 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚 at the layers A, S and C. These are defined with Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 12/21 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt 𝜀𝜀𝜀 (31) 𝜀𝜀𝐴𝐴,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝑠𝑠𝑠𝑠−𝜀𝜀𝑆𝑆,𝑖𝑖 𝜀𝜀𝐶𝐶,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚−𝜀𝜀𝐶𝐶,𝑖𝑖. 𝜀𝜀𝐴𝐴,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝑆𝑆,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝑠𝑠𝑠𝑠−𝜀𝜀𝑆𝑆,𝑖𝑖 𝜀𝜀𝐶𝐶,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚−𝜀𝜀𝐶𝐶,𝑖𝑖. 𝜀𝜀𝐴𝐴,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝑐𝑐𝑐𝑐−𝜀𝜀𝐴𝐴,𝑖𝑖 𝜀𝜀𝜀 (32) 𝜀𝜀𝑆𝑆,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝑠𝑠𝑠𝑠−𝜀𝜀𝑆𝑆,𝑖𝑖 𝜀𝜀𝜀 (33) 𝜀𝜀𝐶𝐶,𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚= 𝜀𝜀𝐶𝐶,𝑡𝑡,𝑚𝑚𝑚𝑚𝑚𝑚−𝜀𝜀𝐶𝐶,𝑖𝑖. Knowing the limits of the strain regions in ULS, it is possible to evaluate in which region the solution of a defined problem is located. This region is obtained by evaluating the sign of the 𝑆𝑆𝑆𝑆 function when the strain in the additional step is equal to the limit between adjacent regions. Observing the behavior of the signal of this function, it is then possible to evaluate the region in which Equation 2 is satisfied, which is the region where the curvature and axial deformation solution to the problem is found. ION OF FACTORED MOMENT RESISTANCE FOR THE COMPOSITE CROSS SECTION For a defined problem, that is, where the cross section and the characteristics of the reinforcements are known, it is now possible to determine the factored moment resistance 𝑀𝑀𝑅𝑅𝑅𝑅 of the composite section. Having determined the region in which the ULS deformation is located, the problem boils down to looking, within the solution space contained in the defined region, for the curvature that determines the root of the Equation 2. The root can be obtained by several iterative numerical methods such as the bisection method, the false position method, the secant method or even the Newton-Raphson method already discussed here. As it is necessary that the solution process does not exceed the limits of the evaluated region and as Equation 2, as it is evaluated here, is not derivable, the false position method becomes the most suitable within the considered methods. 𝑓𝑥𝑎𝑏𝑎𝑏 The false position method fetches the root of the function 𝑓𝑓(𝑥𝑥) contained in a known initial range [𝑎𝑎0, 𝑏𝑏0], where iteratively the search range is reduced so that the root of the function is still contained in the new range [𝑎𝑎𝑘𝑘, 𝑏𝑏𝑘𝑘]. The new range is defined in iteration 𝑘𝑘 by 𝑐𝑏 𝑐𝑐𝑘𝑘= 𝑏𝑏𝑘𝑘− 𝑓𝑓(𝑏𝑏𝑘𝑘)∙(𝑏𝑏𝑘𝑘−𝑎𝑎𝑘𝑘) 𝑓𝑓(𝑏𝑏𝑘𝑘)−𝑓𝑓(𝑎𝑎𝑘𝑘) 𝑐 (34) where 𝑐𝑐𝑘𝑘 is the root of the secant that passes through the points [𝑎𝑎𝑘𝑘, 𝑓𝑓(𝑎𝑎𝑘𝑘)] and [𝑏𝑏𝑘𝑘, 𝑓𝑓(𝑏𝑏𝑘𝑘)]. Once 𝑐𝑐𝑘𝑘 is determined, the response interval is reduced by making 𝑎𝑎𝑘𝑘+1 = 𝑐𝑐𝑘𝑘 and 𝑏𝑏𝑘𝑘+1 = 𝑏𝑏𝑘𝑘, if 𝑓𝑓(𝑎𝑎𝑘𝑘) and 𝑓𝑓(𝑐𝑐𝑘𝑘) have the same sign, otherwise, 𝑎𝑎𝑘𝑘+1 = 𝑎𝑎𝑘𝑘and 𝑏𝑏𝑘𝑘+1 = 𝑐𝑐𝑘𝑘 where 𝑐𝑐𝑘𝑘 is the root of the secant that passes through the points [𝑎𝑎𝑘𝑘, 𝑓𝑓(𝑎𝑎𝑘𝑘)] and [𝑏𝑏𝑘𝑘, 𝑓𝑓(𝑏𝑏𝑘𝑘)]. Once 𝑐𝑐𝑘𝑘 is determined, the response interval is reduced by making 𝑎𝑎𝑘𝑘+1 = 𝑐𝑐𝑘𝑘 and 𝑏𝑏𝑘𝑘+1 = 𝑏𝑏𝑘𝑘, if 𝑓𝑓(𝑎𝑎𝑘𝑘) and 𝑓𝑓(𝑐𝑐𝑘𝑘) have the same sign, otherwise, 𝑎𝑎𝑘𝑘+1 = 𝑎𝑎𝑘𝑘 and 𝑏𝑏𝑘𝑘+1 = 𝑐𝑐𝑘𝑘. Figure 10 illustrates the solution strategy of the false position method, assuming linear behavior for the function in the assigned solution interval, and successively shortens this interval until it reaches a stopping criterion such 𝑓𝑐𝜉 𝑓𝑓(𝑐𝑐𝑘𝑘) < 𝜉𝜉2 𝜉 (35) where 𝜉𝜉2 is the error allowed for the function 𝑓𝑓(𝑥𝑥) in this process. where 𝜉𝜉2 is the error allowed for the function 𝑓𝑓(𝑥𝑥) in this process. Figure 10. Solution strategy of the false position method. Figure 10. Solution strategy of the false position method. 13/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 Rev. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 ION OF FACTORED MOMENT RESISTANCE FOR THE COMPOSITE CROSS SECTION IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt To define the deformation in ULS using the false position method, the function 𝑓𝑓(𝑥𝑥) assumes Equation 2 written in the total step, the variable 𝑥𝑥 is the curvature of the additional step 𝜅𝜅𝑎𝑎 and the search interval [𝑎𝑎0, 𝑏𝑏0] is defined as the limiting curvatures of the strain region in ULS for the section in question ൣ𝜅𝜅𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚, 𝜅𝜅𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚൧. 𝑀𝑑𝑀 Once the additional step strains are defined, the moment 𝑀𝑀𝑑𝑑 that satisfies Equation 3 written in total step is the factored moment resistance 𝑀𝑀𝑅𝑅𝑅𝑅 of the composite cross section. 8 EXAMPLES The method proposed here to design composite cross sections was implemented in a routine in MATLAB. This routine was then used to develop the following examples. 7 DESIGN METHOD OF COMPOSITE CROSS SECTION The factored moment resistance evaluation process, described in item 6, requires a complete definition of the problem, including the area and position of the reinforcement. For design, it is necessary to organize the problem to determine the amount of passive reinforcement needed so that the composite section can withstand the acting factored moment 𝑀𝑀𝑆𝑆𝑆𝑆. 𝑀𝑅𝑅 Using the processes described in items 4, 5 and 6 to determine the 𝑀𝑀𝑅𝑅𝑅𝑅 it is possible to use the false position method to determine the area of passive reinforcement that satisfies the design equation 𝑀𝑀 (36) 𝑀𝑀𝑅𝑅𝑅𝑅−𝑀𝑀𝑆𝑆𝑆𝑆= 0. To design the passive reinforcement area, using the false position method, the function 𝑓𝑓(𝑥𝑥) assumes Equation 36, the independent variable 𝑥𝑥 is the passive reinforcement area 𝐴𝐴𝑠𝑠 and the search interval [𝑎𝑎0, 𝑏𝑏0] is defined as the assumed minimum and maximum reinforcement areas for this cross section ൣ𝐴𝐴𝑠𝑠,𝑚𝑚𝑚𝑚𝑚𝑚, 𝐴𝐴𝑠𝑠,𝑚𝑚𝑚𝑚𝑚𝑚൧. To design the passive reinforcement area, using the false position method, the function 𝑓𝑓(𝑥𝑥) assumes Equation 36, the independent variable 𝑥𝑥 is the passive reinforcement area 𝐴𝐴𝑠𝑠 and the search interval [𝑎𝑎0, 𝑏𝑏0] is defined as the assumed minimum and maximum reinforcement areas for this cross section ൣ𝐴𝐴𝑠𝑠,𝑚𝑚𝑚𝑚𝑚𝑚, 𝐴𝐴𝑠𝑠,𝑚𝑚𝑚𝑚𝑚𝑚൧. Figure 11. Flowchart of the composite section design process. Figure 11. Flowchart of the composite section design process. 14/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt During the design process, summarized in Figure 11, the 𝑆𝑆𝑆𝑆 function is evaluated for different areas of passive reinforcement. The discretization procedure of the reinforcement elements in the cross section is presented in the next item, 7.1, since the positioning of these elements is necessary for the evaluation of Equation 3. 7.1 Discretization model of reinforcement elements with continuous variation The reinforcements are positioned in the cross section in layers that must respect the horizontal and vertical spacing of [31]. Furthermore, as this is a real problem, the number of rebars is defined by an integer number. The problem written with an amount of reinforcement defined by an integer, presents a discontinuous relationship between reinforcement area and moment resistance. This discontinuity makes it difficult to use numerical methods to determine the answer to the design problem. 𝐴𝑠𝑀𝑅𝑅𝑞𝐴 To promote a continuous relationship between 𝐴𝐴𝑠𝑠 and 𝑀𝑀𝑅𝑅𝑅𝑅, the discretization of the individual elements representing the steel rebars is replaced here by 𝑞𝑞 rectangular elements. These elements have the sum of their areas equivalent to 𝐴𝐴𝑠𝑠, they are equally spaced and arranged in a reinforced region of the cross section that estimates the real positioning of the discrete rebars 𝑏𝑏𝑏 The reinforced region is defined by the dimensions 𝑏𝑏 and ℎ, where 𝑏𝑏 is the displacement of the axis of the outermost reinforcement and ℎ the axis of the highest reinforcement layer in the cross section. Dimensions 𝑏𝑏 and ℎ are determined by a linear function in relation to 𝐴𝐴𝑠𝑠 obtained from the extreme values 𝑏𝑏𝑚𝑚𝑚𝑚𝑚𝑚, ℎ𝑚𝑚𝑚𝑚𝑚𝑚, 𝑏𝑏𝑚𝑚𝑚𝑚𝑚𝑚 e ℎ𝑚𝑚𝑚𝑚𝑚𝑚, established by the actual positioning of the minimum and maximum reinforcement in the cross section, Figure 12. Figure 12. Real positioning and continuous representation of passive reinforcement. 𝐴𝑀 Figure 12. Real positioning and continuous representation of passive reinforcement. 𝐴𝑀 The use of rectangular elements makes the relationship between 𝐴𝐴𝑠𝑠 and 𝑀𝑀𝑅𝑅𝑅𝑅 continuous but introduces a difference in the position of the resulting 𝑅𝑅𝑠𝑠𝑠𝑠, in relation to the real positioning of the passive reinforcement rebars. Thus, at the end of the process, the factored moment resistance of the beam must be verified with the real positioning of the reinforcement rebars. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 8.1 Comparison with the experimental results In order to roughly estimate the influence of the simplifying hypothesis of the model, the ultimate moment evaluated by this methodology will be compared with the experimental results obtained by [35] and [36]. In their work, two full- scale reinforced concrete beams were tested to failure in bending. These beams were casted in two steps, with the precast section being loaded prior to the second casting, simulating the effect investigated in this paper. 15/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Figure 13. Cross-section of beam-1 and beam-2. Figure 13. Cross-section of beam-1 and beam-2. The cross sections of the tested beams are shown in Figure 13. Table 1 shows the estimated strength of the concrete of the beams measured by the strength of the cylindrical specimens at different ages coincident with the test steps. The yielding stress of the steel reinforcement, obtained from tensile tests on samples extracted from the rebars, is 𝑓𝑓𝑦𝑦= 490 MPa. The bending moment acting on the precasted beam in the initial step is 𝑀𝑀𝑖𝑖= 531 kN ∙m for the beam-1 and 𝑀𝑀𝑖𝑖= 385 kN ∙m for the beam-2, according to [35] and [36]. Table 1. Strength of the cylindrical specimens at different ages. First cast - initial step First cast - total step Second cast - total step 𝒇𝒇𝒄𝒄,𝒊𝒊 (𝐌𝐌𝐌𝐌𝐌𝐌) 𝒇𝒇𝒄𝒄,𝒕𝒕 (𝐌𝐌𝐌𝐌𝐌𝐌) 𝒇𝒇𝒄𝒄,𝒕𝒕 (𝐌𝐌𝐌𝐌𝐌𝐌) Beam-1 51 59 32 Beam-2 66 68 25 Table 1. Strength of the cylindrical specimens at different ages. Removing the effect of the Rüsch coefficient from the concrete constitutive relationship and using the strength safety factor 𝛾𝛾𝑐𝑐= 𝛾𝛾𝑠𝑠= 1 the ultimate moment of beam-1 and beam-2 is estimated as shown in Table 2. Figures 14 and 15 show the strains of the cross section, as well as the resulting forces in equilibrium in the initial, additional and total steps evaluation of the ultimate moment of beam-1 and beam-2, respectively. Figure 14. Strain diagram in ULS for the beam-1. Figure 14. Strain diagram in ULS for the beam-1. 16/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Figure 15. Strain diagram in ULS for the beam-2. Figure 15. Strain diagram in ULS for the beam-2. Table 2. Summary of test and estimated ultimate moment. Table 2. Summary of test and estimated ultimate moment. 8.1 Comparison with the experimental results Ultimate moment in total step 𝑴𝑴𝒕𝒕 (𝐤𝐤𝐤𝐤∙𝐦𝐦) Error (%) Experimental [35] and [36] Estimated by this work Beam-1 1953,5 1927,7 (limited by the steel strain equal to 10 ‰) 1,3 Beam-2 2346,2 1922,1 (limited by the concrete strain equal to 3,5 ‰) 18,1 The differences shown in Table 2 can be explained by the simplifications of the proposed model. The constitutive relationship used for the concrete limits the maximum strain to 𝜀𝜀𝑐𝑐𝑐𝑐, in the experiment, it is likely that the maximum strain of the concrete was higher. This limitation hinders the development of larger reactions of compressed concrete close to layer A because the curvature is limited by the maximum deformation in layer B. This effect is more pronounced in beam- 2 because, in this beam, the failure was characterized by the maximum strain of the concrete of the layer B. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 8.2 Numerical Example To evaluate the differences introduced in the design when a cross section is casted in a single step and in two steps, it is proposed the design of the cross section in Figure 16. In this example the factored moment in the precast section when the second step is being casted is 15.792 kN ∙m, and the factored moment resistance in ULS of composite beam need to be 42.658 kN ∙m. Figure 16. Proposed beam for numerical example. Figure 16. Proposed beam for numerical example. 17/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Figures 17 and 18 show the strains in the ULS, as well the resulting forces in equilibrium in the cross section when the beam in Figure 16 is design considering single-step and two-step casting, respectively. Figures 17 and 18 show the strains in the ULS, as well the resulting forces in equilibrium in the cross section when the beam in Figure 16 is design considering single-step and two-step casting, respectively. 2 Developing the design of the beam assuming that the entire beam is casted in a single step, 177,1 cm2 of passive reinforcement area is needed. Figure 17. Strain diagram in ULS for the example assuming casting in single step. 𝐴𝐴𝑠𝑠= 177,1 cm2. Figure 17. Strain diagram in ULS for the example assuming casting in single step. 𝐴𝐴𝑠𝑠= 177,1 cm2. In contrast, if the beam in Figure 16 is built by two casting steps, with the precast beam not being shored during the second casting step, the design results in a 189,0 cm2 of passive reinforcement area. In contrast, if the beam in Figure 16 is built by two casting steps, with the precast beam not being shored during the second casting step, the design results in a 189,0 cm2 of passive reinforcement area. Figure 18. Strain diagram in ULS for the example assuming two casting steps. 𝐴𝐴𝑠𝑠= 189,0 cm2. Figure 18. Strain diagram in ULS for the example assuming two casting steps. 𝐴𝐴𝑠𝑠= 189,0 cm2. Analyzing the two designs, it is observed that 6,7% more passive reinforcement is required, when the beam is built in two casting steps, in relation to de construction in a single cast. 9. CONCLUSIONS Here it is proposed a methodology to design prestressed sections constructed in two casting steps, built without additional shoring, where the strain discontinuity in the cross section, introduced by the construction process, is considered. The proposed design process consists of iteratively varying the area of passive reinforcement until the factored moment resistance is equal to the requested factored moment. Initially the strains in the precast section are evaluated and, in sequence, the strains in ULS that satisfy the equilibrium and compatibility equations are found. The damped Newton-Raphson method for evaluating the deformations in the initial step proved to be efficient and numerically stable. The false position method, here used to satisfy the equilibrium and the design equation, is sufficiently efficient and necessary because the search need to be developed within the validity limits of the equations, different from the Newton Raphson method, which does not respect the boundary conditions. Also, within the solutions adopted to make the method viable, the discretization of the reinforcement elements with continuous variation allowed the solution of the design equation using iterative numerical processes seamlessly. With all the solutions proposed here, this method can be easily implemented in computational routines for verification and design of composite elements with precision. With all the solutions proposed here, this method can be easily implemented in computational routines for verification and design of composite elements with precision. Regarding the comparison of the experimental results, the accuracy of the model is high when the failure is characterized by the limit strain of the reinforcement, and lower when the limit strain occurs in concrete. It should be noted that this divergence is due to a simplification of the constitutive relationship of the NBR 6118:2014 and that the estimated resistance is lower than that measured in the test. To obtain an ultimate moment with better experimental and theoretical fitting a more realistic concrete constitutive law is recommended. The numerical example results show that the section built in two steps, resulted in a moment resistance 3,0% lower in relation to the verification considering a single casting step. These results may vary when compared to other cross sections, different ratios between the height of each cast stage, different strain in the precast section and different prestressing level. To evaluate the relation between each of these variables in the factored moment resistance further investigation is needed. 9. CONCLUSIONS For a better understanding of the extent of the effects investigated here, it is very important that more experimental tests of concrete beams casted in two stages with higher loading level in the precast section are carried out due to the lack of tests like this in the literature. It is also proposed that future investigations evaluate experimentally and numerically the effects of time-dependent strain, such as shrinkage and creep, in the moment resistance of concrete composite cross sections. 8.2 Numerical Example 𝐴𝑠 Analyzing the two designs, it is observed that 6,7% more passive reinforcement is required, when the beam is built in two casting steps, in relation to de construction in a single cast. 𝐴𝑠 2 Using the same reinforcement area, 𝐴𝐴𝑠𝑠= 189,0 cm2, it is possible to evaluate and compare the factored moment resistance of the beam in Figure 16, considering the construction in single casting step and in two casting steps. Evaluating the beam constructed in a single step, the factored moment resistance is 43.963 kN ∙m. When the factored moment resistance is evaluated considering two casting steps, it is equal to 42.658 kN ∙m. For the example in Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 18/21 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt Figure 16, the beam with two casting stages provides a factored resistance moment 3,0% lower than the beam with a single casting step with the same reinforcement area. Figure 16, the beam with two casting stages provides a factored resistance moment 3,0% lower than the beam with a single casting step with the same reinforcement area. ACKNOWLEDGEMENTS This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. REFERENCES [1] R. Tomek, "Advantages of precast concrete in highway infrastructure construction," Procedia Eng., vol. 196, pp. 176–180, 2017, http://dx.doi.org/10.1016/j.proeng.2017.07.188. [1] R. Tomek, "Advantages of precast concrete in highway infrastructure construction," Procedia Eng., vol. 196, pp. 176–180, 2017, http://dx.doi.org/10.1016/j.proeng.2017.07.188. [2] Associação Brasileira de Normas Técnicas, Projeto e Execução de Estruturas de Concreto Pré-Moldado, ABNT NBR 9062:2017, 2017. [3] T. I. Campbell, V. K. R. Kodur, S. M. R. Lopes, J. Harrop, and A. E. Gamble, "Study of moment redistribution in prestressed concrete beams," J. Struct. Eng., vol. 125, no. 3, pp. 351–352, Mar 1999, http://dx.doi.org/10.1061/(ASCE)0733- 9445(1999)125:3(351). [4] A. A. Yee and D. Hon, "Structural and economic benefits of precast/prestressed concrete construction," PCI J., vol. 46, no. 4, pp. 34– 42, 2001. [5] J. C. Saemann and G. W. Washa, "Horizontal shear connections between precast beams and cast-in-place slabs," ACI J. Proc., vol. 17, no. 3, 1964, http://dx.doi.org/10.14359/7832. [6] A. H. Mattock, Shear Transfer Under Monotonic Loading, Across an Interface Between Concretes Cast at Different Times. Seattle: Washington Univ., 1976. [Online]. Available: https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/PB275257.xhtml 19/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt [7] J. Wairaven, J. Frenay, and A. Pruijssers, "Influence of concrete strength and load history on the shear friction capacity of concrete members," PCI J., vol. 32, no. 1, pp. 66–84, Jan 1987, http://dx.doi.org/10.15554/pcij.01011987.66.84. [8] R. E. Loov and A. K. Patnaik, "Horizontal shear strength of composite concrete beams with a rough interface," PCI J., vol. 39, no. 1, pp. 48–69, 1994, http://dx.doi.org/10.15554/pcij.01011994.48.69. [9] A. K. Patnaik, "Behavior of composite concrete beams with smooth interface," J. Struct. Eng., vol. 127, no. 4, pp. 359–366, Apr 2001, http://dx.doi.org/10.1061/(ASCE)0733-9445(2001)127:4(359). [10] L. F. Kahn and A. Slapkus, "Interface shear in high strength composite T-beams," PCI J., vol. 49, no. 4, pp. 102–110, 2004, http://dx.doi.org/10.15554/pcij.07012004.102.110. [11] M. A. Mansur, T. Vinayagam, and K.-H. Tan, "Shear transfer across a crack in reinforced high-strength concrete," J. Mater. Civ. Eng., vol. 20, no. 4, pp. 294–302, Apr 2008, http://dx.doi.org/10.1061/(ASCE)0899-1561(2008)20:4(294). [12] A. A. Semendary, W. K. Hamid, E. P. Steinberg, and I. Khoury, "Shear friction performance between high strength concrete (HSC) and ultra high performance concrete (UHPC) for bridge connection applications," Eng. Struct., vol. 205, pp. 110–122, 2020, http://dx.doi.org/10.1016/j.engstruct.2019.110122. [13] American Concrete Institute, Building Code Requirements for Structural Concrete, ACI 318-19, 2019. REFERENCES [14] American Association of State Highway and Transportation Officials, AASHTO Bridge Design Specifications, 9th ed. Washington: AASHTO, 2020. [15] J. P. R. Dantas, “Investigação experimental da fadiga em lajes de pontes armadas com barras ou telas soldadas,” M.S. thesis, Esc. Politéc., Univ. São Paulo, São Paulo, 2010. [16] P. S. P. Cavalcanti, “Investigação experimental da fadiga ao cisalhamento em lajes de pontes com pré-lajes,” M.S. thesis, Esc. Politéc., Univ. São Paulo, São Paulo, 2011. [17] E. C. Caixeta, “Investigação experimental da fadiga em lajes de pontes com ou sem pré-lajes,” M.S. thesis, Esc. Politéc., Univ. São Paulo, São Paulo, 2010. [18] S. A. Kristiawan, "Evaluation of models for estimating shrinkage stress in patch repair system," Int. J. Concr. Struct. Mater., vol. 6, no. 4, pp. 221–230, 2012, http://dx.doi.org/10.1007/s40069-012-0023-y. [19] H. W. Birkeland, "Differential shrinkage in composite beams," ACI J. Proc., vol. 56, no. 5, pp. 1123–1136, 1 http://dx.doi.org/10.14359/8133. [20] D. E. Branson, "Time-dependent effects in composite concrete beams," ACI J. Proc., vol. 61, no. 2, pp. 213– http://dx.doi.org/10.14359/7776. [21] Y. Yuan and M. Marosszeky, "Restrained shrinkage in repaired reinforced concrete elements," Mater. Struct., vol. 27, no. 7, pp. 375– 382, Aug 1994, http://dx.doi.org/10.1007/BF02473440. [22] J. Silfwerbrand, "Stresses and strains in composite concrete beams subjected to differential shrinkage," ACI Struct. J., vol. 94, no. 4, pp. 347–353, 1997, http://dx.doi.org/10.14359/485. [23] Y. Yuan, G. Li, and Y. Cai, "Modeling for prediction of restrained shrinkage effect in concrete repair," Cement Concr. Res., vol. 33, no. 3, pp. 347–352, Mar 2003, http://dx.doi.org/10.1016/S0008-8846(02)00960-2. [24] J. Zhou, G. Ye, E. Schlangen, and K. van Breugel, "Modelling of stresses and strains in bonded concrete overlays subjected to differential volume changes," Theor. Appl. Fract. Mech., vol. 49, no. 2, pp. 199–205, 2008, http://dx.doi.org/10.1016/j.tafmec.2007.11.006. [25] S. Sullivan, “Construction and behavior of precast bridge deck panel systems,” Ph.D. dissertation, Virginia T nstruction and behavior of precast bridge deck panel systems,” Ph.D. dissertation, Virginia Tech, Blacksburg, 2007 [26] H. Beushausen and M. G. Alexander, "Localised strain and stress in bonded concrete overlays subjected to differential shrinkage," Mater. Struct., vol. 40, no. 2, pp. 189–199, Jan 2007, http://dx.doi.org/10.1617/s11527-006-9130-z. [27] R. H. Evans and A. S. Parker, "Behavior of prestressed concrete composite beams," ACI J. Proc., vol. 51, no. 5, pp. 861–878, 1955, http://dx.doi.org/10.14359/11721. [28] N. M. Taha, A Microcomputer Program for the Design of Composite Beam. Cairo: Cairo Univ., 1978. [29] S. Dritsos, K. Pilakoutas, and E. Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 Editors: Leandro Trautwein, Antônio Carlos dos Santos Author contributions: EVWT: conceptualization, development of the computational algorithm, data analysis, writing and translation; GAP and TNB: supervision, revision, translation. [35] C. Mazzotti and N. Buratti, "A design oriented fibre-based model for simulating the long-term behaviour of RC beams: Application to beams cast in different stages," J. Build. Eng., vol. 44, pp. 103176, Dec 2021, http://dx.doi.org/10.1016/j.jobe.2021.103176. [36] M. Bottoni, C. Mazzotti, and M. Savoia, "Long-term experimental tests on precast beams completed with cast in situ concrete," Eur. J. Environ. Civ. Eng., vol. 13, no. 6, pp. 727–744, 2009, http://dx.doi.org/10.1080/19648189.2009.9693148. [34] L. M. Santos, Sub-Rotinas Básicas do Dimensionamento de Concreto Armado, 1a ed. São Paulo: Thot, 1994. REFERENCES Kotsira, "Effectiveness of flexural strengthening of RC members," Constr. B 3, pp. 165–171, Jun 1995, http://dx.doi.org/10.1016/0950-0618(95)00010-D. [30] J.-W. Hwang, J.-H. Kwak, and H.-G. Kwak, "Finite-element model to evaluate nonlinear behavior of posttensioned composite beams with partial shear connection," J. Struct. Eng., vol. 141, no. 8, pp. 04014205, Aug 2015, http://dx.doi.org/10.1061/(ASCE)ST.1943- 541X.0001174. ileira de Normas Técnicas, Projeto de Estruturas de Concreto – Procedimento, ABNT NBR 6118:2014, 2014. 1] Associação Brasileira de Normas Técnicas, Projeto de Estruturas de Concreto – Procedimento, ABNT NBR 611 [32] Associação Brasileira de Normas Técnicas, Cordoalhas de Aço para Estruturas de Concreto Protendido – Especificação, ABNT NBR 7483:2020, 2020. [33] L. M. Silva and R. C. Carvalho “Análise de seções transversais de concreto armado e protendido sujeitas a flexão oblíqua composta em estados-limites último e de serviço por integração analítica,” Rev. Sul-Am. Eng. Estrut., vol. 16, no. 2, pp. 76–97, 2019, http://dx.doi.org/10.5335/rsaee.v16i2.8482. 20/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022 E. V. W. Trentini, G. A. Parsekian, and T. N. Bittencourt [34] L. M. Santos, Sub-Rotinas Básicas do Dimensionamento de Concreto Armado, 1a ed. São Paulo: Thot, 1994. [34] L. M. Santos, Sub-Rotinas Básicas do Dimensionamento de Concreto Armado, 1a ed. São Paulo: Thot, 1994. [35] C. Mazzotti and N. Buratti, "A design oriented fibre-based model for simulating the long-term behaviour of RC beams: Application to beams cast in different stages," J. Build. Eng., vol. 44, pp. 103176, Dec 2021, http://dx.doi.org/10.1016/j.jobe.2021.103176. [35] C. Mazzotti and N. Buratti, "A design oriented fibre-based model for simulating the long-term behaviour of RC beams: Application to beams cast in different stages," J. Build. Eng., vol. 44, pp. 103176, Dec 2021, http://dx.doi.org/10.1016/j.jobe.2021.103176. [36] M. Bottoni, C. Mazzotti, and M. Savoia, "Long-term experimental tests on precast beams completed with cast in situ concrete," Eur. J. Environ. Civ. Eng., vol. 13, no. 6, pp. 727–744, 2009, http://dx.doi.org/10.1080/19648189.2009.9693148. [36] M. Bottoni, C. Mazzotti, and M. Savoia, "Long-term experimental tests on precast beams completed with cast in situ concrete," Eur. J. Environ. Civ. Eng., vol. 13, no. 6, pp. 727–744, 2009, http://dx.doi.org/10.1080/19648189.2009.9693148. [36] M. Bottoni, C. Mazzotti, and M. Savoia, "Long-term experimental tests on precast beams completed with cast in situ concrete," Eur. J. Environ. Civ. Eng., vol. 13, no. 6, pp. 727–744, 2009, http://dx.doi.org/10.1080/19648189.2009.9693148. 21/21 Rev. IBRACON Estrut. Mater., vol. 15, no. 4, e15410, 2022

https://openalex.org/W1993561313

https://europepmc.org/articles/pmc2727344?pdf=render

English

Sometimes the naked taste of potato reminds me of being poor

Emerging infectious diseases

cc-by

ABOUT THE COVER ABOUT THE COVER Vincent van Gogh (1853–1890) The Potato Eaters (1885) (detail) Oil on canvas (81.5 cm × 114.5 cm) Van Gogh Museum, Amsterdam, the Netherlands (Vincent van Gogh Foundation) Vincent van Gogh (1853–1890) The Potato Eaters (1885) (detail) Oil on canvas (81.5 cm × 114.5 cm) Van Gogh Museum, Amsterdam, the Netherlands (Vincent van Gogh Foundation) Author affiliation: Centers for Disease Control and Prevention, At- lanta, Georgia, USA Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 15, No. 6, June 2009 “Sometimes the naked taste of potato reminds me of being poor” “Ac- cording to official statistics just made public for the last six years,” the New York Times reported in 1901, “an average of 150,000 persons have yearly died in France from con- sumption, while in Paris alone the total for that period has been 83,274 deaths …. All classes have suffered from the disease, but it has been particularly fatal in those sections of the city occupied by working families.” cupation with the plight of the poor, whose lives he had ex- perienced from close up. “The point is,” he wrote to Theo, “I’ve tried to bring out the idea that these people eating po- tatoes by the light of their lamp have dug the earth with the self-same hands they are now putting into the dish, and it thus suggests manual labor and a meal honestly earned.” Depicting night scenes was a creative outlet, a way to test technical innovations and explore the relationship be- tween the cycles of nature and rural life. He drew from the traditions of the 17th-century Dutch masters, particularly Rembrandt, and the Barbizon school landscape painters Charles Daubigny and Jules Dupré. He was also influenced by the impressionists, the pointillists, and Japanese print- makers Hiroshige and Hokusai. Before van Gogh’s untime- ly death at age 37, these diverse influences had culminated in a unique style, the blend of striking colors and riveting brushstrokes. When he “took up his pencil” against the values of in- dustrial society, van Gogh made no effort to sugarcoat any- thing. He knew squalor. “Miners, men and women, going to the shaft in the morning through the snow, by a path along a hedge of thorns,” after a day of exhausting labor, they had the color of a “very dusty, unpeeled potato,” their postures showing isolation and resignation. Poverty, with its attendant malnutrition and crowding, so well captured by van Gogh in The Potato Eaters and by Zola in Germinal, always has been a hotbed of emerging infections. No longer referred to as consumption, TB is still a killer, its rates disproportionately high among the poor. Lice and other pests thrive among the homeless, spreading trench fever and other infections. And proximity to domes- tic animals and rodents in crowded areas expands the range of influenza, spotted fevers, and plague. But public health efforts to prevent and control the effects of poverty persist. “Sometimes the naked taste of potato reminds me of being poor” ―Leonard Nathan Polyxeni Potter “I w o way gleans from it I cannot find in it at all.” Nonetheless, Vincent tried to follow in his father’s evangelical footsteps, but his youthful zeal and empathetic ministry were misin- terpreted by the hierarchy of the Dutch Reformed Church. They rejected him, ending his studies in theology and his tenure as missionary to a coal mining community in Bel- gium. “I want to paint men and women with that something of the eternal which the halo used to symbolize and which we seek to convey by the actual radiance and vibra- tion of our coloring,” wrote Vincent van Gogh in one of his celebrated letters. Hundreds of these were written mostly to his brother Theo, an art dealer in Paris who provided him with financial and emotional support throughout his brief but brilliant career. The letters lay out the artist’s philoso- phy of life and reveal ample literary inclinations as well as spiritual depth. “Saying a thing well is as interesting and as difficult as painting it,” he wrote. “Even in that deep misery,” he wrote about his rejec- tion, “I felt my energy revive, and I said to myself, in spite of everything I shall rise again: I will take up my pencil, which I had forsaken in my discouragement, and I will go on with my drawing.” Van Gogh began his artistic career at age 27, while still in Belgium, by painting peasants, whom he perceived as closer to nature than other people, in the manner of his contemporary Jean-François Millet. And with as much zeal as he had pursued his religious mission, he now tried to capture the divine in everyday life. Van Gogh was born in Zundert, the Netherlands, and was raised in a religious albeit not always harmoni- ous household. “Father cannot understand or sympathize with me …. I too read the Bible … as I read Michelet or Balzac or Eliot … and what Father in his little academic The Potato Eaters, on this month’s cover, was van Gogh’s first major work. This painting of a family gathered around the table for the evening meal reflected his preoc- DOI: 10.3201/eid1506.000000 DOI: 10.3201/eid1506.000000 1001 merging Infectious Diseases • www.cdc.gov/eid • Vol. 15, No. 6, June 2009 ABOUT THE COVER “wholly different way of life from ours” continued. Bibliography 1. Bonilla DL, Kabeya H, Henn J, Kramer VL, Kosoy MY. Bartonella quintana in body lice and head lice from homeless persons, San Francisco, California, USA. Emerg Infect Dis. 2009;15:912–5. 2. Nathan L. The potato eaters. Alexandria (VA): Orchises Press; 1998. 3. Serpa JA, Teeter LD, Musser JM, Graviss EA. Tuberculosis dis- parity between US-born blacks and whites, Houston, Texas, USA. Emerg Infect Dis. 2009;15:899–904. 4. The complete letters of Vincent van Gogh, Vols I–III. New York: New York Graphic Society; 1958. 5. Van Gogh and the colors of the night [cited 2009 Apr 16]. Available from http://rawartint.wordpress.com/2008/09/22/the-museum-of- modern-art-moma-presents-van-gogh-and-the-colors-of-the-night 6. Zola É. Germinal. London: Penguin Classics; 2004. Address for correspondence: Polyxeni Potter, EID Journal, Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop D61, Atlanta, GA 30333, USA; email: [email protected] 1. Bonilla DL, Kabeya H, Henn J, Kramer VL, Kosoy MY. Bartonella quintana in body lice and head lice from homeless persons, San Francisco, California, USA. Emerg Infect Dis. 2009;15:912–5. 1. Bonilla DL, Kabeya H, Henn J, Kramer VL, Kosoy MY. Bartonella quintana in body lice and head lice from homeless persons, San Francisco, California, USA. Emerg Infect Dis. 2009;15:912–5. But despite van Gogh’s efforts, this night scene was not well received. It was perceived as not realistic enough, awkward, even technically incorrect. His artistic goals were not understood. “What I try … is not to draw a hand, but the gesture, not a mathematically correct head, but the general expression ….” And his preliminary work was not appreci- ated. He had done extensive drawings and visited a local family regularly, sketching while they ate. “By continually observing peasant life, at all hours of the day, I have be- come so involved in it that I rarely think of anything else.” 2. Nathan L. The potato eaters. Alexandria (VA): Orchises Press; 1998. 2. Nathan L. The potato eaters. Alexandria (VA): Orchises Press; 1998. 3. Serpa JA, Teeter LD, Musser JM, Graviss EA. Tuberculosis dis- parity between US-born blacks and whites, Houston, Texas, USA. Emerg Infect Dis. 2009;15:899–904. 3. Serpa JA, Teeter LD, Musser JM, Graviss EA. Tuberculosis dis- parity between US-born blacks and whites, Houston, Texas, USA. Emerg Infect Dis. 2009;15:899–904. 4. The complete letters of Vincent van Gogh, Vols I–III. New York: New York Graphic Society; 1958. 4. The complete letters of Vincent van Gogh, Vols I–III. New York: New York Graphic Society; 1958. 5. “Sometimes the naked taste of potato reminds me of being poor” This hope recalls the message of Germinal, “Beneath the blazing sun, in that morning of new growth, the countryside rang with song, as its belly swelled with a black and aveng- ing army of men, germinating slowly in its furrows, grow- ing upwards in readiness for harvests to come, until one day soon their ripening would burst open the earth itself.” “When weavers weave that cloth which I think they call cheviot, or those curious multicolored Scottish tartan fabrics,” van Gogh wrote in reference to the coloring in the Potato Eaters, “then they try, as you know, to get strange broken colors and grays into the cheviot and to get the most vivid colors to balance each other in the multicolored che- quered cloth so that instead of the fabric being a jumble, the … pattern looks harmonious from a distance.” The somber hues and harsh texture of The Potato Eat- ers went against convention, as did the exaggerated fea- tures of the peasants. “I’ve held the threads of this fabric in my hands all winter long and searched for the defini- tive pattern,” he wrote, “and although it is now a fabric of rough and coarse appearance, the threads have nonetheless been chosen with care and according to certain rules. And it might just turn out to be a genuine peasant painting. I know that it is.” Address for correspondence: Polyxeni Potter, EID Journal, Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop D61, Atlanta, GA 30333, USA; email: [email protected] Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop D61, Atlanta, GA 30333, USA; email: [email protected] Bibliography Van Gogh and the colors of the night [cited 2009 Apr 16]. Available from http://rawartint.wordpress.com/2008/09/22/the-museum-of- modern-art-moma-presents-van-gogh-and-the-colors-of-the-night The life of the poor was also on the literary minds of van Gogh’s day. In 1884–85, naturalist author Émile Zola wrote Germinal, his novel about a coal miners’ strike in northern France in the 1860s. This famous account of pov- erty and oppression struck a nerve even if it did not end mining strikes or the misery that brought them about. This 6. Zola É. Germinal. London: Penguin Classics; 2004. 1002 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 15, No. 6, June 2009

https://openalex.org/W2950827045

https://europepmc.org/articles/pmc6466807?pdf=render

English

DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis

Genome biology

cc-by

Background quantified and removed using factor analysis provided that the unwanted variation is uncorrelated with the covariates of interest [9, 10]. RNA-seq is currently the most prevailing method for profiling transcriptional activities using high-throughput sequencing technology [1]. The sequencing tag count per unit of transcript length is used to measure the rela- tive abundance of the transcript [2]. Various factors exist that may affect the faithful representation of transcript abundance by RNA-seq read counts. Normalization is a crucial step in post-experiment data processing to en- sure a fair comparison of gene expression in RNA-seq analysis [3, 4]. The most commonly used approach is to normalize the read counts globally by a sample-specific scale factor to adjust the sequencing depth. Choices of the scale factor include the total number of reads (or mean), median, trimmed mean of M values [5], and upper quartile [3]. The second type of normalization aims to remove the read count bias due to physical or chemical features of RNA sequences or uncontrollable technical aspects. The GC content is known to affect the read counts in a nonlinear way [6, 7], and this effect can be sample specific under different culture or library preparation protocols [8]. Systematic bias may also arise due to technical effects such as library preparation and sequencing batches. Such systematic biases can be [ , ] Another type of bias arises from cDNA fragmentation and mRNA degradation. The RNA-seq assay requires fragmenting the cDNA (reversely transcribed from mRNA) or mRNA for high-throughput sequencing. Ideally, for a complete, non-degraded transcript, if the fragmentation is completely random, we expect to see reads uniformly distributed along the transcript. Never- theless, the fragmentation by random priming is not truly random due to primer specificity [11–13]. Conse- quently, read count per unit length of a transcript may not strictly reflect the transcript abundance when com- paring the expression of different genes. For the same gene, assuming the same protocol is applied to different samples, the bias attributable to fragmentation across samples should be similar. Thus, fragmentation bias is less problematic in a gene-by-gene differential expres- sion (DE) analysis. In contrast, mRNA degradation can vary substantially in both extent and pattern between genes and between samples [14, 15]. The mRNA degrad- ation has different pathways and can happen in any re- gion of a transcript [16]. DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis Bin Xiong, Yiben Yang, Frank R. Fineis and Ji-Ping Wang* Bin Xiong, Yiben Yang, Frank R. Fineis and Ji-Ping Wang* © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract RNA degradation affects RNA-seq quality when profiling transcriptional activities in cells. Here, we show that transcript degradation is both gene- and sample-specific and is a common and significant factor that may bias the results in RNA-seq analysis. Most existing global normalization approaches are ineffective to correct for degradation bias. We propose a novel pipeline named DegNorm to adjust the read counts for transcript degradation heterogeneity on a gene-by-gene basis while simultaneously controlling for the sequencing depth. The robust and effective performance of this method is demonstrated in an extensive set of simulated and real RNA-seq data. Keywords: RNA-seq, Normalization, RNA degradation, Degradation normalization, Alternative splicing, Non-negative matrix factorization eq, Normalization, RNA degradation, Degradation normalization, Alternative splicing, Non-negative Xiong et al. Genome Biology (2019) 20:75 https://doi.org/10.1186/s13059-019-1682-7 Xiong et al. Genome Biology (2019) 20:75 https://doi.org/10.1186/s13059-019-1682-7 Background Perfect control of sample degradation during the experiment is difficult, particu- larly when the samples are collected from field studies * Correspondence: [email protected] Department of Statistics, Northwestern University, Evanston, IL 60208, USA Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology (2019) 20:75 Page 2 of 18 Xiong et al. Genome Biology (2019) 20:75 Alternative splicing is frequently observed in higher organisms, and it further complicates the gene expres- sion estimation in RNA-seq [23]. In the gene-level DE analysis, we test the equivalence of relative abundance of transcripts in copy numbers between samples or condi- tions. If the two samples have differential exon usage, read counts need to be adjusted accordingly to better represent the transcript relative abundance in the re- spective samples. Currently, most existing statistical packages for RNA-seq analysis (e.g., DESeq [24] and edgeR [25]) all take the raw read counts as input, while such complexities are completely ignored in practice. or clinical samples. More importantly, different genes may degrade at different rates [17], which makes it im- possible to remove this bias by normalizing the read counts of all genes in the same sample by the same constant. While the major impact of RNA degradation on gene expression analysis has been well recognized [17, 18], methods for correcting the degradation bias have not been fully explored in the literature. A few methods have been proposed to quantify the RNA integrity including RNA integrity numbers (RIN) [19], mRIN [20], and tran- script integrity number (TIN) [13]. The RIN gives a sample-specific overall RNA quality measure, but not at the gene level. In practice, a sample with RIN ≥7 (on a scale of 0 to 10) is often regarded as having good quality. The mRIN and TIN measures were both defined in the gene level by comparing the sample read distribution with reference to the hypothetical uniform distribu- tion. In real data, due to GC content bias, primer specificity, and other complexities, the read count may substantially deviate from the uniform distribu- tion along the transcript [7]. In this paper, we propose a novel data-driven method to quantify the transcript degradation in a generalized sense for each gene within each sample. Using the estimated degradation index scores, we build a normalization pipeline named DegNorm to correct for degradation bias on a gene-by-gene basis while simul- taneously controlling the sequencing depth. Background The per- formance of the proposed pipeline is investigated using simulated data, and an extensive set of real data that came from both cell line and clinical samples sequenced in poly(A)+ or Ribo-Zero protocol. g p To reduce the degradation effect, Finotello et al. pro- posed to quantify the exon-level expression by the max- imum of its per-base counts instead of the raw read counts [21]. If a given exon is in a more degraded re- gion, the local maximum may still be an underestimate of true abundance. On the other hand, the larger vari- ance associated with the local maximum (e.g., spikes) may result in instability in DE analysis. Based on the TIN measure, Wang et al. proposed a degradation normalization method based on loess regression of read counts on the TIN measure for genes within the same sample [13]. However, the uniform baseline assumption and the failure to compare gene-specific degradation across samples appear to be the two major limitations, which may lead to extreme variability and bias in DE analysis (to be shown below). Jaffe et al. proposed a quality surrogate variable analysis (qSVA) to remove the confounding effect of RNA quality in DE analysis [22]. They investigated the degradation of RNA-seq data from dorsolateral prefrontal cortex (DLPFC) tissue under two different RNA-seq protocols, namely, poly(A)+ (mRNA-- seq) vs. ribosomal depletion (Ribo-Zero-seq). Thousands of features significantly associated with degradation were identified under either protocol separately, while no overlap was found between the two protocols. Further- more, comparing the DLPFC samples and the peripheral blood mononuclear cell (PBMC) samples [17] both se- quenced under the same poly(A)+ protocol, they found only four shared features. It is unclear how the sequence features identified in this study can be generalized for degradation bias correction in other RNA-seq data in practice. Results Data sets We consider six RNA-seq data sets that were generated from cell lines or clinical samples under either the main- stream poly(A) enrichment (mRNA-seq) or ribosomal RNA depletion protocol (Ribo-Zero-seq). The first one was from a brain glioblastoma (GBM) cell line study of a human for the impact of RNA degradation on gene expression ana- lysis [26]. Technical replicates of RNA samples were frag- mented under different incubation time and temperature using the NEBNext Magnesium RNA fragmentation mod- ule. We chose to analyze nine mRNA-seq samples in three groups of three, corresponding to three average RNA integ- rity number (RIN) = 10, 6, and 4, respectively (to be re- ferred to as R10, R6, and R4 for simplicity). We will perform DE analysis for R10 vs. R4 and R6 vs. R4. The second set contained 32 single-end mRNA-seq samples from human peripheral blood mononuclear cells (PBMC) of 4 different subjects: S00, S01, S02, and S03 [17]. The extracted RNA sample from each subject was kept in room temperature for 0, 12, 24, 36, 48, 60, 72, and 84 h, respectively, to approximate the natural degradation process. We choose S01 as an illustrating example and will perform DE analysis for 0 + 12 h vs. 24 + 48 h (results for other subjects are similar). The third set was from Sequencing Quality Control (SEQC) Consortium [27, 28] and contained two subsets of mRNA-seq data, namely SEQC-AA and SEQC-AB. The SEQC-AA subset consisted of 16 technical replicates from Stratagene’s universal human reference (UHR) RNA Page 3 of 18 Page 3 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology score at a given location within the transcript is defined as the total number of reads (single-end) or DNA fragments (paired-end) that cover this position (Additional file 1). If mRNA transcripts are complete and the fragmentation is random, we expect to see a flat coverage curve in the entire transcript except in the head and tail region (Fig. 1a). Nevertheless, in real data, the read coverage curves rarely display a uniform pattern; instead, dramatic and gene-specific differences are often observed across samples (Fig. 1b–e). The non-uniformity itself is less concerning as long as the coverage pattern is consistent across samples (Fig. 1b) such that the read counts can still faithfully represent the relative abundance of transcripts. Results Data sets In contrast, hetero- geneous coverage patterns are often observed where some samples show significantly decayed read counts in some regions (Fig.1c–e). One major cause of this hetero- geneity is mRNA degradation, which is clearly shown in the case of ACTN4 gene of R4 samples from the GBM data (Fig. 1d). Different sample preparation methods may lead to distinct read distributions. For example, FFPE samples may show a highly localized discrete read distribution pattern in contrast to a continuous distribu- tion typically observed in FF samples (Fig. 1c). Alterna- tive splicing may also result in depleted read count in the entire region of an exon, as exemplified in the ST7 gene of B samples (region 1900–2900 bp) from the SEQC-AB data (Fig. 1e). For the gene-level DE analysis, loss of read count due to such complexities needs to be compensated to ensure unbiased quantification of gene expression. library with two runs for eight lanes each. We will run DE analysis of the first run vs. the second run. The second subset contained two biological conditions: condition A of five samples from the same Stratagene’s UHR RNA, and condition B of five samples from Ambion’s human brain reference RNA. The first four replicates from both condi- tions were prepared by the same technician while the fifth was by Illumina. We excluded the fifth sample from both conditions because they showed a dramatic difference in coverage curves compared to the rest. The fourth data set contained RNA-seq data from dorso- lateral prefrontal cortex (DLPFC) tissue of five brains— three controls and two schizophrenia cases [22]. Each tissue was left in room temperature (off of ice) for 0, 15, 30, and 60 min for degradation. The RNA sample was extracted and prepared for both mRNA-seq and Ribo-Zero-seq. We chose to analyze one schizophrenia case (Br1729, results for other subjects are similar). We will perform DE analysis T0+T15 min vs. T30+T60 min under the same protocol, i.e., mRNA-seq or Ribo-Zero-seq, and then cross-platform DE analysis between the two protocols. The fifth data set originated from three pairs of matched fresh-frozen (FF) and formalin-fixed paraffin- embedded (FFPE) tissues of three breast tumor patients (namely T1, T2, T3) with a moderate archival time of about 4–5 years [29]. The FFPE samples are typically partially degraded. Results Data sets We will analyze the mRNA-seq data of FF (500 ng) and FFPE (100 ng) to investigate whether the degradation normalization can help improve the DE analysis in fragmented clinical RNA samples. The last data set arose from a clinical study on how AMP kinase (AMPK) promotes glioblastoma bioenergetics and tumor growth [30]. It was shown that cancer cells can acti- vate AMPK and highjack the stress-regulating pathway in cells. Thus, inhibiting AMPK in cancer cells may lead to treatment of GBM. RNA-seq data was collected from two patient-derived GBM stem cell (GSC) lines (GBM9 and GBM10) between control and AMPK knockout to identify differentially expressed genes. We will perform DE analysis between the three control and three knockout samples in GBM10 cell line. Unlike the first five sets where the ground truth of gene expression was known or mostly verified, the AMPK knockout data represents a typical case in clinical studies where only a handful genes of interest, most often suspected as differentially expressed, were PCR verified. For this reason, in the following analysis, we will compare the different normalization methods by benchmarking our ana- lyses using the first five sets and then present the AMPK data as a case study in the last. Generalized degradation and degradation normalization algorithm The degradation we target to normalize is defined in a generalized sense. Any systematic decay of read count in any region of a transcript in one or more samples com- pared to the rest in the same study is regarded as deg- radation. Clearly, mRNA degradation is one main cause, but alternative splicing and other factors may be the confounders that are difficult to deconvolute. To avoid confusion, in the following context, we will reserve the term “mRNA degradation” for the physical degradation of mRNA sequences, and “degradation” or “transcript degradation” for the generalized degradation without specification. We propose DegNorm, a degradation normalization pipeline based on non-negative matrix factorization over-approximation (NMF-OA, see the “Methods” sec- tion and Additional file 1). We assume there is a gene-specific ideal shape of coverage curve, called an “envelope” function, identical across the samples in the given study. Each envelope function is scaled by a sam- ple- and gene-specific abundance parameter to represent Non-uniformity and heterogeneity in read distribution pattern We define a total transcript as the concatenation of all annotated exons from the same gene. The read coverage Xiong et al. Genome Biology (2019) 20:75 Page 4 of 18 (2019) 20:75 Xiong et al. Genome Biology a b d c e Fig. 1 RNA-seq read coverage score shows between-sample heterogeneity in the pattern along transcripts. a The read coverage score, defined as the number of reads that cover each base pair, is expected to have a trapezoidal shape along the transcript if the read start position is uniformly distributed. b An example from SEQC-AB data shows a non-uniform but consistent read coverage pattern, where the average magnitude of coverage score for each sample may faithfully represent the transcript abundance given the sequencing depth is normalized. The diagram in red under the coverage plot shows the total transcript with exon boundaries from genome annotations (same for c, d and e below). c The TMEM229B gene from the breast tumor data shows differential coverage score patterns between FF and FFPE samples. Reads from FF samples are continuously distributed across the entire transcript while those from FFPE samples are highly enriched in a few disjoint blocks or fragments. d The ACTN4 gene from the GBM data with RIN number = 10 vs. RIN = 4 shows clear degraded coverage score towards the 5′ end of the transcripts in the latter group. e An example from the SEQC-AB data shows that alternative splicing likely causes sharply decayed coverage score across the entire alternatively spliced exon region b a b a c d d e e Fig. 1 RNA-seq read coverage score shows between-sample heterogeneity in the pattern along transcripts. a The read coverage score, defined as the number of reads that cover each base pair, is expected to have a trapezoidal shape along the transcript if the read start position is uniformly distributed. b An example from SEQC-AB data shows a non-uniform but consistent read coverage pattern, where the average magnitude of coverage score for each sample may faithfully represent the transcript abundance given the sequencing depth is normalized. The diagram in red under the coverage plot shows the total transcript with exon boundaries from genome annotations (same for c, d and e below). c The TMEM229B gene from the breast tumor data shows differential coverage score patterns between FF and FFPE samples. Non-uniformity and heterogeneity in read distribution pattern Reads from FF samples are continuously distributed across the entire transcript while those from FFPE samples are highly enriched in a few disjoint blocks or fragments. d The ACTN4 gene from the GBM data with RIN number = 10 vs. RIN = 4 shows clear degraded coverage score towards the 5′ end of the transcripts in the latter group. e An example from the SEQC-AB data shows that alternative splicing likely causes sharply decayed coverage score across the entire alternatively spliced exon region the expected coverage curve for the given gene within each sample if no degradation occurs. Degradation may occur in any region of the transcript to cause negative bias in the observed read counts. To illustrate this, we generated four expected coverage curves of identical shape but with different abundance levels (Fig. 2a), among which samples S1 and S2 are subject to degrad- ation in the 5′ end with different patterns. Based on the expected curves, we further simulated a random realization of four complete curves with sampling error imposed (Fig. 2b, to be referred to as latent curves) and two degraded for sample S1 and S2, respectively (Fig. 2c). The NMF-OA algorithm takes the four observed cover- age curves (i.e., two non-degraded (S3 and S4) and two degraded (S1 and S2)) as input and estimates the latent curves by minimizing the squared distance between the observed and latent, subject to the constraint that the la- tent curves must dominate their respective observed curves at all positions (Figs. 2d, e; the “Methods” sec- tion). We define the degradation index (DI) score for each gene within each sample, as the fraction of area the expected coverage curve for the given gene within each sample if no degradation occurs. Degradation may occur in any region of the transcript to cause negative bias in the observed read counts. To illustrate this, we generated four expected coverage curves of identical shape but with different abundance levels (Fig. 2a), among which samples S1 and S2 are subject to degrad- ation in the 5′ end with different patterns. Based on the expected curves, we further simulated a random realization of four complete curves with sampling error imposed (Fig. 2b, to be referred to as latent curves) and two degraded for sample S1 and S2, respectively (Fig. 2c). Non-uniformity and heterogeneity in read distribution pattern The NMF-OA algorithm takes the four observed cover- age curves (i.e., two non-degraded (S3 and S4) and two degraded (S1 and S2)) as input and estimates the latent curves by minimizing the squared distance between the observed and latent, subject to the constraint that the la- tent curves must dominate their respective observed curves at all positions (Figs. 2d, e; the “Methods” sec- tion). We define the degradation index (DI) score for each gene within each sample, as the fraction of area covered by the estimated latent curve, but above the ob- served curve (Fig. 2e). It measures the proportion of missing read count due to degradation given the current sequencing depth. The DegNorm pipeline iteratively corrects for degrad- ation bias while simultaneously normalizing sequencing depth. First, the NMF-OA algorithm is applied to se- quencing depth-normalized read counts for all genes one by one to estimate the DI scores. Second, the result- ing DI scores are then used to adjust the read counts by extrapolation for each gene. The adjusted read counts are used to normalize the raw data for sequencing depth. These two steps are repeated until the algorithm con- verges (the “Methods” section). DI score as sample quality diagnostics p q y g The estimated DI scores provide an overview of the within-sample, between-sample, and between-condition variation of degradation extent and patterns. We plotted the DI scores in three ways: a box plot of DI scores for each sample (Fig. 3a–f), a heatmap of the DI scores Xiong et al. Genome Biology (2019) 20:75 (2019) 20:75 Page 5 of 18 Xiong et al. Genome Biology 0 250 500 750 0 750 1000 250 500 0 250 500 750 0 250 500 750 0 250 500 750 position observed coverage vs normalized coverage e S1 S2 S3 S4 observed normalized normalized observed observed normalized observed normalized 0 250 500 750 observed coverage with degradation 0 250 500 position 750 1000 0 250 500 750 normalized coverage 0 750 1000 250 500 position S1 S2 S3 S4 S1 S2 S3 S4 0 250 500 750 0 750 1000 250 500 0 250 500 750 0 250 500 750 0 250 500 750 position observed coverage vs normalized coverage a b c d e S1 S2 S3 S4 observed normalized 0 250 500 750 1000 0 250 500 750 expected coverage position 0 250 500 750 0 250 500 750 1000 coverage w/o degradation position S1 S2 S3 S4 normalized observed observed normalized observed normalized S1 expected S1 degraded S2 expected S2 degraded S3 expected S4 expected Fig. 2 A proof-of-concept example of the proposed method for normalizing degradation patterns. a Expected or theoretical read coverage curves of one gene from four samples of identical shape (solid lines, without degradation), two of which (S1 and S2) are subject to degradation according to a rate indicated by the dashed lines in the 5′ end. b A realization of the four coverage curves without degradation randomly simulated according to the expected curves in a. These curves are regarded as the latent and unobserved data. c Observed coverage curves after imposing random degradation to S1 and S2 (S3 and S4 stay intact). d Estimates of the non-degraded latent curves from the proposed algorithm solely based on the observed coverage curves in c. DI score as sample quality diagnostics e A sample-by-sample comparison between the observed and estimated latent coverage curves a 0 250 500 750 1000 0 250 500 750 expected coverage position S1 expected S1 degraded S2 expected S2 degraded S3 expected S4 expected b 0 250 500 750 0 250 500 750 1000 coverage w/o degradation position S1 S2 S3 S4 b a e 0 250 500 750 normalized coverage 0 750 1000 250 500 position S1 S2 S3 S4 d 0 250 500 750 observed coverage with degradation 0 250 500 position 750 1000 S1 S2 S3 S4 c d d c Fig. 2 A proof-of-concept example of the proposed method for normalizing degradation patterns. a Expected or theoretical read coverage curves of one gene from four samples of identical shape (solid lines, without degradation), two of which (S1 and S2) are subject to degradation according to a rate indicated by the dashed lines in the 5′ end. b A realization of the four coverage curves without degradation randomly simulated according to the expected curves in a. These curves are regarded as the latent and unobserved data. c Observed coverage curves after imposing random degradation to S1 and S2 (S3 and S4 stay intact). d Estimates of the non-degraded latent curves from the proposed algorithm solely based on the observed coverage curves in c. e A sample-by-sample comparison between the observed and estimated latent coverage curves number (RIN), R = 10, 6, and 4, respectively. The DI scores show a clear escalating pattern of degradation se- verity across the three groups (Fig. 3d, j) with two strong clusters, i.e., R10 vs. R6+R4 (Additional file 2: Figure S1d). The scatter plots of DI scores further exemplify a higher correlation between samples within the same RIN group than across different RIN groups (Additional file 2: Figure S1 g-i). sorted in ascending order of the average scores of the first condition defined in the DE analysis (Fig. 3g–l), and a pairwise correlation matrix of DI scores between sam- ples (Additional file 2: Figure S1a-f). For SEQC-AA data with 16 technical replicates, the median of DI scores is ~ 0.35, consistent across samples (Fig. 3a). While the degradation pattern varies between genes, no systematic between-condition difference is ob- served (Fig. 3g, Additional file 2: Figure S1a). DI score as sample quality diagnostics In contrast, for the SEQC-AB data, the 4 samples from condition B have relatively lower and more homogeneous degrad- ation than that from A samples (Fig. 3b). Many genes show a condition-specific clustered pattern in DI scores (Fig. 3h), resulting in a high within-condition correlation (Additional file 2: Figure S1b). For the DLPFC Br1729 Ribo-Zero-seq data, DegNorm recovered an increasing pattern of degradation from time 0 to 60 min as expected (Fig. 3e, k, Additional file 2: Figure S1e). The three pairs of breast tumor samples were prepared in two different ways—fresh frozen (FF) and formalin-fixed paraffin-embedded (FFPE)—but both sequenced under the mRNA-seq protocol. The DI scores confirm that the mRNA transcripts in FFPE samples tend to be highly degraded compared to the paired FF samples (Fig. 3f, l), and degradation pat- terns are strongly clustered within the same FF or FFPE group (Additional file 2: Figure S1f). The PBMC and GBM data are known to have differen- tial mRNA degradation. The DI scores of PBMC S01 data confirm a progressive deterioration of average deg- radation when samples underwent degradation in room temperature for 0, 12, 24, and 48 h, respectively (Fig. 3c, i, Additional file 2: Figure S1c). The degradation from 24 to 48 h was particularly accelerated compared to the first 24 h. The nine GBM samples were previously classified into three groups according to the RNA integrity In summary, the DI scores from the DegNorm provide meaningful quantification of gene-level degradation be- tween samples for both cell line and clinical samples under both mRNA-seq and Ribo-Zero-seq protocols. Page 6 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology a b g h i c d e f j k l Fig. 3 Degradation index (DI) scores show gene-/sample-/condition-specific degradation heterogeneity. a–f Box plots of DI scores presented in a between-group comparison defined for the differential expression analysis as follows. a SEQC-AA data (16,670 genes): the 8 (1–8) technical replicates from the first run vs. the 8 (9–16) from the second run. b SEQC-AB data (19,061 genes): 4 biological replicates from A condition vs. 4 from B condition. c PBMC data for subject S01 (14,051 genes): 2 samples exposed at room temperature for 0 and 12 h (S01_T1 and S01_T2) vs. 2 for 24 and 48 h (S01_T3 and S01_T4), respectively. DI score as sample quality diagnostics a SEQC-AA data (16,670 genes): the 8 (1–8) technical replicates from the first run vs. the 8 (9–16) from the second run. b SEQC-AB data (19,061 genes): 4 biological replicates from A condition vs. 4 from B condition. c PBMC data for subject S01 (14,051 genes): 2 samples exposed at room temperature for 0 and 12 h (S01_T1 and S01_T2) vs. 2 for 24 and 48 h (S01_T3 and S01_T4), respectively. d GBM data (14,298 genes): 3 replicates each for RIN number = 10, 6, and 4, respectively. e DLPFC data for subject Br1729 from Ribo-Zero-seq (18,634 genes): 2 samples exposed to room temperature for 0 and 15 min (T0, T15) vs. 2 for 30 and 60 min (T30, T60). f Breast tumor data of 3 matched pair (T1, T2, T3) prepared from FF and FFPE methods, respectively (10,996 genes). g–l For each data set presented in a–f, the heatmap presents the DI scores of genes sorted in the ascending order of the average DI score of the first condition (in the GBM case, the R10 samples), where each row corresponds to the same gene across samples The degradation pattern is gene-specific, and the deg- radation extent may vary substantially between samples or conditions. and UQ methods yielded very similar results in all data we analyzed in this paper (Additional file 2: Figure S3a-f for some examples). For visualization purpose, only the UQ results are presented in the main figures. We first examine the five data sets that originated from the samples that had no true biological difference between conditions under test (i.e., SEQC-AA, PBMC-S01, GBM, DLPFC, and breast tumor). RNA deg- radation induces bias and thus may cause extra variance. Severe degradation may even result in a difference of transcript abundance for some genes when they are pre- pared for RNA-seq. Thus, we investigate how different methods may reduce variance by plotting the coefficient of variation (CV) of normalized read count vs. mean read count in log scale (the logged mean was linearly transformed to 0–1 range, Fig. 4a–f). Overall, the TIN method gives a relatively larger CV than the other three methods. When RNA degradation is a major concern such as in GBM-R10vsR4, and breast tumor FF vs. DI score as sample quality diagnostics d GBM data (14,298 genes): 3 replicates each for RIN number = 10, 6, and 4, respectively. e DLPFC data for subject Br1729 from Ribo-Zero-seq (18,634 genes): 2 samples exposed to room temperature for 0 and 15 min (T0, T15) vs. 2 for 30 and 60 min (T30, T60). f Breast tumor data of 3 matched pair (T1, T2, T3) prepared from FF and FFPE methods, respectively (10,996 genes). g–l For each data set presented in a–f, the heatmap presents the DI scores of genes sorted in the ascending order of the average DI score of the first condition (in the GBM case, the R10 samples), where each row corresponds to the same gene across samples i a b g h i h b a g d j d c c k k l l j k l f f e e e Fig. 3 Degradation index (DI) scores show gene-/sample-/condition-specific degradation heterogeneity. a–f Box plots of DI scores presented in a between-group comparison defined for the differential expression analysis as follows. a SEQC-AA data (16,670 genes): the 8 (1–8) technical replicates from the first run vs. the 8 (9–16) from the second run. b SEQC-AB data (19,061 genes): 4 biological replicates from A condition vs. 4 from B condition. c PBMC data for subject S01 (14,051 genes): 2 samples exposed at room temperature for 0 and 12 h (S01_T1 and S01_T2) vs. 2 for 24 and 48 h (S01_T3 and S01_T4), respectively. d GBM data (14,298 genes): 3 replicates each for RIN number = 10, 6, and 4, respectively. e DLPFC data for subject Br1729 from Ribo-Zero-seq (18,634 genes): 2 samples exposed to room temperature for 0 and 15 min (T0, T15) vs. 2 for 30 and 60 min (T30, T60). f Breast tumor data of 3 matched pair (T1, T2, T3) prepared from FF and FFPE methods, respectively (10,996 genes). g–l For each data set presented in a–f, the heatmap presents the DI scores of genes sorted in the ascending order of the average DI score of the first condition (in the GBM case, the R10 samples), where each row corresponds to the same gene across samples Fig. 3 Degradation index (DI) scores show gene-/sample-/condition-specific degradation heterogeneity. a–f Box plots of DI scores presented in a between-group comparison defined for the differential expression analysis as follows. DI score as sample quality diagnostics Results are shown for SEQC-AA, PBMC-S01, GBM R10 + R6, GBM R10 + R4, DLPFC Br1729 Ribo-Zero, and breast tumor data. a–f Coefficient of variation (CV) vs. mean read counts (in log scale): compared are results from proposed DegNorm pipeline and other methods including upper quartile (UQ), RUVr, and TIN. The mean counts for each data were scaled to 0–1 range by a linear transformation (i.e., (Xi −minj(Xj))/maxj(Xj) where Xi is the log of the mean count for gene i). The CV curve was generated using the R built-in function smooth.spline. The beanplot under the CV plot shows the density of log of mean read counts from DegNorm (the densities of read counts from other normalization methods are similar and not shown). g–l Empirical cumulative distribution function (ECDF) of the p value from DE analysis. The RUVr results were generated using the RUV-seq package. For TIN method, we followed Wang et al. [13] with details described in the uploaded R Markdown file PBMC-S01 and GBM-R6vsR4 comparisons with known modest between-condition difference in mRNA degrad- ation, the ECDF curves from different methods were all well above the diagonal line (except TIN for PBMC data), suggesting that differential degradation probably has caused a difference in gene abundance level when the samples were sequenced (Fig. 4h, i). Both DegNorm and TIN methods brought the ECDF curve down to- wards the diagonal line compared to UQ and RUVr, in- dicating that correction for degradation bias helps reduce potential false positives. Nevertheless, the TIN curve in PBMC-S01 data was well below the diagonal line (Fig. 4h), which may indicate a loss of statistical power due to the large variance of the normalized read counts (Fig. 4b). In contrast, the RUVr ECDF curves in both comparisons are significantly higher than that in UQ (regardless that RUVr had lower CV than UQ), compared to other methods except in the very lower or upper end where the CV was inflated due to outliers (Fig. 4d, f). The RUVr approach applies UQ normalization first and then further removes the add- itional variation estimated from the factor analysis. It always reduces CV over the UQ method. We normalized the raw read count by dividing it by 1 - DI score for each gene within each sample. The ad- justed read counts (rounded) were input into the edgeR package [25] for DE analysis. DI score as sample quality diagnostics FFPE comparisons, DegNorm pronouncedly reduced the CV DegNorm improves accuracy in gene expression analysis We set out to evaluate how the proposed DegNorm pipeline may improve differential expression analysis by comparing it with other seven normalization methods including UQ [3], TIN [13], RUVr, RUVg [10], trimmed mean of M values (TMM) [5], relative log expression (RLE) [24], and total read count (TC) [4]. The RUV methods were designed to remove unwanted variation, but it is unclear whether it is effective for correcting degradation bias. We dropped the RUVg method from the main text for its performance can be very sensitive to the choice of empirical control genes (Additional file 2: Figure S2a-f) or the choice of factor(s) from the factor analysis in the estimation of unwanted variation (Additional file 2: Figure S2 g, h). The TMM, RLE, TC, (2019) 20:75 Page 7 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology h a a b g h c d i j e f k l Fig. 4 Differential expression (DE) analysis in data sets that had no true differential expression. Results are shown for SEQC-AA, PBMC-S01, GBM R10 + R6, GBM R10 + R4, DLPFC Br1729 Ribo-Zero, and breast tumor data. a–f Coefficient of variation (CV) vs. mean read counts (in log scale): compared are results from proposed DegNorm pipeline and other methods including upper quartile (UQ), RUVr, and TIN. The mean counts for each data were scaled to 0–1 range by a linear transformation (i.e., (Xi −minj(Xj))/maxj(Xj) where Xi is the log of the mean count for gene i). The CV curve was generated using the R built-in function smooth.spline. The beanplot under the CV plot shows the density of log of mean read counts from DegNorm (the densities of read counts from other normalization methods are similar and not shown). g–l Empirical cumulative distribution function (ECDF) of the p value from DE analysis. The RUVr results were generated using the RUV-seq package. For TIN method, we followed Wang et al. [13] with details described in the uploaded R Markdown file b g b h a g a j i c d i j d c f l k e l f k e Fig. 4 Differential expression (DE) analysis in data sets that had no true differential expression. DI score as sample quality diagnostics When all genes are true nulls, the empirical cumulative distribution function (ECDF) of p value tends to be a diagonal line. Thus, an ECDF curve closer to the diagonal line indicates better performance of the normalization method in correcting the degradation bias. For SEQC-AA data with 16 technical replicates, all 4 methods resulted in expected ECDF curves close to the diagonal line (Fig. 4g). For Page 8 of 18 Xiong et al. Genome Biology (2019) 20:75 Page 8 of 18 Xiong et al. Genome Biology (2019) 20:75 The p value ECDF curve provides a global picture of a false-positive rate at different type-I error rate thresholds when all null hypotheses are true. In practice, as the ground truth is unknown, one typically claims the DE by controlling the false discovery rate (FDR) to correct mul- tiple comparison errors. Thus, we further compared the false-positive rate of different methods by controlling the nominal FDR under the criterion of q value ≤0.05 using q-value package [31–33]. For SEQC-AA data with a little degradation difference between samples, all four methods resulted in very few claimed positives (Additional file 3: Table S1), consistent with the close-to-diagonal-line nature of p value ECDF curves in Fig. 4g. For the rest data with differential degradation, all methods yielded a substantial number of false positives. For comparison, we plotted the ratio of the false-positive rate of UQ, RUVr, and TIN over DegNorm in log2 scale (Fig. 5a). The UQ and RUVr methods consistently yielded more false positives than DegNorm, by a factor ranging from 1 to 3.7 and 1.4–36.8, respectively. The TIN method reduced the false-positive rate over Deg- Norm in the PBMC-S01 with a factor of 1.37. suggesting an ineffective correction of degradation bias or even an adverse effect to cause extra false positives. For the GBM-R10vsR4 comparison, ECDF curves are all far above the diagonal line, likely indicating a substantial change of transcript abundance level for many genes due to drastic degradation in R4 samples (Fig. 4j). The DLPFC and breast tumor FF-FFPE RNA-seq data were both generated from clinical tissue samples. For the DLPFC Br1729 T0+T15 vs. T30+T60 com- parison, the DegNorm resulted in a slightly lower p value than UQ in Ribo-Zero (Fig. 4k) but much lower than UQ in mRNA-seq (Additional file 2: Figure S4a) data. For the breast tumor data, FFPE samples were shown substantially fragmented and degraded than FF samples (Figs. DI score as sample quality diagnostics 1c and 3f). DegNorm resulted in a lower p value curve than all other methods (Fig. 4l). We also did cross-protocol DE analysis by comparing the four Br1729 mRNA-seq with four Ribo-Zero sam- ples. All p value curves were way above the diagonal line (Additional file 2: Figure S4b), suggesting DE analysis across different sequencing protocols should not be recommended. a 6 5 4 3 2 1 0 -1 -2 -3 -4 GBM R10vsR4 GBM R6vsR4 PBMC-S01 DLPFC-Br1729 Ribo-Zero DLPFC-Br1729 Poly(A)+ Breast tumor FFvsFFPE UQ RUVr TIN Log2 ratio of FPR relative to DegNorm a b d e c 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 SEQC-AB p-value ECDF UQ DegNorm RUVr TIN 0.75 0.80 0.85 0.90 ROC curve AUC 1 2 3 4 5 PCR Log2FC threshold 0.95 UQ DegNorm RUVr TIN SEQC-AB PCR verifed genes--set 1 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 TPR FPR UQ DegNorm RUVr TIN SEQC-AB PCR verifed genes--set 2 0.00 0.25 0.50 0.75 1.00 TPR 0.00 0.25 0.50 0.75 1.00 FPR UQ DegNorm RUVr TIN SEQC-AB 0.4 0.5 0.6 0.7 0.8 0.9 CV UQ DegNorm RUVr TIN Log mean counts 0.00 0.25 0.50 0.75 1.00 Density 6 5 4 3 2 1 0 -1 -2 -3 -4 GBM R10vsR4 GBM R6vsR4 PBMC-S01 DLPFC-Br1729 Ribo-Zero DLPFC-Br1729 Poly(A)+ Breast tumor FFvsFFPE UQ RUVr TIN Log2 ratio of FPR relative to DegNorm f SEQC-AB PCR verifed genes--set 2 Fig. 5 Differential expression analysis results. a Log2 ratio of false-positive rate of UQ/RUVr/TIN relative to DegNorm at q value = 0.05 criterion for data sets where genes have no differential expression. b–f DE results for SEQC-AB data. b Coefficient of variation (CV) vs. mean normalized read counts (in log scale). c Empirical cumulative distribution function (ECDF) of the p value from DE analysis. d Receiving operating characteristic curve (ROC) calculated from ~ 500 PCR-verified genes [34]. Genes with absolute log2 fold change value ≥2 were defined as true positives, and absolute log2 fold change value ≤0.1 was defined as true negatives. Genes with log2 fold change in between were disregarded. e ROC calculated based on ~ 17,304 PCR-verified genes published in a separate study [27] for the same SEQC AB samples. The same threshold values of log2 fold change as in c were used in defining the positives and negatives. DI score as sample quality diagnostics f The area under the ROC curve (AUC) statistic as a function of the threshold value of absolute log2 fold change used to define the true positives based on the PCR data from e c 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 SEQC-AB p-value ECDF UQ DegNorm RUVr TIN b SEQC-AB 0.4 0.5 0.6 0.7 0.8 0.9 CV UQ DegNorm RUVr TIN Log mean counts 0.00 0.25 0.50 0.75 1.00 Density b a c p-value Log mean counts d SEQC-AB PCR verifed genes--set 1 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 TPR FPR UQ DegNorm RUVr TIN e SEQC-AB PCR verifed genes--set 2 0.00 0.25 0.50 0.75 1.00 TPR 0.00 0.25 0.50 0.75 1.00 FPR UQ DegNorm RUVr TIN 0.75 0.80 0.85 0.90 ROC curve AUC 1 2 3 4 5 PCR Log2FC threshold 0.95 UQ DegNorm RUVr TIN f SEQC-AB PCR verifed genes--set 2 f d e e Fig. 5 Differential expression analysis results. a Log2 ratio of false-positive rate of UQ/RUVr/TIN relative to DegNorm at q value = 0.05 criterion for data sets where genes have no differential expression. b–f DE results for SEQC-AB data. b Coefficient of variation (CV) vs. mean normalized read counts (in log scale). c Empirical cumulative distribution function (ECDF) of the p value from DE analysis. d Receiving operating characteristic curve (ROC) calculated from ~ 500 PCR-verified genes [34]. Genes with absolute log2 fold change value ≥2 were defined as true positives, and absolute log2 fold change value ≤0.1 was defined as true negatives. Genes with log2 fold change in between were disregarded. e ROC calculated based on ~ 17,304 PCR-verified genes published in a separate study [27] for the same SEQC AB samples. The same threshold values of log2 fold change as in c were used in defining the positives and negatives. f The area under the ROC curve (AUC) statistic as a function of the threshold value of absolute log2 fold change used to define the true positives based on the PCR data from e Page 9 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology Xiong et al. Genome Biology (2019) 20:75 We present three examples that exemplify how DegNorm may improve the accuracy in the DE analysis. We used local false discovery rate (lfdr) from q-value package [31–33] to quantify the significance of DE ana- lysis. DI score as sample quality diagnostics Compared to q value, which quantifies the average FDR for all genes with smaller p value than the given one, lfdr is more appropriate to quantify the false discov- ery rate associated with any individual p value. The MMP14 gene in the GBM-R10vsR4 comparison displayed a clear degradation in the 5′ end in R4 samples (Fig. 6a) and tested positive using UQ method (p value = 1.31e−8, lfdr = 1.1e−8). DegNorm compensated for the degraded portion of R4 samples and returned negative test result (p value = 0.91, lfdr = 0.32, Fig. 6b). The second example is the PIK3C2A gene from SEQC-AB comparison, a PCR-verified negative with log2 fold change = 0.06. It tested positive under UQ (p value = 0.005, lfdr = 0.02) (Fig. 6c) while negative under DegNorm with degradation correction in the 5′ end of A samples (p value = 0.77, lfdr = 0.97, Fig. 6d). The third example is the NDUFV3 gene from the SEQC-AB data, showing nearly depleted cover- age in the entire third exon region from ~ 223 to 1327 nt for B samples likely due to alternative splicing. It tested positive under UQ (p value = 1.94e−37, lfdr = 3.39e−09, Fig. 6e). DegNorm returned a negative result (p value = 0.808, lfdr = 0.942, Fig. 6f), consistent with negative PCR verification (log2 fold change = −0.114). Nevertheless, we will show below that this relatively lower false-positive rate from the TIN method is an indi- cation of undermined power due to the excessively inflated variance (Fig. 4a–f). g The SEQC-AB data presents an atypical example con- taining an unusually large number of differentially expressed genes [27, 34]. DegNorm produced an overall lower CV than UQ, RUVr, and TIN methods (Fig. 5b). The DegNorm ECDF lies below those of UQ and RUVr methods while above TIN (Fig. 5c). With the presence of truly differentially expressed genes, the lower ECDF can be interpreted as a tendency to result in fewer false posi- tives (good) or more false negatives (bad, less power) or a mix in the DE analysis. To investigate this, we utilized 2 sets of PCR-verified genes, 1 from the original MAQC study of 843 genes [34] and the other from SEQC study of 20,801 genes [27], as the ground truth to construct re- ceiver operating characteristic curves (ROC) (Fig. 5d, e). Similar to Rissio et al. DI score as sample quality diagnostics [10], we defined a gene as a posi- tive if the absolute value of log2 fold change ≥2, a nega- tive if ≤0.1, and undefined otherwise. For both sets, the ROC curves suggest that DegNorm achieved better true-positive rate (sensitivity) than UQ, RUVr, and TIN while controlling the false-positive rate (1-specificity). For example, at FPR = 0.05 (specificity = 0.95), the larger PCR set suggested a 78.1% true-positive rate for Deg- Norm in comparison with 73.7%, 67.3%, and 48.3% for UQ, RUVr, and TIN methods, respectively. When we varied the of log2 fold change threshold value from 1 to 5 to define the positives, the area under the ROC curve (AUC) from all methods (except TIN) increased as expected, while the AUC from DegNorm remained the largest and the gap between DegNorm and other methods enlarge (Fig. 5f). As the true-negative set was fixed in this experiment, the true-positive rate drove the change of AUC as the threshold value in- creases. This suggests DegNorm improves the power to identify highly differentially expressed genes over other normalization methods while controlling the false-positive rate. Therefore, we conclude that the lower ECDF curve from DegNorm method (Fig. 5c) manifests a good tendency to reduce false-positive rate or increase specificity without sacrifice of statis- tical power or sensitivity. GBM AMPK knockout data GBM AMPK data presents a typical case in a clinical study where RNA-seq is used to survey a transcriptome for differential expression between control and treat- ment samples. DegNorm uncovered higher degradation in the control than the AMPK knockout samples (Fig. 7a) and heterogeneity in degradation pattern between the two conditions (Fig. 7b). DegNorm also resulted in a lower CV curve and a lower ECDF of p value than UQ and RUVr methods (Fig. 7c, d). We claimed differential expression under q value ≤0.05 and compared the test-positive sets between DegNorm, UQ, and RUVr (Fig. 7e). There were 2798 positive genes shared by all 3 methods, while UQ and RUVr produced 935 and 775 more positives than DegNorm, respectively. We sus- pect degradation may cause an excess of false positives (as suggested by similar plots in Fig. 6 for SEQC-AB data); nevertheless, without a large set of PCR-verified genes, it is impossible to rigorously assess the sensitivity and specificity. Chhipa et al. [30] analyzed the RNA-seq data and concluded that the bioenergetics of cellular me- tabolism was the most significantly downregulated path- way in AMPK-depleted samples. They applied RT-qPCR and verified a small set of downregulated genes using in- dependent GBM cell lines from 3 other patients. Among the 12 genes verified as downregulated, including HIF1a, The RUVr and TIN methods both showed pronoun- cedly lower power than DegNorm and UQ, but due to different reasons (Fig. 5d–f). The RUVr method is guaranteed to reduce more variation based on the UQ-normalized data. Nevertheless, the removed vari- ation may contain true biological difference if it is con- founded with unwanted variation, which will result in more false negatives. On the other hand, the large vari- ance incurred by TIN method appears to severely under- mine the power in the DE analysis (Fig. 5b). Page 10 of 18 Xiong et al. Genome Biology (2019) 20:75 (2019) 20:75 Xiong et al. GBM AMPK knockout data Genome Biology a b c d e f 0 200 400 600 0 500 1000 1500 2000 Coverage SEQC-AB-NDUFV3 Exons: 0 200 400 600 0 500 1000 1500 2000 Coverage SEQC-AB-NDUFV3 Exons: Coverage SEQC-AB-PIK3C2A A01 A02 A03 A04 B01 B02 B03 B04 Coverage 200 100 0 0 2000 4000 SEQC-AB-PIK3C2A 300 Coverage 600 900 0 0 1000 2000 3000 R10 R10 R10 R4 R4 R4 GBM-R10vsR4-MMP14 300 Coverage 600 900 0 0 1000 2000 3000 GBM-R10vsR4-MMP14 6000 8000 300 200 100 0 300 0 2000 4000 6000 8000 A01 A02 A03 A04 B01 B02 B03 B04 A01 A02 A03 A04 B01 B02 B03 B04 A01 A02 A03 A04 B01 B02 B03 B04 R10 R10 R10 R4 R4 R4 Fig. 6 Examples illustrating that DegNorm improves accuracy in RNA-seq DE analysis. a, b Coverage curves under UQ and DegNorm normalization methods side-by-side for gene MMP14 in GBM R10vsR4 comparison. Same coverage plots for PIK3C2A (c, d) and NDUFV3 (e, f) genes, respectively, in the SEQC-AB comparison. All three genes changed from positive to negative after degradation normalization in DE analysis using edgeR. In (e, f), annotated exon boundaries are plotted to show possible alternative splicing in the exon from 223 to 1327 nt b 300 Coverage 600 900 0 0 1000 2000 3000 GBM-R10vsR4-MMP14 R10 R10 R10 R4 R4 R4 a 300 Coverage 600 900 0 0 1000 2000 3000 R10 R10 R10 R4 R4 R4 GBM-R10vsR4-MMP14 b a d Coverage 200 100 0 0 2000 4000 SEQC-AB-PIK3C2A 6000 8000 300 A01 A02 A03 A04 B01 B02 B03 B04 c Coverage SEQC-AB-PIK3C2A A01 A02 A03 A04 B01 B02 B03 B04 200 100 0 300 0 2000 4000 6000 8000 d c f 0 200 400 600 0 500 1000 1500 2000 Coverage SEQC-AB-NDUFV3 Exons: A01 A02 A03 A04 B01 B02 B03 B04 e 0 200 400 600 0 500 1000 1500 2000 Coverage SEQC-AB-NDUFV3 Exons: A01 A02 A03 A04 B01 B02 B03 B04 f e Fig. 6 Examples illustrating that DegNorm improves accuracy in RNA-seq DE analysis. a, b Coverage curves under UQ and DegNorm normalization methods side-by-side for gene MMP14 in GBM R10vsR4 comparison. Same coverage plots for PIK3C2A (c, d) and NDUFV3 (e, f) genes, respectively, in the SEQC-AB comparison. All three genes changed from positive to negative after degradation normalization in DE analysis using edgeR. GBM AMPK knockout data In (e, f), annotated exon boundaries are plotted to show possible alternative splicing in the exon from 223 to 1327 nt LDHA, SLC2A1, HK1, GPI, ALDOA, TPI1, PFKM, ENO1, GABPA, TFAM, and COX20, DegNorm, UQ, and RUVr all successfully identified 8 except HK1, PFKM, GABPA, and COX20, whereas TIN method missed 2 additional genes, GPI and ALDOA (Add- itional file 4: Table S2). The discrepancy between the RNA-seq set and PCR results could be due to cell line-specific variation, lack of power due to small sample size or sample quality (personal communication with Dr. Dasgupta). Clearly, such a small-scale verification is in- sufficient to conclude about the sensitivity, neither can we evaluate the specificity without verification of the negatives. The discrepancy between DegNorm and other LDHA, SLC2A1, HK1, GPI, ALDOA, TPI1, PFKM, ENO1, GABPA, TFAM, and COX20, DegNorm, UQ, and RUVr all successfully identified 8 except HK1, PFKM, GABPA, and COX20, whereas TIN method missed 2 additional genes, GPI and ALDOA (Add- itional file 4: Table S2). The discrepancy between the RNA-seq set and PCR results could be due to cell line-specific variation, lack of power due to small sample size or sample quality (personal communication with Dr. Dasgupta). Clearly, such a small-scale verification is in- sufficient to conclude about the sensitivity, neither can we evaluate the specificity without verification of the negatives. The discrepancy between DegNorm and other methods does provide alerts to users of possible false positives caused by degradation when interpreting the results of such studies. Simulation study As for any gene, all samples are subject to this over-estimation bias; this bias may partially cancel off in DE analysis. As a result, the p value ECDF and ROC curves of DegNorm almost perfectly overlap with the latent and UQ curves (Figs. 8g, h). At FPR = 0.05, DegNorm, RUVr, and TIN had sensitivity decay of 0.6%, 1.1%, and 8.6%, respect- ively, compared to using the latent counts or UQ method (Additional file 5: Table S3). genes had no degradation whereas in the rest 3 settings, 80% of the genes were randomly selected for degrad- ation. In the second setting, for a gene selected for deg- radation, either 3, 4, or 5 samples out of the 8 were randomly chosen for degradation, whereas in the third, either all 4 control samples or 4 treatment samples were randomly chosen for degradation. In both second and third settings, the degradation extent for each de- graded gene was random but following the same distribu- tion. In the fourth setting, for each gene selected to degrade, 2 control samples were randomly selected for degradation with the same expected severity, while all treatment samples had a sample-specific systematic differ- ence in expected severity. The simulation details are de- scribed in the “Methods” section and Additional file 1. When degradation is an issue as in simulations II–IV, the estimated DI scores provide an informative characterization of the overall degradation severity of each sample (Additional file 2: Figure S5a-f) as well as the similarity of gene-specific degradation pattern be- tween samples (Additional file 2: Figure S5 g-l). DegNorm demonstrates consistently better correction of degradation bias than the UQ, TIN, and RUVr methods, as evidenced by higher regression coefficient of determination (R2), and a nearly symmetric distribution of data points around the diagonal line in the scatter plot of normalized vs. latent read counts (Fig. 9a–d, Additional file 2: Figure S6a-h). The UQ method does not correct for any degradation bias but instead only Simulation I presents a scenario where samples have no degradation bias or any other bias but only between-sample variation in sequencing depth. In the following, we shall refer the latent count as the true read count for a gene before degradation is imposed. Deg- Norm is data driven and always returns non-negative DI scores by design. The estimated DI scores have a median ~ 0.11 in all samples (Fig. Simulation study To further systematically investigate the performance of DegNorm, we conducted a simulation study in a two-condition comparison: 4 control (A) vs. 4 treat- ment (B) samples in 4 different degradation settings. Each sample of 20,000 genes was simulated with a ran- dom sequencing depth of 40–60 million reads, 5% of which were chosen to be upregulated and another 5% for downregulated in expression. In the first setting, all Page 11 of 18 (2019) 20:75 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology a a b d e c Fig. 7 DE analysis in GBM AMPK knockout data. a Box plot of estimated DI scores where condition W and T stand for control and knockout samples, respectively. b Correlation plot of DI scores shows within condition replicates had a higher correlation of degradation pattern than between condition samples. c Coefficient of variation (CV) vs. mean normalized read counts (in log scale). The bean pot under the CV plot shows the density of log of mean read counts from DegNorm. d Empirical cumulative distribution function (ECDF) of the p value from DE analysis. e Venn-diagram shows the number of claimed differentially expressed genes and pairwise overlap by DegNorm, UQ, and RUVr methods at q value threshold = 0.05 c b b a c d d d e e Fig. 7 DE analysis in GBM AMPK knockout data. a Box plot of estimated DI scores where condition W and T stand for control and knockout samples, respectively. b Correlation plot of DI scores shows within condition replicates had a higher correlation of degradation pattern than between condition samples. c Coefficient of variation (CV) vs. mean normalized read counts (in log scale). The bean pot under the CV plot shows the density of log of mean read counts from DegNorm. d Empirical cumulative distribution function (ECDF) of the p value from DE analysis. e Venn-diagram shows the number of claimed differentially expressed genes and pairwise overlap by DegNorm, UQ, and RUVr methods at q value threshold = 0.05 Unsurprisingly, the UQ method perfectly normalizes the sequencing depth (Fig. 8c), while all other three methods caused bias or extra variance to different extents (Fig. 8d–f). The positive bias from DegNorm is more pronounced when read counts are low such that read coverage curve cannot be well estimated (Fig. 8e). Simulation study 8a), demonstrating the absence of between-condition heterogeneity (Fig. 8b). To investi- gate how the positive bias in DI scores may impact DE analysis, we first plotted the normalized vs. latent read count from different methods (Fig. 8c–f) (the latent read count in simulation I is just the raw read count). Xiong et al. Genome Biology (2019) 20:75 Page 12 of 18 Xiong et al. Genome Biology Fig. 8 Results for simulation I. a Box plot of estimated DI scores where conditions A and B stand for control and treatment samples, respectively. b Heatmap of DI scores of genes sorted in the ascending order of the average DI score of condition A (control), showing no between-condition clustered pattern. c–f Scatter plot of normalized read count vs. latent read count in log2 scale with diagonal line imposed for simulation I. Compared are results from proposed DegNorm pipeline and other methods including upper quartile (UQ), RUVr, and TIN. g, h ECDF plots of p value and ROC curves for simulation I under different normalization methods. All curves except TIN were well overlapping with each other, suggesting close performance in DE analysis Fig. 8 Results for simulation I. a Box plot of estimated DI scores where conditions A and B stand for control and treatment samples, respectively. b Heatmap of DI scores of genes sorted in the ascending order of the average DI score of condition A (control), showing no between-condition clustered pattern. c–f Scatter plot of normalized read count vs. latent read count in log2 scale with diagonal line imposed for simulation I. Compared are results from proposed DegNorm pipeline and other methods including upper quartile (UQ), RUVr, and TIN. g, h ECDF plots of p value and ROC curves for simulation I under different normalization methods. All curves except TIN were well overlapping with each other, suggesting close performance in DE analysis to different extents. We tabulated the sensitivity of each method at FPR = 0.05 (Additional file 5: Table S3) and plotted it in Fig. 10. In settings II and III, where degrad- ation was randomly chosen among samples (II) or condi- tions (III), the UQ and RUVr methods were both ineffective to correct this non-systematic but gene-specific bias (Figs. 9k, l and 10). Simulation study In particular, in simulation III as degradation was applied to one condition of random choice for a given gene, the degradation bias was com- pletely confounded with the covariate of interest and can- not be removed by UQ or RUVr method. Consequently, many false positives were called due to degradation bias (Fig. 9l). At FPR = 0.05 threshold, DegNorm improved the sensitivity by a factor of 1.28, 1.30, and 2.01 compared to UQ, RUVr, and TIN methods, respectively (Fig. 10, Additional file 5: Table S3). In contrast, the treatment samples in setting IV had a systematic difference in aver- age degradation (Additional file 2: Figure S5c), both UQ and RUVr performed reasonably well (Figs. 9m and 10). In all four degradation settings, the TIN method showed inferior power to detect true DE genes. normalizing sequencing depth (Fig. 9a, Additional file 2: Figure S6a, e). Similar extreme variance issue is also ob- served in the TIN method in simulations II–IV (Fig. 9b, Additional file 2: Figure S6b, f). To gain more insights into the CV plots, we singled out the 18,000 truly non-differentially expressed genes from simulations II–IV and plotted the CV against the mean of normalized count (Fig. 9e–g). Without the con- founding effect of biological difference, the CV of these genes tends to be inflated due to degradation-caused loss of read count. Thus, the CV is an informative measure to assess the effectiveness of degradation normalization. Indeed, the UQ and RUVr curves are both well above the latent curve in all simulations, suggesting degrad- ation bias was not or inadequately corrected. Similar to what we observed in the real data, the TIN CV curve dominates other methods in each setting, echoing the excess of variance observed in the scatter plot (Fig. 9b, Additional file 2: Figure S6b, f). In contrast, the DegNorm curves were the closest to the respective la- tent curves but with a slight underestimation of CV in the lower half. Discussion and conclusions DegNorm improves the accuracy in DE analyses. In both ECDF (Fig. 9h–j) and ROC plots (Fig. 8k–m), the DegNorm curve was the closest to the latent one, demon- strating improvement over other normalization methods In this paper, we showed that RNA degradation pattern and severity are not only sample specific, but also gene-specific, and thus commonly used global Page 13 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology Fig. 9 Comparison of different normalization methods in simulations II, III, and IV. a–d Scatter plot of normalized read count vs. latent read count in log2 scale with diagonal line imposed from DegNorm, UQ, RUVg, RUVr, and TIN methods for simulation II. e–g Coefficient of variation (CV) vs. normalized mean read count (in log scale) for 18,000 true-negative genes (out of a total of 20,000 genes). The bean pot under the CV plot shows the density of log of mean read counts from DegNorm. h–j Empirical CDF of p value from edgeR DE analysis. The latent curve corresponds to the results using the true read counts before the degradation was imposed. k–m Receiver operating characteristic curves of DE analysis Fig. 9 Comparison of different normalization methods in simulations II, III, and IV. a–d Scatter plot of normalized read count vs. latent read count in log2 scale with diagonal line imposed from DegNorm, UQ, RUVg, RUVr, and TIN methods for simulation II. e–g Coefficient of variation (CV) vs. normalized mean read count (in log scale) for 18,000 true-negative genes (out of a total of 20,000 genes). The bean pot under the CV plot shows the density of log of mean read counts from DegNorm. h–j Empirical CDF of p value from edgeR DE analysis. The latent curve corresponds to the results using the true read counts before the degradation was imposed. k–m Receiver operating characteristic curves of DE analysis Page 14 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology Fig. 10 Bar chart of sensitivity (TPR) of different methods in simulations II, III, and IV. True-positive rate (sensitivity) is compared at false-positive rate threshold = 0.05 for each method in each simulation factors. Indeed, in all five benchmark data sets consid- ered in this paper, DegNorm performed consistently bet- ter compared to other methods regardless of whether mRNA degradation was a known concerning issue. Discussion and conclusions Un- like mRIN and TIN measures where degradation was de- fined as the deviation from hypothesized uniform coverage curve, the DI score from DegNorm is defined with reference to an adaptively estimated latent coverage curve that minimizes the distance to the observed cover- age curves. If a gene has 50% degradation but having consistent coverage curves across samples, the estimated DI scores will all be nearly 0. In this case, the read count in each sample can still accurately reflect the relative abundance between samples, and degradation correction is unnecessary. From this perspective, the DI score is de- fined in a relative sense. Normalizing the read counts for degradation bias using DI scores is hoped to minimize the extrapolation needed, thus avoids an excess of vari- ance. The advantage of this strategy was exemplified in all real and simulated data sets in contrast to the TIN method. normalization methods that impose a sample-specific constant adjustment to all genes within the same sample are ineffective to correct for this bias. The RUVr ap- proach is guaranteed to reduce variation, while they failed to show pronounced improvement over the UQ method in the DE analysis in all data sets considered in this study (even got worse in all data under consider- ation). One complexity is that the true biological differ- ence is often confounded with degradation (e.g., SEQC-AB data) and other unwanted variation. The fac- tor analysis cannot well separate the unwanted from the wanted variation, and it may even remove the true biological difference of interest. This confounding issue was illustrated in the SEQC-AB data by the high within-condition and low between-condition correl- ation of DI scores (Additional file 2: Figure S1b). In particular, the RUVg method was sensitive to the selec- tion of empirical control genes or the factor(s) used to estimate the unwanted variation (Additional file 2: Figure S2a-j). More objective criteria in this regard need to be developed for the RUVg method. Although motivated by mRNA degradation, we de- fined the degradation in this paper in a generalized and relative sense. The quantified DI scores may reflect con- founding effects from mRNA degradation, alternative splicing, and other factors. Risso et al. [10] showed that in the SEQC-AB data, samples were clustered due to the difference of sample preparation, experiment protocol, sequencing run batches and flow cell, etc. Discussion and conclusions Such factors could impact the RNA samples by changing the read coverage curves. Thus, normalizing heterogeneity in coverage curves may help reduce bias due to such There are a few limitations of the DegNorm method. First, like any other normalization method, DegNorm is a post hoc approach that is designed to alleviate the is- sues due to degradation heterogeneity between samples/ genes and thus improve the accuracy of DE analysis. It cannot completely remove the bias for every single gene. In particular, cautions must be taken when testing DE for samples that have a dramatic difference in degrad- ation (e.g., GBM R10 vs. R4), as it may change the true abundance level of transcripts of interest and lead to the excess of false positive or false negatives. High-quality Page 15 of 18 Page 15 of 18 Page 15 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology (2019) 20:75 RNA samples are always desirable in RNA-seq. Further- more, cross-platform DE analysis is not recommendable. Second, the core component of DegNorm is a matrix factorization over-approximation algorithm aiming to correct for the degradation bias that commonly exists in RNA samples, even for the high-quality SEQC data. DegNorm tends to result in a pronounced over-estimation bias in DI scores for genes with low read counts regardless of whether degradation is present (simulations II–IV) or absent (simulation I). We showed in simulation I that this bias is not a big concern in DE analysis as it tends to be homogeneous across all sam- ples for all non-degraded genes. Third, DegNorm has only been tested in this paper on the bulk RNA-seq data generated from a cell line or clinical samples (FF or FFPE) under mRNA-seq or Ribo-Zero-seq protocol. The effectiveness of DegNorm for RNA-seq data from 3′ end sequencing (3seq) or other variants needs to be further investigated in the future. Lastly, DegNorm is computing intensive due to parsing read alignment results, calcula- tion of read coverage curves, and repeated non-negative matrix factorization of large matrices. We have imple- mented DegNorm in a Python package available at https://nustatbioinfo.github.io/DegNorm/. Currently, it took about 9 h to run the entire pipeline on the FF vs. FFPE comparison (6 samples) on a 22-core node on the Linux cluster. Refinement of NMF-OA algorithm Refinement of NMF OA algorithm The NMF-OA optimization algorithm provides an ap- proximate solution to this problem. However, in the RNA-seq data, the performance of the solution can be affected by two confounding factors, the degradation ex- tent and the sequencing depth. In our quadratic object- ive function Q(Ki, Ei), an fij of larger magnitude tends to have more influence on the estimation of envelope func- tion. A dominant scale in fij may force the algorithm to fit an envelope function that resembles fij to minimize the loss. Thus, a good scale normalizing factor for se- quencing depth is important to yield a good estimate of the envelope function ei(x). With gene-specific and sample-specific degradation, the total number of reads may not provide a reliable measure of sequencing depth. Second, given Fi that is appropriately normalized for sequencing depth, the scale factor Ki should reflect the relative abundance of the gene in the non-degraded re- gion of each sample (to be referred to as the baseline re- gion below) (Fig. 2a). The non-degraded region must The NMF-OA optimization algorithm provides an ap- proximate solution to this problem. However, in the RNA-seq data, the performance of the solution can be affected by two confounding factors, the degradation ex- tent and the sequencing depth. In our quadratic object- ive function Q(Ki, Ei), an fij of larger magnitude tends to have more influence on the estimation of envelope func- tion. A dominant scale in fij may force the algorithm to fit an envelope function that resembles fij to minimize the loss. Thus, a good scale normalizing factor for se- quencing depth is important to yield a good estimate of the envelope function ei(x). With gene-specific and sample-specific degradation, the total number of reads may not provide a reliable measure of sequencing depth. Discussion and conclusions We have implemented an MPI release through parallel computing that can reduce the time by a factor of n where n is the number of nodes used. cause random fluctuation of fij(x) from the ideal curve kijei(x) (Fig. 2a, b). We assume the random error is negli- gible compared to the major bias arising from degrad- ation. Thus we require kijei(x) ≥fij(x) for all j and x. The difference between kijei(x) and fij(x) provides an estimate of degraded portion of read count. We propose a method that allows to estimate kij and ei(x) to quantify the degradation extent of each gene within each sample while simultaneously controlling the sequencing depth. Estimating degradation via non-negative matrix over- i i Let fij = (fij(1), … , fij(Li))T, j = 1, … , p and Fi = (fi1, … , fip)T. Let Ki = (ki1, … , kip)T, Ei = (ei(1), … , ei(Li))T. We propose to estimate Ki and Ei by minimizing the follow- ing quadratic loss function subject to some constraint: Q Ki; Ei ð Þ ¼ X Li x¼1 X p j¼1 kijei x ð Þ−f ij x ð Þ h i2 s:t:kijei x ð Þ −f ij x ð Þ≥0; kij; ei x ð Þ > 0; ∀j; ∀x: We can configure this problem into a non-negative matrix factorization problem [35, 36] as follows: min Ki;Ei ‖KiEi T−Fi ‖2 s:t:Fi ≤KiEi T; Ki ≥0; Ei ≥0; where ‖∙‖2 stands for the element-wise quadratic norm (i.e., sum of squared elements), and ≤and ≥for element- wise logical comparison. We call this a rank-one non-negative matrix factorization over-approximation (NMF-OA) problem as Ki and Ei both have rank 1 and KiEi T ≥Fi. The proposed iterative algorithm is described in details in the Additional file 1. In summary, we conclude DegNorm provides a pipe- line for informative quantification of gene-/sample-spe- cific transcript degradation pattern and for effective correction of degradation bias in RNA-seq. We intend for DegNorm to serve as a general normalization method to improve the accuracy in the gene expression differentiation analysis. DegNorm algorithm Sequencing depth adjustment: given the current estimate of (Ki,Ei) for i = 1, … , n, define the degradation index (DI) score as: 1. Sequencing depth adjustment: given the current estimate of (Ki,Ei) for i = 1, … , n, define the degradation index (DI) score as: ρij ¼ 1− PLi x¼1 f ij x ð Þ PLi x¼1kijei x ð Þ : 3. Steps 1 and 2 are repeated until the algorithm converges. Graphically, ρij stands for the fraction of the total area under the curve kijei(x) but above fij(x) (Fig. 2e). The DI score is used to calculate an adjusted read count by extrapolation: The DE analysis using edgeR was carried out based on the degradation normalized read count ~Xij (rounded) at convergence with upper quartile (UQ) normalization for sequencing depth. ~Xij ¼ Xij 1−ρij : DegNorm algorithm Suppose we have p samples and n genes, Xij is the read count for gene i in sample j. For simplicity of notation, we first illustrate the proposed method by focusing on one gene. Let fij(x), x = 1, … , Li; i = 1, … , n; j = 1, … , p be the read coverage score for transcript i of length Li from sam- ple j. When different isoforms are present, the Li positions represent the assembly of all expressed exons in the se- quential order. We assume there is an envelope function ei(x) that defines the ideal shape of read coverage curve for gene i if no degradation exists. The actual ideal cover- age curve for the given gene in the jth sample is kijei(x), where kij denotes the confounded effect of sequencing depth and relative abundance of gene i in sample j. Second, given Fi that is appropriately normalized for sequencing depth, the scale factor Ki should reflect the relative abundance of the gene in the non-degraded re- gion of each sample (to be referred to as the baseline re- gion below) (Fig. 2a). The non-degraded region must Degradation causes fij(x) to deviate downward from kijei(x) in degraded region(s). Clearly, sampling error can Page 16 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology (2019) 2 Xiong et al. Genome Biology from the remaining bins is ≤0.1 or if 80% bins have been dropped (see details in Additional file 1). The remaining transcript regions on selected bins are regarded as the baseline. Denote the read coverage curve on baseline region as F i . preserve a similar shape in gene’s coverage curves from different samples. If one can first estimate Ki from the identified baseline region, then NMF-OA algorithm will lead to a better estimate of the envelope function Ei, par- ticularly in the situation when the degradation extent is severe as the GBM data. We account for these two con- siderations by proposing an iterative degradation normalization pipeline (DegNorm) as follows: g i  Run NMF-OA on F i . The resulting K i is a refined estimate of Ki, given which the envelope function can be obtained as: eiðxÞ ¼ max j f ijðxÞ k ij ( ) ; x ¼ 1; …; Li: 1. Simulations We identify a subset of bins on which Fi preserves the most similar shape as the envelope function Ei across all samples by progressively dropping bins that have the largest sum of squares of normalized residuals (Ri/Fi) and repeatedly applying NMF-OA to the remaining bins This step stops if the maximum DI score obtained  Divide each gene into 20 bins. Define the residual matrix as Ri = KiEi T −Fi . We identify a subset of bins on which Fi preserves the most similar shape as the envelope function Ei across all samples by progressively dropping bins that have the largest sum of squares of normalized residuals (Ri/Fi) and repeatedly applying NMF-OA to the remaining bins. This step stops if the maximum DI score obtained Simulations We first simulated the latent read count for each gene within each sample. For a given gene without DE, the la- tent counts (without degradation) in both control and treatment samples were randomly simulated from the negative binomial distribution with the same mean and same dispersion parameters that were randomly chosen from the fitted values of SEQC-AB data. For a gene with DE, the latent counts for the control samples were first simulated as above, and those for the treatment samples were simulated from the negative binomial with mean parameter increased/decreased by a factor of (1.5 + γ) for up- or downregulated genes, respectively, where γ was simulated randomly from an exponential distribution with mean = 1. In the second step, we simulated the degradation given the latent read count (simulations II– IV). For a given gene, we first chose a Gaussian mixture distribution that covers the entire range of total tran- script to model the read start position distribution. For simplicity, we only consider 5′ end degradation. Each read from a gene that was pre-selected for degradation degraded (or was disregarded) with the probability that depended on the start position of the read, defined by a cumulative distribution function (CDF) of a lognormal distribution. The degradation pattern and extent can be tuned by varying the parameters in the lognormal distri- bution (see more details in Additional file 1). Next, we calculate the sequencing depth scaling factor using the degradation-corrected total number of read count: sj ¼ Pn i¼1 ~Xij Medianj Pn i¼1 ~Xij   : In this paper, the results presented were all based on this scale normalization. Alternatively, we can use other normalization methods like TMM or UQ to calculate the normalizing constant sj based on the degradation-adjusted read count ~Xij if the presence of extreme outliers is a concern. 2. Degradation estimation: given the estimated sequencing depth sj from step 1, adjust the coverage curves as follows: fij←fij sj : fij←fij sj :  Let Fi = (fi1, … , fip)T. Run NMF-OA for each gene on updated coverage curves Fi and obtain the estimate of (Ki,Ei).  Divide each gene into 20 bins. Define the residual matrix as Ri = KiEi T −Fi .  Let Fi = (fi1, … , fip)T. Run NMF-OA for each gene on updated coverage curves Fi and obtain the estimate of (Ki,Ei). Additional files Additional file 1: Supplementary methods. (DOCX 109 kb) Additional file 2: Supplementary figures and figure legends. (PDF 69270 kb) Additional file 1: Supplementary methods. (DOCX 109 kb) Additional file 2: Supplementary figures and figure legends. (PDF 69270 kb) Page 17 of 18 Page 17 of 18 (2019) 20:75 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology Funding This work is partially supported by funds by NIH R01GM107177. Funding This work is partially supported by funds by NIH R01GM107177. 10. Risso D, Ngai J, Speed TP, Dudoit S. Normalization of RNA-seq data using factor analysis of control genes or samples. Nat Biotechnol. 2014; 32:896–902. Review history h h 8. Risso D, Schwartz K, Sherlock G, Dudoit S. GC-content normaliza RNA-Seq data. BMC Bioinformatics. 2011;12:480. The review history is included as Additional file 6. 9. Leek JT, Storey JD. Capturing heterogeneity in gene expression studies by surrogate variable analysis. PLoS Genet. 2007;3:1724–35. 9. Leek JT, Storey JD. Capturing heterogeneity in gene expression studies by surrogate variable analysis. PLoS Genet. 2007;3:1724–35. References 1. Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet. 2009;10:57–63. 1. Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet. 2009;10:57–63. Additional file 4: Table S2. This table compares the p values and q values of differential expression analysis by edgeR from different normalization methods for 12 genes which were PCR-verified positives in the AMPK data. (XLSX 12 kb) Additional file 4: Table S2. This table compares the p values and q values of differential expression analysis by edgeR from different normalization methods for 12 genes which were PCR-verified positives in the AMPK data. (XLSX 12 kb) 2. Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol. 2010;28:511–5. 3. Bullard JH, Purdom E, Hansen KD, Dudoit S. Evaluation of statistical methods for normalization and differential expression in mRNA-Seq experiments. BMC Bioinformatics. 2010;11:94. Additional file 5: Table S3. This table compares the sensitivity (true-positive rate) achieved by different normalization methods in the differential expression analysis at FPR (false-positive rate) threshold = 0.05 for simulations I–IV. (XLSX 8 kb) Additional file 5: Table S3. This table compares the sensitivity (true-positive rate) achieved by different normalization methods in the differential expression analysis at FPR (false-positive rate) threshold = 0.05 for simulations I–IV. (XLSX 8 kb) 4. Dillies MA, Rau A, Aubert J, Hennequet-Antier C, Jeanmougin M, Servant N, Keime C, Marot G, Castel D, Estelle J, et al. A comprehensive evaluation of normalization methods for Illumina high-throughput RNA sequencing data analysis. Brief Bioinform. 2013;14:671–83. Additional file 6: Review history. (DOCX 53 kb) 5. Robinson MD, Oshlack A. A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol. 2010;11:R25. Acknowledgements The authors would like to thank Dr. Xiaozhong (Alec) Wang and Dr. Xinkun Wang from Northwestern University, Dr. Biplab Dasgupta from Cincinnati Children’s Hospital and three anonymous reviewers for insightful comments and helpful discussions. 6. Dohm JC, Lottaz C, Borodina T, Himmelbauer H. Substantial biases in ultra- short read data sets from high-throughput DNA sequencing. Nucleic Acids Res. 2008;36:e105. 7. Li J, Jiang H, Wong WH. Modeling non-uniformity in short-read rates in RNA-Seq data. Genome Biol. 2010;11:R50. Availability of data and materials 11. Hansen KD, Brenner SE, Dudoit S. Biases in Illumina transcriptome sequencing caused by random hexamer priming. Nucleic Acids Res. 2010; 38:e131. The DegNorm python package is available for download at Github https:// nustatbioinfo.github.io/DegNorm/ [37] under the MIT license for distribution. The version of R codes used in this paper is available on Zenodo with DOI https://doi.org/10.5281/zenodo.2595528 [38]. 12. Roberts A, Trapnell C, Donaghey J, Rinn JL, Pachter L. Improving RNA- Seq expression estimates by correcting for fragment bias. Genome Biol. 2011;12:R22. The raw sequencing data can be downloaded from the NCBI Sequence Read Archive (SRA) or Gene Expression Omnibus (GEO). The GBM data can be downloaded under SRA accession number SRP023548 [26]. The PBMC data is available under SRA accession number SRP041955 [17]. The SEQC-AA and SEQC-AB data can be downloaded under accession number GSE47774 (SRR898239-SRR898254) [27] under accession number GSE49712 [28] respectively. The DLPFC dataset is available under accession number SRP108559 [22]. The FF and FFPE breast tumor data can be downloaded under accession number GSE11396 [29]. The AMPK kinase knockout data is available under accession number GSE82183 [30]. 13. Wang L, Nie J, Sicotte H, Li Y, Eckel-Passow JE, Dasari S, Vedell PT, Barman P, Wang L, Weinshiboum R, et al. Measure transcript integrity using RNA-seq data. BMC Bioinformatics. 2016;17:58. 14. Wang Y, Liu CL, Storey JD, Tibshirani RJ, Herschlag D, Brown PO. Precision and functional specificity in mRNA decay. Proc Natl Acad Sci U S A. 2002;99: 5860–5. 15. Yang E, van Nimwegen E, Zavolan M, Rajewsky N, Schroeder M, Magnasco M, Darnell JE Jr. Decay rates of human mRNAs: correlation with functional characteristics and sequence attributes. Genome Res. 2003;13:1863–72. The processed data used in this paper has been deposited in Zenodo https://doi.org/10.5281/zenodo.2595303 [39]. The simulated data and R codes used in the analysis of this paper are available on Zenodo with DOI https://doi.org/10.5281/zenodo.2595559 [40]. 16. Houseley J, Tollervey D. The many pathways of RNA degradation. Cell. 2009; 136:763–76. 17. Gallego Romero I, Pai AA, Tung J, Gilad Y. RNA-seq: impact of RNA degradation on transcript quantification. BMC Biol. 2014;12:42. 17. Gallego Romero I, Pai AA, Tung J, Gilad Y. RNA-seq: impact of RNA degradation on transcript quantification. BMC Biol. 2014;12:42. Received: 23 July 2018 Accepted: 27 March 2019 Additional file 3: Table S1. This table summarizes the number of false positives claimed in differential expression analysis by edgeR from different normalization methods at q value threshold = 0.05 for data sets including SEQC-AA, GBM RIN10 vs. RIN4, GBM RIN6 vs. RIN4, PBMC-S01, DLPFC Br1729 Ribo-zero, DLPFC Br1729 poly(A)+, and breast tumor FF vs. (XLSX 9 kb) Competing interests 22. Jaffe AE, Tao R, Norris AL, Kealhofer M, Nellore A, Shin JH, Kim D, Jia Y, Hyde TM, Kleinman JE, et al. qSVA framework for RNA quality correction in differential expression analysis. Proc Natl Acad Sci U S A. 2017;114:7130–5. Authors’ contributions BX d JW d d 18. Copois V, Bibeau F, Bascoul-Mollevi C, Salvetat N, Chalbos P, Bareil C, Candeil L, Fraslon C, Conseiller E, Granci V, et al. Impact of RNA degradation on gene expression profiles: assessment of different methods to reliably determine RNA quality. J Biotechnol. 2007;127:549–59. BX and JW conceived and designed the study. BX, YY, and JW developed the methods. BX and FF developed the DegNorm algorithm and software package. BX, YY, FF, and JW wrote the manuscript. All authors read and approved the final manuscript. 19. Schroeder A, Mueller O, Stocker S, Salowsky R, Leiber M, Gassmann M, Lightfoot S, Menzel W, Granzow M, Ragg T. The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC Mol Biol. 2006;7:3. Consent for publication Not applicable. Consent for publication Not applicable. 21. Finotello F, Lavezzo E, Bianco L, Barzon L, Mazzon P, Fontana P, Toppo S, Di Camillo B. Reducing bias in RNA sequencing data: a novel approach to compute counts. BMC Bioinformatics. 2014;15(Suppl 1):S7. Competing interests The authors declare that they have no competing interests. Ethics approval and consent to participate Ethics approval and consent to participate Not applicable. 20. Feng H, Zhang X, Zhang C. mRIN for direct assessment of genome-wide and gene-specific mRNA integrity from large-scale RNA-sequencing data. Nat Commun. 2015;6:7816. Not applicable. Not applicable. Publisher’s Note 23. Pan Q, Shai O, Lee LJ, Frey BJ, Blencowe BJ. Deep surveying of alternative splicing complexity in the human transcriptome by high-throughput sequencing. Nat Genet. 2008;40:1413–5. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Page 18 of 18 Page 18 of 18 Xiong et al. Genome Biology (2019) 20:75 Xiong et al. Genome Biology Xiong et al. Genome Biology (2019) 20:75 24. Anders S, Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010;11:R106. 25. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40. 26. Sigurgeirsson B, Emanuelsson O, Lundeberg J. Sequencing degraded RNA addressed by 3′ tag counting. PLoS One. 2014;9:e91851. addressed by 3′ tag counting. PLoS One. 2014;9:e91851. 27. Consortium SM-I. A comprehensive assessment of RNA-seq accuracy, reproducibility and information content by the Sequencing Quality Control Consortium. Nat Biotechnol. 2014;32:903–14. 28. Rapaport F, Khanin R, Liang Y, Pirun M, Krek A, Zumbo P, Mason CE, Socci ND, Betel D. Comprehensive evaluation of differential gene expression analysis methods for RNA-seq data. Genome Biol. 2013;14:R95. 29. Bossel Ben-Moshe N, Gilad S, Perry G, Benjamin S, Balint-Lahat N, Pavlovsky A, Halperin S, Markus B, Yosepovich A, Barshack I, et al. mRNA-seq whole transcriptome profiling of fresh frozen versus archived fixed tissues. BMC Genomics. 2018;19:419. 30. Chhipa RR, Fan Q, Anderson J, Muraleedharan R, Huang Y, Ciraolo G, Chen X, Waclaw R, Chow LM, Khuchua Z, et al. AMP kinase promotes glioblastoma bioenergetics and tumour growth. Nat Cell Biol. 2018;20:823–35. 31. Storey JD. A direct approach to false discovery rates. J. R. Stat. Soc. Ser. B Stat Methodol. 2002;64:479–98. 32. Storey JD. The positive false discovery rate: a Bayesian interpretation and the q-value. Ann Stat. 2003;31:2013–35. 32. Storey JD. The positive false discovery rate: a Bayesian interpretation and the q-value. Ann Stat. 2003;31:2013–35. 33. Storey JD, Tibshirani R. Statistical significance for genomewide studies. Proc Natl Acad Sci U S A. 2003;100:9440–5. 33. Storey JD, Tibshirani R. Statistical significance for genomewide studies. Proc Natl Acad Sci U S A. 2003;100:9440–5. 34. Canales RD, Luo Y, Willey JC, Austermiller B, Barbacioru CC, Boysen C, Hunkapiller K, Jensen RV, Knight CR, Lee KY, et al. Evaluation of DNA microarray results with quantitative gene expression platforms. Nat Biotechnol. 2006;24:1115–22. 34. Publisher’s Note Canales RD, Luo Y, Willey JC, Austermiller B, Barbacioru CC, Boysen C, Hunkapiller K, Jensen RV, Knight CR, Lee KY, et al. Evaluation of DNA microarray results with quantitative gene expression platforms. Nat Biotechnol. 2006;24:1115–22. 35. Gillis N, Plemmons RJ: Dimensionality reduction, classification, and spectral mixture analysis using nonnegative underapproximation. Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery Xvi 2010, 7695. 35. Gillis N, Plemmons RJ: Dimensionality reduction, classification, and spectral mixture analysis using nonnegative underapproximation. Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery Xvi 2010, 7695. 36. Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization. Nature. 1999;401:788–91. 36. Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization. Nature. 1999;401:788–91. 37. Xiong B, Yang Y, Fineis F and Wang JP. Python package of DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis. 2019. https://nustatbioinfo.github.io/DegNorm/ 37. Xiong B, Yang Y, Fineis F and Wang JP. Python package of DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis. 2019. https://nustatbioinfo.github.io/DegNorm/ 38. Xiong B, Yang Y, Fineis F and Wang JP. Rcodes of DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis. 2019. https://doi.org/10.5281/zenodo.2595528 39. Xiong B, Yang Y, Fineis F and Wang JP. Processed data to DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis. 2019. https://doi.org/10.5281/zenodo.2595303 40. Xiong B, Yang Y, Fineis F and Wang JP. Simulation data and R codes for analyses to DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis. 2019. https://doi.org/10.5281/ zenodo.2595559 40. Xiong B, Yang Y, Fineis F and Wang JP. Simulation data and R codes for analyses to DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis. 2019. https://doi.org/10.5281/ zenodo.2595559

W2317029373.txt

https://www.biodiversitylibrary.org/partpdf/87115

de

A Bittern Taken in West Florida

˜The œAuk/˜The œauk

public-domain

Me | General Notes. 905 ‘Key to the Water Birds of Florida’ is that “it occasionally wanders to our shores.”’ In Bulletin No. 26 of the Biological Survey, Wells W. Cooke in his article ‘ Distribution and Migration of North American Ducks, Geese and Swans’ says, in respect to the European Widgeon, that ‘fon the Atlantic coast the dates are almost entirely in the fall and winter.... there are only three records after February 5.” If this is the case, and I find no evidence to the contrary, this is the fourth record only after the date he mentions, for the Atlantic coast.—S. Prescorr Fay, Boston, Mass. Two Records of the Golden-eye at DeFuniak Springs, Florida. — On December 5, 1907, an adult female Golden-eye Duck (Clangula clangula americana) was taken on a small fresh-water lake near De Funiak Springs, Florida.. The specimen was taken by M. Harry Moore, and mounted by him. It was sent to the American Museum of Natural History, and identified by W. De W. Miller from an examination of the specimen. On November 20, 1909, an adult female of the same species was seen on the same lake. It was observed under such favorable conditions that there was no doubt in my mind as to the identification. Prof. Wells W. Cooke says that this is ‘‘almost the extreme limit of its southern range, as it has been noted only a few times in Florida.’”’— G. CuypvrE FisHer, De Funiak Springs, Florida. Nesting of the Black Duck in Yates County, N. Y. — May 26, 1907, I found a nest of the Black Duck (Anas rubripes) in Potter Swamp. The locality was a young second growth of maple, beech and ash of four to eight inches in diameter. The ground was nearly dry and covered with a rank growth of ferns and skunk cabbage. The female was flushed from a bunch of six small maples growing from a mound about three feet above the surrounding ground and there in the center of the bunch of trees were six eggs layed on a few broken fern stems and dead leaves. | On my next visit (June 2) there were nine eggs and a nice lot of down had been placed around and among the eggs. A farmer told Mr. C. F. Stone of finding a nest of this duck in Potter Swamp in 1892 or 1893, but this is the first authentic record to my knowledge.— Vrerpi BurrcH, Branchport, N. Y. A Bittern taken in West Florida.— On November 17, 1909, an adult male Bittern (Botaurus lentiginosus) was taken near De Funiak Springs, Florida. This is the first individual of this species that I have seen in this vicinity in the two and one-half years that I have been here, so I conclude that it is probably somewhat rare. This specimen, which was in excellent condition, was taken in a dry ‘ broom-sedge’ field,— a rather unusual place, it seemed to me. Its stomach contained two grasshoppers which it had probably captured in this field. The bird was killed by Wm. F. Jones while out quail hunting. 1910. "Nesting of the Black Duck in Yates County, N. Y." The Auk 27, 205–205. https://doi.org/10.2307/4071116. View This Item Online: https://www.biodiversitylibrary.org/item/54324 DOI: https://doi.org/10.2307/4071116 Permalink: https://www.biodiversitylibrary.org/partpdf/87115 Holding Institution Smithsonian Libraries and Archives Sponsored by Smithsonian Copyright & Reuse Copyright Status: Public domain. The BHL considers that this work is no longer under copyright protection. This document was created from content at the Biodiversity Heritage Library, the world's largest open access digital library for biodiversity literature and archives. Visit BHL at https://www.biodiversitylibrary.org. This file was generated 1 April 2024 at 12:12 UTC 

https://openalex.org/W3049299837

https://qmro.qmul.ac.uk/xmlui/bitstream/123456789/83782/2/Rose%20De%20Novo%20Designed%20Peptide%20and%20Protein%20Hairpins%20Self-Assemble%20into%20Sheets%20and%20Nanoparticles%202021%20Published.pdf

English

<i>De Novo</i> Designed Peptide and Protein Hairpins Self-assemble into Sheets and Nanoparticles

bioRxiv (Cold Spring Harbor Laboratory)

cc-by

De Novo Designed Peptide and Protein Hairpins Self-Assemble into Sheets and Nanoparticles Johanna M. Galloway,* Harriet E. V. Bray, Deborah K. Shoemark, Lorna R. Hodgson, Jennifer Coombs, Judith M. Mantell, Ruth S. Rose, James F. Ross, Caroline Morris, Robert L. Harniman, Christopher W. Wood, Christopher Arthur, Paul Verkade, and Derek N. Woolfson 1. Introduction Coombs Bristol Centre for Functional Nanomaterials School of Physics University of Bristol HH Wills Physics Laboratory Tyndall Avenue Bristol BS8 1TL, UK Dr. R. S. Rose School of Biological and Chemical Sciences Fogg Building Queen Mary University of London Mile End Road, London E1 4QD, UK Dr. C. Morris School of Chemistry University of Glasgow 0/1 125 Novar Drive, Glasgow G12 9TA, UK Dr. C. W. Wood School of Biological Sciences Roger Land Building, King’s Buildings, Edinburgh EH9 3JQ, UK Dr. J. M. Galloway, Dr. H. E. V. Bray, Dr. J. F. Ross, Dr. C. Morris, Dr. R. L. Harniman, Dr. C. W. Wood, Dr. C. Arthur, Prof. D. N. Woolfson School of Chemistry University of Bristol Cantock’s Close, Bristol BS8 1TS, UK E-mail: [email protected] Dr. J. M. Galloway, Dr. J. F. Ross School of Chemistry University of Leeds Leeds LS2 9JT, UK Dr. D. K. Shoemark, Dr. L. R. Hodgson, Dr. J. Coombs, J. M. Mantell, Prof. P. Verkade, Prof. D. N. Woolfson School of Biochemistry University of Bristol Medical Sciences Building University Walk Bristol BS8 1TD, UK The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/smll.202100472. © 2021 The Authors. Small published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 1. Introduction Woolfson School of Biochemistry University of Bristol Medical Sciences Building University Walk Bristol BS8 1TD, UK DOI: 10.1002/smll.202100472 The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/smll.202100472. © 2021 The Authors. Small published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Dr. L. R. Hodgson, Dr. C. Morris, Dr. C. W. Wood, Prof. P. Verkade, Prof. D. N. Woolfson BrisSynBio/Bristol Biodesign Institute University of Bristol Life Sciences Building, Tyndall Avenue, Bristol BS8 1TQ, UK Dr. J. Coombs Bristol Centre for Functional Nanomaterials School of Physics University of Bristol HH Wills Physics Laboratory Tyndall Avenue Bristol BS8 1TL, UK Dr. R. S. Rose School of Biological and Chemical Sciences Fogg Building Queen Mary University of London Mile End Road, London E1 4QD, UK Dr. C. Morris School of Chemistry University of Glasgow 0/1 125 Novar Drive, Glasgow G12 9TA, UK Dr. C. W. Wood School of Biological Sciences Roger Land Building, King’s Buildings, Edinburgh EH9 3JQ, UK Small 2021, 17, 2100472 2100472  (1 of 11) Dr. J. M. Galloway, Dr. H. E. V. Bray, Dr. J. F. Ross, Dr. C. Morris, Dr. R. L. Harniman, Dr. C. W. Wood, Dr. C. Arthur, Prof. D. N. Woolfson School of Chemistry University of Bristol Cantock’s Close, Bristol BS8 1TS, UK E-mail: [email protected] Dr. J. M. Galloway, Dr. J. F. Ross School of Chemistry University of Leeds Leeds LS2 9JT, UK Dr. D. K. Shoemark, Dr. L. R. Hodgson, Dr. J. Coombs, J. M. Mantell, Prof. P. Verkade, Prof. D. N. Woolfson School of Biochemistry University of Bristol Medical Sciences Building University Walk Bristol BS8 1TD, UK DOI: 10.1002/smll.202100472 The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/smll.202100472. © 2021 The Authors. Small published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Small 2021, 17, 2100472 Dr. L. R. Hodgson, Dr. C. Morris, Dr. C. W. Wood, Prof. P. Verkade, Prof. D. N. Woolfson BrisSynBio/Bristol Biodesign Institute University of Bristol Life Sciences Building, Tyndall Avenue, Bristol BS8 1TQ, UK Dr. J. Research Article www.small-journal.com www.small-journal.com 1. Introduction The design and assembly of peptide-based materials has advanced consid- erably, leading to a variety of fibrous, sheet, and nanoparticle structures. A remaining challenge is to account for and control different possible supramo- lecular outcomes accessible to the same or similar peptide building blocks. Here a de novo peptide system is presented that forms nanoparticles or sheets depending on the strategic placement of a “disulfide pin” between two elements of secondary structure that drive self-assembly. Specifically, homodimerizing and homotrimerizing de novo coiled-coil α-helices are joined with a flexible linker to generate a series of linear peptides. The helices are pinned back-to-back, constraining them as hairpins by a disulfide bond placed either proximal or distal to the linker. Computational modeling indi- cates, and advanced microscopy shows, that the proximally pinned hairpins self-assemble into nanoparticles, whereas the distally pinned constructs form sheets. These peptides can be made synthetically or recombinantly to allow both chemical modifications and the introduction of whole protein cargoes as required. Self-assembling proteins form the building blocks of life to control many, if not all, cellular process. Natural self-assembling proteins include: 1D actin filaments for cell structure and motility;[1] 2D S-layers that form protective outer barriers in some bacteria;[2] 3D capsids that protect viral DNA/RNA,[3] and the shells of bacterial microcompartments, which are natural nanoreactors.[4] These natural scaffolds can be engineered to display other pro- teins to create assemblies for biotech- nology.[5] For example, enzyme pathways fused to viral capsid proteins self-assemble to create nanoreactors.[6] Virus-like parti- cles have also been engineered to display antigenic epitopes for use in vaccination,[7] or adorned with targeting motifs to direct them to diseased cells.[8] The toolbox of 2100472  (1 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH Dr. J. M. Galloway, Dr. H. E. V. Bray, Dr. J. F. Ross, Dr. C. Morris, Dr. R. L. Harniman, Dr. C. W. Wood, Dr. C. Arthur, Prof. D. N. Woolfson School of Chemistry University of Bristol Cantock’s Close, Bristol BS8 1TS, UK E-mail: [email protected] Dr. J. M. Galloway, Dr. J. F. Ross School of Chemistry University of Leeds Leeds LS2 9JT, UK Dr. D. K. Shoemark, Dr. L. R. Hodgson, Dr. J. Coombs, J. M. Mantell, Prof. P. Verkade, Prof. D. N. 2.1. Design of α-Helical Hairpin Peptides These modeled “C to Q” hairpin constructs did not remain flat, but instead flexed out of plane with no preferred direction of curvature, indi- cating that the assembly was unstable in the absence of the disulfide pins (Movie S3 and Figure S1c,d, Supporting Information).if Previously, we have used de novo α-helical coiled-coil pep- tides to make self-assembled peptide cages (SAGEs).[14] In SAGEs a homotrimeric coiled-coil peptide (CC-Tri3)[15] is joined back-to-back with a disulfide bond to one of two halves of a het- erodimeric coiled-coil pair (CC-DiA and CC-DiB).[15] This gener- ates two complementary building-blocks or hubs, A and B.[14] When mixed, these self-assemble into a hexagonal lattice that curves,[14] and forms closed structures with the aid of defects.[16] SAGEs have been adapted for uptake by mammalian cells as potential drug delivery vehicles,[17] and decorated with immu- nogenic peptide sequences to make a modular vaccine plat- form.[18] The peptide cages are permeable to small molecules, so are ideal scaffolds for nanoreactors.[19] The above examples all employ synthetically derived peptides and hubs. Natural proteins, such as enzymes and whole protein antigens, are too large to incorporate into SAGEs in this way. However, the system can be adapted to make peptide-protein fusions for recombinant production, leading to functionalized pSAGEs.[20] Others have pioneered approaches to construct large coiled- coil based nanoparticles. For example, Burkhard and co-workers have linked a de novo homotrimeric and modified natural homopentameric coiled-coil sequences with a short GG linker and an inter-helix disulfide bridge proximal to the loop which self-assemble into polyhedral nanoparticles.[12d,g]  The group has been particularly successful at developing these as vac- cine platforms for the presentation of antigenic peptides.[12e-i] In a different concept, Marsh and co-workers combine de novo coiled-coil units and natural oligomeric proteins to render defined protein nanocages.[21] These simulations gave the first indication of differences between the proximally and distally pinned systems. We ration- alize this in terms of different flexibilities in the constructs. For the proximal pin, the helical N and C termini have more freedom to explore space and move apart, resulting in wedge- shaped building-blocks, narrow at the loop and wider at the termini. A distal pin reduces this freedom and constrains the assembly to a flatter topology. Here we present self-assembling hairpin designs that form two distinct supramolecular assemblies; namely, closed nano- particles and extended sheets. These designs combine ideas from the SAGEs and Burkhard’s nanoparticles. 2.1. Design of α-Helical Hairpin Peptides Our design strategy used a GSGSG sequence to link two self- assembling coiled-coil 3 heptad building blocks: homodimeric (CC- Di) and homotrimeric (CC-Tri3).[15] This gives two possible linear peptide sequences: CC-Di–link–CC-Tri3 and CC-Tri3– link–CC-Di, abbreviated to DT and TD respectively. These were pinned back-to- back with a disulfide bond between cysteines placed at one of two complementary f positions either proximal (Cys15 & Cys35) or distal (Cys8 & Cys41) to the linking loop. In combination, this gives four possible constructs illustrated in Figure 1 and their sequences listed in Table 1. To assess any differences in the placement of the disulfide pins, we constructed all-atom models for 19-hexagon patches of assembled arrays for the two different pin positions and subjected these to molecular dynamics (MD) simulations in water (Figure 1c–f, Movies S1 and S2, Supporting Information). Intriguingly, the patch of the proximally pinned hairpins (pepHP–DTprox) curved in the 100 ns simulation. This resulted in the N and C termini of the peptides being presented on the convex face and the loops on the concave side. Projection of this curvature suggested that the arrays could close to form a nanoparticle 71  ±  11  nm in diameter, i.e., like the SAGEs (Figure S1a, Supporting Information). In contrast, simulations of arrays of distally pinned hairpins (pepHP–DTdist), the 19-hexagon patches remained relatively flat throughout the tra- jectories, with no preferred direction or magnitude of curvature (Figure S1b, Supporting Information). This suggested that these peptides might form flat extended sheets. The cysteines in the model assembly were replaced with glutamine, so that the hairpin building blocks could no longer form the disulfide pin. 2.1. Design of α-Helical Hairpin Peptides Specifically, homodimeric (CC-Di) and homotrimeric (CC-Tri3) blocks[15] are joined by a flexible linker, and pinned back-to-back with a disulfide bond (Figure  1). Computational modeling indicates that these should fold into stable hairpins that can be arrayed hexagonally. Moreover, and distinctively, the models suggest that the position of the disulfide may influence the supramo- lecular assembly profoundly: a pin proximal to the loop leads to curved arrays, which could close to form particles, whereas a distal pin restricts curvature, potentially leading to extended sheets. These extremes are confirmed experimentally using a useful self-assembling protein structures has been expanded by mutating natural protein interfaces to induce controlled self- assembly and through de novo design.[9] Such structures are usu- ally computationally designed to form closely packed 2D arrays[10], tubes,[11] or 3D icosahedral particles.[12] However, these beautifully ordered and near crystalline assemblies may not be amenable to decoration with large cargos, or be permeable to small mole- cules due to their close packed nature. Also, these engineered or designed arrays can require thermal annealing to assemble,[10d,e] which may reduce or even destroy the activity of any appended proteins. Alternatively, 1D fiber assemblies can form extensive 3D gels,[13] but controlling the localization of appended motifs and/ or functions is limited because network formation is a stochastic process. Between these two extremes of close-packed order and 3D entangled networks, there is space for the development of room-temperature self-assembling biomolecular systems that can display functional cargos and be permeable to small molecules. variety of advanced microscopy methods for hairpins made by peptide synthesis and when produced recombinantly. The vali- dated designs offer de novo scaffolds with potential to display a range of functionalities for application in imaging, cell tar- geting, nanoreactors, drug delivery, and modular vaccines. © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 DOI: 10.1002/smll.202100472 DOI: 10.1002/smll.202100472 2100472  (1 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 Small 2021, 17, 2100472 www.small-journal.com 2.2. Peptide Hairpins Assemble into Particles and Sheets as Predicted b) Cartoon of the envisaged hexagonal network formed when the hairpins self-assemble via their coiled-coil interfaces, labelled with intervertex dimensions measured from pre-simulated hexagons. c–f) Snapshots from molecular dynamics (MD) simulations of all-atom models of patches of assembled peptides. These were taken c,e) before and d,f) after 100 ns of MD under periodic boundary conditions using explicit TIP3P water, pH 7.0, 298 K, 0.15 m NaCl. Snapshots from assemblies of the c,d) proximally pinned hairpins (Movie S1, Supporting Information) and e,f) distally pinned hairpins (Movie S2, Supporting Information). Figure 1. Rational design and assembly of helical hairpin peptide. a) Schematic of hairpin design. The hairpins are constructed from two 3-heptad de novo coiled-coil peptides based on the homodimer CC-Di (grey, PDB code 4DZM) and the homotrimer CC-Tri3 (green, PDB code 4DZL). These are joined by a GSGSG “loop” (yellow) and an interhelix disulfide bond “pin” (black) between the polar helical facets. This leaves the coiled-coil forming facets exposed to engage in peptide–peptide interactions. The disulfide bond is placed either proximal or distal to the loop, which gives four possible hairpin configurations: with the dimer N terminal (left) or C terminal (right), with proximal (top) and distal (bottom) disulfide bonds. Full amino acid sequences for the designs are in Table 1. b) Cartoon of the envisaged hexagonal network formed when the hairpins self-assemble via their coiled-coil interfaces, labelled with intervertex dimensions measured from pre-simulated hexagons. c–f) Snapshots from molecular dynamics (MD) simulations of all-atom models of patches of assembled peptides. These were taken c,e) before and d,f) after 100 ns of MD under periodic boundary conditions using explicit TIP3P water, pH 7.0, 298 K, 0.15 m NaCl. Snapshots from assemblies of the c,d) proximally pinned hairpins (Movie S1, Supporting Information) and e,f) distally pinned hairpins (Movie S2, Supporting Information). the particles seemed to have a nonuniform size distribution. As the particle edges were difficult to discern in both the nega- tively stained and cryo-TEM images, we assessed particle size distribution by atomic force microscopy (AFM) (see below). All of the analogous peptides with the distal pin, pepHP–DTdist and pepHP–TDdist, formed sheet-like structures, extending for >100  nm  in the xy dimension (Figure  2i–p). Thus, these ini- tial experimental data support the MD simulations. 2.2. Peptide Hairpins Assemble into Particles and Sheets as Predicted All four peptide hairpin designs (Figure 1a and Table 1) were syn- thesized and purified by HPLC. The constructs were oxidized with iodine to form the disulfide pin, and subject to Ellman’s test[22] to confirm disulfide bond formation. The peptides were con- firmed as monomers by nanospray ionization mass spectrometry 2100472  (2 of 11) Small 2021, 17, 2100472 www.advancedsciencenews.com www.small-journal.com Figure 1. Rational design and assembly of helical hairpin peptide. a) Schematic of hairpin design. The hairpins are constructed from two 3-heptad de novo coiled-coil peptides based on the homodimer CC-Di (grey, PDB code 4DZM) and the homotrimer CC-Tri3 (green, PDB code 4DZL). These are joined by a GSGSG “loop” (yellow) and an interhelix disulfide bond “pin” (black) between the polar helical facets. This leaves the coiled-coil forming facets exposed to engage in peptide–peptide interactions. The disulfide bond is placed either proximal or distal to the loop, which gives four possible hairpin configurations: with the dimer N terminal (left) or C terminal (right), with proximal (top) and distal (bottom) disulfide bonds. Full amino acid sequences for the designs are in Table 1. b) Cartoon of the envisaged hexagonal network formed when the hairpins self-assemble via their coiled-coil interfaces, labelled with intervertex dimensions measured from pre-simulated hexagons. c–f) Snapshots from molecular dynamics (MD) simulations of all-atom models of patches of assembled peptides. These were taken c,e) before and d,f) after 100 ns of MD under periodic boundary conditions using explicit TIP3P water, pH 7.0, 298 K, 0.15 m NaCl. Snapshots from assemblies of the c,d) proximally pinned hairpins (Movie S1, Supporting Information) and e,f) distally pinned hairpins (Movie S2, Supporting Information). Figure 1. Rational design and assembly of helical hairpin peptide. a) Schematic of hairpin design. The hairpins are constructed from two 3-heptad de novo coiled-coil peptides based on the homodimer CC-Di (grey, PDB code 4DZM) and the homotrimer CC-Tri3 (green, PDB code 4DZL). These are joined by a GSGSG “loop” (yellow) and an interhelix disulfide bond “pin” (black) between the polar helical facets. This leaves the coiled-coil forming facets exposed to engage in peptide–peptide interactions. The disulfide bond is placed either proximal or distal to the loop, which gives four possible hairpin configurations: with the dimer N terminal (left) or C terminal (right), with proximal (top) and distal (bottom) disulfide bonds. Full amino acid sequences for the designs are in Table 1. © 2021 The Authors. Small published by Wiley-VCH GmbH 2.2. Peptide Hairpins Assemble into Particles and Sheets as Predicted d,h,l, p) Cryo-TEM images (unannotated on the left, peptide structure edges highlighted in yellow dotted lines on the right to distinguish them from the lacey carbon support delineated in pink). Hairpin peptide aliquots (100 × 10–6 m) were assembled for 1 h in HBS (25 × 10–3 m HEPES, 25 × 10–3 m NaCl, pH 7.2). Figure 2. Negative stained TEM, FFTs, and cryo-TEM of peptide hairpin assemblies. Pin proximal to the loop: a–d) pepHP–TDprox and e– h) pepHP–DTprox, formed closed structures. Pin distal to the loop: i–l) pepHP–TDdist and m–p) pepHP–DTdist, formed flat structures. Negative stained samples at a,e) low, b,f,i,m) intermediate, and c,g,j,n) high magnification. Fourier transforms (FFT) from distally pinned hairpins, area highlighted in j) FFT shown k) from pepHP–DTdist, and area highlighted in n) FFT shown in o) for pepHP–TDdist. d,h,l, p) Cryo-TEM images (unannotated on the left, peptide structure edges highlighted in yellow dotted lines on the right to distinguish them from the lacey carbon support delineated in pink). Hairpin peptide aliquots (100 × 10–6 m) were assembled for 1 h in HBS (25 × 10–3 m HEPES, 25 × 10–3 m NaCl, pH 7.2). a multilamellar, stiffened particles, with the same diameter as seen in the earlier assembly. Next, we used AFM to probe the nature and dimensions of the assemblies in more detail, Figure  3. First, at 1  h postas- sembly, the proximally pinned peptide pepHP–DTprox, formed nanoparticles with an average maximum height of 27 ± 22 nm and average diameter 101 ± 58 nm (n = 5400) Figure 3a–c. The large ranges reflected a bimodal distribution of the aspect ratios, centered on ≈0.1 and ≈0.4. After 24 h, the particles had approximately the same diameter (104 ± 79 nm), but only the thicker particles were apparent, with a height average of 39 ± 38 nm and aspect ratio 0.37, Figure 3d–f. Though the dis- tributions were broad, these data were consistent with the par- ticle sizes seen in TEM and cryo-TEM, and estimated from the modeling (71 ± 11 nm), and showed that the particle diameters were stable over time. In addition, the experimental data indi- cated maturation of the particles between 1 and 24 h. 2.2. Peptide Hairpins Assemble into Particles and Sheets as Predicted Further, as there was no discernible difference in assemblies produced by the constructs with CC-Di or CC-Tri3 units placed first in the sequence, all further data presented had the CC-Di sequence at the N terminus, i.e., HP-DT constructs. (Figures S2 and S3, Supporting Information, sequences and mass data in Table S1, Supporting Information). As judged by a change from clear to cloudy, all four peptides began to assemble as soon as they were hydrated in aqueous buffer. As a result, the samples scattered light and it was not possible to record reliable circular dichroism spectra to confirm α-helical assemblies. Instead, we visualized the assemblies by negative-stain trans- mission electron microscopy (TEM) and cryogenic transmis- sion electron microcopy (cryo-TEM) (Figure  2). With either CC-Di or CC-Tri3 as the N-terminal block, the proximally pinned peptides pepHP–DTprox and pepHP–TDprox respectively, formed closed nanoparticle structures (Figure 2a–h). In TEM, Table 1. Peptide sequences synthesized using solid-phase peptide synthesis. Peptide name and corresponding amino acid sequence aligned with numbering of amino acid position (1 to 51) and coiled-coil register (a to g). Sequences based on a basis set of coiled-coils,[15] with the sequence based on the homodimer (CC-Di) shown in grey, and the sequence based on the homotrimer (CC-Tri3) shown in green. Cysteine mutations introduced at f positions in order to form a disulfide bonded hairpin are underlined. Linkers (G, GSGSG) and chromophore tags (GYY) are shown in black. 2100472  (3 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH Name pepHP–DTdist pepHP–DTprox pepHP–TDdist pepHP–TDprox Small 2021, 17, 2100472 2100472  (3 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH Name pepHP–DTdist pepHP–DTprox pepHP–TDdist pepHP–TDprox Small 2021, 17, 2100472 2100472  (3 of 11) 2100472  (3 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Figure 2. Negative stained TEM, FFTs, and cryo-TEM of peptide hairpin assemblies. Pin proximal to the loop: a–d) pepHP–TDprox and e– h) pepHP–DTprox, formed closed structures. Pin distal to the loop: i–l) pepHP–TDdist and m–p) pepHP–DTdist, formed flat structures. Negative stained samples at a,e) low, b,f,i,m) intermediate, and c,g,j,n) high magnification. Fourier transforms (FFT) from distally pinned hairpins, area highlighted in j) FFT shown k) from pepHP–DTdist, and area highlighted in n) FFT shown in o) for pepHP–TDdist. © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 2.2. Peptide Hairpins Assemble into Particles and Sheets as Predicted Proximally pinned pepHP–DTprox (100 × 10–6 m) were assembled in HBS (25 × 10–3 m HEPES, 25 × 10–3 m NaCl, pH 7.2) for a–c) 1 h and d–f) 24 h, then deposited onto mica. Images were recorded using peak force atomic force microscopy (PF-AFM, a,d). Particle height and diameter data were extracted using a particle analysis script (link available in Supporting Informa- tion) and plotted to show height versus diameter b,e) and diameter versus aspect ratio c,f) (areas colored in red were the highest populated, those in blue the lowest populated). In b,c) n = 5400 particles, in e,f) n = 613 particles. Distally pinned pepHP–DTdist was assembled as above for 1 h, deposited on mica and imaged using tapping mode AFM (TM-AFM, g). Height profiles were measured across a tear using Nanoscope analysis v1.5. (positions annotated on (h) and shown in (i)) where dotted lines indicate the area averaged to get the height difference between the mica substrate and the assembled peptide sheet, shown in Table S2 (Supporting Information), n = 5, ± standard deviation. 2.2. Peptide Hairpins Assemble into Particles and Sheets as Predicted We cannot be sure what this is due to, but posit that recruitment of pep- tides to the structures over time may result in organization into AFM of the distally pinned peptide, pepHP–DTdist, revealed thin (5.6  ±  0.2 nm, Table S2, Supporting Information n = 5 ± standard deviation) sheets (Figure 3g–i and Figure S4, Supporting Information). This is similar to the expected height of one hairpin from termini to the loop, which is ≈5  nm. This value, and the tight distribution of the experimental data support the hypoth- esis that these peptides form unilamellar sheets. It was not pos- sible to discern the lattice clearly in AFM (Figure S5, Supporting Information), which could be due to tip resolution, or flexibility and thermal motions in the assembly. However, fast Fourier transform (FFT) analysis of further TEM and cryo-TEM images (Figure 4b,d) revealed regular structures with spacings of 4.8 nm. This corresponds closely to the expected inter-vertex (inter-trimer) distances of the hexagonal lattice, which span four helices each just over ≈1 nm in diameter (Figure 1b). 2100472  (4 of 11) 2100472  (4 of 11) Small 2021, 17, 2100472 www.advancedsciencenews.com www.small-journal.com Figure 3. AFM grainsizing and height measurements of hairpin peptides. Proximally pinned pepHP–DTprox (100 × 10–6 m) were assembled in HBS (25 × 10–3 m HEPES, 25 × 10–3 m NaCl, pH 7.2) for a–c) 1 h and d–f) 24 h, then deposited onto mica. Images were recorded using peak force atomic force microscopy (PF-AFM, a,d). Particle height and diameter data were extracted using a particle analysis script (link available in Supporting Informa- tion) and plotted to show height versus diameter b,e) and diameter versus aspect ratio c,f) (areas colored in red were the highest populated, those in blue the lowest populated). In b,c) n = 5400 particles, in e,f) n = 613 particles. Distally pinned pepHP–DTdist was assembled as above for 1 h, deposited on mica and imaged using tapping mode AFM (TM-AFM, g). Height profiles were measured across a tear using Nanoscope analysis v1.5. (positions annotated on (h) and shown in (i)) where dotted lines indicate the area averaged to get the height difference between the mica substrate and the assembled peptide sheet, shown in Table S2 (Supporting Information), n = 5, ± standard deviation. www.small-journal.com www.advancedsciencenews.com Figure 3. AFM grainsizing and height measurements of hairpin peptides. © 2021 The Authors. Small published by Wiley-VCH GmbH 2.3. Mixed Assemblies Can Be Made, but Preassembled Structures Do Not Exchange Peptide Components were much larger than the nanoparticles formed by the proxi- mally pinned constructs, which made correlating the light and electron microscopy for individual sheets clearer than for the particles. The peptides had N-terminal 5(6)-carboxyfluorescein (green) and 5(6)-carboxytetramethylrhodamine (red) labels, Table S2 (Supporting Information). First, the labeled peptides were mixed when unfolded in 50% aqueous acetonitrile and To test for peptide mixing during assembly and for exchange postassembly, we made two fluorescently labeled variants of pepHP–DTdist and performed correlative light and electron microscopy (CLEM). We focused on this design as the sheets 2100472  (5 of 11) Small 2021, 17, 2100472 2100472  (5 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 www.small-journal.com www.advancedsciencenews.com Figure 4. Negative stain TEM, cryo-TEM and FFTs showing peptide and protein hairpin sheet structures. Distally pinned hairpin peptide pepHP–DTdist a) negatively stained and b) FFT of area indicated in yellow, and c) cryo-TEM and d) FFT of area indicated. Areas highlighted in (a) from negative stain are shown in e), and highlighted in (c) from cryo-TEM are shown in f), (enlarged 2× and brightness and contrast enhanced). Protein version proHP–HTdist g) negatively stained and h) FFT of area indicated, and i) cryo-TEM and j) FFT of area indicated. Areas exhibiting striped features of the assembled protein hairpins highlighted in (g) from negative stain are shown in (k), and highlighted in i) from cryo-TEM are shown in l) (enlarged 2× and brightness and contrast enhanced). Features in FFTs are labelled with their corresponding distance in real space in nm. Hairpin samples (100 × 10–6 m) were hydrated in HBS (25 × 10–3 m HEPES, 25 × 10–3 m NaCl, pH 7.2) for 1 h. Figure 4. Negative stain TEM, cryo-TEM and FFTs showing peptide and protein hairpin sheet structures. Distally pinned hairpin peptide pepHP–DTdist a) negatively stained and b) FFT of area indicated in yellow, and c) cryo-TEM and d) FFT of area indicated. Areas highlighted in (a) from negative stain are shown in e), and highlighted in (c) from cryo-TEM are shown in f), (enlarged 2× and brightness and contrast enhanced). Protein version proHP–HTdist g) negatively stained and h) FFT of area indicated, and i) cryo-TEM and j) FFT of area indicated. 2.3. Mixed Assemblies Can Be Made, but Preassembled Structures Do Not Exchange Peptide Components Areas exhibiting striped features of the assembled protein hairpins highlighted in (g) from negative stain are shown in (k), and highlighted in i) from cryo-TEM are shown in l) (enlarged 2× and brightness and contrast enhanced). Features in FFTs are labelled with their corresponding distance in real space in nm. Hairpin samples (100 × 10–6 m) were hydrated in HBS (25 × 10–3 m HEPES, 25 × 10–3 m NaCl, pH 7.2) for 1 h. cloning site and a His-tag[23] at the 3’ end (Figure S7, Tables S1 and S5, Supporting Information); the prescript “pro” refers to the recombinantly expressed protein constructs. These were overexpressed in SHuffle T7 cells, which are engineered to allow disulfide bond formation in their cytoplasm, then puri- fied and characterized by SDS-PAGE and confirmed as mono- mers by nanospray ionization mass spectrometry (Figure S8a–f, Supporting Information). Ellman’s test[22] confirmed that the molecules contained disulfide bonds. CLEM imaging of cell sections immunolabelled for the His-tag revealed protein dis- persed within the cells rather than forming inclusion bodies (Figure S9, Supporting Information). freeze dried. When hydrated, these samples formed mixed assemblies as judged by the coincidence of the fluorescence from the two labels and electron density in CLEM, Figure S6a–f (Supporting Information). This demonstrates that differ- ently decorated hairpin peptides can be combined into the same assembly. Second, green- and red-labeled hairpins were hydrated for 1  h separately before mixing, incubated for 1  h further, and then prepared for microscopy. In this case, the CLEM revealed distinct regions of green and red fluorescence, indicating that once assembled, the structures were stable and did not exchange peptide modules. Thus, despite their flexible construction, the hairpin peptides form robust and stable sheet assemblies from solution. Whereas the proximally pinned synthetic peptide hairpins formed spherical structures (Figure  2a–h), the recombinant variant, proHP–DTprox formed aggregates (Figure S10a, Sup- porting Information pH 7.2). However, like the peptide variant (Figure 4a–f), the distally pinned protein hairpin, proHP–DTdist formed sheets (Figure  4g–l). Interestingly, these were smaller and had a clear ultrastructure, in the form of stripes. These stripes were apparent in both negative stain TEM (Figure 4e,k) and cryo-TEM (Figure  4f,l), and thus they are not a drying © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 2100472  (6 of 11) Small 2021, 17, 2100472 3. Conclusion and Future Outlook In summary, we have described the design, modeling, synthesis, assembly and characterization of two types of de novo designed peptide hairpins. In these, sequences for coiled-coil dimers and trimers are joined through a flexible linker. The hairpins are then pinned, using disulfide bridges either proximal to or distal from the linker. The aim being to expose the hydrophobic seams of the coiled-coil segments and promote assembly of the hairpins into hexagonal arrays. The two pin positions have profoundly different effects on the topology of the supramo- lecular assemblies formed. The proximal pins lead to closed, spherical objects on the order of ≈100 nm in diameter, whereas the distal pin results in sheet-like assemblies consistent with a monolayer of folded, self-assembled hairpins. These struc- tures are observed in a range of advanced microscopies, and the proposed mechanism of formation are supported by extended molecular dynamics simulations. Specifically, the proximal pin allows splaying of the helical termini, which in turn leads to curved arrays that can close, whereas the distal pins give a more tightly constrained hairpin structure, consistent with building blocks required to make flat sheets. p p The assembly behavior of the protein hairpins was tested by hydration in HBS at pH values between 6.0 and 8.0 (Figure S10, Supporting Information). The distally pinned protein formed extensive sheet like structures at low pH (<pH  6.5), and 3D aggregates at higher pH (8.0). Stripes appeared in the images of the proHP–DTprox structures, but at slightly lower pH (6.5) when compared to the proHP–DTdist (7.2). Aggregates of proHP–DTprox were formed at low and high pH, indicating that the proximally pinned protein was not able to form stable structures. Mod- eling indicated that deprotonation of the His-tag induced the opposite sense of curvature to that caused by the proximal pin position. Thus proHP–DTdist may be too flexible, and the oppo- site sense of curvature induced by protonation versus pin posi- tion destabilized the assembly leading to an aggregation-based failure mechanism. The striped assemblies may represent a transition between the low pH “peptide-like” structure and the high pH “unprotonated curved” structures, probably induced by partial protonation of the His-tag. As its assembly was more stable, only proHP–DTdist was investigated further. As noted above, others have developed similar self-assem- bling peptide- and protein-based nanoparticles[12] or sheets.[10] So what are the differences and advantages to our hairpin system? 3. Conclusion and Future Outlook First, by including two points (loop and pin) that define the relative positions and rotational freedom of the two coiled-coil components, we are able to control the topology of the self-assembled structures by design in a single system to render closed nanoparticles or extended sheets. This level of dual control is encouraging for future biomaterials design using relatively short linear peptides. Second, although this is true for some other systems, the relatively short lengths of our hairpin designs and the minimal postsynthesis processing allows them to be made both synthetically and recombinantly. This has allowed us to decorate assemblies chemically, and through fusions to functional natural proteins; and to explore additional properties of the system. For example, experiments with fluorophore-labeled synthetic hairpin peptides show that coassemblies can be made when starting with mixtures 2.5. pH Alters Protein Hairpin Assembly Structure We constructed all-atom models for 19 hexagon patches of assembled arrays for the distally pinned and His-tagged protein proHP–DTdist. The pKa of the histidine side chain is near physi- ological pH.[26] MD simulations were run for the distally pinned protein hairpin proHP–DTdist with unprotonated and proto- nated His-tags as described above (Figure 5 and Movies S4 and S5, Supporting Information). When unprotonated, the patch curved after 100 ns simulation (Figure 5b). The curvature was opposite to that seen for the proximally pinned peptide, with the loop on the convex side in unprotonated proHP–DTdist rather than on the concave side as was seen for pepHP–DTprox. The protonated constructs maintained a flat trajectory throughout the 100 ns simulation (Figure 5d), as was seen for the distally pinned peptide pepHP–DTdist. 2.4. Hairpin Redesign for Protein Expression Next, we sought to add functional proteins to the hairpin constructs. For this, we turned to the expression of synthetic genes in Escherichia coli. We designed genes for two constructs, proHP–DTprox and proHP–DTdist, that harbored a 5’ multiple 2100472  (6 of 11) 2100472  (6 of 11) Small 2021, 17, 2100472 www.small-journal.com www.advancedsciencenews.com artefact and must reflect some underlying structure. FFTs of these revealed spots, which correspond to lines in real space separated by ≈4.8 nm and ≈9.7 nm in both negative stain and cryo-TEM (Figure 4h,j). As described above, this in consistent with the expected vertex-to-vertex spacing of ≈4.8 nm along the hexagon side, and of ≈9.5 nm across the twofold symmetry axis (Figure 1b). When compared to the peptide design, the protein has a flexible charge neutral region N terminal to the hairpin sequence, and a C-terminal His-tag. It is possible that the small sheets are dimerizing through the His-tag,[24] and forming slightly overlapping sheets, leading to Moiré fringes.[25] Alterna- tively, these protein assemblies may form corrugated or stripy structures. Therefore, we investigated how the protein patches behaved in silico and in vitro under different experimental conditions. Information) were expressed in SHuffle T7 cells and purified as described above (Figure S8, Supporting Information). The fluorescence in CLEM images of cell sections indicates that the sfGFP cargo folds in the cytoplasm (Figure S11, Supporting Information). Again, we were not able to see any assembled structures within the cytoplasm. If the hairpins were able to self-assemble in cells, it may be that their open network struc- ture is not electron dense enough to resolve in thin-section in TEM, so these constructs were also characterized in vitro. Ell- man’s test[22] confirmed that the molecules contained disulfide bonds after purification. Nanospray mass spectrometry and SDS-PAGE indicated that proGFP–HP–DTdist was monomeric (Figure S8g–i, Supporting Information). When hydrated in HBS, we observed flat sheets formed by the cargo laden hair- pins (Figure S12, Supporting Information). This demonstrates that distally pinned hairpins are amenable to being decorated with large proteins. © 2021 The Authors. Small published by Wiley-VCH GmbH 2.6. Successful Self-Assembly of Hairpins When Fused to Large Cargos Fusions of the hairpin to the fluorescent protein sfGFP (≈28  kDa) (Figure S7 and Tables S1 and S5, Supporting 2100472  (7 of 11) Small 2021, 17, 2100472 www.advancedsciencenews.com www.small-journal.com Figure 5. Snapshots from all atom MD simulations of 19 hexagon protein hairpin patches. Distally pinned protein hairpins (proHP–DTdist) with a,b) unprotonated (Movie S4, Supporting Information) and c,d) protonated (Movie S5, Supporting Information) His-tags. Snapshots taken (a,c) before (blue, Movie S4, Supporting Information) and b,d) after (magenta, Movie S5, Supporting Information) 100 nm of MD simulation under periodic boundary conditions using explicit TIP3P water, pH 7.0, 298 K, 0.15 m NaCl. n protein hairpin patches. Distally pinned protein hairpins (proHP–DTdist) with rotonated (Movie S5, Supporting Information) His-tags. Snapshots taken (a,c) nta, Movie S5, Supporting Information) 100 nm of MD simulation under periodic m NaCl. Figure 5. Snapshots from all atom MD simulations of 19 hexagon protein hairpin patches. Distally pinned protein hairpins (proHP–DTdist) with a,b) unprotonated (Movie S4, Supporting Information) and c,d) protonated (Movie S5, Supporting Information) His-tags. Snapshots taken (a,c) before (blue, Movie S4, Supporting Information) and b,d) after (magenta, Movie S5, Supporting Information) 100 nm of MD simulation under periodic boundary conditions using explicit TIP3P water, pH 7.0, 298 K, 0.15 m NaCl. system,[14] the homodimer sequence on the hairpins could be changed to a heterodimer pair.[27,28] This would afford an addi- tional level of control on the assembly. of unfolded peptide variants. However, once formed, there is no interchange between assembled structures. This approach could be used to add other functional moieties, e.g., for drug delivery and targeting diseased cells. Extending this to recom- binantly produced hairpin–protein fusions, which express and purify well, could allow the production of nanoparticles or sheets with combinations of small molecule and protein car- goes. © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 4. Experimental Section N,N’- diisopropylcarbodiimide (DIC) in 7 mL dimethylforamide (DMF) at 25 °C for 2 min, then at 25 W microwave irradiation at 50 °C for 5 min), washed (3 × 7 mL DMF), deprotected (7 mL 20% (v/v) morpholine in DMF at 20 W and 75 °C for 5 min), and washed (3 × 7 mL DMF) before adding the next Fmoc protected amino acid. Cryo-TEM: 5 µL in vitro assembled sample was blotted and plunge frozen onto Lacey carbon grids using a Leica EM GP automatic plunge freezer and stored in liquid N2. Frozen samples were loaded into a Gatan CryoTransfer specimen holder and imaged using a Tecnai 20 FEI 200  kV Twin Lens scanning transmission electron microscope (LaB6 filament, 200 keV) on a FEI Eagle 4k x 4k CCD camera. Higher resolution images were recorded on an FEI Talos Arctica equipped with a 200 kV X-FEG and a Gatan GIF Quantum LS energy filter on a Ceta 16 M CCD detector or Gatan K2 DED. In addition to ImageJ, cryo-TEM images were processed using the Gatan Microscopy Suite (GMS) 3 software.i Peptides were capped using 3  eq. acetic anhydride and 4.5  eq. N,N-diisopropylethylamine (DIPEA) in 10  mL DMF for 30  min at room temperature (see Supporting Information for details on fluorescent labeling). The peptides were washed (3 × 5  mL  DMF, 3 × 5  mL  dichloromethane (DCM)), dried, and cleaved (9.5  mL trifluoroacetic acid (TFA), 250  µL water and 250  µL triisopropylsilane (TIPS), 3  h). The resin was removed by filtration and excess TFA evaporated. The crude peptide was precipitated (30  mL diethyl ether (Et2O), 4 °C), pelleted by centrifugation, the Et2O discarded, the pellet dissolved in 50% (v/v) aqueous acetonitrile (MeCN), and freeze dried. Peptides were purified by reverse phase-high pressure liquid chromatography (RP-HPLC, buffer A = 0.1% (v/v) TFA in water, buffer B  = 0.1% (v/v) TFA in MeCN). ≈10  mg was dissolved in 3  mL (70% buffer A,30% buffer B) and reduced using 10  eq tris(2-carboxyethyl) phosphine (TCEP, ≈8  mg), before injection onto a Phenomenex C8 column. A gradient from 30% to 70% buffer B was applied over 30 min, and absorbance monitored at 220 and 280 nm to monitor elution of the peptide, which was checked for purity, pooled and then freeze dried. 4. Experimental Section A detailed description of the materials and methods is included in the Supporting Information. MD Simulation: The coiled-coil helices were arranged in a 19 hexagon patch.[14] The disulfide pins were constructed by hand and aligned, and a His-tag added if necessary, in InsightII (Accelrys), and minor discrepancies in disulfide bond lengths fixed to 2.0 Å with an in-house Fortran program. A box 4 nm larger than the patch assembly was filled with TIP3P water, sodium and chloride ions (0.15 m NaCl) and H+ consistent with pH 7.0 (≈3  million atoms). This was parameterized with the Amber-99SB9ildn force field, before undergoing 5000  steps of energy minimization. The GROMACS-4.6.7 suite was used to perform simulations as NPT ensembles at 298 K using periodic boundary conditions. The simulations were integrated with a leap-frog algorithm over a 2  fs time-step, constraining bond vibrations with the P-LINCS method. Structures were saved every 0.1 ns for analysis and run over 100 ns. Molecular graphics manipulations and visualizations were performed using InsightII, VMD- 1.9.1 and Chimera-1.10.2 and movies made using VMD.i Cell Sample Thin Sections: 2 µL of cell pellet was high pressure frozen on an EMPACT2 + RTS system, and freeze substituted with minimal stain and resin as per Lee et al.[33] Cured resin was 70 nm sectioned using an EM UC6 microtome and 45° diamond knife and collected onto TEM grids. For immunolabelling, sections were blocked (1% (w/v) bovine serum albumen (BSA) in phosphate buffered saline (PBS, 140 × 10–3 m NaCl, 2.7 × 10–3 m KCl, 10.1 × 10–3 m Na2HPO4, 1.8 × 10–3 m KH2PO4 pH 7.4, 2 × 5 min). Blocked sections were labeled with a primary antibody (1:500 6x-His-tag monoclonal mouse/IgG2b antibody in 1% BSA in PBS, 1 h), washed (3 × 5 min in 1% BSA in PBS), incubated with a fluorescent secondary antibody (DyLight 488 labelled secondary polyclonal goat anti mouse IgG antibody, 1 µg mL−1 in 1% BSA in PBS, 1 h), then washed in 1% BSA in PBS followed by water before drying. Fluorescence imaging (described above) was done before TEM. Peptide Synthesis and Purification: A low-loading rink amide resin support was used in a Liberty microwave peptide synthesizer for solid phase peptide synthesis.[30] Fmoc-protected amino acids (0.1 × 10–3 m) were coupled (4.5  eq chloro-1-hydroxybenzotriazole (Cl-HOBt), 5  eq. 4. Experimental Section The reduced peptide was dissolved (1 mg mL−1) in 20% (v/v) aqueous acetic acid, and 1–2 µL iodine solution mixed in to oxidize the cysteines and form the disulfide pin. After 30–60 min, excess iodine was quenched with 100 µL aliquots of 100 × 10-3 m sodium thiosulfate[31] (color change from yellow to clear). This was dried and purified by RP-HPLC as described above. Peptides were analyzed for purity using analytical RP-HPLC and nanospray mass spectrometry (Waters SYNAPT G2S) injected using an Advion TriVersa Nanomate autoinjector. Disulfide bond formation was confirmed using a quantitative Ellman’s test.[22] Protein Expression and Purification: Sequences were optimized for production in E. coli (Table S5, Supporting Information), ordered from GeneART, and transferred to pET3a (lactose control) and TBAD (arabinose control) expression vectors (Figure S7, Supporting Information). These were transformed into SHuffle T7 Express competent cells and expressed by autoinduction[34] in Luria-Bertani (LB) medium (10 g L-1 NaCl, 10 g L-1 tryptone, 5 g L-1 yeast extract) 400 mL plus 8  mL 50× auto induction solution (for pET3a 100  mL contains 25  mL glycerol, 2.5 g D-glucose and 10 g α-lactose; for TBAD 100 mL contains 25  mL glycerol, 5.0  g D-glucose plus 5.0  g L-arabinose) at 30 °C and 200 rpm for 24 h. Cells were pelleted (10 000 × g, 4 °C, 10 min) and frozen at -20 °C. Pellets were thawed and resuspended in 20  mL sonication buffer (TBS: 50 × 10–3 m Tris, 150 × 10–3 m NaCl, pH  8.0; plus: 0.1  × 10–3 m ethylenediaminetetraacetic acid (EDTA), 0.5  × 10-3 m phenylmethane sulfonyl fluoride (PMSF), and 1% (v/v) triton x100) and disrupted by sonication on ice (1  s burst, 1  s rest, 15  min). After centrifugation (29 000 × g, 4 °C, 10 min) the supernatant was applied to an immobilized metal ion affinity chromatography[23] column (Ni-NTA agarose resin, Qiagen), washed (TBS + 20 × 10–3 m imidazole) and the protein eluted (TBS + 300 × 10–3 m imidazole) in 2 mL fractions. Pooled fractions were purified by anion exchange on an ÄKTAprime plus (GE Healthcare) by loading onto 5 mL columns from GE Healthcare. A HiTrap DEAE FF low bind anion exchange column was used for small proteins (proHP–DTprox and proHP–DTdist) or a HiTrap Q FF high binding column for larger proteins (proGFP–HP–DTdist). www.small-journal.com www.small-journal.com www.advancedsciencenews.com a final peptide/protein concentration of 100 × 10-6 m, and incubated for 1–24 h. design, and that designs must be tested when appending func- tionality to any self-assembling building-block. However, it is also exciting, as it opens considerable possibilities for designing a wider range of biomaterial structures and functions. What- ever your stance, the rational and computational de novo design of biomaterials remains both challenging and full of potential. Atomic Force Microscopy (AFM): 5  µL in vitro assembled sample was dropped onto freshly cleaved mica and washed (3 × 100 µL water) and dried. PeakForce AFM (PF-AFM) imaging was done on Bruker Multimode with a Nanoscope V controller with Picoforce extender, and tapping mode (TM-AFM) on a Bruker Multimode with a Quadrexed Nanoscope 3D controller. Transmission Electron Microscopy (TEM): 5  µL in vitro assembled sample was dropped onto a TEM grid, washed (3 × 10 µL water) and dried. For fluorescence imaging for correlative light and electron microscopy (CLEM), grids were then imaged on a Leica DMI4000B inverted epifluorescence microscope using a 63× oil objective lens with a numerical aperture 1.4. These were then negatively stained with 1% (w/v) uranyl acetate, washed (10  µL water), and dried. Samples were imaged on a Tecnai 12-FEI 120  kV BioTwin Spirit TEM (tungsten filament, 120 keV) and a FEI Eagle 4k × 4k CCD camera. TEM images were processed and aligned for CLEM using ImageJ.[32] © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 2.6. Successful Self-Assembly of Hairpins When Fused to Large Cargos Third, the fabric of the system is potentially modular, and we envisage that other de novo coiled-coil units could be swapped in to change the homomeric components used here.[27] For instance, as we have demonstrated for the foregoing SAGE In all of these respects, the linear peptide hairpins that we describe provide de novo modules to add to the growing toolkit of components for the rational construction of biomaterials.[9,12c,29] In all of these respects, the linear peptide hairpins that we describe provide de novo modules to add to the growing toolkit of components for the rational construction of biomaterials.[9,12c,29] That said, how subtle differences in peptide design translate to relatively small changes in module structure and then to com- pletely different assembly topologies are both alarming and encouraging. It is of concern because it highlights how careful the design process must be in order to arrive at a targeted 2100472  (8 of 11) Small 2021, 17, 2100472 Keywords coiled coil, computational modeling, peptide design, protein design, self-assembly Received: January 25, 2021 Published online: February 15, 2021 p p y gi by nanospray mass spectrometry as described for peptides above. [1] T. D. Pollard, J. A. Cooper, Science 2009, 326, 1208. y y y Statistical Analysis: Particle sizes were determined from the computationally modeled patches by extrapolating the curvature of the patch and assuming this would form a closed sphere. An average size was calculated by measuring the extrapolated diameter at 19 points equally spaced in time between 5 and 50  ns, and ± one standard deviation of these diameters is the quoted error. The mass of the purified peptide or protein was calculated from the recorded nanospray spectra by fitting using the Micromass MassLynx software maximum entropy analysis MaxEnt1 to 1 Da accuracy. AFM height data were flattened using the first order flattening tool in the Nanoscope software (v1.5), and particle height and diameters extracted using a script available here https://github.com/ wells-wood-research/galloway-jg-hairpin-self-assembly-2020. This script extracted the maximum height (above a 5 nm cutoff) and diameter (above a 50 nm cutoff) of each particle in an AFM topography plot. R. P. Fagan, N. F. Fairweather, Nat. Rev. Microbiol. 2014, 12, [3] J. E. Johnson, W. Chiu, Curr. Opin. Struct. Biol. 2000, 10, 229 [4] C. A. Kerfeld, C. Aussignargues, J. Zarzycki, F. Cai, M. Sutter, Nat. Rev. Microbiol. 2018, 16, 277. [5] a) W. M. Aumiller, M. Uchida, T. Douglas, Chem. Soc. Rev. 2018, 47, 3433; b) A. M. Wen, N. F. Steinmetz, Chem. Soc. Rev. 2016, 45, 4074. [6] D. P. Patterson, B. Schwarz, R. S. Waters, T. Gedeon, T. Douglas, ACS Chem. Biol. 2014, 9, 359. [7] B. Schwarz, K. M. Morabito, T. J. Ruckwardt, D. P. Patterson, J. Avera, H. M. Miettinen, B. S. Graham, T. Douglas, ACS Biomater. Sci. Eng. 2016, 2, 2324. [8] M. J. Rohovie, M. Nagasawa, J. R. Swartz, Bioeng. Transl. Med. 2017, 2, 43. [9] J. L. Beesley, D. N. Woolfson, Curr. Opin. Biotechnol. 2019, 58, 175 [10] a) H. Pyles, S. Zhang, J. J. De Yoreo, D. Baker, Nature 2019, 571, 251; b) A. Thomas, J. F. Matthaei, F. Baneyx, Biotechnol. J. 2018, 13, e1800141; c) S. Gonen, F. DiMaio, T. Gonen, D. Baker, Science 2015, 348, 1365; d) H. V. Zhang, F. Polzer, M. J. Haider, Y. Tian, J. A. Villegas, K. L. Kiick, D. J. Pochan, J. G. Acknowledgements This research was supported by funding from a SLoLa (Strategic Longer and Larger) grant from the BBSRC “Development of supramolecular assemblies for enhancing cellular productivity and the synthesis of chemicals and biotherapeutics,” (BB/M002969/1). Dan Mulvihill and Martin Warren, who are based at the University of Kent, and Richard Pickersgill based at Queen Mary’s University of London, contributed to discussion and development of this work as part of this consortium. The authors also thank Evelyne Deery and Maria Stanley for supplying the pET3a and TBAD plasmids, (University of Kent) and Richard Sessions for support with the computational modeling (University of Bristol). The authors thank the EPSRC for awarding HECBiosym and an Archer Leadership Award for providing computer time on the U.K. Supercomputer Archer. The authors thank the BBSRC/EPSRC funded Synthetic Biology Research Centre (BrisSynBio, BB/L01386X/1) for providing funding for researchers and Bluegem. The authors thank the EPSRC for funding the Chemical Imaging Facility (PF-AFM, EP/ K035746/1), the mass spectrometry equipment (EP/K03927X/1), and the Wolfson Bioimaging Facility (fluorescence and electron microscopy, BB/ L014181/1) at the University of Bristol. [11] [11] a) J. F. Ross, G. C. Wildsmith, M. Johnson, D. L. Hurdiss, K. Hollingsworth, R. F. Thompson, M. Mosayebi, C. H. Trinh, E. Paci, A. R. Pearson, M. E. Webb, W. B. Turnbull, J. Am. Chem. Soc. 2019, 141, 5211; b) I. Uddin, S. Frank, M. J. Warren, R. W. Pickersgill, Small 2018, 14, 1704020; c) A. Pang, S. Frank, I. Brown, M. J. Warren, R. W. Pickersgill, J. Biol. Chem. 2014, 289, 22377. [12] a) J. B. Bale, S. Gonen, Y. Liu, W. Sheffler, D. Ellis, C. Thomas, D. Cascio, T. O. Yeates, T. Gonen, N. P. King, D. Baker, Science 2016, 353, 389; b) N. P. King, J. B. Bale, W. Sheffler, D. E. McNamara, S. Gonen, T. Gonen, T. O. Yeates, D. Baker, Nature 2014, 510, 103; c) P.-S. Huang, S. E. Boyken, D. Baker, Nature 2016, 537, 320; d) S. Raman, G. Machaidze, A. Lustig, U. Aebi, P. Burkhard, Nanomed.: Nanotechnol. Biol. Med. 2006, 2, 95; e) R. Dey, Y. Xia, M.-P. Nieh, P. Burkhard, ACS Biomater. Sci. Eng. 2017, 3, 724; f) C. P. Karch, P. Burkhard, Biochem. Pharmacol. 2016, 120, 1; g) T. A. P. F. Pimentel, Z. Yan, S. A. Jeffers, K. V. Holmes, R. S. Hodges, P. Burkhard, Chem. Biol. Drug Des. 2009, 73, 53; h) C. P. Karch, P. Burkhard, G. R. Acknowledgements Matyas, Z. Beck, JoVE 2019, e60103; i) Q. Guo, D. Dasgupta, T. A. P. F. Doll, P. Burkhard, D. E. Lanar, Methods 2013, 60, 242. Conflict of Interest The authors declare no conflict of interest. Keywords Saven, Sci. Adv. 2016, 2, e1600307; e) A. D. Merg, G. Touponse, E. v. Genderen, X. Zuo, A. Bazrafshan, T. Blum, S. Hughes, K. Salaita, J. P. Abrahams, V. P. Conticello, Angew. Chem., Int. Ed. 2019, 58, 13507; f) J. F. Matthaei, F. DiMaio, J. J. Richards, L. D. Pozzo, D. Baker, F. Baneyx, Nano Lett. 2015, 15, 5235; g) Z. Chen, M. C. Johnson, J. Chen, M. J. Bick, S. E. Boyken, B. Lin, J. J. De Yoreo, J. M. Kollman, D. Baker, F. DiMaio, J. Am. Chem. Soc. 2019, 141, 8891; h) A. J. Ben-Sasson, J. L. Watson, W. Sheffler, M. C. Johnson, A. Bittleston, L. Somasundaram, J. Decarreau, F. Jiao, J. Chen, I. Mela, A. A. Drabek, S. M. Jarrett, S. C. Blacklow, C. F. Kaminski, G. L. Hura, J. J. De Yoreo, J. M. Kollman, H. Ruohola-Baker, E. Derivery, D. Baker, Nature 2021, 589, 468. Supporting Information Supporting Information is available from the Wiley Online Library or from the author. Tris, 25 × 10–3 m NaCl) and high salt buffer (50 × 10–3 m Tris, 500 × 10–3 m NaCl), with the pH set to ≈1 above the pI of the protein (pH  9.0 for proHP–DTxxx, and pH  7.5 for proGFP–HP–DTdist), was applied at 2  mL per min for 30 min. Absorbance at 280 nm was monitored as a proxy for protein concentration, and eluted proteins were collected in 2  mL fractions. Fractions were analyzed for purity by sodium dodecyl sulfide polyacrylamide gel electrophoresis (SDS-PAGE, Figure S8, Supporting Information), buffer exchanged (50  × 10–3 m ammonium bicarbonate) and freeze dried. Smaller proteins (proHP–DTxxx) could also be purified by RP-HPLC; and protein purity, mass and oligomeric state was confirmed b t t d ib d f tid b © 2021 The Authors. Small published by Wiley-VCH GmbH 4. Experimental Section A gradient between a low salt (50 × 10–3 m Peptide and Protein Assembly: Peptide aliquots (1–10  nmol) were dissolved in 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffered saline (HBS, 25 × 10-3 m HEPES, 25 × 10-3 m NaCl, pH 7.2) to 2100472  (9 of 11) Small 2021, 17, 2100472 www.small-journal.com [27] F. Thomas, A. L. Boyle, A. J. Burton, D. N. Woolfson, J. Am. Chem. Soc. 2013, 135, 5161. [28] J. Jin, E. G. Baker, C. W. Wood, J. Bath, D. N. Woolfson, A. J. Turberfield, ACS Nano 2019, 13, 9927. [16] a) J. M. Galloway, L. Senior, J. M. Fletcher, J. L. Beesley, L. R. Hodgson, R. L. Harniman, J. M. Mantell, J. Coombs, G. G. Rhys, W. F. Xue, M. Mosayebi, N. Linden, T. B. Liverpool, P. Curnow, P. Verkade, D. N. Woolfson, ACS Nano 2018, 12, 1420; b) M. Mosayebi, D. K. Shoemark, J. M. Fletcher, R. B. Sessions, N. Linden, D. N. Woolfson, T. B. Liverpool, Proc. Natl. Acad. Sci. USA 2017, 114, 9014. [29] a) D. N. Woolfson, Z. N. Mahmoud, Chem. Soc. Rev. 2010, 39, 3464; b) Q. Luo, C. Hou, Y. Bai, R. Wang, J. Liu, Chem. Rev. 2016, 116, 13571; c) W. Zhang, S. Mo, M. Liu, L. Liu, L. Yu, C. Wang, Front. Chem. 2020, 8, 587975. [30] G. B. Fields, R. L. Noble, Int. J. Pept. Protein Res. 1990, 35, 161. [17] J. L. Beesley, H. E. Baum, L. R. Hodgson, P. Verkade, G. S. Banting, D. N. Woolfson, Nano Lett. 2018, 18, 5933. [31] D. Andreu, F. Albericio, N. A. Solé, M. C. Munson, M. Ferrer, G. Barany, in Methods in Molecular Biology: Peptide Synthesis Proto- cols, Vol. 35 (Eds: M. W. Pennington, B. M. Dunn), Humana Press Inc., Totowa, NJ 1994, p. 91.f [18] C. Morris, S. J. Glennie, H. S. Lam, H. E. Baum, D. Kandage, N. A. Williams, D. J. Morgan, D. N. Woolfson, A. D. Davidson, Adv. Funct. Mater. 2019, 29, 1807357. [32] a) M. D. Abramoff, P. J. Magalhaes, S. J. Ram, Biophotonics Int. 2004, 11, 36; b) J. Schindelin, C. T. Rueden, M. C. Hiner, K. W. Eliceiri, Mol. Reprod. Dev. 2015, 82, 518. [19] D. K. Shoemark, A. Avila Ibarra, J. F. Ross, J. L. Beesley, H. E. V. Bray, M. Mosayebi, N. Linden, T. B. Liverpool, S. N. McIntosh-Smith, D. N. Woolfson, R. B. Sessions, Phys. Chem. Chem. Phys. 2019, 21, 137. [33] M. J. Lee, J. Mantell, L. Hodgson, D. Alibhai, J. M. Fletcher, I. R. Brown, S. Frank, W.-F. Xue, P. Verkade, D. N. Woolfson, M. J. Warren, Nat. Chem. Biol. 2018, 14, 142. [20] J. F. Ross, A. Bridges, J. M. Fletcher, D. Shoemark, D. Alibhai, H. E. www.advancedsciencenews.com www.advancedsciencenews.com E. N. G. Marsh, Proc. Natl. Acad. Sci. USA 2016, 113, 8681; b) D. P. Patterson, A. M. Desai, M. M. B. Holl, E. N. G. Marsh, RSC Adv. 2011, 1, 1004; c) D. P. Patterson, M. Su, T. M. Franzmann, A. Sciore, G. Skiniotis, E. N. G. Marsh, Protein Sci. 2014, 23, 190; d) A. S. Cristie-David, J. Chen, D. B. Nowak, A. L. Bondy, K. Sun, S. I. Park, M. M. Banaszak Holl, M. Su, E. N. G. Marsh, J. Am. Chem. Soc. 2019, 141, 9207. B. Song, D. N. Woolfson, ACS Biomater. Sci. Eng. 2015, 1, 431; B. Song, D. N. Woolfson, ACS Biomater. Sci. Eng. 2015, 1, 431; b) E. F. Banwell, E. S. Abelardo, D. J. Adams, M. A. Birchall, A. Corrigan, A. M. Donald, M. Kirkland, L. C. Serpell, M. F. Butler, D. N. Woolfson, Nat. Mater. 2009, 8, 596; c) P. Worthington, S. Langhans, D. Pochan, Adv. Drug Delivery Rev. 2017, 110-111, 127; d) K. Nagy-Smith, Y. Yamada, J. P. Schneider, J. Mater. Chem. B 2016, 4, 1999; e) R. V. Rughani, J. P. Schneider, MRS Bull. 2008, 33, 530. b) E. F. Banwell, E. S. Abelardo, D. J. Adams, M. A. Birchall, d) A. S. Cristie-David, J. Chen, D. B. Nowak, A. L. Bondy, K. Sun, S. I. Park, M. M. Banaszak Holl, M. Su, E. N. G. Marsh, J. Am. Chem. Soc. 2019, 141, 9207. [22] G. L. Ellman, Arch. Biochem. Biophys. 1959, 82, 70 22] G. L. Ellman, Arch. Biochem. Biophys. 1959, 82, 70. [23] J. Porath, Protein Expression Purif. 1992, 3, 263. [14] J. M. Fletcher, R. L. Harniman, F. R. H. Barnes, A. L. Boyle, A. Collins, J. Mantell, T. H. Sharp, M. Antognozzi, P. J. Booth, N. Linden, M. J. Miles, R. B. Sessions, P. Verkade, D. N. Woolfson, Science 2013, 340, 595. [24] J. Wu, M. Filutowicz, Acta Biochim. Pol. 1999, 46, 591. [25] D. B. Williams, C. B. Carter, in Transmission Electron Microscopy: A Textbook for Materials Science, Springer, New York 2009, p. 389. [26] A. L. Hansen, L. E. Kay, Proc. Natl. Acad. Sci. USA 2014, 111, E1705. [15] J. M. Fletcher, A. L. Boyle, M. Bruning, G. J. Bartlett, T. L. Vincent, N. R. Zaccai, C. T. Armstrong, E. H. C. Bromley, P. J. Booth, R. L. Brady, A. R. Thomson, D. N. Woolfson, ACS Synth. Biol. 2012, 1, 240. Data Availability Statement The data that support the findings of this study are openly available from the Leeds University Data Repository at https://doi.org/10.5518/955, reference [35]. [13] a) N. Mehrban, B. Zhu, F. Tamagnini, F. I. Young, A. Wasmuth, K. L. Hudson, A. R. Thomson, M. A. Birchall, A. D. Randall, 2100472  (10 of 11) © 2021 The Authors. Small published by Wiley-VCH GmbH Small 2021, 17, 2100472 www.small-journal.com © 2021 The Authors. Small published by Wiley-VCH GmbH V. Bray, J. L. Beesley, W. M. Dawson, L. R. Hodgson, J. Mantell, P. Verkade, C. M. Edge, R. B. Sessions, D. Tew, D. N. Woolfson, ACS Nano 2017, 11, 7901. [34] F. W. Sturdier, Protein Expression Purif. 2005, 41, 207. [35] J. M. Galloway, H. E. V. Bray, D. K. Shoemark, L. R. Hodgson, J. Coombs, J. M. Mantell, R. S. Rose, J. F. Ross, C. Morris, R. L. Harniman, C. W. Wood, C. Arthur, P. Verkade, D. N. Woolfson, Dataset 2021, https://doi.org/10.5518/955. [21] a) A. Sciore, M. Su, P. Koldewey, J. D. Eschweiler, K. A. Diffley, B. M. Linhares, B. T. Ruotolo, J. C. A. Bardwell, G. Skiniotis, 2100472  (11 of 11) Small 2021, 17, 2100472 © 2021 The Authors. Small published by Wiley-VCH GmbH

https://openalex.org/W2130815378

https://www.ajol.info/index.php/tjpr/article/download/103809/93942

English

Effect of Betong Watercress and Phenethyl Isothiocyanate on N-Demethylation of Caffeine in Rats

Tropical journal of pharmaceutical research

cc-by

Abstract Keywords: Watercress, Phenethyl isothiocyanate, Caffeine, N-demethylation, Fluvoxamine Cytochrome Tropical Journal of Pharmaceutical Research is indexed by Science Citation Index (SciSearch), Scopus, International Pharmaceutical Abstract, Chemical Abstracts, Embase, Index Copernicus, EBSCO, African Index Medicus, JournalSeek, Journal Citation Reports/Science Edition, Directory of Open Access Journals (DOAJ), African Journal Online, Bioline International, Open-J-Gate and Pharmacy Abstracts Abstract Purpose: To investigate the effect of Betong watercress and phenethyl isothiocyanate (PEITC) on the N-demethylation of caffeine (CF) in rats y ( ) Methods: Male Wistar rats were subjected to 2 phases of experiment. Phase I, they received a single oral dose of CF (10 mg/kg), while in phase II, they were pretreated with the following regimens: 10 mg/kg fluvoxamine, i.p.; a single oral dose of 2, 10, and 20 mg/kg PEITC; 2, 10, and 20 mg/kg PEITC, once daily for five days; single oral dose of 800 mg/kg Betong watercress; 800 mg/kg Betong watercress once daily for five days, before receiving the same dose of CF as in phase I. Serum concentrations of CF and its metabolites after 3 h of CF administration were measured. Caffeine metabolic ratios (CMRs) and ratio of serum concentration of metabolites to that of CF were calculated and compared. p Results: CMRs were decreased by a single pretreatment \with fluvoxamine (30 – 40%), PEITC (2 - 20 mg/kg) (40 – 55%), and Betong watercress (9 – 22%). The decreases caused by 10 and 20 mg/kg PEITC were significantly greater than those by fluvoxamine. CMRs were also decreased after five days of pretreatment with all doses of PEITC (43 – 69%) and Betong watercress (28 – 44%). The reduction in metabolic ratio after single- and multiple PEITC pretreatments was dose- and time-independent. Conclusion: Betong watercress and PEITC inhibit N-demethylation of CF in rats. Such an effect indicates that they have inhibitory activity on CYP1A2 and CYP2C. p Results: CMRs were decreased by a single pretreatment \with fluvoxamine (30 – 40%), PEITC (2 - 20 mg/kg) (40 – 55%), and Betong watercress (9 – 22%). The decreases caused by 10 and 20 mg/kg PEITC were significantly greater than those by fluvoxamine. CMRs were also decreased after five days of pretreatment with all doses of PEITC (43 – 69%) and Betong watercress (28 – 44%). The reduction in metabolic ratio after single- and multiple PEITC pretreatments was dose- and time-independent. Conclusion: Betong watercress and PEITC inhibit N-demethylation of CF in rats. Such an effect indicates that they have inhibitory activity on CYP1A2 and CYP2C. Tropical Journal of Pharmaceutical Research April 2014; 13 (4): 559-566 ISSN: 1596-5996 (print); 1596-9827 (electronic) © Pharmacotherapy Group, Faculty of Pharmacy, University of Benin, Benin City, 300001 Nigeria. All rights reserved. Tropical Journal of Pharmaceutical Research April 2014; 13 (4): 559-566 ISSN: 1596-5996 (print); 1596-9827 (electronic) © Pharmacotherapy Group, Faculty of Pharmacy, University of Benin, Benin City, 300001 Nigeria. All rights reserved. Available online at http://www.tjpr.org http://dx.doi.org/10.4314/tjpr.v13i4.11 Available online at http://www.tjpr.org http://dx.doi.org/10.4314/tjpr.v13i4.11 Original Research Article Effect of Betong Watercress and Phenethyl Isothiocyanate on N-Demethylation of Caffeine in Rats Benjamas Janchawee1,2*, Angkana Wongsakul1, Sathaporn Prutipanlai1 and Pritsana Ruangrut3 1Department of Pharmacology, 2Natural Product Research Center, Faculty of Science, 3Department of Biomedical Sciences, Faculty of Medicine, Prince of Songkla University, Songkhla 90112, Thailand *For correspondence: Email: [email protected]; Tel: +66-74-446678; Fax: +66-74-446678 Revised accepted: 9 February 2014 Received: 13 April 2013 Authentication and preparation of Betong watercress CF is a marker for determining any influence of xenobiotics on this enzyme activity and caffeine metabolic ratios (CMRs) are acceptable indicators for determining CYP1A activity in humans and rats [8]. Fig 1: Betong watercress (Phak Num Betong; Nasturtium officinale R.Br.) collected from Betong district, Yala province by B. Janchawee; (A) fresh plant; (B) a voucher specimen (B. Janchawee #1). To prepare a dry powder, fresh Betong watercress (10 kg) was cleansed, roughly chopped and mixed with deionized water before homogenizing with a kitchen blender. The paste was wrapped with cheese cloth and squeezed to release the juice. The residue was mixed with another portion of deionized water and squeezed again. All extracts were pooled then lyophilized. The aim of this study was to investigate the effects of BW and PEITC on caffeine N- demethylation in rats. Experimental design An experiment consisted of phase I, referring to a single oral dose administration of CF (10 mg/kg), and phase II, referring to pretreatment with the study drugs before receiving the same dose of CF as in phase I. The wash-out period was two days. Rats were randomly divided into nine groups (n=6) which were assigned different pretreatments as follows. Group I, the animals were intraperitoneally injected with fluvoxamine at a single dose of 10 mg/kg. Groups II, III, and IV, the animals were pretreated with a single dose of PEITC at the doses of 2, 10, and 20 mg/kg, respectively. Groups V, VI, and VII, the animals were pretreated with multiple doses of PEITC at the doses of 2, 10, and 20 mg/kg, respectively, once daily for five days. The animals in group VIII were administered a single dose of 800 mg/kg of dry powders of Betong watercress juice while those in group IX received INTRODUCTION against chemical carcinogen-induced tumors in animals [2]. Watercress (Nasturtium officinale R.Br.) and other members of the Cruciferae (Brassicaceae) contain glucosinolates which can be hydrolysed by myrosinase to isothiocyanates (ITCs) after chewing. In the crucifer family, watercress contains the largest amounts of ITCs, mostly phenethyl isothiocyanate (PEITC) [1] which is a dietary substance that has a protective effect Watercress and PEITC influence phases I & II drug metabolizing enzymes. Watercress inhibits the CYP2E1-mediated metabolism of some drugs [3, 4], but higher doses enhance the activity of ethoxyresorufin-O-deethylase (EROD; CYP1A), and NAD(P)H-quinonereductase, in a dose-dependent manner [5]. The effect of the Trop J Pharm Res, April 2014; 13(2): 559 Janchawee et al Authentication and preparation of Betong watercress Chemicals Standard CF, TB, TP, PX, PEITC, fluvoxamine (FV), acetaminophen (AP), 1,2-benzenedithiol, and myrosinase were obtained from Sigma- Aldrich, USA. HPLC grade acetonitrile and tetrahydrofuran were obtained from Mallinckrodt Baker Inc., USA. All other chemicals were of analytical reagent grade. Authentication and preparation of Betong watercress PEITC on CYP1A expression and activity is still unclear. An in vitro study has demonstrated that CYP1A2 and other isoforms are inhibited by PEITC [6]. However, an animal study has shown inconsistent effects of PEITC on CYP1A activity [7]. In December 2009, Betong watercress was purchased from a local farm at Betong district, Yala province, Thailand. The whole plant was identified at the Princess Maha Chakri Sirindhorn Natural History Museum, Prince of Songkla University where the voucher specimen (B. Janchawee 1; Fig. 1B) has been deposited. Betong watercress (Phak Num Betong; Fig. 1A) is a variety of watercress cultivated in Thailand, mostly in the South at the Betong District of Yala Province. It is claimed that this plant is originally from France and has been brought to grow in China and then in Thailand by Chinese immigrants. However, its scientific name has never been formally confirmed. Additionally, no evidence of its effects on drug metabolizing enzymes such as CYP1A2 has been reported. Caffeine (CF) is primarily metabolized via N- demethylation to paraxanthine (PX), theobromine (TB) and theophylline (TP) and C-8-hydroxylation to 1,3,7-trimethyluric acid (TMU) in humans and rats. These reactions are mainly mediated by CYP1A2. CF is a marker for determining any influence of xenobiotics on this enzyme activity and caffeine metabolic ratios (CMRs) are acceptable indicators for determining CYP1A activity in humans and rats [8]. Betong watercress (Phak Num Betong; Fig. 1A) is a variety of watercress cultivated in Thailand, mostly in the South at the Betong District of Yala Province. It is claimed that this plant is originally from France and has been brought to grow in China and then in Thailand by Chinese immigrants. However, its scientific name has never been formally confirmed. Additionally, no evidence of its effects on drug metabolizing enzymes such as CYP1A2 has been reported. Fig 1: Betong watercress (Phak Num Betong; Nasturtium officinale R.Br.) collected from Betong district, Yala province by B. Janchawee; (A) fresh plant; (B) a voucher specimen (B. Janchawee #1). enzymes such as CYP1A2 has been reported. Caffeine (CF) is primarily metabolized via N- demethylation to paraxanthine (PX), theobromine (TB) and theophylline (TP) and C-8-hydroxylation to 1,3,7-trimethyluric acid (TMU) in humans and rats. These reactions are mainly mediated by CYP1A2. Chromatographic method Separation was modified from the previous method [12]. The Fortis® C18 column (150×4.6 mm i.d., 5 μm) connected to a guard column, Sunfire® (20×4.6 mm i.d., 5 μm) was used and controlled at 25ºC. The mobile phase was a methanol-water mixture (90:10, v/v) flowed at 1.1 mL/min. The sample (20 μL) was detected at 365 nm. Preparation of standard solutions p Stock solutions of CF and AP (internal standard) were prepared in deionized water. Those of PX and TP were prepared in 0.1 M ammonium hydroxide. That of TB was prepared in 0.1M sodium hydroxide. Working solutions (0.08 - 40 μg/mL) were prepared as a mixture with deionized water. Calibration standard mixtures (0.02 - 10 μg/mL) containing AP (25 μg/mL) were prepared. The amount of PEITC per 1 mg of Betong watercress dry power (Amt. PEITC) was calculated using the equation (1). ……(1) Sample derivatization Determination of PEITC required two steps; 1) hydrolysis of glucosinolates, and 2) ITC derivatization via the cyclocondensation reaction [12]. Samples (n=3) of Betong watercress dry powder (2 mg) were dissolved in deionized water (2 mL), then mixed with myrosinase solution (1 mL) and incubated at 37°C for 2 h. After centrifugation (875 x g, 15 min), the supernatant (100 μL) was mixed with cyclocondensation reagents, then incubated at 65°C for 2 h prior to HPLC injection. Method validation The analytical methods were validated in accordance with the USFDA guidance [13] in terms of linearity, precision, accuracy, recovery of extraction (except for PEITC), and lower limit of quantification (LLOQ). For CF and it metabolites, spiked serum samples (0.02-10 μg/mL, n=5) were used. The LLOQ was determined with a precision of 20% and accuracy of 80-120%. For PEITC, standard samples (0.1- 30 μg/mL, n=5) in the cyclocondensation reagents were used. The peak area of its derivative (1,3-benzenedithiole-2-thione) was Chromatographic method g p The HPLC system consisted of Waters 2695 Separations Module and a Waters 5487 dual λ absorbance detector (Milford, MA, USA). Data were processed using the EmpowerTM Software System. Separation was modified based on the previous method [11]. A Symmetry® C18 column (250×4.6 mm i.d., 5 μm) connected to a SymmetryTM guard column (20×3.9 mm i.d., 5 μm) was used and controlled at 32ºC. A water- acetic acid-tetrahydrofuran mixture (996.5:1:2.5, v/v/v) was the mobile phase A, while acetronitrile was the mobile phase B. The gradient elution was as follows; B, 6% at 0-9 min, increased to 15% from 9 to 10 min (curve 10), decreased to 6% from 10 to 11 min (curve 6), and post run equilibration of 3 min. The flow rate was 1.3 mL/min. The sample (20 μL) was detected at 274 nm. Animals Adult male Wistar rats (195 - 220 g) were obtained from Southern Laboratory Animal Facility, Prince of Songkla University. The animal experiments were approved by the Animal Ethics Committee, Prince of Songkla University (ref. 17/51). The animals were handled in accordance with the guidelines of the National Research Council of Thailand based on the International Guiding Principles for Biomedical Research Involving Animals [9]. Trop J Pharm Res, April 2014; 13(2): 560 Janchawee et al the same dose of Betong watercress once daily for 5 days. working standard solutions in cyclocondensation reagents, i.e., 600 µL of 10 mM 1,2- benzenedithiol in isopropanol and 500 µL of 0.1M phosphate buffer (pH 8.5). The mixture was incubated as described in the next section. PEITC was dissolved in corn oil. The dry powder of Betong watercress was suspended in 20% aqueous acacia. Blood samples were taken before and 3 h after CF administration in phase I and phase II. Serum was separated by centrifugation at 1000 x g for 20 min. Sample extraction Based on the modified method [10], serum (200 μL) was combined with AP (100 μg/mL, 50 μL) and ammonium sulphate (60 mg). A mixture of chloroform-isopropanol (85:15, v/v, 6 mL) was added and mixed (30 s). After centrifugation (875 x g, 15 min), the organic phase was evaporated. The residue was reconstituted in the mobile phase (100 μL) for HPLC analysis. CFinal is the final concentration of PEITC measured by HPLC; V1 is the volume of the aliquot from the myrosinase reaction (0.1 mL); V2 is the total volume of the derivatizing reaction mixture (1.2 mL); V3 is the total volume of the myrosinase reaction mixture (3 mL); and W is the weight of dry powder used in the reaction. Chromatographic profiles CF, TB, PX, and TP either in the spiked serum (Fig. 2A) or in real serum samples (Fig. 2B), were well separated within 13 min. The derivative product of PEITC was eluted at 3.91 min and it was also well separated with no interference (data not shown). Fig 2: Chromatograms of CF and its metabolites; (A) rat serum spiked with internal standard (IS; 25 µg/mL) and standard TB, PX, TP, and CF (4.5 µg/mL); (B) serum sample at 3 h after CF administration. Calibration curve of the peak area of the derivative plotted against PEITC concentrations also showed good linearity (r = 0.9999). The intra- and inter-day precision range was 0.6-4.1% RSD. The intra- and inter-day accuracy range was -7.5-(+) 18.2% DEV. The LLOQ was 0.1 µg/mL. RESULTS Validation showed that calibration curves for CF and its metabolites were linear (r > 0.9990). All analytes were completely extracted from the serum. Ranges of intra- and inter-day precision for CF, PX, TP, and TB were 2.8-13.4, 2.3-9.3, 2.1-7.3, and 3.5-6.5% RSD, respectively. Ranges of intra- and inter-day accuracy were -11.5-(+) 18.5, -5.7-(+) 11.3, -18.2-(+) 8.3, and -4.6-(+) 17.9% DEV, respectively. The LLOQs of CF and PX was 0.02 µg/mL, while those for TP and TB were 0.04 and 0.08 µg/mL, respectively. Determination of PEITC in Betong watercress Preparation of standard solutions A stock solution of PEITC in acetronitrile was diluted to different concentrations (0.1-30 μg/mL). Calibration standard solutions (0.3 - 5 μg/mL) were prepared by adding 100 μL of the Trop J Pharm Res, April 2014; 13(2): 561 Janchawee et al measured. The LLOQ was determined based on signal to noise ratio (S/N) ≥ 5. compared by Student’s t-test. A statistically significant difference was accepted at p<0.05. Data analysis All data were expressed as mean ± SE. CMRs were calculated as the ratios of the serum concentration of each metabolite to that of CF (TB/CF, PX/CF, and TP/CF), and total serum concentrations of metabolites to that of CF (TB+PX+TP/CF). CMRs in phases I and II experiments were compared using Student’s paired t-test. The percentage changes in MRs among different groups of drug pretreatment for each MR parameter was analyzed by using ANOVA followed by LSD test, while those for single- and multiple pretreatments were Fig 2: Chromatograms of CF and its metabolites; (A) rat serum spiked with internal standard (IS; 25 µg/mL) and standard TB, PX, TP, and CF (4.5 µg/mL); (B) serum sample at 3 h after CF administration. single- and multiple pretreatments were Fig 3: CMRs (A-D) in rats (n=6) receiving CF alone (phase I; ) and pretreated with FV, PEITC, and Betong watercress (WC) prior to receiving the same dose of caffeine (phase II; ); *p < 0.02 compared with phase I p p Fig 3: CMRs (A-D) in rats (n=6) receiving CF alone (phase I; ) and pretreated with FV, PEITC, and Betong watercress (WC) prior to receiving the same dose of caffeine (phase II; ); *p < 0.02 compared with phase I Trop J Pharm Res, April 2014; 13(2): 562 Janchawee et al Fig 4: CMRs (A-D) in rats (n = 6) receiving CF alone (phase I; ) and five-day pretreatment with PEITC, and WC prior to receiving the same dose of CF (phase II; ); * p<0.03 compared with phase I Fig 5: Reduction in CMRs (%) in rats (n=6) receiving a single- and multiple pretreatments with FV, PEITC, and WC; a compared with FV; b compared with PEITC 2, PEITC 10, and PEITC 20; c compared with PEITC 2; d compared with PEITC 10; e compared with a single dose Caffeine metabolic ratios The effect of a single pretreatment of FV, PEITC, and Betong watercress on the CMRs are shown i Fi 3 FV (10 /k i ) d i ifi watercress on CMRs are shown in Fig 4. All doses of PEITC caused a significant reduction in phase II CMRs (TB/CF, PX/CF, TP/CF, and TB+PX+TP/CF). Data analysis Multiple pretreatment with Janchawee et al Fig 4: CMRs (A-D) in rats (n = 6) receiving CF alone (phase I; ) and five-day pretreatment with PEITC, and WC prior to receiving the same dose of CF (phase II; ); * p<0.03 compared with phase I Janchawee et al Janchawee et al Fig 4: CMRs (A-D) in rats (n = 6) receiving CF alone (phase I; ) and five-day pretreatment with PEITC, and WC prior to receiving the same dose of CF (phase II; ); * p<0.03 compared with phase I Fig 4: CMRs (A-D) in rats (n = 6) receiving CF alone (phase I; ) and five-day pretreatment with PEITC, and WC prior to receiving the same dose of CF (phase II; ); * p<0.03 compared with phase I Fig 5: Reduction in CMRs (%) in rats (n=6) receiving a single- and multiple pretreatments with FV, PEITC, and WC; a compared with FV; b compared with PEITC 2, PEITC 10, and PEITC 20; c compared with PEITC 2; d compared with PEITC 10; e compared with a single dose Caffeine metabolic ratios watercress on CMRs are shown in Fig 4. All Fig 4: CMRs (A-D) in rats (n = 6) receiving CF alone (phase I; ) and five-day pretreatment with PEITC, and WC prior to receiving the same dose of CF (phase II; ); * p<0.03 compared with phase I Fig 5: Reduction in CMRs (%) in rats (n=6) receiving a single- and multiple pretreatments with FV, PEITC, and WC; a compared with FV; b compared with PEITC 2, PEITC 10, and PEITC 20; c compared with PEITC 2; d compared with PEITC 10; e compared with a single dose Fig 5: Reduction in CMRs (%) in rats (n=6) receiving a single- and multiple pretreatments with FV, PEITC, and WC; a compared with FV; b compared with PEITC 2, PEITC 10, and PEITC 20; c compared with PEITC 2; d compared with PEITC 10; e compared with a single dose DISCUSSION The present study showed that a 5-day pretreatment with PEITC exhibited a similar inhibition of CF demethylation to a single pretreatment, i.e., 43 – 69% vs. 40 – 55%, respectively. This indicates a time-independent manner of CYP1A2 and CYP2Cs inhibition of PEITC. The inhibitory effect of PEITC (2-20 mg/kg) showed dose-independent pattern both for single- and multiple dose-pretreatments. The percentage decrease of CMRs after a single high dose of PEITC tended to increase although no statistical significance was noted. After multiple high doses, the percentage decrease of CMRs appeared to decrease. Such dose-independent inhibition may be due to the non-linear pharmacokinetic behavior of PEITC as demonstrated in the earlier study [14]. The authors investigated the kinetics of a single dose administration of PEITC in rats and described that clearance tended to decrease and the volume of distribution tended to increase after higher doses, i.e., > 10 µmol/kg (1.65 mg/kg). In humans and rats, CF is primarily metabolized via N-demethylation to TB (1-N position), PX (3- N position), and TP (7-N position), and via C-8- hydroxylation to TMU. The recent in vitro experiments using cDNA-expressed P450s, liver microsomes and specific P450 inhibitors have demonstrated the different relative contributions of CYP isoforms to the metabolism of caffeine in both species. In humans, 3-N-demethylation is the main pathway (~70%) compared to 1-N- and 7-N-demethylation (7-8%) and C-8-hydroxylation (~15%). Both 3-N- and 1-N-demethylation are specifically catalyzed by CYP1A2. 7-N- Demethylation is catalyzed mainly by CYP1A2 and with a lesser extent by CYP2C8/9 and CYP3A4. In contrast, C-8-hydroxylation is the major route (~70%) in rats compared to 1-N- and 7-N-demethylation (8-9%) and 3-N-demethylation (~13%). C-8-Hydroxylation is mainly regulated by CYP1A2 (72%). 1-N- and 3-N-Demethylation are contributed by CYP1A2 with the proportion of 37.5% and 47%, respectively. 3-N-Demethylation is also regulated by CYP2C11 (31%). 7-N- Demethylation is regulated mostly by CYP2C subfamily (~66%) particularly CYP2C6 and CYP2C11 [8]. This work used FV as a reference inhibitor although its potent CYP1A2 inhibitory effect has been mostly established in human studies. FV also has inhibitory effects on CYP2C19 in humans [15]. This study shows that FV at the dose of 10 mg/kg, which was lower than that known to exert its pharmacological action in rats, i.e., 25 mg/kg [16] inhibited both CYP1A2 and CYP2Cs as noted by the 30 – 40% decrease in CMRs. Caffeine metabolic ratios watercress on CMRs are shown in Fig 4. All doses of PEITC caused a significant reduction in phase II CMRs (TB/CF, PX/CF, TP/CF, and TB+PX+TP/CF). Multiple pretreatment with Betong watercress also significantly decreased phase II CMRs (TB/CF, TP/CF, and TB+PX+TP/CF, but not PX/CF). watercress on CMRs are shown in Fig 4. All doses of PEITC caused a significant reduction in phase II CMRs (TB/CF, PX/CF, TP/CF, and TB+PX+TP/CF). Multiple pretreatment with Betong watercress also significantly decreased phase II CMRs (TB/CF, TP/CF, and TB+PX+TP/CF, but not PX/CF). The effect of a single pretreatment of FV, PEITC, and Betong watercress on the CMRs are shown in Fig. 3. FV (10 mg/kg, i.p.) caused a significant reduction of CMRs (TB/CF, PX/CF, TP/CF, and TB+PX+TP/CF) in phase II compared with phase I. Similar results were obtained with different doses of PEITC (2, 10, and 20 mg/kg, p.o.). Pretreatment with Betong watercress (800 mg/kg, p.o.) also resulted in a significant decrease in PX/CF, TP/CF, and TB+PX+TP/CF ratios, but not TB/CF ratio. The reductions in CMRs are presented in Fig. 5. After a single pretreatment, FV decreased CMRs by 30-40%. PEITC (2-20 mg/kg) decreased CMRs by 40-55% in a dose-independent pattern. The decreases of TB/CF and TB+PX+TP/CF ratios caused by 10-20 mg/kg PEITC, i.e., 50 – 55%, were significantly greater than those caused by FV. The decreases in CMRs caused The effect of a five-day pretreatment with similar doses of pure PEITC and PEITC in Betong Trop J Pharm Res, April 2014; 13(2): 563 Janchawee et al by Betong watercress were 9-22% and these were significantly less than those caused by FV and PEITC at all doses. Multiple doses of PEITC caused 43-69% MR reduction which was not significantly different when compared with a single dose pretreatment. In a similar way to a single dose, no dose-dependency of PEITC was seen. Multiple pretreatments with Betong watercress caused 28-44% reduction in the TB/CF and TP/CF ratios which were significantly greater than those obtained from a single pretreatment. microsomes from baculovirus-infected insect cells expressing human CYP isoforms, and demonstrated that PEITC competitively inhibited CYP1A2 and noncompetitively inhibited CYP2C9, CYP2C19 but not CYP2C8 [6]. Another work [7] studying the enzyme activity in rats receiving PEITC for 2 weeks showed that only the low dose (1.1 mg/kg) of PEITC decreased CYP1A1 and CYP1A2 activity, and the higher doses (11 and 110 mg/kg) increased CYP3A activity. Caffeine metabolic ratios No data have yet indicated an influence of PEITC on CYP2C6 or CYP2C11. DISCUSSION The present study determined CMRs that can reflect an alteration in CF metabolism rate and the relevant CYP isoform activities. Although the hydroxylation is the major route in rats, only MRs representing the demethylation was investigated in this study due to the lack of the commercial standard of the metabolite TMU. However a change in demethylation MRs indicates the modification of activity of CYP1A2 and CYP2C that contribute to such reactions in this species. To investigate the effect of Betong watercress on CF demethylation, the preparation derived from fresh, not cooked, vegetable was used. It has been recognized that ITCs are products of glucosinolate-myrosinase hydrolysis. Myrosinase located separately from glucosinoates in plant cells [17]. When humans ingest fresh watercress, plant cells are damaged by chewing, and the enzyme is released and rapidly catalyzes conversion of gluconasturtiin to PEITC. By using urinary N-acetylcysteine conjugate as a marker, The decreases of TB/CF, PX/CF, TP/CF ratios after single- and multiple pretreatments of PEITC and Betong watercress were most likely caused by inhibition of CYP1A2 and/or CYP2C. This is consistent with the in vitro study that used Trop J Pharm Res, April 2014; 13(2): 564 Janchawee et al CYP1A2 and CYP2C. Since CYP1A2 plays a role in procarcinogen bioactivation, and interference with bioactivation mediated by cytochrome P450s is proposed as one mechanism of antitumor effect of xenobiotics, Betong watercress may be of benefit in chemoprevention, especially if one ingests a whole plant which contains more glucosinoate precursors. consumption of 57 g of fresh watercress resulted in a 30-67% conversion of gluconasturtiin corresponding to 12-15 mg (72.6 - 90.7 µmole) of PEITC in the body [18]. For cooked watercress, cooking diminishes the concentration of glucosinolates in plant tissue and also inactivates myrosinase [19]. In individuals ingested cooked watercress in which myrosinase is completely inactivated by cooking, glucosinolates are converted to ITCs via colonic microflora. However, the extent of conversion is substantially less than after ingesting the uncooked vegetable [20]. REFERENCES 1. Fenwick GR, Heaney RK, Mullin WJ. Glucosinolates and their breakdown products in food and food plants. Crit Rev Food Sci Nutr 1983; 18: 123-201. 2. Stoner GD, Morse MA. Isothiocyanates and plant polyphenols as inhibitors of lung and esophageal cancer. Cancer Lett 1997; 114: 113-119. 3. Gong YQ, Fan Y, Wu DZ, Yang H, Hu ZB. In vivo and in vitro evaluation of erianin, a novel anti-angiogenic agent. Eur J Cancer, 2004; 40: 1554-1565. Dry powder of Betong watercress administered at a dose of 800 mg/kg, which is equivalent to a low dose of PEITC (2 mg/kg), had an inhibitory effect similar to that seen with pure PEITC. This indicates PEITC absorption probably in intestine. However, the extent of inhibition that resulted from Betong watercress was much lower than that caused by 2 mg/kg PEITC. This may be due to a variation of hydrolysis of the glucosinolates in the dry powders to PEITC in vivo. Such conversion occurs in the colon of rats and humans via microflora possessing myrosinase activity. In addition, glucosinolates may be converted to other classes of metabolites, of which ITCs constitute a small proportion. The released ITCs may be further metabolized to unknown compounds by colonic microflora before their absorption [22]. Dry powder of Betong watercress administered at a dose of 800 mg/kg, which is equivalent to a low dose of PEITC (2 mg/kg), had an inhibitory effect similar to that seen with pure PEITC. This indicates PEITC absorption probably in intestine. 4. Leclercq I, Desager JP, Horsmans Y. Inhibition of chlorzoxazone metabolism, a clinical probe for CYP2E1, by a single ingestion of watercress. Clin Pharmacol Ther 1998; 64: 144-149. 5. Lhoste EF, Gloux K, De Waziers I, Garrido S, Lory S, Philippe C, Rabot S, Knasmuller S. The activities of several detoxication enzymes are differentially induced by juices of garden cress, water cress and mustard in human HepG2 cells. Chem Biol Interact 2004; 150: 211-219. 6. Nakajima M, Yoshida R, Shimada N, Yamazaki H, Yokoi T. Inhibition and inactivation of human cytochrome P450 isoforms by phenethyl isothiocyanate. Drug Metab Dispos 2001; 29: 1110-1113. 7. Konuse N, Ioannides C. Chemoprevention effect of phenethyl isothiocyanate. Toxicology 2007; 240:184. 8. Kot M, Daniel WA. Caffeine as a marker substrate for testing cytochrome P450 activity in human and rat. Pharmacol Rep 2008; 60: 789-797. ACKNOWLEDGEMENT This work was funded by Prince of Songkla University, the Graduate School, and Natural Product Research Center. The authors thank Assoc.Prof.Dr. Kitichate Sridith and Mr. Charoensak Sae Wei, Princess Maha Chakri Sirindhorn Natural History Museum, for their kind assistance in authenticating the plant. The authors acknowledge Amphoe Betong Agricultural Office, Yala Province, Thailand for providing information about Betong watercress and plant samples. Thanks also to Dr Brian Hodgson for assistance with the English. The present work has ensured that the dry powder of fresh juice preparation contained PEITC. During blending of the fresh plant to prepare juice, glucosinolates come into contact with endogenous myrosinase and hydrolysis was rapidly initiated. It was likely that juice contained PEITC as reported earlier, and that PEITC was a component of watercress juice with a concentration of 24 mg/L [21]. In addition, it is likely that glucosinolates, which are water soluble, at the cut surfaces of plant cells leached out during blending process. Hence Betong watercress juice probably also contained glucosinolate precursors that will be changed to PEITC in the gut. After incubation of the dry powders of juice with exogenous myrosinase, it was found that the PEITC content was 2.50 µg/mg dry powders or 15.1 µmole/g dry powders which was equivalent to 68.5 g of fresh vegetable. Trop J Pharm Res, April 2014; 13(2): 566 CONCLUSION 9. CIOMS. The International Guiding Principles for Biomedical Research Involving Animals, 2012 [cited Inhibition of CF demethylation by PEITC and Betong watercress indicates an inhibition of Trop J Pharm Res, April 2014; 13(2): 565 Trop J Pharm Res, April 2014; 13(2): 565 Janchawee et al 2013 December 18]. Available from: http://www.cioms.ch/. from: 17. Hoglund AS, Lenman M, Falk A, Rask L. Distribution of myrosinase in rapeseed tissues. Plant Physiol 1991; 95: 213-221. 18]. 10. Grant DM, Tang BK, Kalow W. Variability in caffeine metabolism. Clin Pharmacol Ther 1983; 33: 591-602. 18. Chung FL, Morse MA, Eklind KI, Lewis J. Quantitation of human uptake of the anticarcinogen phenethyl isothiocyanate after a watercress meal. Cancer Epidemiol Biomarkers Prev 1992; 1: 383-388. 11. Caubet MS, Elbast W, Dubuc MC, Brazier JL. Analysis of urinary caffeine metabolites by HPLC-DAD: the use of metabolic ratios to assess CYP1A2 enzyme activity. J Pharm Biomed Anal 2002; 27: 261-270. 19. Johnson IT. Glucosinolates in the human diet. Bioavailability and implications for health. Phytochem Rev 2002; 1: 183-188. 12. Jiao D, Yu MC, Hankin JH, Low S-H, Chung F-L. Total isothiocyanate contents in cooked vegetables frequently consumed in Singapore. J Agric Food Chem 1998; 46: 1055-1058. 20. Getahun SM, Chung FL. Conversion of glucosinolates to isothiocyanates in humans after ingestion of cooked watercress. Cancer Epidemiol Biomarkers Prev 1999; 8: 447-451. 13. USFDA. Guidance for Industry: Bioanalytical Method Validation, 2001 [cited 2012 November 9]. Available from: http://www.fda.gov/. 21. Kassie F, Laky B, Gminski R, Mersch-Sundermann V, Scharf G, Lhoste E, Kansmuller S. Effects of garden and water cress juices and their constituents, benzyl and phenethyl isothiocyanates, towards benzo(a)pyrene-induced DNA damage: a model study with the single cell gel electrophoresis/Hep G2 assay. Chem Biol Interact 2003; 142: 285-296. 14. Ji Y, Kuo Y, Morris ME. Pharmacokinetics of dietary phenethyl isothiocyanate in rats. Pharm Res 2005; 22: 1658-1666. 15. Yasui-Furukori N, Takahata T, Nakagami T, Yoshiya G, Inoue Y, Kaneko S, Tateishi T. Different inhibitory effect of fluvoxamine on omeprazole metabolism between CYP2C19 genotypes. Br J Clin Pharmacol 2004; 57: 487-494. 22. Rungapamestry V, Duncan AJ, Fuller Z, Ratcliffe B. Effect of cooking brassica vegetables on the subsequent hydrolysis and metabolic fate of glucosinolates. Proc Nutr Soc 2007; 66: 69-81. 16. Claassen V. Review of the animal pharmacology and pharmacokinetics of fluvoxamine. Br J Clin Pharmacol 1983; 15(Suppl 3): 349S-355S. Trop J Pharm Res, April 2014; 13(2): 566

https://openalex.org/W2150390192

http://ira.lib.polyu.edu.hk/bitstream/10397/16377/1/Lai_Who_Are_Healthy.pdf

English

Who are the healthy active seniors? A cluster analysis

BMC geriatrics

cc-by

RESEARCH ARTICLE Open Access * Correspondence: [email protected] 1School of Nursing (SN), The Hong Kong Polytechnic University (PolyU), Hung Hom, Kowloon, Hong Kong Special Administrative Region (HKSAR), China Full list of author information is available at the end of the article Abstract Background: This paper reports a cluster analysis of a sample recruited from a randomized controlled trial that explored the effect of using a life story work approach to improve the psychological outcomes of older people in the community. Methods: 238 subjects from community centers were included in this analysis. After statistical testing, 169 seniors were assigned to the active ageing (AG) cluster and 69 to the inactive ageing (IG) cluster. Results: Those in the AG were younger and healthier, with fewer chronic diseases and fewer depressive symptoms than those in the IG. They were more satisfied with their lives, and had higher self-esteem. They met with their family members more frequently, they engaged in more leisure activities and were more likely to have the ability to move freely. Conclusion: In summary, active ageing was observed in people with better health and functional performance. Our results echoed the limited findings reported in the literature. Keywords: Ageing, Elderly, Community, Cluster analysis © 2014 Lai et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Claudia K Y Lai1*, Engle Angela Chan2 and Kenny C W Chin2 Claudia K Y Lai1*, Engle Angela Chan2 and Kenny C W Chin2 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Background reports and papers, but ideas surrounding the concept are discussed with the assumption that individuals who age healthily are those who are more independent in daily ac- tivities and free from debilitating illnesses. Active ageing is discussed in the literature as a goal that health professionals, policy makers, and the general public would like to attain. The term ‘active ageing’ was first adopted by the World Health Organization (WHO) in the late 1990s [1]. The WHO defined active ageing as the process of optimizing opportunities for health, participa- tion, and security in order to enhance quality of life as people aged (p. 12) [2]. To date, there is no universally accepted definition of ‘active ageing’ [3]. The concepts of active ageing overlap with those of healthy ageing, productive ageing, or successful ageing – some terms that commonly appear in the literature. Rowe and Kahn defined successful aging not only as achieving better outcomes of physical and cognitive health, but also being actively engaged with life [6]. Bowling and Dieppe conducted a systematic review on the models of successful ageing and concluded that there are two main approaches to understanding the concept – the psychosocial school, which defines successful ageing as wellbeing of the mental states such as life satisfaction, and the biomedical school, which interprets the concept as the prevention of diseases and disabilities [7]. A consen- sus has yet to be reached on the definition of successful ageing [8]. Still, success in ageing is commonly defined as how well an older person has maintained or achieved bet- ter health outcomes. Healthy ageing as a concept has been defined in various ways and with different underlying assumptions [4]. In general, the concept of healthy ageing has been described as a complex process of adapting to physical and socio- psychological changes over one’s lifetime [5]. Healthy age- ing is not clearly defined as a concept in many published The concept of productive ageing conceptualizes ageing from a somewhat different perspective. The productive ageing framework transcends the physical or functional realm of an individual’s health and views an older person who is capable of accomplishing his/her goals and tasks as Page 2 of 7 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 would lead to better psychological outcomes (referring to a higher level of life satisfaction, better self-esteem, and improved well-being). Background In the course of our study, the project team observed that there seemed to be some differences in outcomes between older adults who were physically and functionally active as opposed to those who were not. We therefore proceeded to conduct a systematic literature review (as discussed above) and a cluster ana- lysis to examine the validity of our clinical impression. This paper reports the results of a cluster analysis of the data generated from this sample. The objective of this re- port was to identify the profile of active older people in this community study in relation to their psychosocial sta- tus. Ethical approval was obtained from the University’s Human Subjects Ethics Application Review Committee (synonymous with an ethics review board in the West). In- formed written consent was obtained from all of the subjects. having a sense of purpose in life [9]. The viewpoint fo- cuses on the older adults’ contributions to society as well as their internal affective state, which may have a positive impact on their wellbeing. Both the external and internal (to the older person) views emphasize engagement by older adults in their physical, psychological, and socioeco- nomic environment. In summary, there are some common concepts in these terms relating to ageing, but in conceptualization they differ somewhat in focus. The definitions of various terms are arbitrarily used in a variety of contexts [5]. Gerontologists – whether clinicians or researchers – have not come any closer to reaching a consensus after years of deliberation. As a multidimensional concept, the term active ageing includes concepts of activity, health, independence, and productivity in older age [10,11]. The term is intended to convey a more inclusive message than ‘healthy ageing’ and to recognize factors in addition to health care that affect how individuals and populations age [1]. Older persons with chronic disease may still be considered healthy if they are socially and intellectually active [12]. Despite keen discussions of the concept of active ageing, there are few reports in the literature on the determi- nants of active ageing or the factors associated with it. Intervention and control conditions Among these 13 papers, three did not focus on active ageing although the term was mentioned [13-15]; an- other three contained conceptual discussions of the term [16-18]; two were about older people’s perceptions of ac- tive ageing [19,20]; four were about other concepts related to ageing or active ageing – e.g., the concept of quality of life [21], successful ageing [22], and how neighborhood in- fluenced function [23]; and one study reported using focus group interviews to examine whether an education pro- grammed for active ageing would lead to participant em- powerment after training [24]. Only one – Eskurza et al. studied the role of oxidative stress in physically active age- ing in humans [25]. Although not an exhaustive search of all health-related databases, our search showed that there is a dearth of studies on the determinants of active ageing or the factors associated with it. The intervention was the construction of an individual LSB using a life story approach developed by the first au- thor. It was a one-to-one intervention consisting of four to six weekly sessions lasting from 30 to 60 minutes per session. The control condition consisted of social activities unrelated to the discussion of the subjects’ own life stories. Sampling and the sample A systematic search of the literature using the search engine EBSCOHost to access the databases CINAHL, MEDLINE (1965+), and Social Work Abstracts was con- ducted. The following keywords were used without choos- ing any fields to search for relevant papers for the period 1978 to 2013, on the determinants of or factors associated with active ageing - (“active ageing” OR “active aging”) AND (“determinants” OR “factors” OR “associat*”). Eighteen papers were found and 13 papers remained after duplicates were culled. All active members of the 17 centers were used as the sampling frame. A random sample of 244 subjects who (i) were aged 60 or above, (ii) were able to understand and speak Cantonese, (iii) were able to see and hear with or without aids, and (iv) did not have any active major ill- nesses or psychosocial crises, were recruited and randomly assigned into an intervention and a control group. Methods The project team collaborated with a non-profit-making non-governmental organization in Hong Kong. Its eld- erly services section runs 17 community and day care centers for subjects serving the local population. Data analysis Of h 244 Of the 244 randomized subjects, 238 were included in the analysis because of missing data in the GDS, GHQ, LSI-A and RSES scores of the rest of the cases. A two- step clustering approach, namely agglomerative hier- archical cluster analysis and k-means cluster analysis, was used to develop a system of typology for character- izing factors that contribute to those subjects who were more satisfied with their lives and had higher self- esteem, and who were healthier and less depressed. In addition, discriminant analysis was used to develop a screening model that would allow us to allocate the sub- jects into an appropriate grouping based on the range of the characteristics to be tested. Those in the active ageing cluster were more likely to be younger (mean age: 76.53 (AG) versus 78.57 (IG)), healthier with fewer chronic diseases (mean number of chronic diseases: 2.06 (AG) versus 2.51 (IG)), less de- pressed (mean scores: 1.89 (AG) versus 6.68 (IG)), more satisfied with their life (mean scores: 14.64 (AG) versus 8.46 (IG)), and to have higher self-esteem (mean scores: 8.52 (AG) versus 5.62 (IG)). They also enjoyed better sleep quality (less than one sleepless night per week: 71.69% (AG) versus 41.79% (IG)). Most of their income came from the support of their family members (60.84% (AG) versus 30.43% (IG)) and they were more satisfied with their economic condition (enough money for daily expenses: 41.87% (AG) versus 19.12% (IG)). They joined more leisure activities (average number of leisure activ- ities engaged in: 1.99 (AG) versus 1.48 (IG)) and were more likely to have the ability to move freely (climbing stairs without help: 91.72% (AG) versus 62.32% (IG)). Furthermore, the subjects in this cluster were more likely to make friends with others (mean scores: 9.10 (AG) versus 6.93 (IG)), more willing to participate in group activities such as dancing, singing, and playing musical instruments for leisure (26.34% (AG) versus 9.09% (IG)). Their relationships with other family mem- bers were also better (mean scores: 9.98 (AG) versus 6.67 (IG)) and they met their family members more fre- quently (average number of meetings per month: 5.56 (AG) versus 3.84 (IG) (Table 2). Measures including GDS, GHQ, LSI-A, and RSES, which were our primary outcomes, were used to deter- mine to which clusters the subjects belonged. Data analysis Of h 244 Agglom- erative hierarchical cluster analysis was first used to decide on the optimal number of clusters among the subjects, which was based on statistics, namely semi- partial R-squared (SPRSQ), Cubic clustering criteria (CCC), and pseudo F (PSF). These statistics provide in- formation about the cluster solution at any given step (i.e., the new cluster that formed at this step, and the consequences of forming the new cluster). The value with a large percentage decrease in SPRSQ at a given cluster refers to that cluster number as the optimal clus- ter solution, while large values for the CCC and PSF at a given cluster suggest a good stopping point for the clus- ter solution. The results in Table 1 show that the CCC and the PSF have the highest values at cluster 2. A large decrease in SPRSQ is also detectable in that cluster, im- plying that cluster 2 is the best solution. To develop a model that can be used to determine the characteristics that differentiate two groups, a discrimin- ant analysis was performed. A total of 238 cases were randomly split into two sets of data, D1 and D2, which Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Page 3 of 7 Table 1 SPRSQ, CCC & PSF values at given cluster(s) Cluster SPRSQ CCC PSF 10 0.0146 8.13 85.6 9 0.0154 8.28 88.8 8 0.0201 8.2 91.7 7 0.0203 8.67 97 6 0.0257 9.2 103 5 0.0395 9.15 109 4 0.0437 14.4 120 3 0.0531 21.3 146 2 0.1463 23.7 162 1 0.4065 0 Note : SPRSQ = Semi-partial R-squared, CCC = Cubic clustering criteria, PSF = Pseudo F. Table 1 SPRSQ, CCC & PSF values at given cluster(s) Table 1 SPRSQ, CCC & PSF values at given cluster(s) Network Scale and Life Events scale to control for the presence of any psychosocial variables that might have an impact on the outcomes. Network Scale and Life Events scale to control for the presence of any psychosocial variables that might have an impact on the outcomes. The outcome measures included the Life Satisfaction Scale Index A (LSI-A), Rosenberg’s Self-Esteem Scale (RSES), the General Health Questionnaire (GHQ), and the Geriatric Depression Scale (GDS), which were col- lected at baseline (T0), immediately post-intervention (T1), 3 months post-intervention (T2), and 6 months post-intervention (T3). We have also collected informa- tion about the subjects’ leisure activities and exercise patterns. The baseline data at T0 were used for data ana- lysis because the intervention and the control conditions are induced activities. All of these measures have been validated for use in Hong Kong by local researchers, and with good psychometric properties. Note : SPRSQ = Semi-partial R-squared, CCC = Cubic clustering criteria, PSF = Pseudo F. appropriate cluster. As a result, two clusters were formed and named active ageing (AG) and inactive age- ing (IG). One hundred and sixty-nine seniors were assigned to AG and 69 to IG. Their profile is presented in Table 2. Data collection h d h The demographic and clinical characteristics of the partic- ipants, including gender, age, date of birth, education, presence of chronic illness, number of medical diagnoses, medications, income source and level, dwelling status, types of leisure activities engaged in, exercise pattern, and presence of sleep problems were collected. Other clinical and control variables that were collected included hearing and vision, the Modified Barthel Index and the Instrumen- tal Activities of Daily Living scale for assessing the partici- pants’ functional performance, and the Lubben Social We conducted a randomized controlled trial (RCT) to determine whether the production of a life story book (LSB) as an intervention for community-dwelling subjects Results f d After determining the number of the clusters, k-means clustering was then used to allocate the seniors into an Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Page 4 of 7 Table 2 Characteristics of the active and inactive clusters Active cluster Inactive cluster Profile characteristics (N =169) (N = 69) Chi-square test N % N % χ2 p-value Gender 0.098 .755 Male 48 28.4% 21 30.4% Female 121 71.6% 48 69.6% Marital status 1.276 .735 Married 72 42.6% 26 37.7% Widow 82 48.5% 34 49.3% Separate 4 2.4% 3 4.3% Single 11 6.5% 6 8.7% Main income source 24.932 .000 From government 50 30.1% 45 65.2% From family members 101 60.8% 21 30.4% Own savings 15 9.0% 3 4.3% Monthly income 13.161 .041 Less than 2,000 23 15.1% 20 30.8% 2,000 - 3,999 76 50.0% 36 55.4% 4,000 – 5,999 33 21.7% 6 9.2% 6,000 – 7,999 6 3.9% 1 1.5% 8,000 – 9,999 7 4.6% 1 1.5% 10,000 – 14,999 5 3.3% 1 1.5% 15,000 – 19,999 2 1.3% 0 .0% Enough money for daily expense 16.172 .003 Extremely insufficient 4 2.4% 4 5.9% Not enough 16 9.5% 15 22.1% Just enough 78 46.4% 36 52.9% Enough 61 36.3% 13 19.1% In excess 9 5.4% 0 .0% Living status 8.092 .088 Alone 60 35.5% 37 53.6% Live with spouse 38 22.5% 13 18.8% Live with spouse & children 23 13.6% 5 7.2% Live with children 39 23.1% 13 18.8% Live with other people 9 5.3% 1 1.4% Sport habit 2.322 .508 Less than one day per week 15 9.0% 9 13.4% 1 - 3 days per week 33 19.9% 17 25.4% 4 - 5 days per week 18 10.8% 7 10.4% 6 - 7 days per week 100 60.2% 34 50.7% Sleepless night 18.914 .000 Less than one day per week 119 71.7% 28 41.8% 1 - 3 days per week 30 18.1% 22 32.8% 4 - 5 days per week 5 3.0% 5 7.5% 6 - 7 days per week 12 7.2% 12 17.9% Table 2 Characteristics of the active and inactive clusters Page 5 of 7 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 contained 109 and 129 candidates respectively. D1 was the test sample and D2 was the holdout sample, which variables found to be significantly different between the AG and IG were entered into the model. Results f d As a result, Table 2 Characteristics of the active and inactive clusters (Continued) Leisure activities engaged in* Mahjong 33 19.8% 6 9.1% 1.966 0.049 Reading/writing 57 34.1% 11 16.7% 2.642 0.008 Sing/dancing/musical instrument 44 26.3% 6 9.1% 2.891 0.004 Play chess 11 6.6% 2 3.0% 1.066 0.287 Watching TV/listen to radio 102 61.1% 52 78.8% −2.573 0.010 Hearing 1.424 .491 Normal 121 71.6% 44 63.8% Slightly difficult 37 21.9% 19 27.5% Fairly difficult 11 6.5% 6 8.7% Difficult 0 .0% 0 .0% Vision 14.757 .002 Fine 141 83.4% 43 62.3% Slight problem 22 13.0% 16 23.2% Mild problem 5 3.0% 8 11.6% Moderately difficult 1 .6% 2 2.9% Difficult 0 .0% 0 .0% Climbing stairs 30.028 .000 Help with other people 14 8.3% 26 37.7% Independent 154 91.7% 43 62.3% Mean (SD) Mean (SD) t p-value Demographic Variables Age 76.53 (7.10) 78.57 (7.90) −1.944 .053 Years of education 4.66 (4.60) 3.46 (3.21) 2.200 .029 No. of siblings alive 2.55 (2.67) 1.64 (2.69) 2.351 .020 No. of children alive 2.85 (1.82) 2.41 (1.86) 1.668 .097 No. of chronic diseases suffered 2.06 (1.48) 2.51 (1.54) −2.071 .039 No. of leisure activities engaged in 1.99 (1.06) 1.48 (.75) 4.070 .000 Control variables LSNS - family network 9.98 (6.41) 6.67 (5.70) 3.735 .000 LSNS - friend network 9.10 (5.53) 6.93 (5.03) 2.818 .005 Life events (6 Items) .18 (.39) .41 (.65) −2.682 .009 BI 98.63 (3.39) 93.20 (10.50) 4.207 .000 IADL 22.98 (3.36) 20.22 (4.66) 4.475 .000 Outcome measures LSI-A 14.64 (2.15) 8.46 (2.82) 16.377 .000 GDS 1.89 (1.75) 6.68 (2.87) −12.917 .000 RSES 8.52 (1.42) 5.62 (2.17) 10.244 .000 GHS 8.07 (2.60) 13.62 (5.35) −8.214 .000 Note :1. LSI-A = Life Satisfaction Scale, GDS = Geriatric Depression Scale, RSES = Rosenberg’s Self-Esteem Scale, GHQ = General Health Questionnaire, BI = Modified Barthel Index, IADL = Lawton Instrumental Activities of Daily Living Scale, LSNS = Lubben Social Network Scale. 2. *Multiple-response items between clusters were tested with two-proportion z-test. Table 2 Characteristics of the active and inactive clusters (Continued) Table 2 Characteristics of the active and inactive clusters (Continued) Note :1. LSI-A = Life Satisfaction Scale, GDS = Geriatric Depression Scale, RSES = Rosenberg’s Self-Esteem Scale, GHQ = General Health Questionnaire, BI = Modified Barthel Index, IADL = Lawton Instrumental Activities of Daily Living Scale, LSNS = Lubben Social Network Scale. 2. *Multiple-response items between clusters were tested with two-proportion z-test. Results f d variables found to be significantly different between the AG and IG were entered into the model. As a result, variables including GDS, GHQ, LSI-A, RSES, Life Event (LE), the Lubben Social Network Scale (LSNS), the contained 109 and 129 candidates respectively. D1 was the test sample and D2 was the holdout sample, which served as internal validation. We first conducted the dis- criminant analysis on the test sample, and all of the variables found to be significantly different between the AG and IG were entered into the model. As a result, variables including GDS, GHQ, LSI-A, RSES, Life Event (LE), the Lubben Social Network Scale (LSNS), the Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Page 6 of 7 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Modified Barthel Index (BI), the Lawton Instrumental Activities of Daily Living Scale (IADL), the number of leisure activities engaged in, normal vision (including with corrective eyewear), income source, and sleep qual- ity, were retained in the model. Discriminating power was evaluated by several criteria: (a) Wilks’ lambda, (b) variance explained, and (c) percentage correctly classi- fied. The results are shown in Table 3. The discrimin- ation model gave Wilks’ lambda = 0.28, Chi-square = 118.92, df = 13, and p <0.001, implying that the AG and IG clusters were significantly different with respect to the given discriminator variables in the model. In addition, the means of these variables for the two groups were also significantly different at the 5% level. Seventy- two per cent of the variation between the two groups was accounted for by these discriminating variables. Of these variables, LSI-A and GDS were the most signifi- cant factors contributing to the discriminant function. The resultant equation was then used to allocate the se- niors to an appropriate group according to the seniors’ information in the D2 dataset. As a result, 98.51% and 94.12% of those in the AG and IG groups, respectively, could be correctly predicted, indicating that the model has a high level of predictive power. with fewer depressive symptoms than those in the in- active ageing cluster. They were more satisfied with their lives and had higher self-esteem and better sleep quality. They were more satisfied with their financial situation (had enough money for daily expenses), and most of their income came from their family members. Results f d Their re- lationships with family members were also better. They met with their family members more frequently. They engaged in more leisure activities and were more likely to have the ability to move freely (climbing stairs with- out help) and were more willing to participate in group activities such as dancing, singing, and playing musical instruments in their leisure time. Our findings are fairly similar to those reported by López et al. in a study that we located outside of the sys- tematic literature search reported above [26]. López et al. wanted to establish the health and socio-cultural determinants of active ageing in a sample of 456 com- munity living adults aged 54 to 75 years old in Spain. They found more active agers in men than in women, whereas gender was not a significant variable in our re- sults. In their logistic regression model, being a woman and the number of diagnosed diseases were risk factors against active ageing, whereas years of education was a protective factor against the absence of active ageing. Similarly, our study showed that the number of chronic diseases and the number of years of education were sig- nificantly different in the AG and IG. Discussion Those in the active ageing cluster were more likely to be younger and healthier, with fewer chronic diseases and Table 3 Standardized canonical discriminant function coefficients Table 3 Standardized canonical discriminant function coefficients Variables Discriminant coefficients Correlation with discriminant fun Demographic variables Normal vision (with corrective eyewear) - (Yes versus No) .182 -.157 Income source - Government vs from family/own savings .158 .174 Sleepless night - <1 per week vs at least once per week -.124 -.300 Number of leisure activities engaged in -.028 -.123 Control variables Lawton Instrumental Activities of Daily Living Scale (IADL) -.186 -.206 Lubben Social Network Scale (LSNS) - Family subscale .185 -.153 Life events (LE) .142 .216 Lubben Social Network Scale (LSNS) - Friend subscale -.047 -.181 Modified Barthel Index (BI) .025 -.237 Outcome measures Life Satisfaction Scale (LSI-A) -.699 -.737 Geriatric Depression Scale (GDS) .535 .653 General Health Questionnaire (GHS) .267 .385 Rosenberg’s Self-Esteem Scale (RSES) .098 -.515 Note: All variables are significant at p = 0.05. Discriminant coefficients Correlation with discriminant function Correlation with discriminant function Demographic variables Page 7 of 7 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Lai et al. BMC Geriatrics 2014, 14:127 http://www.biomedcentral.com/1471-2318/14/127 Because the sample consisted of participants who were social service centre members and likely to be more ac- tive than those who were not, the findings have to be interpreted with caution in view of the sampling bias. The WHO policy framework paper on active ageing ad- vocates for the adoption of policies by governments to meet the challenges of global ageing [1]. Active ageing is not merely for senior citizens, but rather for people to realize their potential for health throughout the life course. 8. Depp CA, Jeste DV: Definitions and predictors of successful aging: a comprehensive review of larger quantitative studies. Am J Geriatr Psychiatry 2006, 14(1):6–20. y y 9. Kaye LW, Butler SS, Webster NM: Toward a productive ageing paradigm for geriatric practice. Ageing Int 2003, 28(2):200–213. for geriatric practice. Ageing Int 2003, 28(2):200–213. 10. World Health Organization: Health and Ageing: A Discussion Paper. Geneva: World Health Organization; 2001a. Geneva: World Health Organization; 2001a. g 11. World Health Organization: Active Ageing: From Evidence to Action. Geneva: World Health Organization; 2001b. 1. World Health Organization: Active Ageing: From Evidence to Ac 11. World Health Organization: Active Ageing: F Geneva: World Health Organization; 2001b. Geneva: World Health Organization; 2001b. 12. International Council of Nurses: International Council of Nurses on healthy ageing. Nurs Stand 1999, 13(41):31. 13. Aihara Y, Minai J, Kikuchi Y, Aoyama A, Shimanouchi S: Cognition of social capital in older Japanese men and women. J Aging Health 2009, 21(8):1083–1097. 14. Cheng Y, Rosenberg MW, Wang W, Yang L, Li H: Aging, health and place in residential care facilities in Beijing, China. Soc Sci Med 2011, 72(3):365–372. Authors’ contributions CKYL S d i CKYL: Study conception and design, supervision of data acquisition and management, data analysis and interpretation, drafting of the manuscript. EAC: Co-investigator; critical review of the manuscript. KCWC: Data analysis and interpretation, critical review of the manuscript. All authors read and approved the final manuscript. 23. Bowling A, Stafford M: How do objective and subjective assessments of neighbourhood influence social and physical functioning in older age? Findings from a British survey of ageing. Soc Sci Med 2011, 64(12):2533–2549 24. de la Luz Martínez-Maldonado M, Correa-Muñoz E, Mendoza-Núñez VM: Program of active aging in a rural Mexican community: a qualitative approach. BMC Public Health 2007, 7(1):276. 25. Eskurza I, Monahan KD, Robinson JA, Seals DR: Effect of acute and chronic ascorbic acid on flow-mediated dilatation with sedentary and physically active human ageing. J Physiol 2004, 556(1):315–324. Competing interests The authors declare that they have no competing interests. 22. Bowling A, Iliffe S: Psychological approach to successful ageing predicts future quality of life in older adults. Health Qual Life Outcomes 2011, 9:13. doi: 10.1186/1477-7525-9-13. Acknowledgements h This project was supported by the General Research Fund of the Research Grants Council, #543909. The authors are grateful to our collaborator, the Elderly Services Department, Tung Wah Group of Hospitals, and Ms Yu Zhang, research assistant, for their support for this study. 26. López PM, Fernández-Ballesteros R, Zamarrón MD, López SR: Anthropometric, body composition and health determinants of active ageing: a gender approach. J Biosoc Sci 2011, 43(5):597–610. Author details 1 doi:10.1186/1471-2318-14-127 Cite this article as: Lai et al.: Who are the healthy active seniors? A cluster analysis. BMC Geriatrics 2014 14:127. 1School of Nursing (SN), The Hong Kong Polytechnic University (PolyU), Hung Hom, Kowloon, Hong Kong Special Administrative Region (HKSAR), China. 2SN, PolyU, Hung Hom, Kowloon, HKSAR, China. Received: 7 March 2014 Accepted: 19 November 2014 Published: 1 December 2014 Received: 7 March 2014 Accepted: 19 November 2014 Published: 1 December 2014 Received: 7 March 2014 Accepted: 19 November 2014 Published: 1 December 2014 Received: 7 March 2014 Accepted: 19 November 2014 Published: 1 December 2014 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Conclusion 15. Webber SC, Porter MM, Menec VH: Mobility in older adults: a comprehensive framework. Gerontologist 2010, 50(4):443–450. 15. Webber SC, Porter MM, Menec VH: Mobility in older adults: a comprehensive framework. Gerontologist 2010, 50(4):443–450. Our findings add to the current limited knowledge about the correlates of healthy active ageing. Active ageing was observed in people with better health, and functional performance, as well as a more satisfying social network and relationships. Physical health, mental wellbeing, a good relationship with one’s family and a willingness to join in social and group activities – these dimensions in a person’s life are aspects that cannot be easily modified within the duration of a programmed intervention. As such, the promotion of active ageing must occur early in older age, or the sooner the better. 16. Barnett K, Buys L, Lovie-Kitchin J, Boulton-Lewis G, Smith D, Heffernan M: Older women’s fears of violence: the need for interventions that enable active ageing. J Women Aging 2007, 19(3–4):179–193. 17. Williams SJ, Higgs P, Katz S: Neuroculture, active ageing and the ‘older brain’: problems, promises and prospects. Sociol Health Illn 2012, 34(1):64–78. 18. Jeppsson Grassman E, Holme L, Taghizadeh Larsson A, Whitaker A: A long life with a particular signature: life course and aging for people with disabilities. J Gerontol Soc Work 2012, 55(2):95–111. life with a particular signature: life course and aging for people with disabilities. J Gerontol Soc Work 2012, 55(2):95–111. 19. Manthorpe J, Malin N, Stubbs H: Older people’s views on rural life: a study of three villages. J Clin Nurs 2004, 13(s2):97–104. 19. Manthorpe J, Malin N, Stubbs H: Older people’s views on rural life: a study of three villages. J Clin Nurs 2004, 13(s2):97–104. 20. Pavlova MK, Silbereisen RK: Perceived level and appraisal of the growing expectations for active ageing among the young-old in Germany. Res Aging 2012, 34(1):80–99. 21. Bowling A: Perceptions of active ageing in Britain: divergences between minority ethnic and whole population samples. Age Ageing 2009, 38(6):703–710. 21. Bowling A: Perceptions of active ageing in Britain: divergences between minority ethnic and whole population samples. Age Ageing 2009, 38(6):703–710. Competing interests Competing interests The authors declare that they have no competing interests. References 1. Kalache A, Kickbusch I: A global strategy for healthy ageing. World Health 1997, 50(4):4–5. 2. World Health Organization: Active Ageing: A Policy Framework. Geneva; 2002. 3. Bowling A: Enhancing later life: how older people perceive active ageing? Aging Ment Health 2008, 12:293–301. 4. Hansen-Kyle L: A concept analysis of healthy aging. Nurs Forum 2005, 40(2):45–57. 5. Peel N, Bartlett H, McClure R: Healthy ageing: how is it defined and measured? Australas J Ageing 2004, 23(3):115–119. 6. Rowe JW, Kahn RL: Successful aging. Gerontologist 1997, 37(4):433–440. 7. Bowing A, Dieppe P: What is successful ageing and who should define it? BMJ 2005, 331:1548–1551. 1. Kalache A, Kickbusch I: A global strategy for healthy ageing. World Health 1997, 50(4):4–5. 2. World Health Organization: Active Ageing: A Policy Framework. Geneva; 2002. • Convenient online submission • Thorough peer review 3. Bowling A: Enhancing later life: how older people perceive active ageing? Aging Ment Health 2008, 12:293–301. 7. Bowing A, Dieppe P: What is successful ageing and who should define it? BMJ 2005, 331:1548–1551.

https://openalex.org/W2031312462

https://discovery.ucl.ac.uk/id/eprint/1448768/1/nl501565b.pdf

English

Self-Catalyzed Ternary Core–Shell GaAsP Nanowire Arrays Grown on Patterned Si Substrates by Molecular Beam Epitaxy

Nano letters

cc-by

Most are grown on GaAs, InAs, or InP substrates via selective area growth or gold-catalyzed growth.32−36 The self-catalyzed growth method, by using Ga droplets in this work,37 has the advantage of avoiding contamination associated with foreign metal catalytic drop- lets.38−40 The Ga droplet can be crystallized into the NW to realize core−shell heterostructured NWs without any catalyzed vapor−liquid−solid (VLS)41 growth of the shell material in the axial direction. More importantly, Ga-assisted NW growth tends to produce NWs with pure zinc blende phase and low defect densities due to the low surface energy and shape of the Ga droplet.21,42 Despite the aforementioned benefits, there are Received: April 28, 2014 Revised: June 3, 2014 Published: June 27, 2014 4542 dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 T he integration of III−V materials on a Si platform has been pursued for more than 40 years.1,2 By integrating the direct band gap III−V materials that have high absorption coefficients, high carrier mobilities, and large solar spectrum coverage onto the mature and cost-effective Si platform, it would create novel optoelectronic devices for Si photonics.3−9 However, this integration remains challenging due to the large lattice mismatch and the difference in thermal expansion coefficient.3,10,11 In recent years, one-dimensional semiconduc- tor III−V nanowires (NWs) have gained significant attention for integrating III−V materials and devices on a Si platform because of their unique structural, optical, and electronic properties.12−15 The strain formed between III−V NWs and Si substrates can be effectively relieved in an elastic way due to their small interfacial area.16−19 The III−V materials could thus be monolithically grown on Si substrates in the form of NWs. At the same time, the novel device architectures with high performance have been predicted for III−V NWs monolithi- cally grown on a Si platform.15,20−24 For example, it has been predicted that a two-junction tandem solar cell (SC), consisting of a 1.7 eV NW junction and a 1.1 eV Si junction, has a theoretical efficiency of 33.8% at 1 sun AM1.5G and 42.3% under 500 suns AM1.5D concentration.25 In addition, Scofield et al.26 forecasted that a NW laser employing NW photonic crystal structure has the potential to have high-Q optical cavities, small mode volume, high spontaneous emission coupling factor, and low threshold power, which are highly desirable for photonic circuit integration. T The III−V NW growth on unpatterned substrates typically follows a self-assembled mechanism where the NWs are randomly positioned. Letter pubs.acs.org/NanoLett Terms of Use CC-BY Letter pubs.acs.org/NanoLett Terms of Use CC-BY ovel device architectures with high redicted for III−V NWs monolithi- rm.15,20−24 For example, it has been on tandem solar cell (SC), consisting n and a 1.1 eV Si junction, has a 3.8% at 1 sun AM1.5G and 42.3% oncentration.25 In addition, Scofield NW laser employing NW photonic potential to have high-Q optical lume, high spontaneous emission threshold power, which are highly uit integration. The III−V NW growth on unpatterned substrates typically follows a self-assembled mechanism where the NWs are randomly positioned. To improve device performance, such as to achieve high-efficiency NW SCs27−29 and to make NW photonic crystals,30,31 each NW should be precisely located. Controlling the NW position is therefore of critical importance and can be achieved by growing on patterned substrates. Moreover, the parasitic bulk (cluster) deposition, often accompanying the NW growth on unpatterned substrates, can be suppressed by the pattern growth. Consequently, significant attention has been paid to the III−V NW growth on patterned substrates in recent years. Most are grown on GaAs, InAs, or InP substrates via selective area growth or gold-catalyzed growth.32−36 The self-catalyzed growth method, by using Ga droplets in this work,37 has the advantage of avoiding contamination associated with foreign metal catalytic drop- lets.38−40 The Ga droplet can be crystallized into the NW to realize core−shell heterostructured NWs without any catalyzed vapor−liquid−solid (VLS)41 growth of the shell material in the axial direction. More importantly, Ga-assisted NW growth tends to produce NWs with pure zinc blende phase and low defect densities due to the low surface energy and shape of the Ga droplet.21,42 Despite the aforementioned benefits, there are Received: April 28, 2014 Revised: June 3, 2014 Published: June 27, 2014 © 2014 American Chemical Society 4542 dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 The III−V NW growth on unpatterned substrates typically follows a self-assembled mechanism where the NWs are randomly positioned. To improve device performance, such as to achieve high-efficiency NW SCs27−29 and to make NW photonic crystals,30,31 each NW should be precisely located. Controlling the NW position is therefore of critical importance and can be achieved by growing on patterned substrates. Moreover, the parasitic bulk (cluster) deposition, often accompanying the NW growth on unpatterned substrates, can be suppressed by the pattern growth. Consequently, significant attention has been paid to the III−V NW growth on patterned substrates in recent years. pubs.acs.org/NanoLett Terms of Use CC-BY pubs.acs.org/NanoLett Terms of Use CC-BY Self-Catalyzed Ternary Core−Shell GaAsP Nanowire Arrays Grown on Patterned Si Substrates by Molecular Beam Epitaxy Self-Catalyzed Ternary Core−Shell GaAsP Nanowire Arrays Grown on Patterned Si Substrates by Molecular Beam Epitaxy Yunyan Zhang,*,†,∥Jiang Wu,†,∥Martin Aagesen,‡,∥Jeppe Holm,‡ Sabina Hatch,† Mingchu Tang,† Suguo Huo,§ and Huiyun Liu*,† †Department of Electronic and Electrical Engineering, University College London, London WC1E 7JE, United Kingdom ‡Gasp Solar ApS, Gregersensvej 7, Taastrup DK-2630, Denmark §London Centre for Nanotechnology, University College London, London WC1H 0AH, United Kingdom ABSTRACT: The growth of self-catalyzed ternary core−shell GaAsP nanowire (NW) arrays on SiO2 patterned Si(111) substrates has been demonstrated by using solid-source molecular beam epitaxy. A high-temperature deoxidization step up to ∼900 °C prior to NW growth was used to remove the native oxide and/or SiO2 residue from the patterned holes. To initiate the growth of GaAsP NW arrays, the Ga predeposition used for assisting the formation of Ga droplets in the patterned holes, was shown to be another essential step. The effects of the patterned-hole size on the NW morphology were also studied and explained using a simple growth model. A lattice-matched radial GaAsP core−shell NW structure has subsequently been developed with room-temperature photoluminescence emission around 740 nm. These results open up new perspectives for integrating position-controlled III− V NW photonic and electronic structures on a Si platform. O S N d h lf l d G P III V S h h d d KEYWORDS: Nanowire array, patterned growth, self-catalyzed, GaAsP, III−V on Si, high-temperature deoxidization, molecular beam epitaxy −V materials on a Si platform has e than 40 years.1,2 By integrating the materials that have high absorption mobilities, and large solar spectrum e and cost-effective Si platform, it ectronic devices for Si photonics.3−9 remains challenging due to the large e difference in thermal expansion years, one-dimensional semiconduc- Ws) have gained significant attention erials and devices on a Si platform structural, optical, and electronic formed between III−V NWs and Si ly relieved in an elastic way due to 6−19 The III−V materials could thus n Si substrates in the form of NWs. dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 To improve device performance, such as to achieve high-efficiency NW SCs27−29 and to make NW photonic crystals,30,31 each NW should be precisely located. Controlling the NW position is therefore of critical importance and can be achieved by growing on patterned substrates. Moreover, the parasitic bulk (cluster) deposition, often accompanying the NW growth on unpatterned substrates, can be suppressed by the pattern growth. Consequently, significant attention has been paid to the III−V NW growth on patterned substrates in recent years. Most are grown on GaAs, InAs, or InP substrates via selective area growth or gold-catalyzed growth.32−36 The self-catalyzed growth method, by using Ga droplets in this work,37 has the advantage of avoiding contamination associated with foreign metal catalytic drop- lets.38−40 The Ga droplet can be crystallized into the NW to realize core−shell heterostructured NWs without any catalyzed vapor−liquid−solid (VLS)41 growth of the shell material in the axial direction. More importantly, Ga-assisted NW growth tends to produce NWs with pure zinc blende phase and low defect densities due to the low surface energy and shape of the Ga droplet.21,42 Despite the aforementioned benefits, there are be monolithically grown on Si substrates in the form of NWs. At the same time, the novel device architectures with high performance have been predicted for III−V NWs monolithi- cally grown on a Si platform.15,20−24 For example, it has been predicted that a two-junction tandem solar cell (SC), consisting of a 1.7 eV NW junction and a 1.1 eV Si junction, has a theoretical efficiency of 33.8% at 1 sun AM1.5G and 42.3% under 500 suns AM1.5D concentration.25 In addition, Scofield et al.26 forecasted that a NW laser employing NW photonic crystal structure has the potential to have high-Q optical cavities, small mode volume, high spontaneous emission coupling factor, and low threshold power, which are highly desirable for photonic circuit integration. Received: April 28, 2014 Revised: June 3, 2014 Published: June 27, 2014 2 dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 Received: April 28, 2014 Revised: June 3, 2014 Published: June 27, 2014 © 2014 American Chemical Society 4542 Letter Letter Nano Letters Figure 1. SEM images (tilt angle = 25°) of NWs grown on patterned substrates (a,c,e) without and (b,d,f) with the high-temperature deoxidization step. The measured sizes of holes are ∼85 nm for (a,b), ∼185 nm for (c,d), and between 500 nm and 1 μm for (e,f). The insets in (c,e) show a magnified image of each patterned area. Figure 1. SEM images (tilt angle = 25°) of NWs grown on patterned substrates (a,c,e) without and (b,d,f) with the high-temperature deoxidization step. The measured sizes of holes are ∼85 nm for (a,b), ∼185 nm for (c,d), and between 500 nm and 1 μm for (e,f). The insets in (c,e) show a magnified image of each patterned area. The patterned Si(111) substrates were prepared using nanoimprint lithography.46,49 First, a layer of thermal oxide was grown in a high-temperature oven. SF2GS, microresist, was then spin coated over the oxide and imprinted with a jet-and- flash imprint lithography (Imprio 100) system. The imprinted patterns were transferred to the thermal oxide by depositing a 20 nm Al film as shadowing mask followed by both dry and wet etching of the thermal oxide. After the nanoimprint lithography, the patterned substrates were stored in ambient conditions before NW growth. Before being loaded into the ultrahigh- vacuum chamber of MBE system, a 5% aqueous hydrofluoric acid (HF) solution was used to chemically etch the substrates for 2 min in attempt to remove the native oxide from the patterned holes. The final mask SiO2 thickness was ∼30 nm, measured from transmission electron microscopy (TEM). The substrates were then rinsed with deionized water for 1 min and dried with nitrogen. After a thermal degassing at 500 °C for 1 h under ultrahigh vacuum conditions, the substrates were loaded into the MBE growth chamber for NW growth. For NW growth without the HTD step, the temperature was ramped directly to the growth temperature of ∼630 °C. For NW growth with the HTD step, the substrates were heated up to 900 °C for 20 min and then cooled to a growth temperature of ∼630 °C. The NWs were grown with a Ga flux of 1.6 × 10−7 Torr, V/III flux ratios between 3 and 20, and a P/(P + As) flux ratio of 12% throughout the growth duration of 45 min. only a few reports on self-catalyzed NW arrays on patterned Si substrates in the literature,28,43−47 and it remains challenging and difficult to ensure reproducibility.28 Currently, III−V NW research is predominantly performed on binary material systems, such as GaAs and InAs, due to its simple fabrication procedures. dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 An alternative explanation for the phenomenon in Figure 1a,c,e could be the reaction of Ga with the oxide layer to form pinholes in the patterned holes where NW growth initiates.56 Both theories would explain the growth of NWs in Figure 1a,c,e, though a thin oxide layer remained because the Ga droplet could reach the Si substrate through a pinhole in the oxide. Consequently, the holes on the patterned substrate are likely not oxide-free after the HF cleaning. A thin oxide layer in patterned holes could be the reason for reduced repeatability and low yield reported previously for III−V NWs grown on patterned Si substrates.28 This is also supported by the fact that the most successful patterned growths are reported on GaAs,31 InAs,33 or InP32 substrates on which it is much easier to remove native oxide from patterned holes in comparison with Si substrates. For the samples with the HTD step, the native oxide and/or SiO2 residue can be completely removed, leading to a fully exposed Si epitaxial surface for the patterned holes, and hence the complete filling of patterned holes with III−V materials as shown in Figure 1b,d,f. In Figure 1b, Ga consumption by VS epitaxial growth mode is slow because of the small hole size, which will allow a significant percentage of the Ga atoms to accumulate into a droplet and hence to catalyze the NW growth. However, for wider patterned holes in Figure 1d,f, the excessive nucleation area for GaAsP VS epitaxial growth rapidly consumes the Ga atoms and prevents droplet formation. This will inhibit the NW growth, leading to a cluster in each hole in Figure 1d,f. This suggests that an oxide layer on the Si substrate has a 2D nucleation-blocking effect that helps Ga adatoms accumulate into droplets for the self-catalyzed VLS growth.54,55 It also indicates the importance of Ga droplet for self-catalyzed GaAsP NW growth Scanning electron microscope (SEM) measurements of the nanowires were performed with a Zeiss XB 1540 FIB/SEM system. It is believed that the HF treatment can not only remove the native oxide layer but also form stable H-terminated Si surfaces.50−52 The effect of using a HF solution to remove oxide in the patterned holes prior to the growth of GaAsP NWs was first investigated. The NW growth was initialized by simultaneously introducing Ga and group-V fluxes. Despite ternary NWs offering more freedom in adjusting the energy bandgap for different optoelectronic applications, it is a comparatively less touched field. The growth mechanism of ternary NWs is not well understood and many questions remain regarding the critical parameters of NW growth. There is only one report on self- catalyzed ternary III−V NWs grown on patterned Si substrates in the literature.43 The ternary material, GaAsP, has a band gap that covers wavelengths ranging from green (550 nm) to near- infrared (860 nm) at room temperature.48 Consequently, it is one of the most promising III−V compound semiconductors for photovoltaics and visible emitters. In this Letter, we demonstrated that a high-temperature deoxidization (HTD) step prior to NW growth is critical to thoroughly remove the oxide within the patterned holes for the growth of GaAsP NWs. To initiate the NW growth, a Ga predeposition step is another important step to assist the formation of Ga droplet. The influence of patterned-hole size on NW morphology has been further studied and explained. After the core GaAsP NW array growth, the growth of a lattice-matched GaAsP shell was finally demonstrated under vapor−solid (VS) epitaxial mode. 4543 Letter Figure 2. SEM images (tilt angle = 25°) of NWs grown in (a) ∼50, (b) ∼75, and (c) ∼135 nm holes. (d) NW length as a function of NW base diameter for NWs show in (a−c). The insets in panels a−c show the magnified images of a single NW grown in different-size patterned holes. Nano Letters Letter Letter Nano Letters Figure 2. SEM images (tilt angle = 25°) of NWs grown in (a) ∼50, (b) ∼75, and (c) ∼135 nm holes. (d) NW length as a function of NW base diameter for NWs show in (a−c). The insets in panels a−c show the magnified images of a single NW grown in different-size patterned holes. small patterned holes with the similar size of Ga droplets, the NW growth is limited to one, whereas multiple NWs could occupy the wider holes. Because of the inhibited III−V growth on the oxide layer, the position of the grown NW is determined by the pinhole position. Therefore, NWs only partially cover the holes, as shown in the insets of Figure 1c,e. dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 Nano Letters one correspondence between holes and droplets. If the hole radius, r, is larger than that of the droplet, there will be a Si region that is not covered by the Ga droplet in the oxide-free hole, referred to as “vacant oxide-free area”. After the Ga predeposition in step 1, the group V flux is introduced with the group III flux to commence the NW growth at step 2. If the patterned-hole size is smaller than or equal to the droplet size, as shown in case A in Figure 3, all the Ga adatoms within the collection radius, R, are diffusing to the Ga droplet, which provides abundant Ga replenishment for the NW VLS growth. This is in accordance with the NW growth in ∼50 nm holes observed in Figure 2a. When the size of the patterned hole is slightly greater than the Ga droplet size, as shown in case B in Figure 3, there will be a vacant oxide-free Si area in the hole. In the presence of group V flux, the Ga adatoms are not only diffusing and incorporating into the droplet but also deposited together with As and P in the vacant oxide-free area in the form of VS epitaxial growth mode. The vacant oxide-free area will partially consume the Ga adatoms, leading to reducing the number of Ga adatoms incorporated into the droplet. As the NW growth proceeds, the VS epitaxial growth in the vacant oxide-free area grows along the NW length, which leads to the formation of the low energy side facets shown in Figure 2b. If the patterned-hole size is much greater than the catalytic droplet (see the case C in Figure 3), the vacant oxide-free area is much larger. As a result, most of the Ga adatoms diffusing from the SiO2 mask will be consumed by the VS epitaxial growth in the vacant oxide-free area. This will severely reduce the flux of Ga adatoms that reach the droplet for NW growth, leading to the formation of NWs with wide base and short length as observed in Figure 2c. This model also explains the reason why NWs that have a wider base diameter are shorter in length, as shown in Figure 2d. Nano Letters To initiate the NW growth after the HTD step, a 60 s Ga predeposition step with a flux of 1.6 × 10−7 Torr was introduced to form Ga droplets before introducing group V fluxes. At the same time, the effects of hole sizes on the NW morphology were also studied, by having different patterned hole sizes (50, 75, and 135 nm) on the same Si growth substrate. For the small hole size (∼50 nm), the NWs have an uniform morphology with a droplet on the tip of each NW (see Figure 2a), which confirms that the NW growth follows the catalyzed VLS growth mode. Increasing the hole size to ∼75 nm results in NW growth without uniform morphology and parasitic crystal growth observed at the NW base, as shown in the inset of Figure 2b. Further increasing hole size to ∼135 nm promotes the parasitic crystal growth and forms NWs with a wider base and stunted length, as shown in Figure 2c. The relationship between the NW base diameter and the NW length is summarized in Figure 2d. The NW length is decreased with increasing NW base diameter, which indicates that the parasitic crystal growth at the base of the NW is competitive with the NW growth. g The effects of hole size on the morphology of GaAsP nanowires could be explained by the size difference between the Ga droplets and patterned holes, which is illustrated by the three stages of NW growth in Figure 3. The first stage shows Figure 3. Illustration of the NWs grown in holes with different sizes at three stages: stage 1, the formation of Ga droplets; stage 2, the early stage of the NW growth; and stage 3, the final stage of the NW growth. The hole size in (A) is equal to or less than, (B) slightly greater than, and (C) much larger than the catalytic droplet size in stage 1. g g Core−shell growth is highly desirable for NW applications as it not only adjusts the NW diameter but also permits the growth of an advanced structure by radial composition and doping modulation, such as p-i-n lateral junction and lateral quantum wells. Following the core NW growth, the Ga shutter was closed while keeping the group V fluxes to consume the Ga droplets. This leads to the termination of the axial growth. For the samples without the HTD step, the patterned holes are only partially occupied by NW growth, as shown in Figure 1a,c,e. When the size of the patterned hole is ∼85 nm, a significant number of patterned holes are vacant. For the occupied holes, there is only a single NW present, as shown in Figure 1a. For the wider holes (∼185 nm), more holes are occupied by NWs and none is vacant as observed in Figure 1c. Further increasing the hole size to 0.5−1 μm, at least four NWs occupy each hole as observed in Figure 1e. A similar phenomenon has been reported previously.44,45,47 These results suggest that the patterned holes in Figure 1a,c,e might be partly covered by the native oxide and/or SiO2 residue prior to NW growth. To verify this, an in situ HTD step was used prior to the NW growth by heating the substrates up to ∼900 °C to remove the native oxide and/or SiO2 residue after the HF etch.3,5,53 Figure 1b,d,f shows the effects of HTD step on NW growth. Although there are a few NWs grown in the smaller holes in Figure 1b, most of the holes are covered by clusters, and none is vacant as shown in Figure 1b,d,f. Considerable difference between the samples with and without the HTD step is that the patterned holes are completely occupied by III−V materials in the form of clusters or NWs for the samples with the HTD step, while for the samples without the HTD the patterned holes are partly occupied or vacant. This difference strongly suggests that native oxide and/or SiO2 residue remain after the HF etch. If a thin oxide layer coats the patterned hole, pinholes, through which the Si lattice could be reached, may exist. The small pinhole size slows the Ga deposition rate and allows Ga adatoms to accumulate into a droplet and catalyze NW growth.54,55 For dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 4544 Letter Nano Letters The GaAsP shell was then grown around the as-grown NWs by VS epitaxial growth mode at ∼485 °C for 1 h with a Ga flux of 1.6 × 10−7 Torr, a V/III flux ratio of 60 and a P/(P + As) flux ratio of 30%. Figure 4a shows the SEM image of the core−shell GaAsP NWs. The core−shell NW array has uniform morphology with smooth sidewalls. The irregular tip is attributed to the core defects that originated during the unoptimized droplet consumption step for terminating the core growth. Figure 4b shows the X-ray diffraction of both the core and the core−shell NWs. The single-peak and close proximity Figure 3. Illustration of the NWs grown in holes with different sizes at three stages: stage 1, the formation of Ga droplets; stage 2, the early stage of the NW growth; and stage 3, the final stage of the NW growth. The hole size in (A) is equal to or less than, (B) slightly greater than, and (C) much larger than the catalytic droplet size in stage 1. the formation of Ga droplets in the patterned holes using the Ga predeposition step. The droplet initiates NW VLS growth in the patterned holes.54 Because the patterned holes in our research are oxide-free and their size is much less than the Ga adatom diffusion length at this temperature, there is a one-to- Figure 4. (a) SEM image of the core−shell GaAsP NW arrays (tilt angle = 25°), (b) X-ray diffraction results of core and core−shell GaAsP NWs, and (c) room-temperature photoluminescence of the core−shell GaAsP NW arrays. Figure 4. (a) SEM image of the core−shell GaAsP NW arrays (tilt angle = 25°), (b) X-ray diffraction results of core and core−shell GaAsP NWs, and (c) room-temperature photoluminescence of the core−shell GaAsP NW arrays. 4545 Nano Letters Figure 5. (a) Low-magnification TEM image of a core−shell NW standing vertically on the Si substrate. (b) TEM image of the interface between the NW and the Si substrate highlighted by the red circle in (a). (c) The magnified TEM image spliced from four images shows the interface area highlighted by the yellow box in (b). (d) High-resolution TEM image that shows the interface area highlighted by the blue box in (c). (e) Illustration of the high temperature desorption of oxide. Nano Letters Letter Figure 5. (a) Low-magnification TEM image of a core−shell NW standing vertically on the Si substrate. Notes The authors declare no competing financial interest. The authors declare no competing financial interest. ■ACKNOWLEDGMENTS H.L. would like to thank The Royal Society for funding his University Research Fellowship. H.L. would like to thank The Royal Society for funding his University Research Fellowship. Nano Letters (b) TEM image of the interface between the NW and the Si substrate highlighted by the red circle in (a). (c) The magnified TEM image spliced from four images shows the interface area highlighted by the yellow box in (b). (d) High-resolution TEM image that shows the interface area highlighted by the blue box in (c). (e) Illustration of the high temperature desorption of oxide. take place in the holes. The effects of the size of the patterned holes were also studied and shown to have a significant influence on the NW morphology. Finally, the growth of lattice-matched core−shell GaAsP NW arrays has been demonstrated with room-temperature PL emission around 740 nm. These results provide an essential step for the growth of ternary NW arrays on patterned Si substrates and offer us freedom in designing novel photonic devices by controlling both the ternary NW position and composition. of the angles of their peak intensities indicate a lattice-matched core and shell for GaAsP NWs.37 The optical properties of these core−shell NWs were also characterized at room temperature. The photoluminescence measurement was carried out with excitation from 635 nm diode-pumped solid-state laser at a power density of 500 mW/cm2. Figure 4c shows the room- temperature emission from the core−shell NW arrays with peak intensity at ∼740 nm, which exhibits their good crystal quality and demonstrates their potential application in photovoltaics, visible emitters, and photonic crystals. ■AUTHOR INFORMATION The interface between the core−shell NW and the Si substrate was studied with the use of TEM and is shown in Figure 5. As shown in Figure 5a,b, the NW is growing out of the patterned hole and standing vertically on the substrate. From Figure 5b−d, it is interesting to find out that the NW base is extended vertically into Si substrate and laterally wedged into the interface between the SiO2 mask and the Si substrate. This phenomenon could be explained by the illustration in Figure 5e. During the HTD process, the high temperature will cause oxide evaporation. At the same time, the reaction between the Si and the oxide can also be triggered to produce SiO, which is much easier to evaporate than SiO2 at this temperature.57 The reaction started from the pinholes in the thin oxide layer in the holes and extended laterally along the Si/ oxide interface, which caused an empty space wedged into the Si substrate below the patterned hole and the interface region between the SiO2 mask and Si substrate. This wedged space into the Si substrate observed in Figure 5 indicates that the patterned holes are not free of the native oxide and/or SiO2 residue prior to the HTD. During the NW growth, this empty wedged space will be filled by GaAsP. Corresponding Authors Corresponding Authors *E-mail: [email protected] (Y.Z.). ( ) *E-mail: [email protected] (H.L.). dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 Nano Letters Letter (42) Zhang, Y.; Aagesen, M.; Holm, J. V.; Jørgensen, H. I.; Wu, J.; Liu, H. Nano Lett. 2013, 13, 3897. ( ) (10) Choi, C.; Otsuka, N.; Munns, G.; Houdre, R.; Morkoc, H.; Zhang, S. L.; Levi, D.; Klein, M. V. Appl. Phys. Lett. 1987, 50, 992. g pp y (11) Fischer, R.; Morkoc, H.; Neumann, D. A.; Zabel, H.; Choi, C.; (43) Conesa-Boj, S.; Kriegner, D.; Han, X.-L.; Plissard, S.; Wallart, X.; Stangl, J.; Fontcuberta i Morral, A.; Caroff, P. Nano Lett. 2014, 14, 326. (43) Conesa-Boj, S.; Kriegner, D.; Han, X.-L.; Plissard, S.; Wallart, X.; ( ) j g Stangl, J.; Fontcuberta i Morral, A.; Caroff, P. Nano Lett. 2014, 14, 326. ( ) Otsuka, N.; Longerbone, M.; Erickson, L. P. J. Appl. Phys. 1986, 60, 1640. (44) Plissard, S.; Larrieu, G.; Wallart, X.; Caroff, P. Nanotechnology 2011, 22, 275602. (12) Lieber, C. M. MRS Bull. 2011, 36, 1052. (12) Lieber, C. M. MRS Bull. 2011, 36, 1052. (45) Plissard, S.; Dick, K. A.; Larrieu, G.; Godey, S.; Addad, A.; Wallart, X.; Caroff, P. Nanotechnology 2010, 21, 385602. (13) Li, Y.; Qian, F.; Xiang, J.; Lieber, C. M. Mater. Today 2006, 9, 18. g (46) Munshi, A. M.; Dheeraj, D. L.; Fauske, V. T.; Kim, D.-C.; Huh, (46) Munshi, A. M.; Dheeraj, D. L.; Fauske, V. T.; Kim, D.-C.; Huh, J.; Reinertsen, J. F.; Ahtapodov, L.; Lee, K. D.; Heidari, B.; van Helvoort, A. T. J.; Fimland, B. O.; Weman, H. Nano Lett. 2014, 14, 960. (14) Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, (14) Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C. M. Nature 2002, 415, 617. C. M. Nature 2002, 415, 617. (15) Yang, P.; Yan, R.; Fardy, M. Nano Lett. 2010, 10, 1529. ( ) (16) Chuang, L. C.; Moewe, M.; Chase, C.; Kobayashi, N. P.; Chang- Hasnain, C.; Crankshaw, S. Appl. Phys. Lett. 2007, 90, 043115. (47) Gibson, S. J.; Boulanger, J. P.; LaPierre, R. R. Semicond. Sci. Technol. 2013, 28, 105025. pp y (17) Cirlin, G. E.; Dubrovskii, V. G.; Soshnikov, I. P.; Sibirev, N. V.; (48) Vurgaftman, I.; Meyer, J. R.; Ram-Mohan, L. R. J. Appl. Phys. 2001, 89, 5815. Samsonenko, Y. B.; Bouravleuv, A. D.; Harmand, J. C.; Glas, F. Phys. Status Solidi RRL 2009, 3, 112. (49) Haas, F.; Sladek, K.; Winden, A.; von der Ahe, M.; Weirich, T. Nano Letters gy (24) Wu, J.; Li, Y.; Kubota, J.; Domen, K.; Aagesen, M.; Ward, T.; Sanchez, A.; Beanland, R.; Zhang, Y.; Tang, M.; Hatch, S.; Seeds, A.; gy (24) Wu, J.; Li, Y.; Kubota, J.; Domen, K.; Aagesen, M.; Ward, T.; (24) Wu, J.; Li, Y.; Kubota, J.; Domen, K.; Aagesen, M.; Ward, T.; Sanchez, A.; Beanland, R.; Zhang, Y.; Tang, M.; Hatch, S.; Seeds, A.; Liu H Nano Lett 2014 14 2013 Sanchez, A.; Beanland, R.; Zhang, Y.; Tang, M.; Hatch, S.; Seeds, A.; Liu, H. Nano Lett. 2014, 14, 2013. (55) Ambrosini, S.; Fanetti, M.; Grillo, V.; Franciosi, A.; Rubini, S. AIP Adv. 2011, 1, 042142. Liu, H. Nano Lett. 2014, 14, 2013. (25) LaPierre, R. R. J. Appl. Phys. 2011, 110, 014310. ( ) (56) Fontcuberta i Morral, A.; Colombo, C.; Abstreiter, G.; Arbiol, J.; Morante, J. R. Appl. Phys. Lett. 2008, 92, 063112. (26) Scofield, A. C.; Kim, S.-H.; Shapiro, J. N.; Lin, A.; Liang, B.; Scherer, A.; Huffaker, D. L. Nano Lett. 2011, 11, 5387. (57) Madsen, M. H.; Aagesen, M.; Krogstrup, P.; Sørensen, C.; Nygård, J. Nanoscale Res. Lett. 2011, 6, 516. (27) Lin, C.; Povinelli, M. L. Opt. Express 2009, 17, 19371. (28) Heiss, M.; Russo-Averchi, E.; Dalmau-Mallorquí, A.; Tütüncüoğlu, G.; Matteini, F.; Rüffer, D.; Conesa-Boj, S.; Demichel, O.; Alarcon-Lladó, E.; Fontcuberta i Morral, A. Nanotechnology 2014, 25, 014015. (29) Kelzenberg, M. D.; Boettcher, S. W.; Petykiewicz, J. A.; Turner- Evans, D. B.; Putnam, M. C.; Warren, E. L.; Spurgeon, J. M.; Briggs, R. M L i N S A H A N t M t 2010 9 239 M.; Lewis, N. S.; Atwater, H. A. Nat. Mater. 2010, 9, 239. (30) Diedenhofen, S. L.; Janssen, O. T. A.; Hocevar, M.; Pierret, A.; Bakkers, E. P. A. M.; Urbach, H. P.; Rivas, J. G. ACS Nano 2011, 5, 5830. (31) Scofield, A. C.; Shapiro, J. N.; Lin, A.; Williams, A. D.; Wong, P.- S.; Liang, B. L.; Huffaker, D. L. Nano Lett. 2011, 11, 2242. Å (32) Wallentin, J.; Anttu, N.; Asoli, D.; Huffman, M.; Åberg, I.; Magnusson, M. H.; Siefer, G.; Fuss-Kailuweit, P.; Dimroth, F.; Witzigmann, B.; Xu, H. Q.; Samuelson, L.; Deppert, K.; Borgström, M. T. Science 2013, 339, 1057. (33) Tomioka, K.; Mohan, P.; Noborisaka, J.; Hara, S.; Motohisa, J.; Fukui, T. J. Cryst. Growth 2007, 298, 644. (34) Nakai, E.; Yoshimura, M.; Tomioka, K.; Fukui, T. Jpn. J. Appl. Phys. ■REFERENCES (1) Mathine, D. L. IEEE J. Sel. Top. Quantum Electron. 1997, 3, 952. ( ) (1) Mathine, D. L. IEEE J. Sel. Top. Quantum Electron. 1997, 3, 952. ( ) (2) Roelkens, G.; Liu, L.; Liang, D.; Jones, R.; Fang, A.; Koch, B.; Bowers, J. Laser Photonics Rev. 2010, 4, 751. (3) Lee, A. D.; Qi, J.; Tang, M.; Zhang, Y.; Seeds, A. J.; Liu, H. IEEE J. Sel. Top. Quantum 2013, 19, 1701510. (3) Lee, A. D.; Qi, J.; Tang, M.; Zhang, Y.; Seeds, A. J.; Liu, H. IEEE J. Sel. Top. Quantum 2013, 19, 1701510. (4) Lee, A.; Jiang, Q.; Tang, M.; Seeds, A.; Liu, H. Opt. Express 2012, 20, 22181. (4) Lee, A.; Jiang, Q.; Tang, M.; Seeds, A.; Liu, H. Opt. Express 2012, 20, 22181. In summary, the growth of self-catalyzed core−shell GaAsP NW arrays on patterned Si(111) substrates has been demonstrated by solid-source MBE. The chemical etching via HF solution is not sufficient to obtain oxide-free holes on the patterned substrates for the growth of III−V nanowires. By introducing an in situ high-temperature deoxidization step up to ∼900 °C, the patterned holes can be thoroughly cleaned. Furthermore, the Ga predeposition step was demonstrated to be essential to the catalytic droplet formation in the oxide-free holes and hence ensure the NW growth via VLS growth mode (5) Wang, T.; Liu, H.; Lee, A.; Pozzi, F.; Seeds, A. Opt. Express 2011, 19, 11381. (5) Wang, T.; Liu, H.; Lee, A.; Pozzi, F.; Seeds, A. Opt. Express 2011, 19, 11381. (6) Lee, A.; Liu, H.; Seeds, A. Semicond. Sci. Technol. 2013, 28, 015027. (7) Grassman, T. J.; Carlin, A. M.; Grandal, J.; Ratcliff, C.; Yang, L.; Mills, M. J.; Ringel, S. A. Proc. SPIE 2012, 82560R. g (8) Grassman, T. J.; Brenner, M. R.; Gonzalez, M.; Carlin, A. M.; Unocic, R. R.; Dehoff, R. R.; Mills, M. J.; Ringel, S. A. IEEE Trans. Electron Devices 2010, 57, 3361. (9) Geisz, J. F.; Friedman, D. J. Semicond. Sci. Technol. 2002, 17, 769. 4546 Nano Letters E.; Rieger, T.; Lüth, H.; Grützmacher, D.; Schäpers, T.; Hardtdegen, H. Nanotechnology 2013, 24, 085603. (18) Glas, F. Phys. Rev. B 2006, 74, 121302. (19) Joyce, H. J.; Gao, Q.; Hoe Tan, H.; Jagadish, C.; Kim, Y.; Zou, J.; Smith, L. M.; Jackson, H. E.; Yarrison-Rice, J. M.; Parkinson, P.; J.; Smith, L. M.; Jackson, H. E.; Yarrison-Rice, J. M.; Parkinson, P.; J h t f M B P Q t El t 2011 35 23 J.; Smith, L. M.; Jackson, H. E.; Yarrison-Rice, J. M.; Parkinson, P.; Johnstonf, M. B. Prog. Quantum Electron. 2011, 35, 23. gy (50) Morita, Y.; Miki, K.; Tokumoto, H. Appl. Phys. Lett. 1991, 59, 1347. Johnstonf, M. B. Prog. Quantum Electron. 2011, 35, 23. (20) Yan, R.; Gargas, D.; Yang, P. Nat. Photonics 2009, 3, 569. (51) Higashi, G. S.; Chabal, Y. J.; Trucks, G. W.; Raghavachari, K. Appl. Phys. Lett. 1990, 56, 656. (21) LaPierre, R. R.; Chia, A. C. E.; Gibson, S. J.; Haapamaki, C. M.; (21) LaPierre, R. R.; Chia, A. C. E.; Gibson, S. J.; Haapamaki, C. M.; Boulanger, J.; Yee, R.; Kuyanov, P.; Zhang, J.; Tajik, N.; Jewell, N.; Boulanger, J.; Yee, R.; Kuyanov, P.; Zhang, J.; Tajik, N.; Jewell, N.; Rahman, K. M. A. Phys. Status Solidi RRL 2013, 7, 815. (52) Ma, D. D. D.; Lee, C. S.; Au, F. C. K.; Tong, S. Y.; Lee, S. T. Science 2003, 299, 1874. (22) Jiang, X.; Tian, B.; Xiang, J.; Qian, F.; Zheng, G.; Wang, H.; Mai, L.; Lieber, C. M. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 12212. (53) Tang, M.; Chen, S.; Wu, J.; Jiang, Q.; Dorogan, V. G.; Benamara, M.; Mazur, Y. I.; Salamo, G. J.; Seeds, A.; Liu, H. Opt. Express 2014, 22, 11528. (23) Kempa, T. J.; Day, R. W.; Kim, S.-K.; Park, H.-G.; Lieber, C. M. Energy Environ. Sci. 2013, 6, 719. (54) Ambrosini, S.; Fanetti, M.; Grillo, V.; Franciosi, A.; Rubini, S. J. Appl. Phys. 2011, 109, 094306. dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 Nano Letters 2013, 52, 055002. (35) Ikejiri, K.; Sato, T.; Yoshida, H.; Hiruma, K.; Motohisa, J.; Hara, S.; Fukui, T. Nanotechnology 2008, 19, 265604. gy (36) Heiß, M.; Riedlberger, E.; Spirkoska, D.; Bichler, M.; A (36) e ß, .; ed be ge , .; Sp os a, D.; c e , .; bst G.; Fontcuberta i Morral, A. J. Cryst. Growth 2008, 310, 1049. G.; Fontcuberta i Morral, A. J. Cryst. Growth 2008, 310, 1049. (37) Holm, J. V.; Jørgensen, H. I.; Krogstrup, P.; Nygård, J.; Liu, H.; Aagesen, M. Nat. Commun. 2013, 4, 1498. g (38) Jun, W. K.; Willens, R. H.; Duwez, P. O. L. Nature 1960, 187 (4740), 869−870. (39) Bar-Sadan, M.; Barthel, J.; Shtrikman, H.; Houben, L. Nano Lett. 2012, 12, 2352. (40) Wang, Y.; Schmidt, V.; Senz, S.; Gosele, U. Nat. Nanotechnol. 2006, 1, 186. (41) Wagner, R. S.; Ellis, W. C. Appl. Phys. Lett. 1964, 4, 89. dx.doi.org/10.1021/nl501565b | Nano Lett. 2014, 14, 4542−4547 4547

W4386187260.txt

https://zenodo.org/records/8245060/files/pp%20156-170%20V004IS04N001214%20Lassina%20Togola.pdf

fr

Effets sociopsychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako

Zenodo (CERN European Organization for Nuclear Research)

cc-by

ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND Effets sociopsychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako Socio-psychological effects of passenger road transport: Case of road congestion and road transport accidents in the city of Bamako Yapégué BAYOGO, (Enseignant-chercheur) Institut Universitaire de Gestion Université des Sciences Sociales et de Gestion de Bamako République du Mali Lassina TOGOLA, (Enseignant-chercheur) Faculté des Sciences Economiques et de Gestion Université des Sciences Sociales et de Gestion de Bamako République du Mali Adresse de correspondance : Institut Universitaire de Gestion Université des Sciences Sociales et de Gestion Mali (Bamako) 00223 76 46 01 77. Déclaration de divulgation : Les auteurs n'ont pas connaissance de quelconque financement qui pourrait affecter l'objectivité de cette étude. Conflit d’intérêts : Les auteurs ne signalent aucun conflit d'intérêts. Citer cet article BAYOGO, Y., & TOGOLA, L. (2023). Effets sociopsychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako. International Journal of Accounting, Finance, Auditing, Management and Economics, 4(4-2), 156170. https://doi.org/10.5281/zenodo.8245060 Licence Received: July 10, 2023 Cet article est publié en open Access sous licence CC BY-NC-ND Accepted: August 13, 2023 International Journal of Accounting, Finance, Auditing, Management and Economics - IJAFAME ISSN: 2658-8455 Volume 4, Issue 4-2 (2023) 156 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako Effets sociopsychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako Résumé Le transport routier assure la quasi-totalité des déplacements de personnes et de leurs biens dans les mégapoles des pays en développement et la ville de Bamako n’échappent pas à cette règle. La congestion du trafic et les accidents dus au transport routier ont des effets sociaux et psychologiques sur les citoyens. L’objectif de cet article est d’analyser les effets sociopsychologiques en prenant en compte la congestion et les accidents du transport routier dans la ville de Bamako. L’approche méthodologique est basée sur la revue de littérature des thèses de Doctorat, des articles scientifiques, des rapports d’activités et des sites web. Une étude quantitative a été faite sur la base d’enquête auprès de 50 acteurs. Une étude qualitative menée à l’aide d’un guide d’entretien auprès de 11 personnes a permis de vérifier les résultats de l’étude quantitative. Les résultats de l’étude ont montré que la congestion routière et les accidents du transport routier impactent fortement le bien-être social et psychologique des citoyens de la ville de Bamako. De même, les résultats ont indiqué que plus de 90% des personnes enquêtées pensent que les majeurs effets socio – psychologiques du transport routier sont : les pertes en vie humaine, 100% des enquêtés ; le traumatisme dû aux accidents, 95,65% ; les accidents corporels et l’invalidité des accidentés, 91,3% des enquêtés. Les mesures d’atténuation de ces effets sont : le respect des normes de circulation, 100% des enquêtés ; l’installation des appareils de surveillance sur les routes, 91,3% ; la libération des abords des routes et l’application stricte de la loi, 91,3%. La principale limite de la recherche demeure la non-analyse des coûts de la congestion et des accidents de circulation dans le District de Bamako. Mots clés: effets sociaux ; effets psychologiques ; congestion ; accidents ; Bamako. JEL Classification : B55, D91, E71, R41 Type du papier : Recherche empirique Abstract Road transport ensures almost all the movement of people and their goods in the megacities of developing countries and the city of Bamako is no exception to this rule. Traffic congestion and road transport accidents have social and psychological effects on citizens. The objective of this paper is to analyze the socio-psychological effects of transport taking into account road transport accidents and congestion in the Bamako City. The methodological approach is based on a literature review of doctoral theses, scientific articles, activity reports and websites. A quantitative study was carried out on the basis of a survey of 50 stakeholders. A qualitative study was carried out using an interview guide with 11 people to verify the results of the quantitative study. The results of the study showed that road congestion and road transport accidents have a major impact on the social and psychological well-being of the citizens of the city of Bamako. Similarly, the results indicated that more than 90% of those surveyed thought that the major socio-psychological effects of road transport were loss of human life, 100% of respondents; trauma due to accidents, 95.65%; personal injury and disability of accident victims, 91.3% of respondents. The measures to mitigate these effects are compliance with traffic regulations, 100% of respondents; installation of surveillance equipment on the roads, 91.3%; clearing the roadsides and strict application of the law, 91.3%. The main limitation of the research remains the failure to analyze the costs of congestion and traffic accidents in the District of Bamako. Key words: social effects; psychological effects; congestion; accidents; Bamako. Classification JEL : B55, D91, E71, R41 Paper type : Empirical Research 157 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND 1. Introduction La croissance des villes, tant démographique que spatiale, pose de nombreux problèmes, en matière de transport des personnes comme des biens. Le souci de se déplacer, l’évolution des modes de vie, est autant des facteurs qui poussent les populations urbaines à utiliser les véhicules motorisés, qu’ils soient individuels ou collectifs (Murray et al. 1996). Au plan sanitaire, cette tendance se traduit par une augmentation des accidents de la circulation. Selon l’Organisation Mondiale de la Santé, les accidents de la circulation constituent la principale cause de mortalité dans les pays à revenu faible ou intermédiaire. Elle représente la première cause de décès chez les jeunes de 15-29 ans, engendre la mort des usagers vulnérables, la moitié des personnes tuées sur la route sont des piétons et motocyclistes. L’insécurité routière est l’un des facteurs aggravants de la pauvreté et doit être inscrite parmi les priorités des différents gouvernements des pays au sud du Sahara. Elle génère des souffrances physiques et psychologiques au plan individuel et collectif et ses conséquences font d’elle un enjeu majeur pour le développement socio-économique (Mané, 2019). Avec la concentration de la population et des activités de plus en plus nombreuses dans les agglomérations à caractère urbain, la ville occupe de plus en plus des superficies importantes, et l’étalement urbain prend de l’ampleur. Aujourd’hui près de la moitié de la population mondiale vit en zone urbaine (Banque Mondiale, 2005). Le District de Bamako est la capitale de la République du Mali. C’est la plus grande ville du pays avec 3 000 000 d’âmes en 2021 (Direction Nationale de la population du Mali, 2009). Avec la croissance démographique galopante, la ville de Bamako se développe à un rythme extraordinaire. Cette forte croissance de sa population exerce une pression sur le réseau routier. Le besoin de se déplacer pousse les populations de la ville Bamako à utiliser les moyens de transport. Le transport routier des voyageurs dans le District de Bamako est confronté à un certain nombre de difficultés. La circulation est tellement dense qu’elle constitue un problème pour la mobilité des personnes occasionnant des nombreuses congestions et pertes de temps considérables, compte tenu de la croissance démographique et de l’augmentation du parc automobile. Les accidents de circulation constituent, aujourd’hui, un problème de santé publique en République du Mali en général et dans le District de Bamako, en particulier. Le nombre d’accidents et de tués est important, le taux d’accidents de transport voyageur augmente chaque jour. Le District de Bamako, étant le pôle de développement économique majeur du Mali avec 60 % du potentiel industriel et plus de 60 % de l’activité économique formelle, concentre la majorité des véhicules motorisés du pays (Rapport BAD, 2018). Les statistiques des accidents de la circulation routière des cinq dernières années ont montré que le District de Bamako enregistre à lui seul, près de la moitié des victimes des accidents de la route, soit, 4 880 personnes en moyenne par an. Ces accidents, souvent graves et qui peuvent être évités, sont la première cause d’encombrement des services d’urgence dans les hôpitaux (Rapport BAD, 2018). C’est un phénomène qui entraine des dommages en termes de pertes en vie humaine et des dégâts matériels. Chaque mort ou blessé grave imputable à un accident de la route a des conséquences désastreuses, non seulement pour les personnes directement concernées, mais aussi, pour leurs familles, amis et collègues. Un accident grave de la circulation routière constitue dans la majorité des cas des désastres émotionnels et économiques pour les transporteurs et les voyageurs. D’où, la nécessité de trouver des solutions afin de réduire la congestion et les accidents du transport des voyageurs dans le District de Bamako. L’objet de l’article est d’analyser les effets sociopsychologiques du transport en prenant en compte la congestion et les accidents du transport routier dans la ville de Bamako. Pour répondre à cette problématique, après l’introduction, le reste de la communication sera structuré 158 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako en quatre parties. La première partie mettra en évidence la revue de la littérature. La deuxième mettra l’accent sur le cadre méthodologique de l’étude. Dans les troisième et quatrième parties, nous présenterons et interprèterons les résultats. Enfin, nous terminerons par une conclusion. 2. Revue de littérature De multiples études ont été faites concernant la congestion et les accidents du transport routier. Ainsi selon (Moustakbal, 2009), la congestion routière est un phénomène caractérisant la majorité des routes et autoroutes des grandes villes et métropoles mondiales. Elle constitue, de plus en plus, un facteur pesant sur l’ensemble des usagers du réseau routier. Ses répercussions se manifestent dans l’augmentation du temps de déplacements, la surconsommation du carburant, le stress et les risques d’accident. (Cohen, 2006) a montré clairement comment se créer un embouteillage. Selon ses analyses, « sur une route au trafic assez dense, un conducteur ralentit au-dessous de la vitesse limite par exemple si la limite est 130 km/h, mais que le conducteur roule à 120 km/h ; ce conducteur déclenche une réaction en chaîne derrière lui ; puis chaque conducteur freine un peu plus fort que celui qui le précède. Enfin, cette vague de ralentissements va finir par provoquer, quelques kilomètres en arrière, un fort ralentissement, puis, mettre des véhicules à l’arrêt, c’est la formation du bouchon. Selon (MEITE, 2014), la mauvaise utilisation de l’espace public où les piétons et les véhicules en circulation se partagent la chaussée et les trottoirs, favorisant l’insécurité routière à cause du non-respect du Code de la route. Une des principales conséquences de cette situation est la congestion du trafic. (Wiel, 2004) dans son ouvrage ‘’Ville et mobilité : un couple infernal ?’’ montre que le principe de la ville est de faciliter les échanges en nombre et en qualité ; la mobilité, l’accessibilité sont donc essentielles à la qualité de vie urbaine. Mais, la facilité de se mouvoir induit des congestions, des nuisances et des mutations urbaines incontrôlées. De ce fait, Marc WIEL défend la thèse que mobilité et organisation urbaine ne peuvent être dissociées l’une de l’autre et font un système. Car, la mobilité et la morphologie urbaine interagissent l’une sur l’autre, et sont à la fois complémentaires et concurrentes l’une de l’autre. « La ville équilibrée et équilibrante est celle de la vitesse maîtrisée, de la vitesse accordée aux densités de l’occupation du sol, mais aussi, à la fréquence des échanges. L’équilibre entre les transports individuels et collectifs pourra alors être trouvé, mais pas de la même façon selon la taille des agglomérations. (CETE Méditerranée, 2009) a réalisé une étude de trafic sur une partie des voies rapides urbaines de l’aire marseillaise en étudiant plus particulièrement la congestion dite récurrente sur les axes autoroutiers. Le niveau de congestion récurrente peut être défini comme le surplus de demandes qui amène la congestion : c'est-à-dire, le volume de trafic qui doit être enlevé aux sections d'autoroutes pour rétablir une circulation fluide pendant les heures de pointe. Il existe plusieurs formes de congestion, selon leur cause : la congestion récurrente, la congestion « prévisible » (travaux, manifestation, météo) et la congestion due aux incidents et accidents par définition imprévisibles. Chaque cause de congestion est traitée par une stratégie différente. La congestion dite récurrente est liée à un excès de demande observé quotidiennement par rapport à la capacité, c'est-à-dire au débit maximal que peut supporter le réseau. Les congestions récurrentes sont un préjudice collectif d'heures perdues annuelles qui peuvent être chiffrées. (Sané, 2014) ‘’ la mobilité urbaine à Dakar, un frein à l’émergence’’ se focalise sur le problème des embouteillages à Dakar, car, les automobilistes y passent 3 à 4 heures, voire plus par jour, non sans conséquence sur l’ensemble de l’activité économique. L’auteur fait allusion aux comportements des conducteurs qui occasionnent les embouteillages par le non-respect des règles de conduite ; et de l’augmentation du parc automobile. Pour (Sangala, 2017), la critique apportée à cette exiguïté des routes, mais aussi, la présence de nombreuses intersections (ronds-points et carrefours). En principe, si le parc automobile augmente, on devra aussi tenir compte du réseau routier, et l’agrandir ou créer des voies de 159 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND contournement. (Eviar O, 2023) a montré que pour faire face à la congestion et aux embouteillages dans la ville d’Abidjan, l’Etat ivoirien a commandé la construction d’un certains nombres d’infrastructures routières. Selon (Godard et Teurnier, 1992), la croissance urbaine en Afrique subsaharienne est la plus rapide dans le monde. Cette croissance urbaine sans précédent que connaît l’Afrique accroît de manière exponentielle les besoins de déplacements ; alors que le système de transport a de plus en plus des difficultés à les satisfaire ». En effet, c’est le cas à Dakar où on note les problèmes de mobilité avec une demande de transport assez forte. Ainsi, certains ont recours à la marche à pied. Aussi, ce que nous pouvons ajouter, c’est qu’il faut qu’on adapte l’offre de transport en fonction du trafic routier et de l’augmentation de la population, car la configuration des villes change (Sangala, 2017). Les causes des accidents sont multiples et dépendent en grande partie des usagers de la route, des conducteurs, de l’état des infrastructures et du matériel roulant. Les projections sur les cinquante futures années indiquent une hausse considérable du nombre des véhicules et des accidents et l’urgence d’appliquer les mesures de réductions des accidents de la route (Bayogo et al, 2022). En République du Mali, le parc de véhicule est passé de 78 108 véhicules en 2000 à 245 567 en 2010, soit une hausse de 214,39% en dix ans (DNTTMF, 2011 - 2020). Selon la même source, le parc a atteint 418 634 véhicules en 2017, soit une augmentation de 435,97% et 70,48% par rapport à 2000 et 2010. Les traumatismes dus aux accidents de la route représentent un problème majeur pour le développement en Afrique. Le Rapport mondial note que plus de la moitié des personnes tuées sur les routes sont de jeunes adultes âgés entre 15 et 44 ans, et il s’agit souvent des soutiens de famille (Rapport SSATP, 2014). Par ailleurs, la recherche faite par (Yuki K et al, 2011) porte sur un contrôle effectif du trafic en temps réel pour permettre de faire un secourisme rapide dans le but de fluidifier le trafic du réseau de transport. Une comparaison entre les accidents et les décès en Finlande et d’autres pays de l’Europe a été faite par (Paakkonen R et Korpinen L., 2016) les résultats de leur étude ont montré que grâce aux mesures prises par le gouvernement finlandais, ce pays a moins d’accidents de la circulation par rapport à la moyenne européenne. Leur recherche montre la nécessité d’améliorer les mesures déjà utilisées pour le bien-être de la population. Selon eux, les accidents de circulation ont des conséquences sociales, environnementales ainsi qu’économiques, et donc, il est important de chercher les voies et les moyens permettant de les réduire. Les effets d’une forte luminosité de la route sur la performance de la conduite de nuit pour les adultes ont été étudiés. L’étude menée par (Korpinen L et al, 2016) porte sur les accidents dus aux appels téléphoniques, lors de la conduite ou du travail. Selon les résultats de cette étude, environ 26% des personnes accidentées et près de 50% des répondants de leur étude affirment que l’utilisation du téléphone influence le travail ou la conduite. L’étude conclut que le téléphone provoque une forme de distraction dans beaucoup de cas et pas seulement lors de la conduite. Selon cette étude, l’une des causes principales des accidents de la route est l’utilisation du téléphone cellulaire, lors de la conduite. Dans le but de faire face à cette situation, une loi fixant des sanctions lourdes pour l’utilisation du téléphone au volant a été votée. De même, les recherches de ( Waard F et al, 2015) ont conclu que l’utilisation du téléphone au volant constitue un danger très grave en analysant des cyclistes téléphonant lors de leur conduite alors que des observations faites sur 1360 cyclistes dans la ville de Hague en Hollande ont permis de montrer que près de 3,5% seulement ont eu recourt au cellulaire causant des cas d’accidents ( Terzano K., 2013). Un système de sécurité adapté au véhicule avec pour but principal, le contrôle et éventuellement, la limitation de la vitesse des véhicules a été proposée par (Shahram J et al, 2007) alors que (Sitihawa H et Norashikin M., 2018) ont analysé les causes des accidents de la route et ont conclu que ces accidents sont des facteurs de mortalité dans le monde entier. Selon leur étude, les causes des accidents sont multiples et peuvent être dues, soit aux infrastructures, soit aux matériels de transport, soit aux conducteurs. 160 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako Pour y parvenir aux résultats, nous avons formulé l’hypothèse de recherche qui a été établie sur la base de la revue de la littérature. Cette hypothèse de recherche est la suivante : la congestion et les accidents du transport routier des voyageurs ont des effets sociopsychologiques dans la ville de Bamako, à savoir : la dépendance économique des accidentés, les pertes en vie humaine, la perte de temps, etc. 3. Méthodologie de recherche Dans notre étude, nous avons adopté la démarche hypothético-déductive selon le paradigme positiviste. Cette démarche nous a permis de formuler l’hypothèse de recherche que nous allons vérifier dans le cadre d’une étude empirique. 3.1. Terrain et données de l’étude L’élaboration de cet article a nécessité la lecture et l’analyse des ouvrages, des rapports d’activités, des thèses de Doctorat, des articles scientifiques, des mémoires universitaires. Les sites Internet ont été également exploités. Pour les enquêtes de terrain, nous avons choisi la méthode mixte, car, elle nous apporte des informations à la fois quantitatives et qualitatives. Des questionnaires préalablement élaborés ont été administrés à 50 personnes selon les techniques du choix raisonné pour avoir des données quantitatives descriptives. Les enquêtes ont concerné les agents de la police de la Compagnie de circulation routière : 10, les agents de la Direction Générale des Transports : 05, les usagers du transport de voyageurs urbain : 10, les agents de l’Agence Nationale de la Sécurité Routière : 10, les conducteurs : 10, et les agents de santé : 05. Le guide d’entretien a été conçu pour les personnes ressources. Pour cela, nous avons fait un entretien semi-directif avec 11 agents de la Direction Générale des Transports. L’exploitation des différents documents, ainsi que les résultats obtenus au cours des enquêtes auprès des acteurs concernés, ont permis de cerner les effets sociopsychologiques de la congestion et des accidents de circulation du transport voyageur dans le District de Bamako. Tableau 1 : codification des personnes interviewées et leur fonction Initial de l’enquêté Code Fonction Dr F. S, T. K, M. T D1, D2, D3 Agents de santé à Bamako K. K, A. C tous ingénieurs BTP D4, D5 Direction Générale des Transports (DGT) A. G, H. T D6, D7 Agent de la Compagnie de Circulation Routière (CCR) T. G, K, K D8, D9 Agent de la protection civile M. D, L. C D10, D11 Agent de l’Agence Nationale de la Sécurité Routière (ANASER) Source : auteurs, 2022 L’analyse du tableau 1 indique que le guide d’entretien a été conçu de façon aléatoire pour certaines personnes qui sont : trois agents de santé travaillant dans les structures de santé à Bamako, deux ingénieurs des bâtiments et des travaux publics, tout travaillant à la Direction Générale des Transports, deux agents de la police relevant de la Compagnie de Circulation Routière, travailleurs de la protection civile et deux agents de l’Agence Nationale de Sécurité Routière. 3.2. Modèle de recherche Nous présentations notre modèle avec la définition des variables qui vise à montrer les effets sociopsychologiques de la congestion et des accidents du transport routier des voyageurs dans 161 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND la ville de Bamako. La figure 1 représente notre modèle de recherche. Figure 1 : schéma présentant le modèle Variable dépendante Variable indépendante Perte en vie humaine Effets sociopsychologiques de la congestion et des accidents routiers du transport voyageur Dépendance des accidentés Perte de temps Perte matérielle et économique Source: auteurs, 2023 Afin de texter notre hypothèse, nous avons collecté des données. Cela nous a renseignés sur deux types de variables: la variable dépendante et celle indépendante. Les variables dépendantes sont celles susceptibles d’expliquer certaines caractéristiques des effets sociopsychologiques de la congestion et des accidents du transport routier des voyageurs dans la ville de Bamako. 3.3. Traitement des données Quant aux données quantitatives, elles ont été traitées et analysées à l’aide du logiciel SPSS et Exel. Cela a abouti aux résultats. Pour l’analyse qualitative, nous avons procédé à l’analyse du contenu des différents discours tenus. 4. Résultats et discussion Cette quatrième partie de notre étude se focalise sur l’analyse et l’interprétation des résultats obtenus après l’exploration du terrain d’étude. Nos recherches nous ont permis d’obtenir les opinions des personnes enquêtées. Les enquêtés ont souligné les causes de la congestion ainsi que des accidents dus au transport routier dans la ville de Bamako. De même, les effets sociopsychologiques des accidents ont fait l’objet d’étude. Cette section se termine par des propositions d’atténuation de ces effets. 4.1. Causes de la congestion dans la ville de Bamako Nos recherches ont indiqué que les causes de la congestion dans la ville de Bamako sont multiples et s’apprécient différemment suivant les enquêtés. L’analyse de la figure ci-dessous a indiqué que les causes majeures de cette congestion sont : l’insuffisance des infrastructures routières (100% des personnes enquêtées), le mauvais état des infrastructures existantes 91,3%, le non développement des transports collectifs et la mauvaise organisation des transports collectifs 82,61% des personnes enquêtées. Cependant, nos enquêtés pensent que certaines causes, bien qu’elles existent, paraient moins importantes que les causes majeures ci-dessus indiquées. Ces causes mineures sont: le mauvais état et la vétusté du matériel roulant, le nombre élevé des véhicules, l’étroitesse des routes 162 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako existantes et le non-respect des normes de circulation avec respectivement 52, 17%, 52, 52%, 69, 56% et 78, 26% des personnes enquêtées. Figure 2 : opinions des enquêtés sur les causes de la congestion routière dans la ville de Bamako 120 100 100 80 91,3 82,61 69,56 56,52 52,17 60 21,74 8,7 8,7 Non RAS 17,39 8,7 Oui 43,48 39,13 40 20 82,61 78,26 4,35 13,04 4,35 13,04 4,35 0 Insuffisance des Etroitesse des infrastructures routes routières existantes Mauvais état Mauvais état et Non-respect des Nombre élevé Non Mauvaise des vétusté du normes des véhicules développement organisation des infrastructures matériel roulant des transports transports existantes collectifs collectifs Source: auteurs, 2022 Ces résultats se confirment avec l’analyse qualitative. Selon D1, D2 ou D11, la congestion a des causes dépendantes de l’homme et d’autres dépendants du matériel et des infrastructures routières. D1 témoigne « les causes de la congestion sont essentiellement : la multiplicité des intersections, le manque de passage à niveaux différents, d’échangeur, la mauvaise planification urbaine et la prolifération des moyens de transport du personnel en mauvais état, puis, le mauvais et le manque de routes ». De même, D2 raconte ce constat « je pense que trois aspects majeurs sont à la base de la congestion dans la ville de Bamako, ces aspects sont : le manque d’infrastructures, les heures de pointe qui mobilisent un grand nombre d’usagers, la localisation de la plupart des activités à Bamako ».Selon D11, « Les causes de la congestion dans la ville de Bamako sont : le manque d’infrastructures routières, la vétusté du parc l’absence de couloir pour le transport collectif et le manque de transport multimodal ». Les études ont montré que cette congestion est souvent source d’accidents. 4.2. Causes des accidents du transport routier dans la ville de Bamako Les causes des accidents sont multiples. Le tableau 2 spécifie quelques causes des accidents du transport routier dans la ville de Bamako. Tableau 2 : opinions des enquêtés sur les causes des accidents du transport routier dans la ville de Bamako (en %) Propositions Nombre élevé des véhicules Etroitesse des routes Insuffisance d’infrastructures routières Mauvais état des routes Mauvais état et vétusté du matériel roulant Non-respect des normes de circulation Alcoolisme et stupéfiants Fatigue au volant Utilisation du téléphone au volant Oui 26,08 65,21 56,52 82,61 56,52 95,65 91,3 91,3 95,65 Non 69,57 26,09 39,13 13,04 30,44 4,35 8,7 8,7 4,35 RAS 4,35 8,7 4,35 4,35 13,04 Source : auteurs, 2022 163 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND L’analyse du tableau 2 permet de conclure que les causes des accidents sont multiples et sont en général dépendantes des Hommes. Nos enquêtés ont mis l’accent sur: le non-respect des normes de circulation et l’utilisation du téléphone au volant, soit 95,65% des enquêtés. De même, l’accent est mis aussi sur l’alcoolisme et la fatigue au volant 91,3% d’opinions alors que, 26,08% seulement des personnes enquêtées pensent que les accidents sont dus au nombre élevé des véhicules. Les témoignages de D1, D4, D6 et D8 illustrent les mêmes causes majeures des accidents du transport routier dans la ville de Bamako. Selon D1, « Les accidents sont surtout dus à la non-maitrise du Code de la route l’excès de vitesse, le mauvais état des infrastructures routières, la surcharge et le problème de signalisation. D4 met l’accent sur l’incivisme des citoyens. « Moi je pense que les accidents sont surtout causés par le non-respect de la règlementation, l’incivisme, la vétusté des routes, du matériel ainsi que le manque d’entretien routier ». Pour D6 d’autres causes existent, il s’agit de la courtoisie, l’éducation ou l’analphabétisme de certains usagers de la route « À l’instar de la congestion, les accidents sont causés par l’incivisme, la courtoisie, l’éducation, les stupéfiants l’analphabétisme, etc. ». Les enquêtes de terrain nous ont permis de dégager les effets socio – psychologiques des accidents de la route. 4.3. Effets sociopsychologiques des accidents Les transports sont nécessaires certes pour la mobilité urbaine, mais les transports causent souvent des accidents ayant des effets socio – psychologiques sur les populations. Ces effets sont multiples et s’apprécient différemment suivant les enquêtés. Le tableau 3 récapitule les opinions des personnes enquêtées. Tableau 3 : opinions des enquêtés sur les effets socio – psychologiques des accidents du transport routier dans la ville de Bamako (en %) Propositions Isolement des personnes accidentées Dépendance économique des accidentés Chute des revenus des accidentés Pertes en vie humaine Pertes matérielles et économiques Invalidité des accidentés Maladies Invalidité et perte de membre Traumatisme dû aux accidents Accidents corporels Oui Non 56,52 86,96 82,6 100 91,3 78,26 86,96 100 95,65 91,3 RAS 39,13 13,04 8,7 4,35 4,35 13,04 13,04 4,35 8,7 4,35 0 8,7 8,7 Source : auteurs, 2022 À travers le tableau 3, on constate que, les enquêtés se sont appesantis sur certains effets socio – psychologiques qui sont : les pertes en vie humaine 100% des personnes enquêtées, le traumatisme dû aux accidents 95,65%, les accidents corporels et l’invalidité des accidentés ayant 91,3% des enquêtés. Selon notre étude, l’isolement des personnes accidentées a un faible taux (soit 56,52%) dépassant légèrement la moyenne. Les témoignages des personnes interviewées illustrent aussi les effets ci-dessus. Selon D3, « Les effets sociopsychologiques des accidents sont : les cortèges de mariage, l’occupation illicite des chaussés par certains commerçants, l’incivisme, les problèmes familiaux, professionnels, l’isolement dû à la chute des revenus ». Si ce dernier met l’accent sur les cortèges de mariages ou d’autres événements, D5, D7 et D2 se focalisent sur divers effets. D5 témoigne « je pense que les accidents ont des effets sociopsychologique qui sont : les pertes en vie humaine, les maladies, la dépendance économique, 164 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako le traumatisme, le handicap, les troubles psychologiques ; l’isolement, etc. ». De même ce discours est appuyé par celui de D7, « je pense que les deux phénomènes qui sont, les accidents et la congestion ont des effets sociaux et psychologiques comme : le handicap, les pertes économiques, le traumatisme, la méfiance des voyageurs », quant à D2, il met l’accent sur la souffrance, le handicap, le traumatisme « je pense que les accidents du transport routier ont des effets sociaux comme la perte de temps, d’argent, les cas de décès, d’handicap mental ou physique ; la souffrance. De même les accidents ont des effets psychologiques comme le traumatisme, les dépressions psychiatriques, la chute du moral des accidentés ». D9 pense que les accidents provoquent la dépendance sociale et économique, le handicap ou les blessures « je pense que les accidents provoquent des effets sociopsychologiques comme : les blessures, le handicap, ou la dépendance sociale et économique, le traumatisme, le stress, etc. ». 4.4. Propositions d’atténuation des effets d’accidents de transport routier à Bamako Figure 3 : opinions des enquêtés sur les mesures d’atténuation des effets socio – psychologiques des accidents du transport routier dans la ville de Bamako (en %) 120 100 100 91,3 91,3 91,3 86,95 80 73,91 73,91 60 Oui Non 40 RAS 17,39 17,39 8,7 4,35 8,7 8,7 Respect des Installation de Libération des Elargissement normes de matériel de bordures des des voies circulation surveillance aux routes abords des routes Intervention rapide des agents de sécurité et de santé 20 8,7 4,35 4,35 4,35 4,35 0 Prise en charge Application rapide des strict de la loi accidentés pour les fautes établies Source : auteurs, 2022 Nos enquêtes ont permis de faire des propositions de mesures d’atténuation des effets socio– psychologiques des accidents du transport routier dans la ville de Bamako. Les mesures permettant d’atténuer ces effets sont: ✓ le respect des normes de circulation : 100% de nos enquêtés pensent qu’en respectant les normes de circulation, il y aura moins d’accidents et par conséquent, moins d’effets socio– psychologiques du au transport. Le respect de ces normes passent par l’application du code de la route par les usagers et le respect des différentes formes de signalisations qu’elles soient verticales ou horizontales ; ✓ l’installation des appareils de surveillance sur les routes : nos enquêtés approuvent cette mesure à 91,3%. Ces appareils permettront d’identifier les fauteurs et les mauvais conducteurs. De même, ils servent d’appareils de dissuasion pour amener les usagers à une bonne conduite de la route ; 165 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND ✓ la libération des abords des routes et l’application stricte de la loi : ces deux mesures approuvées à 91,3% par les enquêtés permettront de fluidifier le trafic et d’assurer une utilisation correcte des routes. La réglementation prévoit des pénalités à payer pour les différentes infractions ; ✓ l’élargissement des voies : 86,95% des personnes soumises au questionnaire pensent qu’en l’élargissement des routes permettent de fluidifier le trafic. Cette mesure est certes coûteuse, car, elle nécessite des investissements lourds, mais elle permettra d’avoir plus de routes à deux ou trois voies dans le même sens ; ✓ L’intervention et la prise en charge rapide des accidentés sont des mesures proposées et qui permettent d’atténuer les effets socio – psychologiques des accidents du transport routier. ✓ Outre, les mesures citées ci-dessus, d’autres propositions ont été faites par les personnes interviewées comme témoigne D3, D5, D6 et D10. Selon D3, il faut faire une sensibilisation. D3 témoigne « Pour atténuer ces effets, je pense qu’il faut faire une large sensibilisation, une large information, une formation adéquate de tous les acteurs sur la sécurité et les interventions rapides en cas d’accidents ». D5 propose la vigilance et le port du casque par les usagers d’engins à deux roues ; « Selon moi (D5), les mesures envisageables pour atténuer les effets socio – psychologiques des accidents dans la ville de Bamako sont : la vigilance, le port du casque pour les utilisateurs d’engins à 2 roues, la maitrise du code de route ». Selon D6, on doit interdire l’utilisation des stupéfiants, « Pour réduire ces effets, je propose les mesures suivantes : la sensibilisation de la population, l’interdiction des stupéfiants dans la société, le respect des normes de circulation », alors D10 met l’accent sur le changement de comportement des usagers de la route « Pour atténuer ces effets, il faut : respecter les normes de circulation, changer les comportements des usagers, maitriser la conduite, interdire le téléphone au volant et l’utilisation de stupéfiant au volant ». Notre étude révèle que la prise en compte des mesures proposées peut permettre de réduire considérablement la congestion, les accidents et leurs effets sur le mode de vie sociale et psychologique des populations. La figure ci-dessous illustre les opinions des enquêtés concernant le taux de réduction des effets sociopsychologiques après la mise en œuvre des mesures d’atténuation. Notre étude révèle que la prise en compte des mesures proposées peut permettre de réduire considérablement la congestion, les accidents et leurs effets sur le mode de vie sociale et psychologique des populations. La figure ci-dessous illustre les opinions des enquêtés concernant le taux de réduction des effets sociopsychologiques après la mise en œuvre des mesures d’atténuation. Figure 4: opinions des enquêtés concernant le taux de réduction des effets sociopsychologiques des accidents du transport routier à Bamako après la mise en œuvre des mesures d’atténuation 0 10% 30% 60% 0 - 20% 20% - 40% 40% - 60% 60% et plus Source : auteurs, 2022 166 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako L’analyse de la figure 3 permet de conclure que les enquêtés approuvent les mesures d’atténuation proposées dans notre étude. Ceux-ci pensent que la mise en œuvre des mesures ci-dessus citées, permettra de réduire de plus de 60% les accidents, la congestion et leurs effets socio – psychologiques ; seulement 10% des enquêtés pensent que la mise en œuvre des mesures réduirait les effets d’un taux compris entre 20% et 40%. 30% des personnes enquêtées pensent que la mise en œuvre des mesures citées permettra de réduire les effets socio – psychologiques des transports de 40 à 60%. 4.5. Discussion À travers cette étude, le débat et les recherches restent ouverts autour de quatre axes qui sont : les causes de la congestion routière dans une grande agglomération, les causes des accidents de la route, les effets socio – psychologiques des accidents et de la congestion dans les grandes agglomérations et enfin les mesures d’atténuation de ces effets. Les résultats de notre étude ont montré que la congestion routière et les accidents du transport routier impactent fortement le bien-être social et psychologique des citoyens de la ville de Bamako. De même, les résultats indiquent que plus de 90% des personnes enquêtées pensent que les majeurs effets socio – psychologiques du transport routier sont : les pertes en vie humaine, 100% des enquêtés ; le traumatisme dû aux accidents, 95,65% ; les accidents corporels et l’invalidité des accidentés, 91,3% des enquêtés. Les mesures d’atténuation de ces effets sont : le respect des normes de circulation, 100% des enquêtés ; l’installation des appareils de surveillance sur les routes, 91,3% ; la libération des abords des routes et l’application stricte de la loi, 91,3%. Dans le même registre, un (Rapport BAD, 2018) sur le District de Bamako en 2015, la Direction de Régulation de la Circulation et des Transports Urbains a enregistré 156 tués (50 % 2RM, 40 % piétons), 2 018 blessés graves et 1 434 blessés légers (70 % 2RM, 25 % piétons), 2 645 accidents. Le nombre d’accidents sur les axes s’élève à 646, soit environ un quart du total enregistré et le nombre de tués est estimé autour de 25. Une étude sur les accidents de circulation dans le district sanitaire de la ville de Diéma, régions de Kayes, aborde dans le même sens que les accidents de la circulation restent aujourd'hui de véritables problèmes de santé publique en raison des graves conséquences qu'ils engendrent. Pour réduire ce problème, il est nécessaire d'avoir des routes bien adaptées et d'assurer une application rigoureuse du Code de la route. Les blessures étaient les lésions, les plus retrouvées avec 86,1% et la tête étaient la plus touchée avec 48,3% et 4,9 % de décès enregistrés (Traoré et al, 2016). Par ailleurs, selon une étude (Kafando, 2006) sur le transport urbain et la santé des populations dans la ville de Ouagadougou indique que les accidents de la circulation constituent un phénomène qui touche une part importante de la population. En effet, plus de 32% des ménages enquêtés comptent un ou plusieurs membres qui, à un moment ou à un autre de leur vie, ont eu un accident de la circulation (Kafando, 2006). Comme pour la majorité des problèmes de santé, la répartition des accidents de la circulation va de pair avec la pauvreté. Ce sont les ménages vivant dans les quartiers périphériques et appartenant aux groupes socio-économiques les moins favorisés qui sont davantage exposés aux accidents de la circulation plus que ceux vivant dans les quartiers centraux. Aujourd’hui, la congestion constitue un véritable problème dans le Distinct de Bamako. Notre étude montre qu’elle est la principale cause des pertes de temps, c’est-à-dire, des retards. Une étude sur les embouteillages dans la ville de Cotonou aborde dans le même sens. Cette étude montre que la concentration des services administratifs, l’exode rural et les moteurs à deux roues constituent les principales causes de la congestion dans la ville de Cotonou (Akiyo et al, 2016). Par ailleurs, une étude Kumar et Barrett (2008) sur les embouteillages dans les villes africaines indique que dans toutes les villes, le réseau routier est inférieur aux normes. La capacité est insuffisante, il n’y a ni bandes d’urgences ni voies de service, le revêtement est 167 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND dégradé et l’éclairage des rues réduit au minimum. Le mauvais état des routes limite la vitesse des véhicules, réduit considérablement la productivité du parc d’autobus et alourdit les coûts d’entretien des véhicules. Il favorise également l’utilisation des minibus, taxis et motocyclettes qui présentent une plus grande maniabilité que les grands autobus, mais ne sont pas aussi efficaces en tant que moyen de transport public urbain. La plupart des villes ignorent les besoins des piétons. Environ, 65 % du réseau routier manquent de trottoirs, les piétons et véhicules motorisés doivent partager le même espace. Lorsqu’ils existent, les trottoirs sont mal entretenus, comportent des caniveaux à ciel ouvert, et sont grignotés par les propriétés qu’ils bordent. Il n’y a ni passages pour piétons ni ponts, sauf dans les centres villes. À cause de la mauvaise gestion de la circulation, les accidents sont fréquents. Les piétons représentent deux tiers des victimes d’accidents mortels. 5. Conclusion L’objectif de cet article était d’analyser les effets sociopsychologiques en prenant compte de la congestion et les accidents du transport routier dans le District de Bamako. Pour y parvenir, d’abord, nous avons mis en exergue la revue des écrits sur la congestion et les accidents de circulation. Ensuite, nous avons orienté notre réflexion sur la méthodologie basée sur l’approche fixe et la revue documentaire. Par suite, les résultats ont montré que la congestion routière et les accidents du transport routier impactent fortement le bien-être social et psychologique des citoyens de la ville de Bamako. Plusieurs facteurs ont expliqué la congestion et les accidents de circulation dans la ville de Bamako. Il s’agit entre autres, du nombre élevé des véhicules, l’étroitesse des routes, le mauvais état des routes, le non-respect des normes de circulation, l’utilisation des téléphones au volant, l’alcoolisme et le stupéfiant, la mauvaise organisation du transport collectif, etc. Les congestions et les accidents de circulation ont des effets au niveau socio-psychologique et sanitaire. C’est pourquoi notre étude s’intéressait à la congestion et les accidents de circulation dans la ville de Bamako. L’étude a permis de formuler des recommandations qui vont à l’endroit des autorités et des usagers de la route à savoir : le respect des normes de circulation, la libération des bordures des routes, la reprise rapide des accidentés, l’application de la loi pour les fautes établies, etc. Les principales limites de la recherche demeurent la non-analyse des coûts de la congestion et des accidents de circulation dans le District de Bamako. L’étude a permis de cerner dans toutes ses dimensions, les effets sociopsychologiques de la congestion et des accidents de circulation dans le District de Bamako. En termes de perspective, il s’agira pour nous d’orienter notre réflexion sur les coûts de la congestion et des accidents dans le District de Bamako, en vue de cerner leurs pertes sur le plan économique. On ne peut pas généraliser les résultats de cette étude sur d’autres contextes, car l’étude a particulièrement attrait au cas de la ville de Bamako en République du Mali. 168 www.ijafame.org Yapégué BAYOGO & Lassina TOGOLA. Effets socio-psychologiques du transport routier des voyageurs : Cas de la congestion routière et des accidents du transport routier dans la ville de Bamako Références (1). (2). (3). (4). (5). (6). (7). (8). (9). (10). (11). (12). (13). (14). (15). (16). (17). (18). Akiyo, et al. (2016). L’embouteillage dans les grandes villes de l’Afrique de l’Ouest et ses problèmes : cas de Cotonou au Benin, Revue Africaine de Sociologie, volume 20 n°1, pp 87-110. Banque Mondiale. (2005).Transports urbains : étude de politique générale de la Banque Mondiale. Bayogo, Y., et al. (2022). Analyse du transport routier et ses accidents : cas des RN6 et RN7 du Mali, Revue Internationale du chercheur, « Volume 2 : Numéro 2» pp : 1-22 Cohen, S. (2006). Considération sur les courbes débit-vitesse. In Séminaire vitesse. Apport récents de la recherche en matière de vitesse. Arcueil, INRETS no 105, pp.5766. Direction Nationale de la Population du Mali. (2009). Projection et perspective de la population suivant le recensement Général de la Population et de l’Habitats du Mali. DNTTMF. (2011 - 2020) : Observatoire des Transports, Annuaire statistique des transports du Mali. Eviar, O. (2023). Les nouvelles infrastructures routières : un défi pour la durabilité des conditions de vie des populations dans le grand Abidja, Revue Internationale du Chercheur, Vol 4 : N°2, PP: 445-464. Godard, X., et Teurnier, P. (1992). Les transports urbains en Afrique à l’heure de l’ajustement : redéfinir le service public, Paris, Karthala. Kafando, Y. (2006). Transport urbain et santé des populations : le cas de Ouagadougou mémoire de Master, Cotonou, Université d’Abomey Calavi. Korpinen, L., et al. (2016). Accidents and Close Call Situations Due to Cell Phone Use While Moving, Driving, and Working. World Academy of Science, Engineering and Technology; International Journal of Humanities and Social Sciences, Vol: 10, N°6, PP: 2121- 2125. Kumar, A., et Barrett, F. (2008). Coincés dans les embouteillages : le transport urbain en Afrique, rapport Banque Mondiale et le SSATP. Mane, O. (2019). Les accidents de la circulation dans la Commune de Mbour au Sénégal : état des lieux et propositions de solutions, mémoire universitaire, université Senghor. Meite, Y. (2014). Gouvernance du transport urbain et mobilité durable dans le District d’Abidjan, thèse de Doctorat, université de Strasbourg. Moustakbal, A. (2009). L’impact de la congestion routière sur l’industrie du camionnage dans la région de Montréal, mémoire universitaire, université de Québec. Murray, C., J., L., et al. (1996). The global burden of disease : a comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020, Published by the Harvard School of Public Health on behalf of the World Health Organization and the World Bank; Distributed by Harvard University Press. OMS, (2004). Rapport mondial sur la prévention des traumatismes liés aux accidents de circulation. Paakkonen, R., et Korpinen L. (2016). Comparison of injuries and accidents globally and in Finland. World Academy of Science, Engineering and Technology; International Journal of Humanities and Social Sciences, Vol : 10, No:3; PP: 814 – 817. Rapport BAD (Banque Africaine de Développement). (2018). Etude pour l’amélioration de la sécurité routière dans le District de Bamako. 169 www.ijafame.org ISSN: 2658-8455 Volume 4, Issue 4-2 (2023), pp. 156-170. © Authors: CC BY-NC-ND (19). (20). (21). (22). (23). (24). (25). (26). (27). (28). (29). Rapport CETE Méditerranée. (2009).Étude de la capacité et de la congestion sur le réseau MARIUS. Rapport SSATP, (2014). La gestion de la sécurité routière en Afrique. Sane, P. (2014). La mobilité urbaine à Dakar, un frein à l’émergence, Dakar, Enquête. Sangala, S., E. (2017). Problématique de l’encombrement du réseau routier à Dakar : cas de l’avenue Cheikh Anta DIOP (Rond-point Cheikh BETHIO au rond-point UCAD), mémoire de Master, Dakar, Université Cheikh Anta DIOP. Shahram, J., et al. (2007). Adaptive Car Safety System. World Academy of Science, Engineering and Technology International Journal of Computer and Information Engineering, Vol:1, No:1, PP: 13 – 31. Sitihawa, H., et Norashikin M. (2018). Technical Determinant of Road Accident: A Systematic Review. International Journal of Engineering & Technology, 7 (3.36) PP: 34-39. Terzano, K. (2013). Bicycling safety and distracted behavior in The Hague, the Netherlands. Accident. Analysis. Prevention, vol. 57, pp. 87–90. Traore, et al. (2016). Accidents de la circulation routière dans le district sanitaire de Diéma, Mali Médical, Vol. 31 Numéro 2, p15-19. Waard, F., et al. (2015): More screen operation than calling: the results of observing cyclists’ behavior while using mobile phones. Accident Analysis and Prevention, vol. 76, pp. 42–48 Wiel, M. (2004).Ville et mobilité : un couple infernal, Paris, les éditions de l’aube. Yuki, K., et al. (2011). An Effective Traffic Control for both Real-time Bursts and Reliable Bursts in OBS Networks. World Academy of Science, Engineering and Technology International Journal of Electronic Engineering, Vol : 5, N°11, PP : 814 817. 170 www.ijafame.org 

https://openalex.org/W3109752955

https://research.chalmers.se/publication/521682/file/521682_Fulltext.pdf

English

UV-to-IR Absorption of Molecularly p-Doped Polythiophenes with Alkyl and Oligoether Side Chains: Experiment and Interpretation Based on Density Functional Theory

˜The œjournal of physical chemistry. B

cc-by

UV-to-IR Absorption of Molecularly p-Doped Polythiophenes with Alkyl and Oligoether Side Chains: Experiment and Interpretation Based on Downloaded from: https://research.chalmers.se, 2024-10-24 05:44 UTC Citation for the original published paper (version of record): Sahalianov, I., Hynynen, J., Barlow, S. et al (2020). UV-to-IR Absorption of Molecularly p-Doped Polythiophenes with Alkyl and Oligoether Side Chains: Experiment and Interpretation Based on Density Functional Theory. Journal of Physical Chemistry B, 124(49): 11280-11293. http://dx.doi.org/10.1021/acs.jpcb.0c08757 N.B. When citing this work, cite the original published paper. research.chalmers.se offers the possibility of retrieving research publications produced at Chalmers University of Technology. It covers all kind of research output: articles, dissertations, conference papers, reports etc. since 2004. research.chalmers.se is administrated and maintained by Chalmers Library (article starts on next page) This is an open access article published under a Creative Commons Attribution (CC-BY) License, which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited. This is an open access article published under a Creative Commons Attribution (CC-BY) License, which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited. Article pubs.acs.org/JPCB UV-to-IR Absorption of Molecularly p‑Doped Polythiophenes with Alkyl and Oligoether Side Chains: Experiment and Interpretation Based on Density Functional Theory Ihor Sahalianov, Jonna Hynynen, Stephen Barlow, Seth R. Marder, Christian Müller and Igor Zozoulenko* Read Online ABSTRACT: The UV-to-IR transitions in p-doped poly(3-hexylthiophene) (P3HT) with alkyl side chains and polar polythiophene with tetraethylene glycol side chains are studied experimentally by means of the absorption spectroscopy and computationally using density functional theory (DFT) and tight-binding DFT. The evolution of electronic structure is calculated as the doping level is varied, while the roles of dopant ions, chain twisting, and π−π stacking are also considered, each of these having the effect of broadening the absorption peaks while not significantly changing their positions. The calculated spectra are found to be in good agreement with experimental spectra obtained for the polymers doped with a molybdenum dithiolene complex. As in other DFT studies of doped conjugated polymers, the electronic structure and assignment of optical transitions that emerge are qualitatively different from those obtained through earlier “traditional” approaches. In particular, the two prominent bands seen for the p-doped materials are present for both polarons and bipolarons/polaron pairs. The lowest energy of these transitions is due to excitation from the valence band to orbitals located in the gap between the bands. The higher-energy band is a superposition of excitation from the val spin-resolved orbitals in the gap and an excitation between bands. © 2020 American Chemical Society Received: September 25, 2020 Revised: November 4, 2020 Published: November 25, 2020 The Journal of Physical Chemistry B At the same time, a peak at intermediate energies ≈1.3−1.7 eV (700−900 nm) accompanied by a peak at low energies ≈0.3−0.7 eV (>1500 nm) develop. As the oxidation level increases further, the absorbance of the intermediate-energy peak reaches a maximum and then decreases, while that of the low-energy peak increases and becomes dominant in the spectra. views,48−51 is still dominated by the early traditional picture. Hence, we argue that DFT studies that elucidate the electronic structure of conjugated polymers are vital to provide more accurate descriptions of these transitions, despite the limitation of various levels of DFT (for example, the tendency to over- delocalize electrons). As briefly reviewed above, many DFT-based studies of polythiophenes and related polymers addressing various aspects of the electronic structure and the nature of charge carriers have been reported during past decades. However, a number of crucial questions remain unanswered. One such question is the effect of doping on the electronic structure and optical transitions. Most DFT-based studies focusing on the absorption spectra for the systems at hand consider only the isolated charged oligomers and neglect the influence of compensating charge associated with the anions that results from the dopants and which modifies the electrostatic potential around the positively charged π-systems. Providing such an understanding of their effect on the electronic structure and corresponding optical transitions represents one of the central aims of this study. In the literature, one can find various interpretations of the nature of these peaks in the electronic absorption spectra and the explanation of their evolution during oxidation and reduction of thin films. An early interpretation, which is still often invoked in the current literature, was developed in the 1980s and was based on various methods, including Su− Schrieffer−Heeger Hamiltonian,27 valence effective Hamilto- nian,28 various Hartree−Fock methods,17,29,30 Holstein Ham- iltonian,31 and others. Following Heimel,32 we will refer to these approaches and corresponding interpretations as “tradi- tional” ones. The Journal of Physical Chemistry B In recent years, density functional theory (DFT) and time-dependent (TD)-DFT methods have been widely utilized to study various aspects of electronic structure and optical absorption in polythiophenes and related poly- mers.22,32−44 These methods have also been applied to n- doped polymers32,45 and p-type/n-type blends.46 It has been realized that DFT-based approaches lead to a qualitatively different electronic structure and provide a distinct inter- pretation of the absorption peaks and the evolution of the spectra as compared to the above-mentioned traditional picture (see Section III.I for discussion of the differences in the traditional and DFT descriptions).22,32,36,41 The essential reason for these differences is that the traditional model fails to capture some very basic and essential physics; specifically, it ignores the Coulomb interactions leading to the lifting of the spin degeneracy that DFT calculations take into account effectively. These effects have been discussed more recently in papers by Png et al.,43 Winkler et al.,47 and Heimel32 in a very accessible manner, based upon a combination of experimental studies aided by DFT studies. y Other questions that are not clarified yet within the DFT approach concern the effect of a realistic nanostructure on the electronic absorption spectra. In particular, polythiophenes tend to form π−π-stacked aggregates and in some cases crystallites composed of several chain segments.25,31,44,48,52 Therefore, the effect of π−π stacking on the absorption spectra needs to be quantified. Further, at finite temperatures, the chain segments and crystallites that constitute the solid-state nanostructure are in constant movement, leading to twisting of main-chain bonds and changes in torsion angles and movements of the side chains. While a standard DFT calculation is suitable for accurate simulations of individual polymer chains, it is too computationally expensive to simulate π−π stacks when one needs to average over many dopant configurations or temperature-induced motion of oligomers with long side chains. In the present paper, we utilize a molecular dynamics (MD) simulation based on tight-binding DFT (DFTB) and provide a systematic treatment of the above-mentioned effects. Other questions that are not clarified yet within the DFT approach concern the effect of a realistic nanostructure on the electronic absorption spectra. I. INTRODUCTION favors the hole injection.13 Molecular doping of polythio- phenes (PTs) and related polymers has been a subject of intensive research (see, e.g., refs 13 and 14 for a review). Molecular p-doping of thiophene-based polymers with differ- ent side chains such as poly(3-hexylthiophene) (P3HT), with aliphatic side chains,14,15 and poly(13,13′-([2,2′:5′,2″-terthio- phene]-3,3′-diylbis(oxy))bis(2,5,8,11-tetraoxatridecane)) P- (g42T-T)15,16 with tetraethylene glycol side chains have been studied (see Figure 1 for chemical structure). The studies referred to above focused on electronic properties including conductivity and Seebeck coefficient. Also, the optical spectra of these materials were reported. It should be noted, however, that a detailed theoretical understanding of the effect of doping and side chains on the optical absorption by these polymers is not complete. Molecular doping and side-chain engineering of conducting polymers represent powerful tools to enhance and tune the electronic and optical properties of polymer-based devices. Doping leads to electron transfer between dopants and the polymer backbone, thus increasing the charge carrier concentration. This, in turn, can strongly affect the mobility and the polymer morphology due to changes in the mechanical stiffness/planarity of backbones and the enhancement of backbone−dopant interactions.1−4 The utilization of oli- goether side chains greatly improves polymer solubility in most organic solvents.5 Because of their hydrophilicity, oligoether side chains enhance the intake of hydrated ions and water swelling.6−8 The interplay between molecular doping and the character of the side chains makes it possible in some cases to alter morphological and electrical properties of polymer thin films, enhancing their crystalline order, thermal stability, conductivity, Seebeck coefficients, and other important material parameters. These changes, can in turn, lead to improved performance of polymer-based devices such as organic solar cells,9 field-effect transistors,10 electrochemical transistors,11 actuators, and robotic devices.12 Optical absorption spectroscopy of polythiophenes and related conjugated polymers with a thiophene-like backbone Received: September 25, 2020 Revised: November 4, 2020 Published: November 25, 2020 0 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 Received: September 25, 2020 Revised: November 4, 2020 Published: November 25, 2020 Most of the conducting polymers are p-doped because the Fermi level in the electrodes on the metal/polymer interface is typically aligned with the valence band of polymer, which © 2020 American Chemical Society https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 © 2020 American Chemical Society 020 American Chemical Society 11280 11280 11280 pubs.acs.org/JPCB Article The Journal of Physical Chemistry B Article pubs.acs.org/JPCB Figure 1. Chemical structure of (a) P3HT, (b) P(g42T-T), and (c) Mo(tfd-COCF3)3. Polymers are visualized with two oligomers consisting of N = 18 thiophene units each. Figure 1. Chemical structure of (a) P3HT, (b) P(g42T-T), and (c) Mo(tfd-COCF3)3. Polymers are visualized with two oligomers consisting of N 18 thi h it h Figure 1. Chemical structure of (a) P3HT, (b) P(g42T-T), and (c) Mo(tfd-COCF3)3. Polymers are visualized with two oligomers consisting of N = 18 thiophene units each. such as poly(3,4-ethylenedioxythiophene) (PEDOT) has received tremendous attention during past decades.16−26 Section III.I), they remain largely ignored by many experimental studies on polythiophenes and related polymers. The current literature,19,20,24−26 including many recent re- views,48−51 is still dominated by the early traditional picture. Hence, we argue that DFT studies that elucidate the electronic structure of conjugated polymers are vital to provide more accurate descriptions of these transitions, despite the limitation of various levels of DFT (for example, the tendency to over- delocalize electrons). Section III.I), they remain largely ignored by many experimental studies on polythiophenes and related polymers. The current literature,19,20,24−26 including many recent re- 48 51 ll d d b h l d l such as poly(3,4-ethylenedioxythiophene) (PEDOT) has received tremendous attention during past decades.16−26 When the oxidation level of the polymer is varied, the optical absorption spectra evolve and show features that are similar for all the above-mentioned polymers. Namely, undoped polymer films show absorption at ≈2.0−2.3 eV (450−600 nm) due to the electronic transition from the valence to the conduction band. The absorption peaks of the neat polymer gradually disappear as the doping level increases. At the same time, a peak at intermediate energies ≈1.3−1.7 eV (700−900 nm) accompanied by a peak at low energies ≈0.3−0.7 eV (>1500 nm) develop. As the oxidation level increases further, the absorbance of the intermediate-energy peak reaches a maximum and then decreases, while that of the low-energy peak increases and becomes dominant in the spectra. When the oxidation level of the polymer is varied, the optical absorption spectra evolve and show features that are similar for all the above-mentioned polymers. Namely, undoped polymer films show absorption at ≈2.0−2.3 eV (450−600 nm) due to the electronic transition from the valence to the conduction band. The absorption peaks of the neat polymer gradually disappear as the doping level increases. https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B Through the utilization of the DFT-based approaches, we are able to systematically study and quantify the influence of the above-mentioned key factors on the absorption spectra, such as the effect of doping, the presence of the counterions, the effect of the π−π stacking of the backbones, the effect of the side chains, and the effect of the bending and twisting of the chains. In addition, we attempt to provide a clear guideline for the use of correct language for the assignment of polaron, bipolaron, and polaron pair, as they apply to the electronic state and optical spectra of ionized π systems. We also point out that, despite the fact that the terminology of polaron and bipolaron (as well as polaron pairs) has been discussed in the literature accurately for several decades, the descriptions using the earlier models and inaccurate assignment of spectral features based upon these models continue to be widely used, propagating misconception and confusion. We hope that this paper will serve to provide additional clarity to these issues and encourage the organic electronic community working on doping to accept and adopt the more modern and physically correct picture of the states and the transition that occur upon doping. II.II.II. Density Functional Tight-Binding Simulations. Density functional tight binding58 allows the simulation of larger systems compared to DFT. To describe the effects of π−π stacking and temperature-induced motion of oligomers with long side chains on the absorption spectra of P3HT, we performed molecular dynamics (MD) DFTB simulations. Studied systems consisted of a single oligomer, two π−π- stacked oligomers, and three π−π-stacked oligomers without any explicit or implicit solvents. The initial optimization of the oligomer geometry was performed with the DFT. Therefore, the molecular dynamics simulations were performed without initial equilibration. Atomic motions were calculated with velocity Verlet dynamics under the Nose−Hoover thermostat with a targeted temperature of 300 K, a time step of 1 fs, and a coupling strength of 1500 cm−1. Fermi electron filling at a temperature of 300 K was used to speed up convergence, and the adaptive filling was not considered during MD. A self- consistent charge procedure was performed with tolerance 1 × 10−5. Individual oligomers were simulated for up to 20 ps. Crystallites consisting of two or three stacked oligomers were simulated for 1 ps. The DFTB studies were performed for isolated polymer chains, without periodic boundary conditions. II. METHODS II.I. Experimental Methods. II.I.I. Materials. P3HT was purchased from Ossila Ltd. (number-average molecular weight Mn ∼29 kg mol−1; polydispersity index (PDI) ∼2.2; regioregularity ∼96%), and P(g42T-T) (Mn ∼24 kg mol−1; PDI ∼3.3) was synthesized according to ref 15. Mo(tfd- COCF3)3 was synthesized according to ref 53. The solvents chlorobenzene, o-dichlorobenzene, chloroform, and acetoni- trile with purity >99% were purchased from Sigma-Aldrich. g II.I.II. Sample Preparation and Doping. Thin films of P3HT and P(g42T-T) were spin-coated onto BaF2 substrates from 1:1 w/w chlorobenzene/o-dichlorobenzene at 60 °C and chloroform at room temperature, respectively (7.5 g L−1). Sequential doping with Mo(tfd-COCF3)3 was done by drop- casting 2.5 g L−1 solutions of the dopant in acetonitrile onto polymer films and spinning offthe remaining solution after 3 min. II.I.III. UV−Vis−Near-Infrared (NIR) and Infrared Spectros- copy. Absorption measurements were performed in trans- mission with a PerkinElmer Lambda 900 spectrophotometer in the UV−vis−NIR region and a PerkinElmer FTIR Spectrom- eter “Frontier” in the infrared region. The resulting spectra were stitched together. y g g ( ) II.II.III. Model for Polymer Chains. Both P3HT and P(g42T- T) have rather long polymer chains, where the number of thiophene rings in the chains is ca. 100 or more.15 In our calculation, we choose the number of thiophene rings N = 18 (≈6.9 nm) for the DFT calculations of oligomers and N = 12 for the DFTB calculations of π−π-stacked chains and chains with long side chains. This is justified because the calculated absorption is practically unchanged when the oligomer length increases above N ≈12 (see Figure S6). It should also be noted that much shorter chain lengths in calculations are justified because the coherence length is not expected to extend over the entire chain but rather over the distance comparable to the persistence length, which, for the system at hand, is around 3 nm.66−68 II.II. Theoretical Methods. During the simulations, we used two methods: single oligomers, without and with counterions, were simulated using DFT and time-dependent DFT (TD-DFT). Larger systems, including stacked polymer chains, as well as oligomers with side chains during temperature-induced motion, were simulated using tight- binding density functional theory (DFTB) and TD-DFTB. II.II.I. Density Functional Theory Simulations. Polymer chains with counterions were simulated with DFT, imple- mented in the Gaussian16 software package.54 For the calculation of the optical transitions (vertical excitations), we utilized the TD-DFT approach. The Journal of Physical Chemistry B Calculations were conducted with the DFTB+ software package58 on the SCC-DFTB3 level.59 3ob-3-1 parameter sets60−62 were utilized during all simulations, as well as the D3 dispersion correction63 with the Becke−Johnson dumping.62,64 After obtaining geometries of the polymer chains, UV-to-IR absorption spectra were calculated with the time-dependent density functional tight-binding (TD-DFTB) method.65 The Journal of Physical Chemistry B The Journal of Physical Chemistry B Article in the Supporting Information. While we find that both functionals provide a similar description of the systems, B3LYP gives a better agreement with the experimental data. Hence, all of the DFT calculations presented in the main text correspond to the functional B3LYP. As a basis set, we chose 6-31G(d). This basis set proves itself as a reasonable balance between accuracy and cost-efficiency in simulations of electronic properties and UV−vis absorption spectra.41 Diffuse functions were omitted without the loss of accuracy for both self-charged and counterion-charged oligomers (see Figure S3). Self- consistent calculations are performed in 50 iterations steps (see Figure S4 for details). The dispersion correction is not included as its effect on the absorption spectra for the case of a single oligomer with counterions is negligible (see Figure S5). For the visualization of the absorption spectra, the oscillator strengths were fitted by a Lorentzian with half-width at half- maximum, hwhm = 0.15 eV. using DFT and molecular dynamics simulations utilizing tight- binding DFT. Through the utilization of the DFT-based approaches, we are able to systematically study and quantify the influence of the above-mentioned key factors on the absorption spectra, such as the effect of doping, the presence of the counterions, the effect of the π−π stacking of the backbones, the effect of the side chains, and the effect of the bending and twisting of the chains. In addition, we attempt to provide a clear guideline for the use of correct language for the assignment of polaron, bipolaron, and polaron pair, as they apply to the electronic state and optical spectra of ionized π systems. We also point out that, despite the fact that the terminology of polaron and bipolaron (as well as polaron pairs) has been discussed in the literature accurately for several decades, the descriptions using the earlier models and inaccurate assignment of spectral features based upon these models continue to be widely used, propagating misconception and confusion. We hope that this paper will serve to provide additional clarity to these issues and encourage the organic electronic community working on doping to accept and adopt the more modern and physically correct picture of the states and the transition that occur upon doping. using DFT and molecular dynamics simulations utilizing tight- binding DFT. The Journal of Physical Chemistry B In particular, polythiophenes tend to form π−π-stacked aggregates and in some cases crystallites composed of several chain segments.25,31,44,48,52 Here, we report on a combined experimental and theoretical study of absorption spectroscopy of doped P3HT and P(g42T- T) and interpret the observed spectra based on the electronic structure and corresponding electronic transitions calculated In spite of the qualitatively and quantitatively different picture that modern DFT-based approaches provide (see 11281 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 pubs.acs.org/JPCB The Journal of Physical Chemistry B III. RESULTS AND DISCUSSION polymer blends. II.II.IV. Simulation of Dopants (Counterions). Charging of P3HT and P(g42T-T) was conducted in two ways: with the addition of positive charges to the chain without counterions and with the addition of charges and further balancing of the positively charged polymer chains by negatively charged counterions. Counterions are initially placed at an average distance of 4 Å from the thiophene rings and at least 8 Å apart from neighboring counterions, as was suggested by molecular dynamics studies of related systems.71,72 The geometry of Cl3 − was obtained after the DFT optimization. More details concerning the placement of the counterions are given in the captions of Figure S8. During geometrical optimization of the chains, the ion positions were frozen. This was done because a single oligomer represents only a small part of a realistic thin film. Positions of counterions in the film are affected by all oligomers around, which are not present in the current model. From this point of view, performing a full geometry optimization of an individual oligomer with nonfrozen counterions would completely disregard the effect of the surroundings. Note that a similar placement of counterions was recently used for DFT calculations of the intrinsic capacitance of conducting polymers.73 For each oxidation level, five different counterion positions were calculated, as shown in Figure S8. Due to the complexity of the system, there are many factors affecting the electronic structure and UV−vis and IR absorption spectra of P3HT and P(g42T-T) thin films. We first emphasize the differences in traditional and DFT pictures of the origin of the optical transitions of oxidized polythiophenes and define the terms that we will use in the following sections. Next, we examine a small-molecule oligomeric model system in more detail. The impact of counterions on the doped polymers is then investigated using DFT and TD-DFT simulations. As a next step, DFTB-based molecular dynamics simulations and TD-DFTB simulations of P3HT are conducted to investigate the influence of torsional deformation of backbones and the side-chain orientation. Further, the impact of π−π stacking is studied using TD- DFTB. Finally, experimental data are analyzed based on the calculated spectra. p III.I. Differences in Traditional and DFT Descriptions. The term “polaron” is defined here to indicate a singly charged (Q = +1) charge carrier that has doublet spin multiplicity. II. METHODS The DFT calculations in the present study were performed using two different popular functionals, the global hybrid exchange−correlation functional B3LYP55,56 and the range-separated hybrid functional ωB97XD.57 A comparison of these two functionals for the present system is discussed in Section S1 (Figures S1 and S2) P3HT and P(g42T-T) oligomers consist of thiophene rings (backbone) decorated with long side chains. Including the side chains explicitly into the DFT calculations represents a P3HT and P(g42T-T) oligomers consist of thiophene rings (backbone) decorated with long side chains. Including the side chains explicitly into the DFT calculations represents a 11282 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 pubs.acs.org/JPCB The Journal of Physical Chemistry B Article Article absorption spectra overlap with those of the polymers. This is illustrated in Figure S9 showing the absorption spectra of Cl3 −. The main effect of the counterions is that they give rise to an electrostatic potential that affects localization of charge carriers in the polymeric chains. Note also that the calculated positions of the absorption peaks strongly depend on the choice of the functional (see Section S1), and therefore, DFT-based predictions should be considered to be of semiquantitative character. Hence, the exact type of counterions is not expected to be important for the electronic and optical properties of oligomers, and our choice of counterions is motivated by the computational efficiency. Note that we also tried other counterions often used in conductive polymers, including Cl−, NO3 −, tosylate, and F4TCNQ−(2,3,5,6-tetrafluoro- 7,7,8,8-tetracyanoquinodimethane). We find that for some of these counterions (NO3 −, tosylate), the self-interaction errors of DFT are too high (leading to the charge Q ≈−0.5 to −0.85 e instead of the expected −1 e). Note that the delocalization errors caused by incomplete self-interaction corrections represent one of the known issues of the DFT leading to the inaccuracies in the description of the charge transfer.75,76 Also, the calculations Cl−and F4TCNQ−were plagued by convergency problems. On the other hand, Cl3 −demonstrated a good convergence and an acceptable level of self-interaction error, typically leading to the charge on the counterions Q ≈ −0.95 e. An average charge on each of the counterions was carefully controlled during the geometrical optimization. If the absolute value of the counterion charge was lower than 90% of the elementary charge, the simulation was discarded as unsuccessful. formidable task in terms of the computational time. II. METHODS At the same time, electronic and optical properties of P3HT and P(g42T-T) are primarily determined by the π-electrons residing on the carbon and sulfur atoms in the polymer backbone and are much less affected by the side chains. We performed calculations for oligomers where we subsequently cut the side chain from the full length to no side chains at all (see Figure S7). We find that the side chains with only two carbon atoms give practically the same absorption spectra as the full structures. Hence, in all DFT calculations of the oligomers reported in this paper, the side chains in P3HT and P(g42T-T) are represented as CH2CH3 and OCH3 groups, respectively. The DFT calculations of PT without side chains were performed in implicit dichloromethane solvent with the CPCM model69 with a default set of parameters in Gaussian16. The van der Waals radii of the atoms were chosen according to the universal force field. Cavities were simulated as scaled van der Waals surfaces with scaling parameter 1.1. Charged oligomers can be in several different spin configurations. For instance, for the total charge Q = +2, the oligomer can be in two configurations, singlet (S = 0) or triplet (S = 1). We calculated and compared the ground-state energies of all possible spin configurations for charged oligomers in the range from neutral polymer Q = 0 to +4. All polymer geometries were optimized. For singlet states, we used restricted (closed-shell) spin calculations. For all other cases, we performed unrestricted (open-shell) calculations. A similar approach to the calculation of the ground-state energy for different spin configurations has previously been used for p- type41 and n-type conjugated polymers70 and n-type/p-type polymer blends.46 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B Figure 2. Electronic structure of polythiophenes according to (a, b) traditional approaches (refs 17, 18, 27, 29, and 30) and (c−e) DFT calculations22,32,36,41 for different oxidation levels (Q indicates the charge of a chain, and S indicates the spin). The small blue arrows (spin-up and spin-down states) indicate occupied electronic states. The blue and red lines are occupied and unoccupied states in conduction and valence bands, respectively. (The conduction and valence bands are marked by yellow and blue background colors.) The dashed red lines indicate polaronic/ bipolaronic states in the gap. The black solid lines identify possible electronic transitions. The Journal of Physical Chemistry B pubs.acs.org/JPCB Article Figure 2. Electronic structure of polythiophenes according to (a, b) traditional approaches (refs 17, 18, 27, 29, and 30) and (c−e) DFT calculations22,32,36,41 for different oxidation levels (Q indicates the charge of a chain, and S indicates the spin). The small blue arrows (spin-up and spin-down states) indicate occupied electronic states. The blue and red lines are occupied and unoccupied states in conduction and valence bands, respectively. (The conduction and valence bands are marked by yellow and blue background colors.) The dashed red lines indicate polaronic/ bipolaronic states in the gap. The black solid lines identify possible electronic transitions. carriers, there are two empty spin orbitals in the gap. For higher oxidation levels, charge carriers are polarons and/or bipolarons and are also associated with empty levels in the gap (see Figure 2e). The TD-DFT-based interpretation of the absorption peaks also differs qualitatively from the traditional approaches described above. Namely, the intermediate-energy peak (“polaronic peak”) is assigned to excitation between the valence and conduction bands or a superposition of interband and valence-band-to-empty-level excitation; in contrast to the traditional assignment, it is present not only for polarons (Q = +1) but also for bipolarons and polaron pairs (Q = +2). The Journal of Physical Chemistry B The low-energy peak is attributed to excitation from the valence band to the empty polaron/bipolaron (spin)orbital(s) in the gap; again in contrast to the traditional assignment, this transition is also present for all oxidation levels (i.e., polarons (Q = +1), bipolarons/polaron pair (Q = +2), etc.).41 case on an infinite periodic structure, it de facto became common in the literature on the electronic and optical properties of conducting polymers already from the early days.79 In the present study, we will also follow this nomenclature as it captures and well reflects the essential features of the energy structure even for oligomers of a finite length. As discussed in Section I, p-doped polythiophenes exhibit “intermediate-” and “low-energy” absorption peaks, with the latter being more dominant at high doping levels. In the traditional model, both the polaron and bipolaron are characterized by the presence of two spin-degenerate levels in the gap between valence and conduction bands, the lowest of which is singly occupied for the polaron and empty for the bipolaron.17,18,27,29,30 The intermediate-energy peak is attrib- uted to a combination of excitations from the half-filled “polaronic” energy levels in the gap to the conduction band and from half-filled to empty “polaronic levels” (see Figure 2a). The low-energy peak is attributed to transitions from the valence band to the empty “bipolaronic” energy levels in the gap, as shown in Figure 2b. In other words, the intermediate- energy peak is attributed solely to a doublet of singly charged carriers, i.e., polarons, whereas the low-energy peak is attributed solely to singlet dicationic species, i.e., bipolarons; accordingly, these two peaks are often referred to as polaronic and bipolaronic peaks, respectively.20 III.II. UV-to-IR Absorption in a Reference System: PT N = 6 Oligomers. A quantitative agreement between theory and experiment is a prerequisite for the validation of the theoretical approach. However, because of the complexity of the present system of P3HT and P(g42T-T) films, a possible discrepancy between theory and experiment may reflect shortcomings of the DFT-based approach utilized and/or can be related to a number of factors not included in the model, such as inhomogeneity of thin film doping, chain bending, effects of dielectric environment and water, presence of adsorbed atoms, short-chain segments, and surface effects. Therefore, to validate the method and demonstrate its predictive power, we first focus on a well-defined reference system, in which such experimental uncertainties should not be present. III. RESULTS AND DISCUSSION In the literature, one can find different definitions of a bipolaron (see, e.g., ref 41); however, here we follow refs 34, 77, and 78 in defining a bipolaron as a doubly charged (Q = +2) charge carrier in which two electrons of opposite spins occupy the same spin-degenerate energy level resulting in a singlet spin multiplicity. It is noteworthy that in the literature, in particular in organic chemistry, one often uses notations “radical cations” and “dications” for, respectively, polarons and bipolarons. Also, in physical and quantum chemistry literature, the notion of valence and conduction bands is widely used for the occupied and unoccupied manifolds of states in the oligomers. While the notion of the “band” is, strictly speaking, justified only for the In the experimental films, Mo(tfd-COCF3)3 complexes were used as dopants and thus the counterions are Mo(tfd- COCF3)3 −. These are large molecules, and their utilization in the calculations requires significant computational efforts. Hence, in the calculations, we instead used ions Cl3 −that are much smaller. Note that in most cases (an exception has been recently reported74) the orbitals on counterions do not appreciably mix with those of polymer chains nor their 11283 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B It should be stressed that the spin multiplicity of the ground state and the structure of the energy levels of P3HT shown in Figure 4a,b are the same as those for PEDOT oligomers with the same number of repeat units (N = 18).41 This is not surprising because PEDOT is also composed of thiophene-type rings accommodating π-electrons, primarily responsible for the electronic structure and the absorption spectrum. III.III. Effects of Counterions on the Electronic Structure and Optical Absorption of P3HT and P- (g42T-T) Oligomers. III.III.I. Electronic Structure. We start our discussion with a reference case of N = 18 oligomers without counterions. In this subsection, we present results for P3HT only because P(g42T-T) exhibits practically the same features. Figure 4a shows the ground-state spin multiplicity calculated for each oxidation level (Q), while Figure 4b shows the corresponding energy diagrams. For the polaron (Q = +1), the spin degeneracy is lifted and the ground state is a doublet. Figure 3. UV-to-IR absorption spectra (to the left) and electronic structure (to the right) of polythiophene oligomers consisting of six monomers (shown in the insets) with the oxidation levels (a) Q = +1 (polaron) and (b) Q = +2 (bipolaron). Experimental data were taken from van Haare et al.17 and converted from absorbance to absorption. The blue and red lines are occupied and unoccupied states in conduction and valence bands, respectively. (The conduction and valence bands are marked by yellow and blue background colors; the small blue arrows (spin-up and spin-down states) indicate occupied electronic states in the valence band.) The dashed red lines indicate (unoccupied) polaronic/bipolaronic states in the gap. The black solid lines identify dominant transitions contributing to the peaks TP/B and TC in the absorption spectra. p p Let us now discuss the effect of counterions. For each of the charged oligomers (Q = +1, +2, +3, +4), we calculated the electronic structure for five representative counterion positions relative to the oligomer (see Figure S8 depicting these spatial realizations). For each of these spatial realizations, we calculated energies of the ground states for different spin multiplicities M = 2S +1. The red dots in Figure 4c show the spins of the ground state with the lowest energy for each of these configurations. A comparison with the case without counterions (cf. The Journal of Physical Chemistry B Figure 3. UV-to-IR absorption spectra (to the left) and electronic structure (to the right) of polythiophene oligomers consisting of six monomers (shown in the insets) with the oxidation levels (a) Q = +1 (polaron) and (b) Q = +2 (bipolaron). Experimental data were taken from van Haare et al.17 and converted from absorbance to absorption. The blue and red lines are occupied and unoccupied states in conduction and valence bands, respectively. (The conduction and valence bands are marked by yellow and blue background colors; the small blue arrows (spin-up and spin-down states) indicate occupied electronic states in the valence band.) The dashed red lines indicate (unoccupied) polaronic/bipolaronic states in the gap. The black solid lines identify dominant transitions contributing to the peaks TP/B and TC in the absorption spectra. The Journal of Physical Chemistry B Article pubs.acs.org/JPCB Article provides a validation of the utilized theoretical approach based on DFT and TD-DFT calculations. III.III. Effects of Counterions on the Electronic Structure and Optical Absorption of P3HT and P- (g42T-T) Oligomers. III.III.I. Electronic Structure. We start our discussion with a reference case of N = 18 oligomers without counterions. In this subsection, we present results for P3HT only because P(g42T-T) exhibits practically the same features. Figure 4a shows the ground-state spin multiplicity calculated for each oxidation level (Q), while Figure 4b shows the corresponding energy diagrams. For the polaron (Q = +1), the spin degeneracy is lifted and the ground state is a doublet. We stress that in the DFT approach the spin of the polaron (as well as of all other S > 1 charge carriers) originates from the imbalance between the number of up and down spin electrons in the valence band and not from the empty levels in the gap (which cannot have spins because they correspond to the absence of electrons). The ground state for the case Q = +2 is a triplet and thus corresponds to a polaron pair. The bipolaron (i.e., the singlet state with Q = +2) is energetically higher than the polaron pair. The Q = +3 and +4 systems are, respectively, doublet and triplet, and, therefore, they do not have any bipolaron levels (as defined above, a bipolaron is a spin- degenerate charge carrier where two electrons of opposite spins occupy the same spin-degenerate energy level). The Journal of Physical Chemistry B Figure 4a) shows that in most cases the spin multiplicity remains the same, i.e., doublets for Q = +1 and +3 and triplets for Q = +2 and +4. The only exceptions were one counterion position for Q = +2 and one for Q = +4, which both resulted in the singlet ground states (see Figure 4h,i and related discussion in the text below). The calculations predict a localization of polarons and bipolarons in the vicinity of counterions with localization length extending over several monomer units, as illustrated in Figure 4e−i, which shows the change in the bond length alteration as compared to the neutral oligomers. (Note that the bond length alteration changes from the aromatic to the quinoid character in the oligomer backbone where the charge is localized41). It is also noteworthy that in the absence of counterions, B3LYP calculations do not predict polaron localization, as illustrated in Figure 4e. formation of a polaron pair is energetically unfavorable compared to that of the bipolaron. The polaron pair becomes the ground state of the Q = +2 species for N ≳10. The calculated absorption is in an excellent quantitative agreement with the experimental data.17 For the case of polarons, the absorption spectrum exhibits two peaks, at E ≈0.8 eV (λ ≈ 1550 nm) and E ≈1.6 eV (λ ≈780 nm). The first polaronic absorption peak (denoted as TP/B) is due to transitions from the “valence band” to the polaron energy level. In molecular language, this can be described as a highest occupied molecular orbital (HOMO)(β)-to-lowest unoccupied molecular orbital (LUMO)(β) transition, where HOMO(β) and LUMO(β) denote the highest occupied and lowest unoccupied spin orbitals for the minority spin (the “down” spin in Figure 3), respectively. The second peak TC is due to excitation from the valence band into the “conduction band”, consisting of a superposition of HOMO(α) to LUMO(α), HOMO −1(α) to LUMO + 1(α), and HOMO(β) and LUMO + 2(β) excitations, where α refers to the “up” spin (see Figure 3a). For the case of the bipolaron, only one peak, TP/B, is present, E ≈1.25 eV (λ ≈1000 nm), which, as in the case of the polaron, corresponds to transitions from the valence band into the empty orbital in the gap (i.e., from the HOMO to the LUMO of the dication). The Journal of Physical Chemistry B We chose to model a diluted system of polythiophene (PT) oligomers of fixed, well-controlled lengths and oxidation levels whose absorption spectra were studied by van Haare et al.17 p p p y The schematic diagram of the electronic structure and optical transitions at different oxidation levels according to DFT predictions22,32,36,41 is outlined in Figure 2c−e. For the case of a polaron (Q = +1, S = 1/2), the spin degeneracy is lifted, and there is a single empty level (i.e., empty spin orbital) in the gap; this should be contrasted with the occupation of two spin-degenerate levels by a single electron in the “early” approach (cf. Figure 2a,c). For a bipolaron (Q = +2, S = 0), there is a single spin-degenerate orbital in the gap, also differing from the traditional picture, where there are two such levels. We also consider the possibility of two polarons interacting to give a triplet doubly charged charge carrier (i.e., Q = +2, S = 1) and term these polaron pairs; for these charge p p y The calculated electronic structure and the absorption spectra (superimposed on the experimental spectra from ref 17) of PT oligomers consisting of N = 6 thiophene rings with the oxidation levels Q = +1 and +2 are shown in Figure 3. The ground state for Q = +1 is a polaron (spin S = 1/2) and that for Q = +2 is a bipolaron (spin S = 0). Note that for N = 6 the 11284 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B nal of Physical Chemistry B pubs.acs.org/JPCB Article Article pubs.acs.org/JPCB Figure 4. Total electron spin of the lowest-energy ground states of the P3HT oligomers without and with counterions ((a) and (c), respectively). For the case of oligomers with counterions, five different realizations (shown in Figure S8) were calculated; the red dots in (c) indicate the spin of the ground state with the lowest energy for each of these configurations. (b, d) Energy diagram for P3HT oligomers at different oxidation levels for the cases without and with counterions, respectively. For the case with counterions, the energy diagrams are chosen for representative realizations where the ground energy spin state is the same as for the case without counterions, namely, realization 1,1,1,2 for the cases Q = +1, +2, +3, +4 in Figure S8. The blue and red lines are occupied and unoccupied energy levels. The dashed red lines represent (unoccupied) polaronic states in the gap. The yellow and blue backgrounds mark the conduction and valence bands, respectively. (e−g) Bond length alteration for the case of Q = +1 without counterions and for different positions of counterions, as illustrated in the upper panel; the counterions are highlighted in green. The bond numbers and the bond length alteration are defined in Figure S10. Examples of two different counterion configurations for the case Q = +2 leading to the formation of (h) a bipolaron and (i) a polaron pair. The corresponding energy diagram and the bond length alterations are shown. All calculations (a−i) correspond to the number of thiophene units N = 18. In all energy diagrams, the highest occupied (spin-resolved) orbital levels of the valence bands are aligned at E = 0. counterions,73,80,81 but the overall energetic structure remains similar to that without counterions. case for Q = +3 in Figure 4d, for which there are three empty spin orbitals in the gap. The lowest of these are spin- degenerate with spin-up and spin-down having the same energy levels and identical orbitals, which correspond to the bipolaron, whereas the remaining empty spin-down spin orbital in the gap is associated with the polaron. The Journal of Physical Chemistry B To conclude, the excellent agreement between the experimental and theoretical absorption for the well- controlled reference system of polythiophene oligomers formation of a polaron pair is energetically unfavorable compared to that of the bipolaron. The polaron pair becomes the ground state of the Q = +2 species for N ≳10. The calculated absorption is in an excellent quantitative agreement with the experimental data.17 For the case of polarons, the absorption spectrum exhibits two peaks, at E ≈0.8 eV (λ ≈ 1550 nm) and E ≈1.6 eV (λ ≈780 nm). The first polaronic absorption peak (denoted as TP/B) is due to transitions from the “valence band” to the polaron energy level. In molecular language, this can be described as a highest occupied molecular orbital (HOMO)(β)-to-lowest unoccupied molecular orbital (LUMO)(β) transition, where HOMO(β) and LUMO(β) denote the highest occupied and lowest unoccupied spin orbitals for the minority spin (the “down” spin in Figure 3), respectively. The second peak TC is due to excitation from the valence band into the “conduction band”, consisting of a superposition of HOMO(α) to LUMO(α), HOMO −1(α) to LUMO + 1(α), and HOMO(β) and LUMO + 2(β) excitations, where α refers to the “up” spin (see Figure 3a). For the case of the bipolaron, only one peak, TP/B, is present, E ≈1.25 eV (λ ≈1000 nm), which, as in the case of the polaron, corresponds to transitions from the valence band into the empty orbital in the gap (i.e., from the HOMO to the LUMO of the dication). To conclude, the excellent agreement between the experimental and theoretical absorption for the well- controlled reference system of polythiophene oligomers Figure 4d shows the spin-resolved orbitals for different oxidation levels of the oligomers with counterions. (For each oxidation level, we chose a representative counterion position corresponding to a typical ground-state multiplicity as indicated in the figure.) As expected, the positions of the individual energy levels are affected by the presence of 11285 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 Figure 4. Total electron spin of the lowest-energy ground states of the P3HT oligomers without and with counterions ((a) and (c), respe For the case of oligomers with counterions, five different realizations (shown in Figure S8) were calculated; the red dots in (c) indicate th the ground state with the lowest energy for each of these configurations. The Journal of Physical Chemistry B (b, d) Energy diagram for P3HT oligomers at different oxidation l the cases without and with counterions, respectively. For the case with counterions, the energy diagrams are chosen for representative rea where the ground energy spin state is the same as for the case without counterions, namely, realization 1,1,1,2 for the cases Q = +1, +2, + Figure S8. The blue and red lines are occupied and unoccupied energy levels. The dashed red lines represent (unoccupied) polaronic stat gap. The yellow and blue backgrounds mark the conduction and valence bands, respectively. (e−g) Bond length alteration for the case of without counterions and for different positions of counterions, as illustrated in the upper panel; the counterions are highlighted in green. T numbers and the bond length alteration are defined in Figure S10. Examples of two different counterion configurations for the case Q = +2 to the formation of (h) a bipolaron and (i) a polaron pair. The corresponding energy diagram and the bond length alterations are sh calculations (a−i) correspond to the number of thiophene units N = 18. In all energy diagrams, the highest occupied (spin-resolved) orbi of the valence bands are aligned at E = 0. The Journal of Physical Chemistry B pubs.acs.org/JPCB https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B Note that the coexistence of polarons and bipolarons on the same polymer chain was also predicted for n-doped polymers, such as the ladder polymer poly(benzimidazobenzophenanthroline) (BBL).70 The coexistence of multiple polarons on a single Figure 4h,i illustrates the cases for which two different counterion positions lead to different ground-state spin multiplicities, singlet and triplet, corresponding to, respectively, a bipolaron and a polaron pair. Bipolaron, rather than polaron pair, ground states typically arise when two counterions are relatively close to each other, which helps to overcome Coulombic repulsion between two holes. Note that bipolarons can coexist with polarons on the same oligomers. This is the 11286 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 Figure 5. Evolution of the absorption spectra of (a) P3HT and (b) P(g42T-T) with the oxidation level for the case without counterions. The transition from dark to bright colors corresponds to the increase in the oxidation level. Comparison of the absorption spectra of (c) P3HT and (d) P(g42T-T) for different oxidation levels for the cases with (blue lines) and without (red lines) counterions. For the sake of clarity, only the oscillator strengths are shown (without the Lorentzian fit). (e) Illustration of the electronic transitions contributing to the main peaks in the absorption spectra of the oxidized polymers, TP/B, TP/B+C, and TC. The peak TN corresponds to the absorption spectrum of a neutral polymer chain. (f) Absorption spectra of the P3HT oligomers for two different counterion realizations for Q = +1in position 1, the counterion is located near the oligomer side, while in position 3 the counterion is above its middle thiophene units (the exact geometry can be found in Figure S8). (g) Absorption spectra of the P3HT oligomers for five different counterion realizations for Q = +3 and their combined averaged spectrum (solid line). Panels (f) and (g) show both oscillatory strength and their Lorentzian fit. The Journal of Physical Chemistry B pubs.acs.org/JPCB Article The Journal of Physical Chemistry B Article pubs.acs.org/JPCB pubs.acs.org/JPCB Article pubs.acs.org/JPCB Figure 5. Evolution of the absorption spectra of (a) P3HT and (b) P(g42T-T) with the oxidation level for the case without counterions. The transition from dark to bright colors corresponds to the increase in the oxidation level. Comparison of the absorption spectra of (c) P3HT and (d) P(g42T-T) for different oxidation levels for the cases with (blue lines) and without (red lines) counterions. The Journal of Physical Chemistry B For the sake of clarity, only the oscillator strengths are shown (without the Lorentzian fit). (e) Illustration of the electronic transitions contributing to the main peaks in the absorption spectra of the oxidized polymers, TP/B, TP/B+C, and TC. The peak TN corresponds to the absorption spectrum of a neutral polymer chain. (f) Absorption spectra of the P3HT oligomers for two different counterion realizations for Q = +1in position 1, the counterion is located near the oligomer side, while in position 3 the counterion is above its middle thiophene units (the exact geometry can be found in Figure S8). (g) Absorption spectra of the P3HT oligomers for five different counterion realizations for Q = +3 and their combined averaged spectrum (solid line). Panels (f) and (g) show both oscillatory strength and their Lorentzian fit. chain was experimentally shown for the case of poly(3- decylthiophene).23 corresponds to a superposition of excitations from the valence band into polaronic/bipolaronic levels and from the valence band into the conduction band. Calculated spectra exhibit a clear pattern with increasing oxidation level, as shown in Figure 5a,b: TP/B becomes stronger and shifts to higher energies, whereas TC and TP/B+C become progressively weaker, with TC disappearing at higher oxidation levels (16% and more). This behavior reflects the evolution of the electronic structure with the oxidation level (Figure 4b) where progressively more polaronic/bipolaronic states in the gap emerge. This also explains the shift of TP/B to higher energies because more high- lying energy polaronic/bipolaronic levels become available in the gap, as shown in Figure 4b. Note that for Cox ≈14% the heights of the peaks TP/B and TP/B+C become approximately equal. This can be used to estimate the oxidation level of a polymer film from its absorption spectrum if the films are doped homogeneously. It is noteworthy that the calculated decylthiophene).23 III.III.II. UV-to-IR Absorption Spectra. Figure 5 shows the simulated absorption spectra of P3HT and P(g42T-T) without and with counterions for different oxidation levels: Cox = 5.5, 11.1, 16.6, and 22.2%. (The oxidation level is defined as the number of charges per thiophene ring.) These values correspond to the charge Q = +1, +2, +3, and +4 per chain. Let us first discuss the spectra for the system without the counterions (Figure 5a−d). The spectra of charged polymers show absorption peaks that can be divided into three main groups, as outlined in Figure 5e. https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B The peaks TP/B and TC have the same origin as discussed in Section III.I for polythiophenes in general and in Section III.II for an oligothiophene, i.e., they correspond to the transitions from the valence band to the empty polaronic/bipolaronic levels and from the valence band to the conduction band, respectively. A third peak, TP/B+C, III.III.II. UV-to-IR Absorption Spectra. Figure 5 shows the simulated absorption spectra of P3HT and P(g42T-T) without and with counterions for different oxidation levels: Cox = 5.5, 11.1, 16.6, and 22.2%. (The oxidation level is defined as the number of charges per thiophene ring.) These values correspond to the charge Q = +1, +2, +3, and +4 per chain. Let us first discuss the spectra for the system without the counterions (Figure 5a−d). The spectra of charged polymers show absorption peaks that can be divided into three main groups, as outlined in Figure 5e. The peaks TP/B and TC have the same origin as discussed in Section III.I for polythiophenes in general and in Section III.II for an oligothiophene, i.e., they correspond to the transitions from the valence band to the empty polaronic/bipolaronic levels and from the valence band to the conduction band, respectively. A third peak, TP/B+C, 11287 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 Figure 6. (a) Geometrical structure of a P3HT oligomer with N = 12 (Q = +2) after an MD-DFTB run and (b) corresponding C−C backbone bond alteration at t = 0, 1, 15, and 20 ps. (c) Absorption spectra of a neutral and a doped P3HT oligomer with Q = +2 (Cox = 16.6%) at t = 1, 15, and 20 ps. The bold black line shows the averaged spectrum. (a) Band diagrams of P3HT at 1, 15, and 20 ps. (e) Geometrical structure of bi- and trilayer P3HT after a 1 ps DFTB MD run; Cox = 16.6% (Q = +2). The absorption spectra of neutral (f) and doped (g) P3HT multilayers with Cox = 16.6% at t = 1, 15, and 20 ps. The bond numbers and the bond length alteration are defined in Figure S10. The Journal of Physical Chemistry B pubs.acs.org/JPCB Article The Journal of Physical Chemistry B Article pubs.acs.org/JPCB Figure 6. (a) Geometrical structure of a P3HT oligomer with N = 12 (Q = +2) after an MD-DFTB run and (b) corresponding C−C backbone bond alteration at t = 0, 1, 15, and 20 ps. The Journal of Physical Chemistry B (c) Absorption spectra of a neutral and a doped P3HT oligomer with Q = +2 (Cox = 16.6%) at t = 1, 15, and 20 ps. The bold black line shows the averaged spectrum. (a) Band diagrams of P3HT at 1, 15, and 20 ps. (e) Geometrical structure of bi- and trilayer P3HT after a 1 ps DFTB MD run; Cox = 16.6% (Q = +2). The absorption spectra of neutral (f) and doped (g) P3HT multilayers with Cox = 16.6% at t = 1, 15, and 20 ps. The bond numbers and the bond length alteration are defined in Figure S10. corresponding peaks for the counterion-free chains with the same charge, as shown in Figure 5a. We performed calculations for chains with different oxidation levels and find that the same conclusion holds for all considered systems for both polymers P3HT and P(g42T-T) (see Figure 5c,d). Thus, the effect of counterions on the absorption spectra of polythiophenes with one or several charges can be summarized as follows: the presence of counterions leads to a significant broadening of the absorption peaks, whereas the positions of the peaks remain close to those without counterions. evolution of the absorption spectra is similar to that reported for PEDOT.41 Let us now discuss the influence of counterions on the absorption spectra. The positions of counterions significantly affect the electronic structure of the polymers, which, in turn, affects the absorption spectra. An example is depicted in Figure 5f, where the absorption spectra for the P3HT chains with the Cox ≈5.6% (Q = +1) are calculated for two different counterion positions. In position 1, the counterion is located at the end of the polymer chain, whereas in position 3, the counterion is located at the center of the chain (see Figure S8). This change in the geometry of the system results in different absorption spectra for two oligomers with the same oxidation level. Absorption peaks of P3HT chains with three counterions (Q = +3) for five different spatial counterion orientations are depicted in Figure 5g. As for the case of Q = +1, each arrangement of counterions leads to a different spectrum. However, averaging over these five arrangements results in two broad absorption peaks residing at the positions close to the III.IV. Absorption Spectra in Polymer Chains That Display Backbone Torsion and π−π Stacking. https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B (Note that absorption spectra are obtained by super- position of the Lorentzian-fitted oscillator strengths in Figure 5c,d.) The transition from dark to bright colors corresponds to the increase in the oxidation level. Comparison of experimentally obtained absorption spectra with the theoretical ones for pristine (b) P3HT and (e) P(g42T-T) (note that the pristine polymer is in fact slightly oxygen-doped due to its low ionization energy) and (c, f) the same polymers sequentially doped with Mo(tfd-COCF3)3. The calculated curves include the averaged spectra for all calculated doping levels (Cox = 5.6, 11.1, 6.7, 22.2%). The black curves are experimental studies; the red and blue curves are simulation results for P3HT and P(g42T-T). The calculations are performed for oligomers with N = 18 monomer units. Figure 7. Evolution of the absorption spectra of (a) P3HT and (d) Figure 7. Evolution of the absorption spectra of (a) P3HT and (d) P(g42T-T) with the increase in the oxidation level for the case with counterions. (Note that absorption spectra are obtained by super- position of the Lorentzian-fitted oscillator strengths in Figure 5c,d.) The transition from dark to bright colors corresponds to the increase in the oxidation level. Comparison of experimentally obtained absorption spectra with the theoretical ones for pristine (b) P3HT and (e) P(g42T-T) (note that the pristine polymer is in fact slightly oxygen-doped due to its low ionization energy) and (c, f) the same polymers sequentially doped with Mo(tfd-COCF3)3. The calculated curves include the averaged spectra for all calculated doping levels (Cox = 5.6, 11.1, 6.7, 22.2%). The black curves are experimental studies; the red and blue curves are simulation results for P3HT and P(g42T-T). The calculations are performed for oligomers with N = 18 monomer units. To analyze the effect of π−π stacking on UV−vis absorption spectra, we simulated bilayer and trilayer P3HT crystallites for t = 1 ps, for both neutral and 16.6% oxidized states. From the nanostructure snapshots shown in Figure 6d, the π−π stacking is clearly observed. Absorption spectra in Figure 6f show a broadening of the main peaks as the number of chains in the layer is increased. Unfortunately, TD-DFTB simulation of the doped trilayer sample is too computationally demanding (due to a large number of excited states), and, therefore, Figure 5f shows the spectra for a bilayer only. The Journal of Physical Chemistry B P(g42T-T) thin films are mostly disordered with some degree of crystallinity, while regioregular P3HT films have semicrystal- line character.25,31,48,52,82 The chains of both P3HT and P(g42T-T) are nonplanar; both the backbones and side chains are in constant temperature-induced movement, which causes twisting of the side chains and changes of the torsion angles along the backbones. In this section, we will study the effect of III.IV. Absorption Spectra in Polymer Chains That Display Backbone Torsion and π−π Stacking. P(g42T-T) thin films are mostly disordered with some degree of crystallinity, while regioregular P3HT films have semicrystal- line character.25,31,48,52,82 The chains of both P3HT and P(g42T-T) are nonplanar; both the backbones and side chains are in constant temperature-induced movement, which causes twisting of the side chains and changes of the torsion angles along the backbones. In this section, we will study the effect of 11288 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B pubs.acs.org/JPCB Article Figure 7. Evolution of the absorption spectra of (a) P3HT and (d) P(g42T-T) with the increase in the oxidation level for the case with counterions. (Note that absorption spectra are obtained by super- position of the Lorentzian-fitted oscillator strengths in Figure 5c,d.) The transition from dark to bright colors corresponds to the increase in the oxidation level. Comparison of experimentally obtained absorption spectra with the theoretical ones for pristine (b) P3HT and (e) P(g42T-T) (note that the pristine polymer is in fact slightly oxygen-doped due to its low ionization energy) and (c, f) the same polymers sequentially doped with Mo(tfd-COCF3)3. The calculated curves include the averaged spectra for all calculated doping levels (Cox = 5.6, 11.1, 6.7, 22.2%). The black curves are experimental studies; the red and blue curves are simulation results for P3HT and P(g42T-T). The calculations are performed for oligomers with N = 18 monomer units. twisting and the effect of π−π stacking on the absorption spectra. In the treatment of chain twisting, it is important to account for the effect of the side chains, as they strongly affect the chain movement and bending. The TD-DFT simulation of polythiophenes decorated with side chains and/or several π−π- stacked chains is computationally challenging due to a large size of the systems. The Journal of Physical Chemistry B Therefore, to study the effect of backbone torsions and π−π stacking on the absorption spectra of polythiophenes, we use the TD-DFTB approach combined with DFTB-based molecular dynamics (MD) simulations, which allows us to calculate much larger systems than conventional DFT. Figure 6a shows snapshots of the geometry of a P3HT oligomer with charge Q = +2 calculated using MD simulations on the DFTB level of theory during a time interval of 0−20 ps. The initial geometry of the thiophene rings and side chains is close to planar. With time, the side chains start to move, changing the torsion angles and bond lengths in the thiophene rings. At t = 15 ps, side chains bend relative to the oligomer backbone, and at t = 20 ps, some of the side chains twist so strongly, so that they are in part located on the top and under the thiophene rings. Twisting of the backbone results in significant changes in the bond lengths, as illustrated in Figure 6b. Figure 6c shows the absorption spectra of a neutral P3HT oligomer and the oligomer with Q = +2 calculated for different times t = 1, 10, 15, and 20 ps. All of them show similar behavior where, however, the positions and the strengths of the absorption peaks vary for different spatial conformations of the chains. The black solid line shows the spectrum averaged over spatial conformations at different moments of time. It shows a broadening relative to a spectrum corresponding to a given confirmation. This is similar to the case of a chain with counterions discussed in Section III.III, in which the broadening was caused by averaging over different counterion positions. This similarity is not a coincidence. Indeed, Figure 6d shows the electronic structure of a chain with Q = +2 at different times. While the structure of the energy levels is similar, the exact position of the bipolaron levels in the gap is different for different moments of time. This is related to the changes in the bond length and angles that take place during the torsional motion of the chains (see Figure 6b). The difference in the energy levels leads to different absorption spectra, in a similar way as discussed above for the case of counterions. Figure 7. Evolution of the absorption spectra of (a) P3HT and (d) P(g42T-T) with the increase in the oxidation level for the case with counterions. The Journal of Physical Chemistry B To estimate the oxidation level from the experimental spectra, we used a molar attenuation coefficient of ε = (4.1 ± 0.2) × 103 m2 mol−1 for the TP/B+C transition of oxidized P3HT with the peak at 1.55 eV (800 nm), which we have obtained from spectroelectrochemistry measurements.83 The Beer−Lambert law A = Nv·ε·d relates the peak absorbance A of doped P3HT at 1.55 eV to the molar attenuation coefficient ε, where d = 54 nm is the film thickness. We obtain a polaron density of Nv = (4.2 ± 0.5) × 1026 m−3, which translates to an oxidation level of (11 ± 1)% per thiophene for Mo(tfd- COCF3)3-doped P3HT (assuming a density of 1.1 g cm−3). We note that, as in other DFT studies of polythiophene materials, the calculations presented here lead to electronic orbital structures for polarons and bipolarons that are fundamentally different from those in the earlier traditional approaches from the 1980s that still dominate the current literature. DFT indicates a single unoccupied spin orbital in the band gap of the polymer for a polaron, whereas in the traditional interpretation there are two spin-degenerate levels in the gap, one of which is singly occupied and is associated with the polaronic band. DFT indicates a single spin- degenerate level in the gap for a bipolaron, whereas there are two such levels in the traditional model. A qualitatively different interpretation of the absorption of polythiophenes also emerges: traditionally, the two most prominent peaks are referred to as polaron (at ca. 1.5 eV) and “bipolaron” (at ca. 0.4 eV) peaks,20 but the DFT results show that both major bands are seen for both polarons and bipolarons. This suggests that the existence of polarons or bipolarons cannot be reliably inferred from experimental absorption spectra, although the doping level is indicated by the relative strength of the two major bands. p g y g Let us now proceed to the P(g42T-T) film. The low-energy and the intermediate-energy peaks for the doped sample are very close to the corresponding peaks for P3HT (0.4 and 1.5 eV vs 0.4 and 1.4 eV). Therefore, the assignment of these peaks is the same as that for the P3HT film. In contrast to the doped P3HT, the high-energy peak at ca. 2.3 eV is not present in the doped P(g42T-T) film; we, therefore, conclude that the P(g42T-T) film does not contain undoped regions. The Journal of Physical Chemistry B The Journal of Physical Chemistry B The Journal of Physical Chemistry B pubs.acs.org/JPCB Article corresponding theoretical peaks (TP/B and TP/B+C). The low- energy peak TP/B (at ca. 0.4 eV) is the most pronounced, which suggests that the doping level of the sample is rather high (see Figure 7a). Although calculated spectra for high doping levels are dominated by only two peaks, TP/B and TP/B+C, the third (high-energy) experimental peak (at ca. 2.3 eV) matches exactly the absorption peak of the neutral (undoped) film, excluding vibronic spectrum. This leads us to conclude that some regions remain undoped. Therefore, based on our DFT calculations, the assignment of the experimentally observed peaks for a doped P3HT is as follows. The low- energy peak at ca. 0.4 eV corresponds to TP/B, the intermediate-energy peak at ca. 1.5 eV corresponds to TP/B+C, and the high-energy peak at ca. 2.3 eV corresponds to TN due to the presence of undoped polymer segments in the film (see Figure 7a). simulations, paying particular attention to the understanding of the effect of counterions, side chains, π−π stacking, and temperature-induced bending and twisting of the chains on the absorption spectra of the systems at hand. The calculated results are used to interpret the peaks in the measured spectra, where a good agreement between the theoretical and experimental data is found, especially when using the B3LYP, rather than ωB97XD, functional. DTF calculations predict that the absorption spectra of doped polythiophenes show three distinct peaks TP/B, TC (weak), and TP/B+C (see Figure 5a,b or 7a,d). The peaks TP/B and TC correspond to the excitations from the valence band to the polaronic/bipolaronic levels and the conduction band, respectively, whereas the peak TP/B+C corresponds to a superposition of excitations from the valence band to both polaronic/bipolaronic levels and the conduction band. The evolution of the calculated spectra exhibits a clear pattern with the increase in the oxidation level (see Figure 5a,b): the strength of TP/B increases, whereas TC and TP/B+C become progressively weaker, with TC disappearing at higher oxidation levels. This behavior reflects the evolution of the electronic structure with the oxidation level (Figure 4b) where progressively more polaronic/bipolaronic states in the gap emerge. The Journal of Physical Chemistry B We estimated the oxidation level from the experimental spectra assuming that the molar attenuation coefficient of the TP/B+C transition of oxidized P(g42T-T) with the peak at 1.38 eV (900 nm) is the same as that of oxidized P3HT. We obtain a value of Nv = (4.6 ± 0.5) × 1026 m−3 for a film thickness of d = 80 nm, which indicates an oxidation level of (15 ± 2)% per thiophene for Mo(tfd-COCF3)3-doped P(g42T-T). As dis- cussed above, the ratio of peak heights of TP/B and TP/B+C depends on the doping level (see Figure 5a). A comparison of the experimental peak ratio to the theoretical one is consistent with an oxidation level of ca. 16−20% for P(g42T-T), which is in agreement with the above estimation taking into account experimental uncertainties and systematic errors in DFT. The presence of counterions leads to a significant broad- ening of the absorption peaks, whereas the positions of the peaks remain close to those without counterions. The effect of π−π stacking and temperature-induced bending and twisting of the chains is qualitatively similar to that of dopants and also leads to peak broadening. The origin of the above broadening is analyzed based on the calculated electronic structure. At the same time, we show that the absorption spectra are practically unaffected by the side chains because the electron density of the π-orbitals is localized along the polymer backbones. As already mentioned in Sections I and III.I, the polymers with a thiophene-like backbone share the same features of the absorption spectra. In particular, similar spectral features are also observed for electrochemical doping,19,84 photoinduced de-doping,16 acid doping,85 self-doping,86 and doping with different redox dopants.49 Hence, this confirms our conclusion that the position of the absorption is largely independent of the counterions. https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B We conclude that the effect of twisting of the polymer chains and the effect of π−π coupling on the absorption spectra of polythiophenes with side chains is qualitatively similar to that of counterions and manifests itself in a broadening of the absorption peaks. The calculated spectra accounting for the broadening due to the effect of counterions are shown in Figure 5c,d and are also displayed in Figure 7a,d. We note that, as will be discussed below, the experimental samples are likely to contain both doped or nondoped domains. Hence, for the sake of convenience of the comparison to the experiment, each displayed theoretical spectrum in Figure 7 represents the averaged spectrum for all calculated doping levels (Cox = 5.6, 11.1, 16.7, 22.2%). This corresponds to the average doping level of ca. 14%, close to the experimental one (see below). g p p III.V. Comparison of Simulation and Experimental Studies. Figure 7 shows the measured absorption spectra for undoped P3HT and P(g42T-T), as well as thin films doped with Mo(tfd-COCF3)3 (see Figure 1c for chemical structure). Let us start with the case of P3HT. For the doped films, the spectrum shows three peaks. The comparison with the calculated spectra shows that the first two peaks match the 11289 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 Notes The authors declare no competing financial interest. ■REFERENCES cases with and without counterions (dopants) calculated using functionals B3LYP and ωB97XD; S2 comparison of the experimentally obtained absorption spectra with the theoretical ones for P3HT and P(g42T-T) calculated using ωB97XD; S3 absorption spectra of a charged P3HT monomer with and without diffuse functions in the basis set; S4 convergence of the absorption spectra; S5 absorption spectra of a charged P3HT monomer with and without dispersion correc- tion; S6 dependence of the main absorption peak in the pristine P3HT with reduced side chains on oligomer length; S7 absorption spectra of P3HT and P(g42T-T) for different lengths of sidechains; S8 five different counterion positions considered for each oxidation level; S9 absorption spectra of the Cl3 counterion; S10 definition of the bond numbers (PDF) (1) Lüssem, B.; Keum, C.-M.; Kasemann, D.; Naab, B.; Bao, Z.; Leo, K. Doped Organic Transistors. Chem. Rev. 2016, 116, 13714−13751. (2) Chew, A. R.; Ghosh, R.; Shang, Z.; Spano, F. C.; Salleo, A. Sequential Doping Reveals the Importance of Amorphous Chain Rigidity in Charge Transport of Semi-Crystalline Polymers. J. Phys. Chem. Lett. 2017, 8, 4974−4980. (3) Kiefer, D.; et al. Double Doping of Conjugated Polymers with Monomer Molecular Dopants. Nat. Mater. 2019, 18, 149−155. (4) Aubry, T. J.; et al. Dodecaborane-Based Dopants Designed to Shield Anion Electrostatics Lead to Increased Carrier Mobility in a Doped Conjugated Polymer. Adv. Mater. 2019, 31, No. 1805647. (5) Moia, D.; Giovannitti, A.; Szumska, A. A.; Maria, I. P.; Rezasoltani, E.; Sachs, M.; Schnurr, M.; Barnes, P. R. F.; McCulloch, I.; Nelson, J. Design and Evaluation of Conjugated Polymers with Polar Side Chains as Electrode Materials for Electrochemical Energy Storage in Aqueous Electrolytes. Energy Environ. Sci. 2019, 12, 1349− 1357. (6) Flagg, L. Q.; Bischak, C. G.; Onorato, J. W.; Rashid, R. B.; Luscombe, C. K.; Ginger, D. S. Polymer Crystallinity Controls Water Uptake in Glycol Side-Chain Polymer Organic Electrochemical Transistors. J. Am. Chem. Soc. 2019, 141, 4345−4354. Corresponding Author (7) Cendra, C.; Giovannitti, A.; Savva, A.; Venkatraman, V.; McCulloch, I.; Salleo, A.; Inal, S.; Rivnay, J. Role of the Anion on the Transport and Structure of Organic Mixed Conductors. Adv. Funct. Mater. 2019, 29, No. 1807034. Igor Zozoulenko −Laboratory of Organic Electronics, Department of Science and Technology, Linköping University, 60174 Norrköping, Sweden; orcid.org/0000-0002-6078- 3006; Email: [email protected] (8) Gladisch, J.; Stavrinidou, E.; Ghosh, S.; Giovannitti, A.; Moser, M.; Zozoulenko, I.; McCulloch, I.; Berggren, M. Reversible Electronic Solid-Gel Switching of a Conjugated Polymer. Adv. Sci. 2020, 7, No. 1901144. * sı Supporting Information * sı Supporting Information * sı Supporting Information The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcb.0c08757. The electronic absorption of polythiophenes decorated with aliphatic and oligo ethylene glycol side chains (P3HT and P(g42T-T), respectively) and molecularly doped with Mo(tfd- COCF3)3 −counterions was studied and analyzed based on DFT-, TD-DFT-, and DFTB-based molecular dynamics S1 evolution of the band gap and absorption spectra of P3HT with the increase of the oxidation level for the 11290 11290 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B Article pubs.acs.org/JPCB Article Authors Ihor Sahalianov −Laboratory of Organic Electronics, Department of Science and Technology, Linköping University, 60174 Norrköping, Sweden; orcid.org/0000-0002-0609- 471X (9) Veysel Tunc, A.; De Sio, A.; Riedel, D.; Deschler, F.; Da Como, E.; Parisi, J.; von Hauff, E. Molecular Doping of Low-Bandgap- Polymer:Fullerene Solar Cells: Effects on Transport and Solar Cells. Org. Electron. 2012, 13, 290−296. Jonna Hynynen −Department of Chemistry and Chemical Engineering, Chalmers University of Technology, 41296 Göteborg, Sweden g (10) Nielsen, C. B.; McCulloch, I. Recent Advances in Transistor Performance of Polythiophenes. Prog. Polym. Sci. 2013, 38, 2053− 2069. Stephen Barlow −School of Chemistry and Biochemistry and Center for Organic Photonics and Electronics, Georgia Institute of Technology, Atlanta, Georgia 30332-0400, United States; orcid.org/0000-0001-9059-9974 (11) Giovannitti, A.; Sbircea, D.-T.; Inal, S.; Nielsen, C. B.; Bandiello, E.; Hanifi, D. A.; Sessolo, M.; Malliaras, G. G.; McCulloch, I.; Rivnay, J. Controlling the Mode of Operation of Organic Transistors through Side-Chain Engineering. Proc. Natl. Acad. Sci. U.S.A. 2016, 113, 12017. Seth R. Marder −School of Chemistry and Biochemistry and Center for Organic Photonics and Electronics, Georgia Institute of Technology, Atlanta, Georgia 30332-0400, United States; orcid.org/0000-0001-6921-2536 (12) Melling, D.; Martinez, J. G.; Jager, E. W. H. Conjugated Polymer Actuators and Devices: Progress and Opportunities. Adv. Mater. 2019, 31, No. 1808210. (13) Berggren, M.; Malliaras, G. G. How Conducting Polymer Electrodes Operate. Science 2019, 364, 233. Christian Müller −Department of Chemistry and Chemical Engineering, Chalmers University of Technology, 41296 Göteborg, Sweden; orcid.org/0000-0001-7859-7909 Christian Müller −Department of Chemistry and Chemical Engineering, Chalmers University of Technology, 41296 Göteborg, Sweden; orcid.org/0000-0001-7859-7909 p (14) Hynynen, J.; Kiefer, D.; Yu, L.; Kroon, R.; Munir, R.; Amassian, A.; Kemerink, M.; Müller, C. Enhanced Electrical Conductivity of Molecularly P-Doped Poly(3-Hexylthiophene) through Understand- ing the Correlation with Solid-State Order. Macromolecules 2017, 50, 8140−8148. Complete contact information is available at: https://pubs.acs.org/10.1021/acs.jpcb.0c08757 (15) Kroon, R.; Kiefer, D.; Stegerer, D.; Yu, L.; Sommer, M.; Müller, C. Polar Side Chains Enhance Processability, Electrical Conductivity, and Thermal Stability of a Molecularly P-Doped Polythiophene. Adv. Mater. 2017, 29, No. 1700930. The Journal of Physical Chemistry B The Journal of Physical Chemistry B pubs.acs.org/JPCB Article Article (39) Dai, Y.; Wei, C.; Blaisten-Barojas, E. Density Functional Theory Study of Neutral and Oxidized Thiophene Oligomers. J. Chem. Phys. 2013, 139, No. 184905. (19) Enengl, C.; Enengl, S.; Pluczyk, S.; Havlicek, M.; Lapkowski, M.; Neugebauer, H.; Ehrenfreund, E. Doping-Induced Absorption Bands in P3ht: Polarons and Bipolarons. ChemPhysChem 2016, 17, 3836−3844. (40) Salzner, U. Electronic Structure of Conducting Organic Polymers: Insights from Time-Dependent Density Functional Theory. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2014, 4, 601−622. (20) Massonnet, N.; Carella, A.; Jaudouin, O.; Rannou, P.; Laval, G.; Celle, C.; Simonato, J.-P. Improvement of the Seebeck Coefficient of Pedot:Pss by Chemical Reduction Combined with a Novel Method for Its Transfer Using Free-Standing Thin Films. J. Mater. Chem. C 2014, 2, 1278−1283. (41) Zozoulenko, I.; Singh, A.; Singh, S. K.; Gueskine, V.; Crispin, X.; Berggren, M. Polarons, Bipolarons, and Absorption Spectroscopy of Pedot. ACS Appl. Polym. Mater. 2019, 1, 83−94. (21) Sonmez, G. Polymeric Electrochromics. Chem. Commun. 2005, 5251−5259. pp y (42) Farouil, L.; Alary, F.; Bedel-Pereira, E.; Heully, J.-L. Revisiting the Vibrational and Optical Properties of P3ht: A Combined Experimental and Theoretical Study. J. Phys. Chem. A 2018, 122, 6532−6545. (22) Bruno, C.; Paolucci, F.; Marcaccio, M.; Benassi, R.; Fontanesi, C.; Mucci, A.; Parenti, F.; Preti, L.; Schenetti, L.; Vanossi, D. Experimental and Theoretical Study of the P- and N-Doped States of Alkylsulfanyl Octithiophenes. J. Phys. Chem. B 2010, 114, 8585−8592. (23) Takeda, N.; Miller, J. R. Poly(3-Decylthiophene) Radical Anions and Cations in Solution: Single and Multiple Polarons and Their Delocalization Lengths in Conjugated Polymers. J. Phys. Chem. B 2012, 116, 14715−14723 (22) Bruno, C.; Paolucci, F.; Marcaccio, M.; Benassi, R.; Fontanesi, C.; Mucci, A.; Parenti, F.; Preti, L.; Schenetti, L.; Vanossi, D. Experimental and Theoretical Study of the P- and N-Doped States of Alkylsulfanyl Octithiophenes. J. Phys. Chem. B 2010, 114, 8585−8592. (43) Png, R.-Q.; Ang, M. C. Y.; Teo, M.-H.; Choo, K.-K.; Tang, C. G.; Belaineh, D.; Chua, L.-L.; Ho, P. K. H. Madelung and Hubbard Interactions in Polaron Band Model of Doped Organic Semi- conductors. Nat. Commun. 2016, 7, No. 11948. p y p Alkylsulfanyl Octithiophenes. J. Phys. Chem. B 2010, 114, 8585−8592. (23) Takeda, N.; Miller, J. R. Poly(3-Decylthiophene) Radical Anions and Cations in Solution: Single and Multiple Polarons and Their Delocalization Lengths in Conjugated Polymers. J. Phys. Chem. B 2012, 116, 14715−14723. , , (44) Kaloni, T. The Journal of Physical Chemistry B P.; Schreckenbach, G.; Freund, M. S. Structural and Electronic Properties of Pristine and Doped Polythiophene: Periodic Versus Molecular Calculations. J. Phys. Chem. C 2015, 119, 3979− 3989. , , (24) Voss, M. G.; Scholes, D. T.; Challa, J. R.; Schwartz, B. J. Ultrafast Transient Absorption Spectroscopy of Doped P3ht Films: Distinguishing Free and Trapped Polarons. Faraday Discuss. 2019, 216, 339−362. (45) Ghosh, S.; Gueskine, V.; Berggren, M.; Zozoulenko, I. V. Electronic Structures and Optical Absorption of N-Type Conducting Polymers at Different Doping Levels. J. Phys. Chem. C 2019, 123, 15467−15476. (25) Yamamoto, J.; Furukawa, Y. Electronic and Vibrational Spectra of Positive Polarons and Bipolarons in Regioregular Poly(3- Hexylthiophene) Doped with Ferric Chloride. J. Phys. Chem. B 2015, 119, 4788−4794. (46) Ghosh, S.; Berggren, M.; Zozoulenko, I. Electronic Structures and Optical Properties of P-Type/N-Type Polymer Blends: Density Functional Theory Study. J. Phys. Chem. C 2020, 124, 9203−9214. ( ) (26) Koyama, T.; Nakamura, A.; Kishida, H. Microscopic Mobility of Polarons in Chemically Doped Polythiophenes Measured by Employing Photoluminescence Spectroscopy. ACS Photonics 2014, 1, 655−661. (47) Winkler, S.; Amsalem, P.; Frisch, J.; Oehzelt, M.; Heimel, G.; Koch, N. Probing the Energy Levels in Hole-Doped Molecular Semiconductors. Mater. Horiz. 2015, 2, 427−433. (27) Brédas, J. L.; Wudl, F.; Heeger, A. J. Polarons and Bipolarons in Doped Polythiophene: A Theoretical Investigation. Solid State Commun. 1987, 63, 577−580. (48) Kaloni, T. P.; Giesbrecht, P. K.; Schreckenbach, G.; Freund, M. S. Polythiophene: From Fundamental Perspectives to Applications. Chem. Mater. 2017, 29, 10248−10283. (28) Cornil, J.; Brédas, J. L. Nature of the Optical Transitions in Charged Oligothiophenes. Adv. Mater. 1995, 7, 295−297. , , (49) Heinze, J.; Frontana-Uribe, B. A.; Ludwigs, S. Electrochemistry of Conducting PolymersPersistent Models and New Concepts. Chem. Rev. 2010, 110, 4724−4771. (29) Stafström, S.; Brédas, J. L. Evolution of the Electronic Structure of Polyacetylene and Polythiophene as a Function of Doping Level and Lattice Conformation. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 38, 4180−4191. (50) Ibanez, J. G.; Rincón, M. E.; Gutierrez-Granados, S.; Chahma, M.; Jaramillo-Quintero, O. A.; Frontana-Uribe, B. A. Conducting Polymers in the Fields of Energy, Environmental Remediation, and Chemical−Chiral Sensors. Chem. Rev. 2018, 118, 4731−4816. (30) Cornil, J.; Beljonne, D.; Brédas, J. L. Nature of Optical Transitions in Conjugated Oligomers. Ii. Theoretical Characterization of Neutral and Doped Oligothiophenes. J. Chem. Phys. 1995, 103, 842−849. (51) Bryan, A. M.; Santino, L. M.; Lu, Y.; Acharya, S.; D’Arcy, J. M. The Journal of Physical Chemistry B Conducting Polymers for Pseudocapacitive Energy Storage. Chem. Mater. 2016, 28, 5989−5998. (31) Ghosh, R.; Pochas, C. M.; Spano, F. C. Polaron Delocalization in Conjugated Polymer Films. J. Phys. Chem. C 2016, 120, 11394− 11406. (52) Wang, L.; Beljonne, D. Optical Properties of Regioregular Poly(3-Hexylthiophene) Aggregates from Fully Atomistic Investiga- tions. CrystEngComm 2016, 18, 7297−7304. (32) Heimel, G. The Optical Signature of Charges in Conjugated Polymers. ACS Central Science 2016, 2, 309−315. y g (53) Paniagua, S. A.; Baltazar, J.; Sojoudi, H.; Mohapatra, S. K.; Zhang, S.; Henderson, C. L.; Graham, S.; Barlow, S.; Marder, S. R. Production of Heavily N- and P-Doped Cvd Graphene with Solution- Processed Redox-Active Metal−Organic Species. Mater. Horiz. 2014, 1, 111−115. (33) Dkhissi, A.; Beljonne, D.; Lazzaroni, R.; Louwet, F.; Groenendaal, L.; Brédas, J. L. Density Functional Theory and Hartree−Fock Studies of the Geometric and Electronic Structure of Neutral and Doped Ethylenedioxythiophene (Edot) Oligomers. Int. J. Quantum Chem. 2003, 91, 517−523. Zhang, S.; Henderson, C. L.; Graham, S.; Barlow, S.; Marder, S. R. Production of Heavily N- and P-Doped Cvd Graphene with Solution- Processed Redox-Active Metal−Organic Species. Mater. Horiz. 2014, 1, 111−115. Processed Redox-Active Metal−Organic Species. Mater. Horiz. 2014, 1, 111−115. Neutral and Doped Ethylenedioxythiophene (Edot) Oligomers. Int. J. Quantum Chem. 2003, 91, 517−523. (54) Frisch, M. J. et al. Gaussian 16, revision B.01; Gaussian, Inc.: Wallingford, CT, 2016. Q , , (34) Zade, S. S.; Bendikov, M. Theoretical Study of Long Oligothiophene Dications: Bipolaron Vs Polaron Pair Vs Triplet State. J. Phys. Chem. B 2006, 110, 15839−15846. (55) Becke, A. D. Density-Functional Exchange-Energy Approx- imation with Correct Asymptotic Behavior. Phys. Rev. A 1988, 38, 3098−3100. y (35) Salzner, U. Theoretical Investigation of Excited States of Oligothiophenes and of Their Monocations. J. Chem. Theory Comput. 2007, 3, 1143−1157. (56) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle- Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785− 789. (36) Salzner, U. Investigation of Charge Carriers in Doped Thiophene Oligomers through Theoretical Modeling of Their Uv/ Vis Spectra. J. Phys. Chem. A 2008, 112, 5458−5466. (57) Chai, J.-D.; Head-Gordon, M. Long-Range Corrected Hybrid Density Functionals with Damped Atom-Atom Dispersion Correc- tions. Phys. Chem. Chem. Phys. 2008, 10, 6615−6620. p J y (37) Zamoshchik, N.; Salzner, U.; Bendikov, M. Nature of Charge Carriers in Long Doped Oligothiophenes: The Effect of Counterions. J. Phys. Chem. ■ACKNOWLEDGMENTS (16) Wei, Q.; Mukaida, M.; Kirihara, K.; Naitoh, Y.; Ishida, T. Photoinduced Dedoping of Conducting Polymers: An Approach to Precise Control of the Carrier Concentration and Understanding Transport Properties. ACS Appl. Mater. Interfaces 2016, 8, 2054− 2060. This work was supported by the Knut and Alice Wallenberg Foundation, the Swedish Research Council (2016-05990 and 2018-03824), and the National Science Foundation (through the DMREF program, DMR-1729737). I.Z. thanks the Advanced Functional Material Center at Linköping University for support. We thank Yadong Zhang for the synthesis of Mo(tfd-COCF3)3 and Renee Kroon for the synthesis of P(g42T-T). The computations were performed on resources provided by the Swedish National Infrastructure for Comput- ing (SNIC) at NSC and HPC2N. (17) van Haare, J. A. E. H.; Havinga, E. E.; van Dongen, J. L. J.; Janssen, R. A. J.; Cornil, J.; Brédas, J.-L. Redox States of Long Oligothiophenes: Two Polarons on a Single Chain. Chem. −Eur. J. 1998, 4, 1509−1522. (18) Hwang, J.; Tanner, D. B.; Schwendeman, I.; Reynolds, J. R. Optical Properties of Nondegenerate Ground-State Polymers: Three Dioxythiophene-Based Conjugated Polymers. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 67, No. 115205. 11291 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B Article pubs.acs.org/JPCB Article Conducting Polymers. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 94, No. 205202. Conducting Polymers. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 94, No. 205202. (59) Gaus, M.; Cui, Q.; Elstner, M. Dftb3: Extension of the Self- C Ch D F l T h B d M h d (S (59) Gaus, M.; Cui, Q.; Elstner, M. Dftb3: Extension of the Self- Consistent-Charge Density-Functional Tight-Binding Method (Scc- Dftb). J. Chem. Theory Comput. 2011, 7, 931−948. Consistent-Charge Density-Functional Tight-Binding Method (Scc- Dftb). J. Chem. Theory Comput. 2011, 7, 931−948. (81) Munoz, W. A.; Crispin, X.; Fahlman, M.; Zozoulenko, I. V. Understanding the Impact of Film Disorder and Local Surface Potential in Ultraviolet Photoelectron Spectroscopy of Pedot. Macromol. Rapid Commun. 2018, 39, No. 1700533. (60) Gaus, M.; Goez, A.; Elstner, M. Parametrization and Benchmark of Dftb3 for Organic Molecules. J. Chem. Theory Comput. 2013, 9, 338−354. (82) Alessandri, R.; Uusitalo, J. J.; de Vries, A. H.; Havenith, R. W. A.; Marrink, S. J. Bulk Heterojunction Morphologies with Atomistic Resolution from Coarse-Grain Solvent Evaporation Simulations. J. Am. Chem. Soc. 2017, 139, 3697−3705. (61) Gaus, M.; Lu, X.; Elstner, M.; Cui, Q. Parameterization of Dftb3/3ob for Sulfur and Phosphorus for Chemical and Biological Applications. J. Chem. Theory Comput. 2014, 10, 1518−1537. ̌ pp y p (62) Kubillus, M.; Kubař, T.; Gaus, M.; Řezáč, J.; Elstner, M. Parameterization of the Dftb3 Method for Br, Ca, Cl, F, I, K, and Na in Organic and Biological Systems. J. Chem. Theory Comput. 2015, 11, 332−342. (83) Untilova, V.; et al. High Thermoelectric Power Factor of Poly(3-Hexylthiophene) through in-Plane Alignment and Doping with a Molybdenum Dithiolene Complex. Macromolecules 2020, 53, 6314−6321. (63) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (Dft-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010, 132, No. 154104. (84) Bruchlos, K.; Trefz, D.; Hamidi-Sakr, A.; Brinkmann, M.; Heinze, J.; Ruff, A.; Ludwigs, S. Poly(3-Hexylthiophene) Revisited − Influence of Film Deposition on the Electrochemical Behaviour and Energy Levels. Electrochim. Acta 2018, 269, 299−311. (64) Becke, A. D.; Johnson, E. R. A Density-Functional Model of the Dispersion Interaction. J. Chem. Phys. 2005, 123, No. 154101. (85) Hofmann, A. I.; Kroon, R.; Yu, L.; Müller, C. Highly Stable Doping of a Polar Polythiophene through Co-Processing with Sulfonic Acids and Bistriflimide. J. Mater. Chem. C 2018, 6, 6905− 6910. The Journal of Physical Chemistry B (65) Niehaus, T. A.; Suhai, S.; Della Sala, F.; Lugli, P.; Elstner, M.; Seifert, G.; Frauenheim, T. Tight-Binding Approach to Time- Dependent Density-Functional Response Theory. Phys. Rev. B: Condens. Matter Mater. Phys. 2001, 63, No. 085108. (86) Volkov, A. V.; Singh, S. K.; Stavrinidou, E.; Gabrielsson, R.; Franco-Gonzalez, J. F.; Cruce, A.; Chen, W. M.; Simon, D. T.; Berggren, M.; Zozoulenko, I. V. Spectroelectrochemistry and Nature of Charge Carriers in Self-Doped Conducting Polymer. Adv. Electron. Mater. 2017, 3, No. 1700096. y (66) Heffner, G. W.; Pearson, D. S. Molecular Characterization of Poly(3-Hexylthiophene). Macromolecules 1991, 24, 6295−6299. (67) McCulloch, B.; Ho, V.; Hoarfrost, M.; Stanley, C.; Do, C.; Heller, W. T.; Segalman, R. A. Polymer Chain Shape of Poly(3- Alkylthiophenes) in Solution Using Small-Angle Neutron Scattering. Macromolecules 2013, 46, 1899−1907. (68) Nagai, M.; Huang, J.; Zhou, T.; Huang, W. Effect of Molecular Weight on Conformational Characteristics of Poly(3-Hexyl Thio- phene). J. Polym. Sci., Part B: Polym. Phys. 2017, 55, 1273−1277. ( ) (69) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. Energies, Structures, and Electronic Properties of Molecules in Solution with the C-Pcm Solvation Model. J. Comput. Chem. 2003, 24, 669−681. J p (70) Ghosh, S.; Gueskine, V.; Berggren, M.; Zozoulenko, I. V. Electronic Structures and Optical Absorption of N-Type Conducting Polymers at Different Doping Levels. J. Phys. Chem. C 2019, 123, 15467−15476. (71) Franco-Gonzalez, J. F.; Zozoulenko, I. V. Molecular Dynamics Study of Morphology of Doped Pedot: From Solution to Dry Phase. J. Phys. Chem. B 2017, 121, 4299−4307. y (72) Modarresi, M.; Franco-Gonzalez, J. F.; Zozoulenko, I. Computational Microscopy Study of the Granular Structure and Ph Dependence of Pedot:Pss. Phys. Chem. Chem. Phys. 2019, 21, 6699− 6711. (73) Sahalianov, I.; Singh, S. K.; Tybrandt, K.; Berggren, M.; Zozoulenko, I. The Intrinsic Volumetric Capacitance of Conducting Polymers: Pseudo-Capacitors or Double-Layer Supercapacitors? RSC Adv. 2019, 9, 42498−42508. (74) Jacobs, I. E.; Cendra, C.; Harrelson, T. F.; Bedolla Valdez, Z. I.; Faller, R.; Salleo, A.; Moulé, A. J. Polymorphism Controls the Degree of Charge Transfer in a Molecularly Doped Semiconducting Polymer. Mater. Horiz. 2018, 5, 655−660. (75) Cohen, A. J.; Mori-Sánchez, P.; Yang, W. Insights into Current Limitations of Density Functional Theory. Science 2008, 321, 792. (76) Yu, H. S.; Li, S. L.; Truhlar, D. G. Perspective: Kohn-Sham D F l Th D d S J Ch Ph (76) Yu, H. S.; Li, S. L.; Truhlar, D. G. https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B C 2008, 112, 8408−8418. (58) Aradi, B.; Hourahine, B.; Frauenheim, T. Dftb+, a Sparse Matrix-Based Implementation of the Dftb Method. J. Phys. Chem. A 2007, 111, 5678−5684. J y (38) Salzner, U.; Aydin, A. Improved Prediction of Properties of Π- Conjugated Oligomers with Range-Separated Hybrid Density Func- tionals. J. Chem. Theory Comput. 2011, 7, 2568−2583. 11292 https://dx.doi.org/10.1021/acs.jpcb.0c08757 J. Phys. Chem. B 2020, 124, 11280−11293 The Journal of Physical Chemistry B The Journal of Physical Chemistry B Perspective: Kohn-Sham Density Functional Theory Descending a Staircase. J. Chem. Phys. 2016, 145, No. 130901. (77) Salaneck, W. R.; Friend, R. H.; Brédas, J. L. Electronic Structure of Conjugated Polymers: Consequences of Electron−Lattice Cou- pling. Phys. Rep. 1999, 319, 231−251. p g y p (78) Geskin, V. M.; Brédas, J.-L. Polaron Pair Versus Bipolaron on Oligothiophene Chains: A Theoretical Study of the Singlet and Triplet States. ChemPhysChem 2003, 4, 498−505. p y (79) Bredas, J. L.; Street, G. B. Polarons, Bipolarons, and Solitons in Conducting Polymers. Acc. Chem. Res. 1985, 18, 309−315. ( ) Conducting Polymers. Acc. Chem. Res. 1985, 18, 309−315. (80) Muñoz, W. A.; Singh, S. K.; Franco-Gonzalez, J. F.; Linares, M.; Crispin, X.; Zozoulenko, I. V. Insulator to Semimetallic Transition in 11293

https://openalex.org/W2994740670

http://cds.cern.ch/record/2725295/files/1912.09433.pdf

English

Characterizing the cosmological gravitational wave background: Anisotropies and non-Gaussianity

Physical review. D/Physical review. D.

cc-by

Characterizing the Cosmological Gravitational Wave Background: Anisotropies and non-Gaussianity arXiv:1912.09433v1 [astro-ph.CO] 19 Dec 2019 09433v1 [astro-ph.CO] 19 Dec 2019 Nicola Bartolo a,b,c , Daniele Bertacca a,b , Sabino Matarrese a,b,c,d Marco Peloso a,b , Angelo Ricciardone b , Antonio Riotto e,f , Gianmassimo Tasinato g arXiv:1912.09433v1 [astro-ph.CO] aDipartimento di Fisica e Astronomia “Galileo Galilei”, Universit´a di Padova, 35131 Padova, Italy aDipartimento di Fisica e Astronomia “Galileo Galilei”, Universit´a di Padova, 35131 Padova, Italy cINAF - Osservatorio Astronomico di Padova, I-35122 Padova, Italy dGran Sasso Science Institute, I-67100 L’Aquila, Italy eDepartment of Theoretical Physics and Center for Astroparticle Physics (CAP), CH-1211 Geneva 4, Switzerland f eDepartment of Theoretical Physics and Center for Astroparticle Physics (CAP), CH-1211 Geneva 4, Switzerland fCERN,Theoretical Physics Department, Geneva, Switzerland gDepartment of Physics, Swansea University, Swansea, SA2 8PP, UK gDepartment of Physics, Swansea University, Swansea, SA2 8PP, UK Abstract. A future detection of the Stochastic Gravitational Wave Background (SGWB) with GW experiments is expected to open a new window on early universe cosmology and on the astrophysics of compact objects. In this paper we study SGWB anisotropies, that can offer new tools to discriminate between different sources of GWs. In particular, the cosmological SGWB inherits its anisotropies both (i) at its production and (ii) during its propagation through our perturbed universe. Concerning (i), we show that it typically leads to anisotropies with order one dependence on frequency. We then compute the effect of (ii) through a Boltzmann approach, including contributions of both large-scale scalar and tensor linearized perturbations. We also compute for the first time the three-point function of the SGWB energy density, which can allow one to extract information on GW non-Gaussianity with interferometers. Finally, we include non-linear effects associated with long wavelength scalar fluctuations, and compute the squeezed limit of the 3-point function for the SGWB density contrast. Such limit satisfies a consistency relation, conceptually similar to what found in the literature for the case of CMB perturbations. Contents 1 Introduction 1 2 Boltzmann equation for gravitational waves 3 3 Spherical harmonics decomposition 7 3.1 Initial condition term and q−dependent anisotropies 7 3.2 Scalar sourced term 8 3.3 Tensor sourced term 9 3.4 Summary of the three contributions 9 4 Correlators of GW anisotropies and SGWB non-Gaussianity 10 4.1 Angular power spectrum of GW energy density 11 4.2 Angular bispectrum of GW energy density 11 4.3 Reduced Bispectrum and estimation 13 5 An example: the axion-inflation case 14 6 Squeezed limit and consistency relations of the SGWB 16 6.1 Long wavelength modes as coordinate transformations 17 6.2 Coordinate transformations and the GW distribution function 20 6.3 The squeezed limit of 3-point correlation functions 21 7 Conclusion 24 A Computation of the tensor sourced term 26 B Tensor contribution to the GW bispectrum 28 C Comparison with the CMB 29 C Comparison with the CMB 1 Introduction The current ground based interferometers are close to reach the expected sensitivity to de- tect the Stochastic Gravitational Wave Background (SGWB) from unresolved astrophysical sources [1]. Future space-based (such as LISA [2] and DECIGO [3]) and earth-based (like Einstein Telescope [4, 5] and Cosmic Explorer [6]) interferometers have the potential to detect the SGWB of cosmological origin (see [7–10] for reviews of possible cosmological sources). It is likely that a detection of a cosmological SGWB background will require the ability to dis- criminate it against the astrophysical signal. Astrophysical GW background (AGWB) arises from the superposition of the signals emitted by a large population of unresolved sources that are mainly dominated by two types of events: (i) the periodic long lived sources (e.g. the early inspiraling phase of binary systems) where the frequency is expected to evolve very slowly compared to the observation time; (ii) the short-lived burst sources, e.g. core collapse to neutron stars or black holes, oscillation modes, r-mode instabilities in rotating neutron stars, magnetars and super-radiant instabilities (for example, see [11, 12]). Several techniques – 1 – have been developed to distinguish among the various backgrounds. The most obvious tool for this component separation is the frequency dependence [13], as several cosmological mech- anisms are peaked at some given characteristic scale. However, future detectors will allow for a better angular resolution of anisotropies of the astrophysical background. Therefore, another tool could be the directionality dependence of the SGWB [14–19] and, as we explore here, its statistics. In this work, we discuss graviton propagation through a Boltzmann approach [15] as it is typically done for the CMB. Specifically, we construct and evolve the equation for the distribution f of gravitons in a FLRW background, plus first order scalar and tensor perturbations (we also consider how non-linear effects for the specific case of squeezed non- Gaussianity, as we discuss at the end of this Introduction). At the unperturbed level, following the isotropy and homogeneity of the background, the distribution depends only on time and on the GW frequency p/2π (where ⃗p is the physical momentum of the gravitons) through the combination q ≡p a, where a is the scale factor of the universe. Namely, the gravitons freely propagate, and their physical momentum redshifts during the propagation. This property is shared by any free massless particles, and, in particular, also by the CMB photons. 1 Introduction On the other hand, differently from the photon distribution, the initial population of gravitons is not expected to be thermal (as we have in mind production mechanisms, such as inflation [20, 21], phase transitions [22], or enhanced density perturbations leading to primordial black holes (PBH) [23, 24, 31], which occur at energies well below the Planck scale) which leaves in the distribution a sort of “memory” of the initial state. As we show, the fact that the spectrum is non thermal generically results in angular anisotropies that have an order one dependence on the GW frequency. This is in contrast with the CMB case, for which this dependence only arises at second order in perturbation theory. This initial state will in general be anisotropic, as no mechanism of GW production can be perfectly homogeneous. Additional anisotropies are induced by the GW propagation in the perturbed universe. As we are interested in large scale, we work in a regime of a large hierarchy q ≫k between the GW (comoving) momentum q and the (comoving) momentum k of the large scale perturbations. We confirm that in the angular power spectrum, the Sachs Wolfe (SW) effect is dominating on large scales also for gravitons, while the Integrated Sachs-Wolfe (ISW) contribution is subdominant. We employ this approach to study the non-Gaussianity of the SGWB energy density. Although we are not aware of any dedicated analysis in this sense, it is reasonable to expect that the SGWB produced by incoherent astrophysical sources is Gaussian, due to the central limit theorem. In light of this fact, a measurement of non-Gaussianity would be a signal of large scale coherency, that would likely point to a cosmological origin of the signal. Previous works showed that inflation can result in a sizeable an nonvanishing 3−point function h3 for the graviton wave function, but that this is generically non observable in interferometers [23, 24], due to the decoherence of the phase the GW wave-function h induced by the GW propagation, and due to the finite duration of the measurement (see [26] for a possible exception to this conclusion, occurring for a very specific shape of the bispectrum). Since the phase does not affect the GW energy density, we argue that the energy density is a much better variable to study the statistics of the SGWB. 1 Introduction Also in this case, the of non-Gaussianity can be induced both by the production mechanism and the propagation. As an example of the former, in ref. [31] we recently computed the 3−point function of the SGWB energy density that arises in presence of non-Gaussianity of the scalar perturbations of the local shape (in presence of this non-Gaussianity, a long-scale mode of momentum k can modulate the power – 2 – of the short-scalr scalar perturbations that are responsible for the PBH formation). Here we study the 3−point function induced by the GW propagation. This is also proportional to the non-Gaussianity of the scalar perturbations. In this sense, the SGWB can be used as a novel probe (beyond the CMB and the LSS) of the non-Gaussianity of the scalar perturbations. Although in most of this work we limit our attention to linearized fluctuations, in Sec- tion 6 we consider non-linear effects induced by long-wavelength scalar perturbations, which modulate correlation functions involving short-wavelength modes. We make use of a powerful method first introduced by Weinberg in [27], which focusses on adiabatic systems, and iden- tifies the effects of long modes with an appropriate coordinate transformation. Applying this method to our set-up, we compute how non-linearities induce a non-vanishing squeezed limit of the 3-point function for the SGWB density contrast. We determine how such squeezed limit depends on the scale-dependence of the spectrum of primordial scalar fluctuations; on the momentum dependence of the background SGWB distribution; and on the time, scale, and direction dependence of the scalar transfer functions connecting primordial to late-time adiabatic scalar fluctuations. The paper is organized as follows. In Section 2 we present the computation and the formal solution of the Boltzmann equation for GW propagation. In Section 3 we decompose the formal solution in spherical harmonics, paralleling a treatment that is familiar in the study of CMB perturbations. In Section 4 we compute the angular power spectrum and bispectrum of the SGWB perturbations. In Section 5 we review one physical mechanism that can result in a sizeable cosmological SGWB with some degree of anisotropy. In Section 6 we study non-linear effects on the squeezed bispectrum. These results are discussed and summarized in Section 7. The paper is concluded by three appendices. Appendix A contains the details of the computation of the anisotropies due to the large-scale tensor perturbations. 1 Introduction Appendix B provides some intermediate steps on the computation of the GW bispectrum induced by tensor modes. Finally, Appendix C presents an immediate connection between our formal solutions and the CMB results obtained in the case of initial thermal state. Part of the results contained in the present work were also summarized in the Rapid Communication [58]. 2 Boltzmann equation for gravitational waves We consider first order perturbations around a Friedmann-Lemaitre-Robertson-Walker (FLRW) background in the Poisson gauge ds2 = a2(η)  −e2Φdη2 + (e−2Ψδij + χij)dxidxj , (2.1) (2.1) where a(η) is the scale factor as a function of the conformal time η. Φ and Ψ are scalar perturbations while χij represent the transverse-traceless (TT) tensor perturbations. We neglect linear vector modes since they are not produced at first order in standard mechanisms for the generation of cosmological perturbations (as scalar field inflation), and we consider tensor modes at linearised order. Given the statistical nature of the GW we can define a distribution function of gravitons as f = f(xµ, pµ), which is function of their position xµ and momentum pµ = dxµ/dλ, where λ is an affine parameter along the GW trajectory. As we will see, observables as number density, spectral energy density, and flux (directions) can be derived from the distribution function. The graviton distribution function obeys the Boltzmann equation L[f] = C[f(λ)] + I[f(λ)] , (2.2) (2.2) – 3 – where L ≡d/dλ is the Liouville term, while C and I account, respectively, for the collision of GWs along their path, and for their emissivity from cosmological and astrophysical sources [15]. The collision among GWs affects the distribution at higher orders (in an expansion series in the gravitational strength 1/MP , where MP is the Planck mass) with respect to the ones we are considering, and they can be disregarded in our analysis (see [29] and references therein for a discussion of collisional effects involving gravitons). The emissivity can be due to astrophysical processes (such as black hole merging) in the relatively late universe, as well as cosmological processes, such as inflation or phase transitions. In this work we are only interested in the stochastic GW background of cosmological origin, so we treat the emissivity term as an initial condition on the GW distribution. This leads us to study the free Boltzmann equation, df/dη = 0 in the perturbed universe df dη = ∂f ∂η + ∂f ∂xi dxi dη + ∂f ∂q dq dη + ∂f ∂ni dni dη = 0 , (2.3) (2.3) where ˆn ≡ˆp is the GW direction of motion, and where we have used the comoving momentum q ≡|⃗p|a (as opposed to the physical one, used in [15, 30]). This simplifies the equations by factorizing out the universe expansion. 2 Boltzmann equation for gravitational waves The first two terms in (2.3) encode free streaming, that is the propagation of perturbations on all scales. At higher order this term also includes gravitational time delay effects. The third term causes the red-shifting of gravitons, including the Sachs-Wolfe (SW), integrated Sachs-Wolfe (ISW) and Rees-Sciama (RS) effects. The fourth term vanishes to first order, and describes the effect of gravitational lensing. We shall refer to these terms as the free-streaming, redshift and lensing terms, respectively, as customarily done in CMB physics. y p y Keeping only the terms up to first order in the perturbations, Eq. (2.3) gives ∂f ∂η + ni ∂f ∂xi + ∂Ψ ∂η −ni ∂Φ ∂xi + 1 2ninj ∂χij ∂η  q ∂f ∂q = 0 , (2.4) (2.4) where we have followed the standard procedure developed for the CMB in [30, 39]. The distribution function f can be expanded as , q, ni
= ¯f (q) + f(1) η, xi, q, ni
+ .... ≡¯f (q) −q ∂¯f ∂q Γ η, xi, q, ni
+ .... , (2.5) (2.5) where the dominant, homogeneous and isotropic contribution ¯f(q) solves the zeroth order Boltzmann equation. The function f(1)(η, xi, q, ni) is the solution of the first order equa- tion, and the ellipses denote the higher order solutions in a perturbative expansion. In this expression we have parameterized the first order solution in terms of the function Γ, so to simplify the first order Boltzmann equation [15]. For a thermal distribution with temper- ature T, one finds Γ = δT/T. This is particularly the case for the CMB, for which, due to the thermalization, the temperature anisotropies are frequency-independent up to second order in the perturbations. For gravitons, as we already mentioned, the collisional term is extremely small, and, for a generic production mechanism, Γ generically retains an order one dependence on frequency (as we show below, also for the GW case the propagation effects induce frequency-independent perturbations at linear order). The zeroth order homogeneous Boltzmann equation simply reads ∂¯f/∂η = 0, and it is solved by any distribution that is function only of the comoving momentum q, namely f = ¯f (q). In our approach this solution is simply given as the homogeneous part of the – 4 – initial condition. 2 Boltzmann equation for gravitational waves As a consequence, the physical momentum of the individual gravitons redshifts proportionally to 1/a, and the physical graviton number density n ∝ R d3p ¯f(q) is diluted as a−3 as the universe expands. This is also the case for CMB photons, whose distribution function ¯fCMB = (ep/T −1)−1 is only controlled by the ratio p/T ∝a p = q, where T is the temperature of the CMB bath. We see that these rescalings with a are a consequence of the free particle propagation in the expanding FLRW background, and they do not rely on the distribution being thermal. As anticipated, from the graviton distribution function, evaluated at the present time η0, we can compute the SGWB energy density ρGW (η0, ⃗x) = 1 a4 0 Z d3q q f (η0, ⃗x, q, ˆn) ≡ρcrit,0 Z d ln q ΩGW (⃗x, q) , (2.6) (2.6) where we have introduced the spectral energy density ΩGW and the critical density ρcrit = 3H2M2 p . Here H ≡(1/a2) da/dη is the Hubble rate. Following standard conventions, the suffix 0 denotes a quantity evaluated at the present time. Contrary to most of the studies of the SGWB, that assume a homogeneous ΩGW, in our case the GW energy density depends on space. We denote the homogeneous component of ΩGW as ¯ΩGW (q) ≡ 4π ρcrit,0  q a0 4 ¯f (q) , (2.7) (2.7) For the full spectral energy density, we define For the full spectral energy density, we define ΩGW ≡1 4π Z d2ˆn ωGW(⃗x, q, ˆn) , (2.8) (2.8) and we introduce the SGWB density contrast δGW ≡ωGW(⃗x, q, ˆn) −¯ΩGW(q) ¯ΩGW(q) =  4 −∂ln ¯ΩGW (q) ∂ln q  Γ (η0, ⃗x, q, ˆn) . (2.9) (2.9) In terms of the function Γ, the first order Boltzmann equation reads [15] ∂Γ ∂η + ni ∂Γ ∂xi = S η, xi, ni
, (2.10) (2.10) where where S η, xi, ni
= ∂Ψ ∂η −ni ∂Φ ∂xi −1 2ninj ∂χij ∂η is the source function which includes the physical effects due to cosmological scalar and tensor inhomogeneities. We note that the source is q−independent (thus showing that the anisotropies arising at first order from propagation effects are frequency-independent, as we anticipated). 2 Boltzmann equation for gravitational waves To solve this equation, it is convenient to Fourier transform with respect to spatial coordinates, 3 Γ ≡ Z d3k (2π)3 ei⃗k·⃗xΓ  η, ⃗k, q, ˆn  , (2.11) (2.11) and analogously for the other variables (we use the same notation for a field and for its Fourier transform, as the context always clarifies which object we are referring to). This leads to Γ′ + i k µ Γ = S(η,⃗k, ˆn) , (2.12) (2.12) – 5 – where from now on prime denotes a derivative with respect to conformal time, and where we denote by ˆ where from now on prime denotes a derivative with respect to conformal time, and where we denote by ˆ µ ≡ˆk · ˆn , (2.13) (2.13) µ ≡ˆk · ˆn , the cosine of the angle between the Fourier variable ⃗k and the direction of motion ˆn of the GW. In Fourier space the source term reads the cosine of the angle between the Fourier variable ⃗k and the direction of motion ˆn of the GW. In Fourier space the source term reads S = Ψ′ −ik µ Φ −1 2ninj χ′ ij . (2.14) (2.14) rmation in mind, Eq. (2.12) is readily integrated to give  = eikµ(ηin−η) Γ  ηin, ⃗k, q, ˆn  Γ  η, ⃗k, q, ˆn  = eikµ(ηin−η) Γ  ηin, ⃗k, q, ˆn  + Z η ηin dη′ eikµ(η′−η)   dΨ  η′, ⃗k  dη′ −ikµΦ  η′, ⃗k  −1 2ninj ∂χij  η′, ⃗k  ∂η′  . + Z η ηin dη′ eikµ(η′−η)   dΨ  η′, ⃗k  dη′ −ikµΦ  η′, ⃗k  −1 2ninj ∂χij  η′, ⃗k  ∂η′  . 2 Boltzmann equation for gravitational waves (2.15) (2.15) We integrate the second term in the second line by parts, and obtain We integrate the second term in the second line by parts, and obtain Γ  η, ⃗k, q, ˆn  = eikµ(ηin−η) h Γ  ηin, ⃗k, q, ˆn  + Φ  ηin, ⃗k i −Φ  η, ⃗k  + Z η ηin dη′ eikµ(η′−η)    d h Ψ  η′, ⃗k  + Φ  η′, ⃗k i dη′ −ikµΦ  η′, ⃗k  −1 2ninj ∂ˆχij  η′, ⃗k  ∂η′   , (2.16)  (2.16) with the last two terms in the first line being the boundary terms of this integration. In the following section, we decompose the ˆn-dependence of the solution (representing the arrival direction of the GW on our sky) in spherical harmonics. As we are not interested in the monopole term, we can disregard the −Φ(η, ⃗k) contribution to the solution, and write Γ  η, ⃗k, q, ˆn  ≡ Z η ηin dη′ eikµ(η′−η) × ( h Γ  η′, ⃗k, q, ˆn  + Φ  η′, ⃗k i δ η′ −ηin
+ ∂ h Ψ  η′, ⃗k  + Φ  η′, ⃗k i ∂η′ −1 2ninj ∂ˆχij  η′, ⃗k  ∂η′ ) . (2.17) (2.17) The first term, which was disregarded in [15], carries the “memory” of the initial conditions. Due to this term, the GW energy density anisotropies are generically dependent on the frequency q. We discuss an example of this fact in Section 5, where we study the SGWB produced in axion inflation. Generally, this term has also a dependence on ˆn. This implies that the solution has a dependence on the direction ˆn, which is more general than the one arising from the projec- tion of ⃗k on the line of sight ˆn. (Indeed, the remaining terms in Eq. (2.17) depend on ˆn only through the µ ≡ˆk · ˆn combination. Thanks to this fact, they result in angular corre- lators that are statistically isotropic (as we show in the next two sections).) On the other hand, the angular dependence present in the first term of (2.17) could result in statistically anisotropic correlators (specifically, 2-point and 3-point correlators) that have a more general – 6 – dependence on the multipoles coefficients ℓi and mi than Eqs. (4.3). 2 Boltzmann equation for gravitational waves This would indicate an overall anisotropy of the mechanism responsible for the GW across the entire universe, and, ultimately, a departure from an exact FLRW geometry. While we believe that this can be an interesting topic for future exploration, the present work focuses on the statistically isotropic case, and we assume an initial condition of the form Γin = Γ(ηin, ⃗k, q), which guarantees such a condition. 3 Spherical harmonics decomposition We separate the solution (2.17) in three terms Γ  η, ⃗k, q, ˆn  = ΓI  η, ⃗k, q, ˆn  + ΓS  η, ⃗k, ˆn  + ΓT  η, ⃗k, ˆn  , (3.1) (3.1) where I, S, and T stand for Initial, Scalar and Tensor sourced terms respectively and they are given by ΓI  η, ⃗k, q, ˆn  = eikµ(ηin−η)Γ  ηin, ⃗k, q  , ΓS  η, ⃗k, ˆn  = Z η ηin dη′ eikµ(η′−η)  Φ  η′, ⃗k  δ η′ −ηin
+ ∂ h Ψ  η′, ⃗k  + Φ  η′, ⃗k i ∂η′  , ΓT  η, ⃗k, ˆn  = −ni nj 2 Z η ηin dη′ eikµ(η′−η) ∂ˆχij  η′, ⃗k  ∂η′ . (3.2) (3.2) Similarly to what is usually done for the CMB, in order to compute the angular power spectrum, in an all-sky analysis we decompose the fluctuations using spin-0 or spin-2 spherical harmonics. Since Γ is a scalar, we can express it as Similarly to what is usually done for the CMB, in order to compute the angular power spectrum, in an all-sky analysis we decompose the fluctuations using spin-0 or spin-2 spherical harmonics. Since Γ is a scalar, we can express it as Γ (ˆn) = X ℓ ℓ X m=−ℓ Γℓm Yℓm (ˆn) , inverted by Γℓm = Z d2n Γ (ˆn) Y ∗ ℓm (ˆn) , (3.3) (3.3) where we recall that ˆn is the direction of motion of the GWs. More specifically: where we recall that ˆn is the direction of motion of the GWs. More specifically: Γℓm = Z d2n Y ∗ ℓm (ˆn) Z d3k (2π)3 ei⃗k·⃗x h ΓI  η, ⃗k, q, ˆn  + ΓS  η, ⃗k, ˆn  + ΓT  η, ⃗k, ˆn Γℓm = Z d2n Y ∗ ℓm (ˆn) Z d3k (2π)3 ei⃗k·⃗x h ΓI  η, ⃗k, q, ˆn  + ΓS  η, ⃗k, ˆn  + ΓT  η, ⃗k, ˆn i ≡Γℓm I + Γℓm S + Γℓm T . (3 Γℓm = Z d2n Y ∗ ℓm (ˆn) Z d k (2π)3 eik·⃗x h ΓI  η, ⃗k, q, ˆn  + ΓS  η, ⃗k, ˆn  + ΓT  η, ⃗k, ˆn i ≡Γℓm,I + Γℓm,S + Γℓm,T . 3 Spherical harmonics decomposition (3.4) (3.4) ≡Γℓm,I + Γℓm,S + Γℓm,T . 3.1 Initial condition term and q−dependent anisotropies 3.1 Initial condition term and q−dependent anisotropies Let us first evaluate the initial condition term Let us first evaluate the initial condition term Let us first evaluate the initial condition term Γℓm,I = Z d3k (2π)3 ei⃗k·⃗x0 Γ  ηin, ⃗k, q  Z d2n Y ∗ ℓm (ˆn) e−ik(η0−ηin)ˆk·ˆn . (3.5) (3.5) Following the standard treatment for CMB anisotropies [30], we make use of the ident Following the standard treatment for CMB anisotropies [30], we make use of the identity ℓ e−ik·y = X ℓ (−i)ℓ(2ℓ+ 1) jℓ(ky) Pℓ  ˆk · ˆy  = 4π X ℓ ℓ X m=−ℓ (−i)ℓjℓ(ky) Yℓm(ˆk) Y ∗ ℓm (ˆy) , (3.6) – 7 – (where jℓand Pℓare, respectively, spherical Bessel functions and Legendre polynomial) so to obtain Z d3k Γℓm,I = 4π (−i)l Z d3k (2π)3 ei⃗k·⃗x0 Γ  ηin, ⃗k, q  Y ∗ ℓm  ˆk  jℓ(k (η0 −ηin)) . (3.7) (3.7) Here ⃗x0 denotes our location (that can be set to the origin), η0 denotes the present time, and ηin the initial time. Once again we stress the peculiar property of the initial condition, namely its dependence on the frequency q. In Section 4 we discuss how this imprints the SGWB angular spectrum. 3.2 Scalar sourced term A second source of anisotropy is due to the GW propagation in the large-scale scalar pertur- bations of the universe (the wavenumber of these perturbations k is many order of magnitudes smaller than the GW frequency q, and the GW acts as a probe of this large-scale background). As long as the scalar perturbation is in the linear regime (which is the case for the large-scale modes that leave an impact on the large-scale anisotropies of our interest), we can express it [30] as a transfer function (a deterministic function that encodes the time-dependence of the perturbations) times a stochastic variable ζ. This assumes the absence of isocurvature modes, and, in particular, of anisotropic stresses, as for example those due to the relic neu- trinos. This also assumes that the statistical properties of ζ have been set well before the propagation stage that we are considering (for instance during inflation, or during some early phase transition). Therefore, the scalar perturbations are Φ  η, ⃗k  = TΦ (η, k) ζ  ⃗k  , Ψ  η, ⃗k  = TΨ (η, k) ζ  ⃗k  . (3.8) (3.8) Under the above assumptions, TΦ(η, k) = TΨ(η, k). However, we keep these two terms as distinct in our intermediate computations, so that the present analysis can be most easily generalized, if one wishes to introduce more general sources. his in mind, the scalar sourced term becomes With this in mind, the scalar sourced term becomes ΓS  η0, ⃗k, ˆn  = Z η0 ηin dη′ eikµ(η′−η0)  TΦ η′, k
δ η′ −ηin
+ ∂[TΨ (η′, k) + TΦ (η′, k)] ∂η′  ζ  ⃗k  ≡ Z η0 ηin dη′ e−ikµ(η0−η′)TS η′, k
ζ  ⃗k  , (3.9) (3.9) and we note that we are assuming a single adiabatic mode (i.e. ζ(⃗k) is the operator associated with the conserved curvature perturbation at super-horizon scales). 3.3 Tensor sourced term Finally, the third contribution Γℓm,T is due to the GW propagation in the large-scale tensor modes Finally, the third contribution Γℓm,T is due to the GW propagation in the large-scale tensor modes Γℓm,T = − Z d2n Y ∗ ℓm (ˆn) Z d3k (2π)3 ei⃗k·⃗x0 ni nj 2 Z η ηin dη′ eikµ(η′−η0) ∂χij  η′, ⃗k  ∂η′ . (3.11) (3.11) To evaluate such term we decompose the tensor modes in right and left-handed (respectively λ = ±2) circular polarizations (see e.g. [43]), χij ≡ X λ=±2 eij,λ  ˆk  χ (η, k) ξλ ki
. (3.12) (3.12) The three factors involved in each term are, respectively, the tensor circular polarization operator, the tensor mode function (equal for the two polarizations), and the stochastic variable for that tensor polarization (that is the analog of ζ we discussed in the scalar case). Inserting this decomposition in Eq. (3.11), a lengthy algebra, that we report in Appendix A, leads to A, leads to Γℓm,T = π (−i)ℓ s (ℓ+ 2)! (ℓ−2)! Z d3k (2π)3 ei⃗k·⃗x0 X λ=±2 −λY ∗ ℓm (Ωk) ξλ  ⃗k  Z η0 ηin dη χ′ (η, k) jℓ(k (η0 −η)) k2 (η0 −η)2 , (3.13) (3.13) which is formally analogous to Eq. (3.10), with the product ˆζ Y ∗ ℓm replaced by the combina- tion P λ=±2 ˆξλ(⃗k) −λY ∗ ℓm(Ωk), involving the spin-2 spherical harmonics, and with the scalar transfer function replaced by the tensor one. 3.2 Scalar sourced term Proceeding as above, Γℓm,S = 4π (−i)l Z d3k (2π)3 ei⃗k·⃗x0 ζ  ⃗k  Y ∗ ℓm  ˆk  ( TΦ (ηin, k) jℓ(k (η0 −ηin)) Γℓm,S = 4π (−i)l Z d3k (2π)3 ei⃗k·⃗x0 ζ  ⃗k  Y ∗ ℓm  ˆk  ( TΦ (ηin, k) jℓ(k (η0 −ηin)) + Z η0 ηin dη′ ∂[TΨ (η′, k) + TΦ (η′, k)] ∂η′ jℓ k η0 −η′
+ Z η0 ηin dη′ ∂[TΨ (η′, k) + TΦ (η′, k)] ∂η′ jℓ k η0 −η′

) . (3.10) As we can see, also the SGWB, feels, similarly to the CMB, a Sachs-Wolfe and integrated Sachs-Wolfe effect, which are represented by the first and the second term in (3.10), respec- tively. – 8 – 4 Correlators of GW anisotropies and SGWB non-Gaussianity We now compute the 2-point ⟨ΓℓmΓ∗ ℓ′m′⟩and the 3-point ⟨Γℓ1m1Γℓ2m2Γℓ3m3⟩angular correla- tors of the solutions (3.14). The statistical operators entering in these solutions are the four momentum-dependent quantities Γ(ηin, ⃗k, q), ζ(⃗k), ξR(⃗k), and ξL(⃗k), while the other terms encode deterministic effects such has the time evolution of the large-scale modes (in the lin- earized theory of the cosmological perturbations) and the projection of the GW anisotropies in the harmonic space. In this study, we assume that the stochastic variables are nearly Gaussian, with the 2-point functions D Γ  ηin, ⃗k, q  Γ∗ ηin, ⃗k′, q E = 2π2 k3 PI (q, k) (2π)3 δ  ⃗k −⃗k′ , D ζ  ⃗k  ζ∗ ⃗k′E = 2π2 k3 Pζ (k) (2π)3 δ  ⃗k −⃗k′ , D ξλ  ⃗k  ξ∗ λ′  ⃗k′E = 2π2 k3 Pλ (k) δλ,λ′ (2π)3 δ  ⃗k −⃗k′ , (4.1) D ξλ  ⃗k  ξ∗ λ′  ⃗k′E = 2π2 k3 Pλ (k) δλ,λ′ (2π)3 δ  ⃗k −⃗k′ , (4.1) (4.1) and a subdominant 3-point component and a subdominant 3-point component D Γ  ηin, ⃗k, q  Γ∗ ηin, ⃗k′, q  Γ∗ ηin, ⃗k′′, q E = BI q, k, k′, k′′
(2π)3 δ  ⃗k + ⃗k′ + ⃗k′′ D ζ  ⃗k  ζ  ⃗k′ ζ  ⃗k′′E = Bζ k, k′, k′′
(2π)3 δ  ⃗k + ⃗k′ + ⃗k′′ D ξλ  ⃗k  ξλ′  ⃗k′ ξλ′′  ⃗k′′E = Bλ  ⃗k, ⃗k′, ⃗k′′ δλ,λ′ δλ,λ′′ (2π)3 δ  ⃗k + ⃗k′ + ⃗k′′ (4 2  (4.2) The assumption of nearly Gaussian modes is experimentally verified for the large-scale per- turbations of ζ and of ξλ, as obtained from the CMB data [44]. We assume that this is the case also for the initial condition term. The expressions (4.1) and (4.2) can be readily used to compute the angular correlators of the solutions in (3.14). Moreover, for simplicity of exposition, we have here assumed that the various terms are not cross-correlated. These results in separate sets of correlators for the three terms in (3.14). This assumption can be easily relaxed, and in fact, we did so in [31] where we studied the anisotropic distribution of the GW originated in models with primordial black holes, as we review in Section 5. 3.4 Summary of the three contributions The results derived in the three previous subsections can be written in the (slightly) more compact form Γℓm,I (q) = 4π (−i)l Z d3k (2π)3 ei⃗k·⃗x0 Γ  ηin, ⃗k, q  Y ∗ ℓm  ˆk  jℓ(k (η0 −ηin)) , Γℓm,S = 4π (−i)l Z d3k (2π)3 ei⃗k·⃗x0 ζ  ⃗k  Y ∗ ℓm  ˆk  T S ℓ(k, η0, ηin) , Γℓm,T = 4π (−i)l Z d3k (2π)3 ei⃗k·⃗x0 X λ=±2 −λY ∗ ℓm (Ωk) ξλ  ⃗k  T T ℓ(k, η0, ηin) , (3.14) (3.14) where we have introduced the linear transfer function T X(z) ℓ , with X = S, T which represents the time evolution of the graviton fluctuations originated from the primordial perturbation in) ≡TΦ (ηin, k) jℓ(k (η0 −ηin)) + Z η0 ηin dη′ ∂[TΨ (η, k) + TΦ (η, k)] ∂η jℓ(k (η −ηin)) , ≡ s (ℓ+ 2)! (ℓ−2)! 1 4 Z η0 ηin dη ∂χ (η, k) ∂η jℓ(k (η0 −η)) k2 (η0 −η)2 . (3.15) T T ℓ(k, η0, ηin) ≡ s (ℓ+ 2)! (ℓ−2)! 1 4 Z η0 ηin dη ∂χ (η, k) ∂η jℓ(k (η0 −η)) k2 (η0 −η)2 . (3.15) – 9 – 4 Correlators of GW anisotropies and SGWB non-Gaussianity The computations performed so far assume statistical isotropy (recall the discussion at the end of Section 2). Correspondingly, when we combine (4.1) and (4.2) with (3.14) we obtain angular correlators with well specific dependence on the multipole indices. Specifically, the two point correlators have the dependence ⟨ΓℓmΓ∗ ℓ′m′⟩≡δℓℓ′ δmm′ eCℓ, ⟨Γℓ1m1Γℓ2m2Γℓ3m3⟩≡  ℓ1 ℓ2 ℓ3 m1 m + 2 m3  ebℓℓ′ℓ′′ , (4.3) ⟨ΓℓmΓ∗ ℓ′m′⟩≡δℓℓ′ δmm′ eCℓ, ⟨Γℓ1m1Γℓ2m2Γℓ3m3⟩≡  ℓ1 ℓ2 ℓ3 m1 m + 2 m3  (4.3) while, under the above assumption, the angular power spectrum and the reduced bispectrum consists of the three separate contributions eCℓ= eCℓ,I (q) + eCℓ,S + eCℓ,T , ebℓ1ℓ2ℓ3 = ebℓ1ℓ2ℓ3,I (q) + ebℓ1ℓ2ℓ3,S + ebℓ1ℓ2ℓ3,T (q) + eCℓ,S + eCℓ,T , ebℓ1ℓ2ℓ3 = ebℓ1ℓ2ℓ3,I (q) + ebℓ1ℓ2ℓ3,S + ebℓ1ℓ2ℓ3,T . (4.4) ebℓ1ℓ2ℓ3 = ebℓ1ℓ2ℓ3,I (q) + ebℓ1ℓ2ℓ3,S + ebℓ1ℓ2ℓ3,T . (4.4) (4.4) We recall that the form of the bispectrum factorizes the Wigner-3j symbols [59], which are nonvanishing only provided that P i mi = 0 and that the three ℓi satisfy the triangular inequalities. We recall that the form of the bispectrum factorizes the Wigner-3j symbols [59], which are nonvanishing only provided that P i mi = 0 and that the three ℓi satisfy the triangular inequalities. q In the following we provide the explicit expression for the various contributions to the power spectrum and the reduced bispectrum. – 10 – 4.1 Angular power spectrum of GW energy density In that case, the final temperature anisotropy is 1/3 times the scalar perturbation at the last scattering surface, while Φ at that moment decreased by a factor 9/10 in the transition from radiation to matter domination [30]. With this in mind, the second term in (4.8) leads to CSW ℓ = (3/10)2 eCℓ,S, in agreement with the CMB literature. On the other hand, if the two contributions are correlated, as it would be the case for adiabatic initial condition for ΓI, then both terms in (4.8) contribute to the SW effect for the SGWB. The second term can be compared to the SW contribution to the CMB anisotropies. In that case, the final temperature anisotropy is 1/3 times the scalar perturbation at the last scattering surface, while Φ at that moment decreased by a factor 9/10 in the transition from radiation to matter domination [30]. With this in mind, the second term in (4.8) leads to CSW ℓ = (3/10)2 eCℓ,S, in agreement with the CMB literature. On the other hand, if the two contributions are correlated, as it would be the case for adiabatic initial condition for ΓI, then both terms in (4.8) contribute to the SW effect for the SGWB. 4.1 Angular power spectrum of GW energy density We start with the computation of the two-point function of the initial condition term. From the first of (3.14) we can write We start with the computation of the two-point function of the initial condition term. From the first of (3.14) we can write Γℓm,I (q) Γ∗ ℓ′m′,I (q)  = (4π)2 (−i)ℓ−ℓ′ Z d3k (2π)3 ei⃗k·⃗x0 Z d3k′ (2π)3 e−i⃗k′·⃗x0 D Γ  ηin, ⃗k, q  Γ∗ ηin, ⃗k′, q E × Y ∗ ℓm  ˆk  Yℓ′m′  ˆk′ jℓ(k (η0 −ηin)) jℓ′ k′ (η0 −ηin)
. (4.5) (4.5) The correlator of the initial condition term is then given by the first of (4.1). Using this, and the orthonormality condition of the spherical harmonics, R d2ˆn sYℓm sY ∗ ℓ′m′ = δℓℓ′ δmm′, leads to The correlator of the initial condition term is then given by the first of (4.1). Using this, and the orthonormality condition of the spherical harmonics, R d2ˆn sYℓm sY ∗ ℓ′m′ = δℓℓ′ δmm′, leads to Γℓm,I (q) Γ∗ ℓ′m′,I (q)  = δℓℓ′δmm′ 4π Z dk k [jℓ(k (η0 −ηin))]2 PI (q, k) , (4.6) (4.6) which indeed is of the form dictated by statistical isotropy. The other two terms are obtained analogously. Altogether, we find eCℓ,I (q) = 4π Z dk k [jℓ(k (η0 −ηin))]2 PI (q, k) , eCℓ,S = 4π Z dk k T (S) 2 ℓ (k, η0, ηin) Pζ (k) , eCℓ,T = 4π Z dk k T (T) 2 ℓ (k, η0, ηin) X λ=±2 Pλ (k) . (4.7) (4.7) We know from the CMB that the large-scale tensor modes have a power smaller than the scalar ones. At large scale, the scalar contribution is dominated by the term proportional to the initial value of Φ in T (0) ℓ , which is the analog of the SW contribution for the CMB. The large-scale modes that we are considering re-entered the horizon during matter domination. For these modes, ignoring the late time dark energy domination, TΦ = TΨ = 3/5 [30]. So, for scale invariant power spectra, eCℓ≃eCℓ,I (q) + eCℓ,S ≃ 2π ℓ(ℓ+ 1) " PI (q) + 3 5 2 Pζ # . (4.8) (4.8) The second term can be compared to the SW contribution to the CMB anisotropies. 4.2 Angular bispectrum of GW energy density The characterization of the non-Gaussian properties of the SGWB is a potential tool to discriminate whether a SGWB has a primordial or astrophysical origin. The primoridal 3- point function of the GW field, h3 , is unobservable, due to the decoherence of the associated phase (because of the propagation, and the finite duration of the measurement [23, 24]), with, – 11 – possibly, the exception of very specific shapes [26, 33]. It is more convenient to consider the non-gaussianity associated to the GW energy density angular distribution, which is not affected by this problem [31]. This gives rise to the bispectra in (4.4), which we evaluate now. As we did for the power spectrum, also in this case we start from the initial condition term. Combining the first of (3.14) and the first of (4.2) leads to * 3 Y i=1 Γℓimi,I (q) + = 3 Y i=1  4π (−i)ℓi Z d3ki (2π)3 Y ∗ ℓimi  ˆki  jℓi (ki (η0 −ηin))  × BI (q, k1, k2, k3) (2π)3 δ(3)  ⃗k1 + ⃗k2 + ⃗k3  . 4.2 Angular bispectrum of GW energy density (4.14) (4.14) (4.14) – 12 – Following [60], in Appendix B we show that also this contribution can be cast in a similar form to the previous two terms: Following [60], in Appendix B we show that also this contribution can be cast in a similar form to the previous two terms: * 3 Y i=1 Γℓimi,T + = Gm1m2m3 ℓ1ℓ2ℓ3 " 3 Y i=1 4π (−i)ℓi Z k2 i dki (2π)3 T T ℓ,i (ki, η0, ηin) # X λ=±2 ˜Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) , ( ) (4.15) (4.15) where where ≡ √ 4π  ℓ1 ℓ2 ℓ3 0 0 0 −1 X m1,m2,m3  ℓ1 ℓ2 ℓ3 m1 m2 m3  ˜Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) ≡ √ 4π  ℓ1 ℓ2 ℓ3 0 0 0 −1 X m1,m2,m3  ℓ1 ℓ2 ℓ3 m1 m2 m3  × " 3 Y i=1 Z dΩki −λY ∗ ℓimi (Ωki) √2ℓi + 1 # D ξλ  ⃗k1  ξλ  ⃗k2  ξλ  ⃗k3 E . (4 16 × " 3 Y i=1 Z dΩki −λY ∗ ℓimi (Ωki) √2ℓi + 1 # D ξλ  ⃗k1  ξλ  ⃗k2  ξλ  ⃗k3 E . (4.16) We remark once again that we have neglected for simplicity all the mixed scalar-tensor correlators. 4.2 Angular bispectrum of GW energy density (4.9) (4.9) We then use the representation of the Dirac δ−function, We then use the representation of the Dirac δ−function, We then use the representation of the Dirac δ−function, We then use the representation of the Dirac δ−function, δ(3)  ⃗k1 + ⃗k2 + ⃗k3  = Z d3y (2π)3 ei(⃗k1+⃗k2+⃗k3)·⃗y = Z ∞ 0 dy y2 Z dΩy 3 Y i=1  2 X LiMi iLi jLi (ki y) Y ∗ LiMi (Ωy) YLiMi  ˆki   , (4.10) = Z ∞ 0 dy y2 Z dΩy 3 Y i=1  2 X LiMi iLi jLi (ki y) Y ∗ LiMi (Ωy) YLiMi  ˆki   , (4.10) and the orthonormality of the spherical harmonics, to arrive to and the orthonormality of the spherical harmonics, to arrive to * 3 Y i=1 Γℓimi,I (q) + = Gm1m2m3 ℓ1ℓ2ℓ3 Z ∞ 0 dr r2 3 Y i=1  2 π Z dki k2 i jℓi (ki (η0 −ηin)) jℓi (ki r)  BI q, k, k′, k′′
, ( ) (4.11) where we have introduced the Gaunt integrals where we have introduced the Gaunt integrals Gm1m2m3 ℓ1ℓ2ℓ3 ≡ Z d2ˆn Yℓ1m1 (ˆn) Yℓ2m2 (ˆn) Yℓ3m3 (ˆn) = r (2ℓ1 + 1) (2ℓ2 + 1) (2ℓ3 + 1) 4π  ℓ1 ℓ2 ℓ3 0 0 0   ℓ1 ℓ2 ℓ3 m1 m2 m3  . (4.12) (4.12) We remark that also the bispectrum from the initial condition also generally as an O (1) dependence on the GW frequency. We remark that also the bispectrum from the initial condition also generally as an O (1) dependence on the GW frequency. An analogous computation leads to the contribution from the scalar modes * 3 Y i=1 Γℓimi,S + = Gm1m2m3 ℓ1ℓ2ℓ3 Z ∞ 0 dr r2 3 Y i=1  2 π Z dki k2 i T S ℓi (ki, η0, ηin) jℓi (ki r)  Bζ k, k′, k′′
. (4.13) (4.13) For the tensor sourced contribution we have For the tensor sourced contribution we have For the tensor sourced contribution we have * 3 Y i=1 Γℓimi,T + = X λ=±2 3 Y i=1  4π (−i)ℓi Z k2 i dki (2π)3 T T ℓ,i (ki, η0, ηin) Z dΩki −λY ∗ ℓimi (Ωki)  * 3 Y i=1 ξλ  ⃗ki + . 4.3 Reduced Bispectrum and estimation The three contributions to the bispectrum found above have the correct form (4.3) as dictated by statistical isotropy. For convenience, we collect here the explicit form of the reduced bispectra contributing to (4.4) ebℓ1ℓ2ℓ3,I = Z ∞ 0 dr r2 3 Y i=1  2 π Z dki k2 i jℓi [ki (η0 −ηin)] jℓi (ki r)  BI (q, k1, k2, k3) , ebℓ1ℓ2ℓ3,S = Z ∞ 0 dr r2 3 Y i=1  2 π Z dki k2 i T S ℓi (ki, η0, ηin) jℓi (ki r)  Bζ k, k′, k′′
, ebℓ1ℓ2ℓ3,T = 4 π2 X λ=±2 X mi  ℓ1 ℓ2 ℓ3 0 0 0 −2 Gm1m2m3 ℓ1ℓ2ℓ3 " 3 Y i=1 (−i)ℓi 2ℓi + 1 Z d3kiT T ℓ,i(ki) −λY ∗ ℓimi (Ωki) # × δ  ⃗k + ⃗k′ + ⃗k′′ Bλ  ⃗k, ⃗k′, ⃗k′′ . (4.17) (4.17) To estimate the SGWB bispectrum, we consider only the scalar source contribution eBℓ1ℓ2ℓ3,S and we assume the simplest small non-linear coupling local ansatz for the curvature perturbation 3 To estimate the SGWB bispectrum, we consider only the scalar source contribution eBℓ1ℓ2ℓ3,S and we assume the simplest small non-linear coupling local ansatz for the curvature perturbation ζ (⃗x) = ζg (⃗x) + 3 5 fNL ζ2 g (⃗x) , (4.18) (4.18) where ζg (⃗x) denotes the linear Gaussian part of the perturbation. With the local ansatz, the bispectrum of the scalar perturbations assumes the form [46, 47] where ζg (⃗x) denotes the linear Gaussian part of the perturbation. With the local ansatz, the bispectrum of the scalar perturbations assumes the form [46, 47] Bζ (k1, k2, k3) = 6 5fNL 2π2 k3 1 Pζ (k1) 2π2 k3 2 Pζ (k2) + 2 permutations  . (4.19) (4.19) We insert this in the second of (4.17) and we assume a matter transfer function TΦ (η, k) = TΨ (η, k) = 3/5 g (η) with the growth factor g(η) = 1 and a scale invariant spectrum for the – 13 – primordial curvature fluctuations. We can then integrate over one of the internal momenta ki, 2 π Z dk k2jℓ(k η0) jℓ(k r)  ℓ≫1 = δ (η0 −r) η2 0 . 4.3 Reduced Bispectrum and estimation (4.20) (4.20) The relation (4.20) is exact if k ranges up to infinity, which is not the case for the innermost momentum (as the integral (4.20) is performed first, this will necessarily be the the momen- tum that we order to be the innermost one), due to the triangular inequalities associated to the bispectrum. The condition ℓ≫1 ensures that the support of the integration occurs at sufficiently small k, so that the relation (4.20) becomes exact at large ℓ. The result then allows to immediately perform the integral over r. We then find that the reduced bispectrum from the scalar contribution, assuming that the SW is the dominant contribution, is = 162 625 fNL  4π Z dk1 k1 j2 ℓ1 (k1 η0) Pζ (k1)   4π Z dk2 k2 j2 ℓ2 (k2 η0) Pζ (k2)  ebℓ1ℓ2ℓ3,S = 162 625 fNL  4π Z dk1 k1 j2 ℓ1 (k1 η0) Pζ (k1)   4π Z dk2 k2 j2 ℓ2 (k2 η0) Pζ (k2)  +2 permutations (4 2 (4.21) +2 permutations . This result can also be written in terms of the two-point functions found in Eq. (4.7): t can also be written in terms of the two-point functions found in Eq. (4.7): ebℓ1ℓ2ℓ3,S ≃2 fNL h eCℓ1,S eCℓ2,S + eCℓ1,S eCℓ3,S + eCℓ2,S eCℓ3,S i , (4.22) (4.22) which resembles the one for the CMB angular bispectrum in the Sachs-Wolfe regime [46]. So, the SGWB bispectrum is specified by the fNL parameter and the angular spectrum. Also in this estimate we neglected a possible correlation between the initial and scalar source contributions that should be taken into account when, for instance, ΓI is controlled by the adiabatic scalar perturbation (see [31] for an example). 5 An example: the axion-inflation case In writing this relation, we have assumed that the parameter ξ is in turn controlled by a dynamical field (the rolling axion, in the example of [42]), which results in the background value ¯ξ, and in the perturbation δξ. In this relation, ξ is the value that this parameter had during inflation at the location ⃗x+d ˆn, where d is the distance covered by the gravitons between the initial and the present time (equal for all directions, since we are disregarding the effect of the long scale modes ζ; we note that these modes will contribute to the term ΓS, that we are not discussing in this section). In writing this relation, we have assumed that the parameter ξ is in turn controlled by a dynamical field (the rolling axion, in the example of [42]), which results in the background value ¯ξ, and in the perturbation δξ. ξ, p ξ We then generalize the relation (5.1) to We then generalize the relation (5.1) to ωGW (η0, ⃗x, q, ˆn) = constant × X λ Pλ (q, ξ (η0, ⃗x, ˆn)) , (5.4) (5.4) which has the background value ¯ΩGW (η0, q) = constant × P λ Pλ q, ¯ξ
. The constant factor drops in the ratio 4 −∂ln ¯ΩGW (η0, q) ∂ln q = 4 −∂ln P λ Pλ q, ¯ξ
 ∂ln q , (5.5) (5.5) as well as in as well as in δGW (η0, ⃗x, q, ˆn) = P Pλ (q, ξ (η0, ⃗x, ˆn)) −P Pλ q, ¯ξ
P Pλ q, ¯ξ
= ∂ln P λ lnPλ q, ¯ξ
 ∂¯ξ δξ (⃗x + d ˆn) , (5 6) (5.6) (5.6) ( ) where we have expanded the GW primordial power spectrum to linear order in δξ. In this wat, the relation (2.9) can be recast in the form ΓI(η0, ⃗x0, q, ˆn) ≡F(q, ¯ξ) δξ(⃗x0 + d ˆn) , (5.7) (5.7) with F q, ¯ξ
≡ 1 4 −nT ∂P λ  ln Pλ q, ¯ξ
 ∂¯ξ , nT ≡∂ln P λ Pλ q, ¯ξ
 ∂ln q , (5.8) (5.8) where we have also made use of the standard definition of the tensor spectral tilt nT . The question of whether we have or have not spectral distortion depends on whether the quantity F(q, ¯ξ) is or is not q−dependent. 5 An example: the axion-inflation case The goal of this section is to understand under which conditions the initial term ΓI (q) has a nontrivial q−dependence, that distinguishes it from the other contributions to the anisotropy. There are several mechanisms for the generation of a cosmological GW signal visible at interferometer scales (see [8–10] for recent review). In this section we focus on a specific mechanism: we consider the case where an axion inflaton φ sources gauge fields, which in turn generates a large GW background. In particular we consider the specific evolution shown in Figure 4 of [42], where the inflaton potential is chosen so to lead to a peak in the GW signal at LISA frequencies, without overproducing scalar perturbations and primordial black holes. The amount of GWs sourced in this mechanism is controlled by the parameter ξ ≡( ˙φ/2fφH), where fφ is the decay constant of the axion inflaton. The present fractional energy in GW, ΩGW (η0, q), is related to the primordial GW power spectrum Pλ (ηin, q) by ΩGW (η0, q) = 3 128 Ωrad X λ Pλ (ηin, q) " 1 2 qeq q 2 + 4 9 √ 2 −1 # . (5.1) (5.1) This relation, taken from [10], interpolates between large and small scales. Since we are interested in the modes with q ≫qeq, that entered the horizon during radiation domination, we consider only the second term in the square bracket, and we find ΩGW (η0, q) = constant × X λ Pλ (ηin, q) , (5.2) (5.2) λ – 14 – and, as we will see, the constant term is not relevant for our computation. We are interested in the contribution from the initial condition Γin. So we can set the long modes ζ(⃗k) = hλ(ˆk) = 0 in this discussion. We therefore assume that the value of the energy density that arrives to the location ⃗x from the direction ˆn is controlled by the parameter (5.3) ξ = ¯ξ + δξ (⃗x + d ˆn) . (5.3) In this relation, ξ is the value that this parameter had during inflation at the location ⃗x+d ˆn, where d is the distance covered by the gravitons between the initial and the present time (equal for all directions, since we are disregarding the effect of the long scale modes ζ; we note that these modes will contribute to the term ΓS, that we are not discussing in this section). 5 An example: the axion-inflation case This provides an immediate criterion for evaluating whether and how much the GW anisotropies depend on frequency (as, in principle, one could imagine a GW power spectrum for which the dependence on q of F vanishes, or is extremely suppressed). This conclusion only assumes that the primordial GW signal is function of some additional parameter ξ which has small spatial inhomogeneities, and therefore it likely applies to several other mechanisms. We show in Figure 1 the evolution of the function F corresponding to the GW production shown in Figure 4 of [42]. We see that indeed this quantity presents a nontrivial scale dependence, and therefore the correlators of the anisotropies will be different at different frequencies. – 15 – 10-15 10-10 10-5 1 105 0 1 2 3 4 f Hz ℱ(f) Figure 1. Quantity F as a function of the frequency f = q/2π of the GW signal for the model of axion inflation described in the text. Figure 1. Quantity F as a function of the frequency f = q/2π of the GW signal for the model of axion inflation described in the text. 6 Squeezed limit and consistency relations of the SGWB Then the contribution ΓS reads (see eq (3.9)) ΓS(η,⃗k, ˆn) = 3 5 Z η ηin dη′ e−ikµ(η−η′)  δ(η′ −ηin) g(η′) + 1 2 ∂η′g(η′)  ζ(⃗k) , ≡T S(η, k, µ) ζ  ⃗k  , ΓS(η,⃗k, ˆn) = 3 5 Z η ηin dη′ e−ikµ(η−η′)  δ(η′ −ηin) g(η′) + 1 2 ∂η′g(η′)  ζ(⃗k) , ) = 3 5 Z η ηin dη′ e−ikµ(η−η′)  δ(η′ −ηin) g(η′) + 1 2 ∂η′g(η′)  ζ(⃗k) , ≡T S(η, k, µ) ζ  ⃗k  , (6.4) 5 Z ηin  2 η  ≡T S(η, k, µ) ζ  ⃗k  , (6.4) ≡T S(η, k, µ) ζ  ⃗k  , (6.4) (6.4) where µ = ˆn · ˆk and TS is the definition for the scalar transfer function we adopt here. In matter domination this becomes TS = 3 5 e−ikµ(η−ηin) . (6.5) (6.5) Notice that ΓS does not depend on q. Assembling the definitions above, we can t δGW(η, ⃗k, q, ⃗n) = −∂ln ¯f(q) ∂ln q TS(η, k, µ) ζ(⃗k) . (6.6) (6.6) Indicating with PΓ the power spectrum, we can write the 2-point correlators in momentum space as Indicating with PΓ the power spectrum, we can write the 2-point correlators in momentum space as D Γ(η, ⃗k1, q , ˆn) Γ(η, ⃗k2, q , ˆn) E′ = 2π2 k3 1 PΓ(η, k1, q, ˆn) = 2π2 k3 1 |TS(η, k1, µ1)|2 Pζ(k1) , (6.7) (6.7) where a prime ′ corresponds to correlators understanding the (2π)3 δ(P⃗ki) factor. Then: PΓ(η, k, µ) = ⟨ΓS(η, k, µ)ΓS(η, k′, µ)⟩′ 2π2/k3 = |TS(η, k, µ)|2 Pζ(k) , (6.8) PδGW(η, k, q, µ) =  ∂ln ¯f(q) ∂ln q TS(η, k, µ)  2 Pζ(k) . (6.9) PΓ(η, k, µ) = ⟨ΓS(η, k, µ)ΓS(η, k′, µ)⟩′ 2π2/k3 = |TS(η, k, µ)|2 Pζ(k) , (6.8) (6.8) PδGW(η, k, q, µ) =  ∂ln ¯f(q) ∂ln q TS(η, k, µ)  2 Pζ(k) . (6.9) (6.9) In matter domination, as we learned above, |TS|2 = 9/25, but in general |TS|2 can depend on η, k, ˆn. In what follows, we study how the two-point correlation functions of SGWB anisotropies, when evaluated at small scales k, are modulated by the presence of a long-scale mode ζL ≡ ζ(⃗kL), with |⃗kL| ≪|⃗k|. Such modulation induces a non-vanishing squeezed limit for the three-point function of δGW. 6 Squeezed limit and consistency relations of the SGWB The anisotropies δGW depend on various quantities, (η, k, q, µ), which can be sensitive in a different way to the long mode. We use the systematic approach pionereed by Weinberg [27] that unambiguously associates the effects of a long mode with an appropriate coordinate transformation. We shall closely follow the treatment of [56], which develops the arguments of [27] for the case of CMB, applying it to the SGWB (for similar approaches see also [55, 57]). 6 Squeezed limit and consistency relations of the SGWB Non-linear effects associated with the propagation of interacting GWs in a non-linear universe lead to non-vanishing connected n-point functions even in absence of intrinsic, primordial non-Gaussianity. In particular, the squeezed limit of bispectra associated with GW observ- ables should acquire a non-vanishing value, and satisfy consistency relations that resemble Maldacena’s consistency relations [54]. This is analogous to what happens for CMB [55–57]. In this Section we compute the squeezed limit of the bispectrum for the graviton distri- bution function in the case of adiabatic fluctuations. As in Section 2, we write in momentum space ωGW η, ki, q, nj
= ¯ωGW (η, q)  1 + δGW η, ki, q, ni
 , (6.1) (6.1) where ¯ωGW(η, q) is associated with the energy density of the isotropic SGWB. This quantity depends on time η and on the GW momentum q. Small anisotropies of the SGWB are controlled by the quantity δGW given in Eq. (2.9). We re-write it here, expressing it in terms of the function ¯f(q): δGW(η, ⃗k, q, ⃗n) = −∂ln ¯f(q) ∂ln q ΓS  η, ⃗k, q, ⃗n  , ( ) (6.2) where recall that ΓS controls the fluctuations in the distribution function (see the definitions in Section 2). In this Section we focus on the contribution due to scalar fluctuations. We assume there is no anisotropic stress, and that scalar perturbations in Newtonian gauge satisfy the adiabaticity condition: Φ(η,⃗k) = Ψ(η,⃗k) = 3 5 g(η) ζ(⃗k) , (6.3) (6.3) where g(η) is a function mapping the superhorizon seed (controlled by ζ(⃗k)) to the scalar fluctuations at small scales (see e.g. [61, 62]). It is generally time dependent although it is where g(η) is a function mapping the superhorizon seed (controlled by ζ(⃗k)) to the scalar fluctuations at small scales (see e.g. [61, 62]). It is generally time dependent although it is – 16 – equal to unity in pure matter domination. 6.1 Long wavelength modes as coordinate transformations We discuss how to identify the effects of a long mode with an appropriate coordinate trans- formation. We limit our attention to effects due to scalar fluctuations. The metric including long-wavelength scalars in Poisson gauge is ds2 = a2(η)  −(1 + 2 ΦL) dη2 + (1 −2ΨL) δij dxidxj . (6.10) ds2 = a2(η)  −(1 + 2 ΦL) dη2 + (1 −2ΨL) δij dxidxj . (6.10) – 17 – We assume that the long-scale mode depends on a momentum ⃗kL, with magnitude much smaller than that of the momentum of the short-scale modes introduced in eq. (6.17), but with a certain direction, and we discuss how the quantities (η, k, q, µ), transform under a coordinate redefinition adsorbing the long modes. We start by noticing that the following coordinate transformation preserves the Poisson gauge structure (ζL indicates the long mode of curvature fluctuations at large scales, responsible for the modulation effects): (6.11) (6.12) ˆη = η + ϵ(η) ζL , (6.11) ˆxi = xi (1 −λ ζL) , (6.12) (6.12) with λ constant. After performing such gauge transformation, with λ constant. After performing such gauge transformation, with λ constant. After performing such gauge transformation, with λ constant. After performing such gauge transformation, ˆΦL = ΦL −ϵ′ ζL −H ϵ ζL , ˆΨL = ΨL −λ ζL + H ϵ ζL , (6.13) (6.13) we can ‘gauge away’ the long wavelength scalar modes making the gauge choice ΦL = ϵ′ + Hϵ
ζL , ΨL = (λ −Hϵ) ζL , (6.14) (6.14) so that in the hat coordinates the metric is purely FRW with no long-wavelength perturba- tions. As explained in [27, 56], in order to be consistent with the small k limit of Einstein equations, we need to impose the conditions (in absence of anisotropic stress) λ = 1 , ϵ(η) = 1 a2(η) Z η η⋆ dη′ a2(η′) , (6.15) (6.15) where η∗is some initial reference time. Eq (6.15) immediately leads to the equality ϵ′ = −2H ϵ + 1 . 6.1 Long wavelength modes as coordinate transformations (6.16) (6.16) After performing the coordinate redefinition (6.11), (6.12), we can write a metric containing short-wavelength scalar fluctuations in Poisson gauge ‘on top’ of long fluctuations: After performing the coordinate redefinition (6.11), (6.12), we can write a metric containing short-wavelength scalar fluctuations in Poisson gauge ‘on top’ of long fluctuations: ds2 = a2(ˆη) h −  1 + 2 ˆΦS  dˆη2 +  1 −2ˆΨS  δij dˆxidˆxji . (6.17) (6.17) In fact, such metric contains the long-scale modes within the definition of the hat coordinates. We can then express the perturbations in terms of the original coordinates (η, xi) using again relations (6.11), (6.12). Such operation teaches us how the short wavelength modes are modulated by the long wavelength ones: In fact, such metric contains the long-scale modes within the definition of the hat coordinates. We can then express the perturbations in terms of the original coordinates (η, xi) using again relations (6.11), (6.12). Such operation teaches us how the short wavelength modes are modulated by the long wavelength ones: ˆΦS = ΦS + ΦL + 2ΦSΦL + ϵ ζL ∂ΦS ∂η −λ ζL xi ∂ΦS ∂xi , (6.18) ˆΨS = ΨS + ΨL −2ΨSΨL + ϵ ζL ∂ΨS ∂η −λ ζL xi ∂ΨS ∂xi . (6.19) (6.18) (6.19) Importantly, the short modes acquire a second order correction due to long modes. As we shall discuss in what comes next, these non-linear, higher-order corrections modulate the – 18 – 2-point function for short modes, and lead to a non-vanishing squeezed limit for the 3-point function. 2-point function for short modes, and lead to a non-vanishing squeezed limit for the 3-point function. As a concrete example, that we shall use in what follows, we can consider the case of constant proportionality between pressure and energy density, p = wρ. Being in this case a(η) ∝η2/(1+3w), H = 2/[η(1 + 3w)] we get ϵ(η) ζL = 1 + 3w 5 + 3w η ζL , (6.20) (6.20) and and H ϵ = 2 5 + 3w , (6.21) and H ϵ = 2 5 + 3w , (6.21) which, for matter domination, gives H ϵ = 2/5. which, for matter domination, gives H ϵ = 2/5. We also need to evaluate how the Fourier transform of a function f(xi) changes under a rescaling of spatial coordinates, as in eq (6.12). 6.1 Long wavelength modes as coordinate transformations We find that if we apply a constant rescaling of spatial coordinates f(xi) →f xi(1 −λ ζL)
f(xi) →f xi(1 −λ ζL)
(6.22) (6.22) to a function f, then its Fourier transform, given by to a function f, then its Fourier transform, given by f(xi) = Z d3k (2π)3 ei⃗k⃗x ˜f(kj), transforms as (at first order in a ζL expansion) f(xi(1 −λ ζL)) = Z d3k (2π)3 ei⃗k⃗x(1−λ ζL) ˜f(kj) = Z d3k (2π)3 ei⃗k⃗x h (1 + 3λ ζL) ˜f(kj(1 + λ ζL)) i . (6 23) (6.23) This implies that up to first order in ζL, under the coordinate transformation we are interested in, we have: ˜f(kj) →(1 + 3λ ζL) ˜f kj(1 + λ ζL)
⇒ ˜f(kj) →˜f(kj) + 3λ ζL ˜f(kj) + λ ζL km ∂˜f(kj) ∂km . (6.24) (6.24) As a last step, we now investigate how to transform the coordinates (q, ˆni) that con- trol the GW four-momentum. At first order, neglecting tensors, the GW four-momentum components are given by P 0 = q a2(η) e−Φ, P i = q a2(η) ni eΨ . (6.25) P i = q a2(η) ni eΨ . (6.25) (6.25) We wish to express the previous quantities in terms of hat coordinates, including the effects of the long modes. In particular, we are interested in determining the quantities ˆq and ˆni that are contained into the GW four-momentum, when it is expressed in terms of hat coordinates. We use the fact that P µ is a vector, transforming in the usual way under coordinate transformations (in particular transformations (6.11), (6.12)). Using this fact, we find ˆq a2(ˆη) = 1 + ϵ′ ζL
q a2(η) e−ΦL , (6.26) ˆq a2(ˆη) ˆni = (1 −λ ζL ) q a2(η) ni eΨL . (6.27) (6.26) (6.27) – 19 – Condition (6.26) gives, at first order in the long-scale modes, Condition (6.26) gives, at first order in the long-scale modes, Condition (6.26) gives, at first order in the long-scale modes, ˆq = a2(ˆη) a2(η) 1 + ϵ′ ζL
(1 −ΦL) q = 1 + 2 H ϵ ζL + ϵ′ ζL −ΦL
q =  1 +  1 −3 5g(η)  ζL  q . 6.1 Long wavelength modes as coordinate transformations (6.36) (6.36) The quantity ΓS is mapped to The quantity ΓS is mapped to The quantity ΓS is mapped to The quantity ΓS is mapped to ΓS  ˆη, ˆki, ˆq, ˆni = (1 + 3ζL) ΓS  η + ϵ(η) ζL , ⃗k (1 + ζL) , (1 + βq(η) ζL) q, (1 + βn(η)ζL) ni , (6.37) ΓS  ˆη, ˆki, ˆq, ˆni = (1 + 3ζL) ΓS  η + ϵ(η) ζL , ⃗k (1 + ζL) , (1 + βq(η) ζL) q, (1 + βn( ΓS  ˆη, ˆki, ˆq, ˆni = (1 + 3ζL) ΓS  η + ϵ(η) ζL , ⃗k (1 + ζL) , (1 + βq(η) ζL) q, (1 + βn(η)ζL) ni , (6.37)  (6.37) that, expanded at linear order in ζL, becomes that, expanded at linear order in ζL, becomes ΓS η, ki, q, ni
⇒ΓS  ˆη, ˆki, ˆq, ˆni = (1 + 3ζL) ΓS η, ki, q, ni
+∂ΓS ∂η ϵ(η) ζL + ki ∂ΓS ∂ki ζL + βq(η) ∂ΓS ∂ln q ζL + βn(η) nj ∂ΓS ∂nj ζL . ( (6.38) We now assemble the results obtained. The SGWB energy density, including anisotropies, is modulated by the long mode ζL as We now assemble the results obtained. The SGWB energy density, including anisotropies, is modulated by the long mode ζL as ωGW  ˆη, ˆki, ˆq, ˆni = ¯ωGW (η, q) " 1 + δGW +  4βq −4 H ϵ + βq ∂ln ¯f(q) ∂ln q  ζL +  3 + βq(η) ∂ln q ∂ln ¯f(q) ∂2 ln ¯f(q) ∂(ln q)2  ζL δGW + + ∂ln Γ ∂η ϵ(η) + ki ∂ln Γ ∂ki + βn(η) ∂ln Γ ∂ln µ  ζL δGW # . (6.39) ωGW  ˆη, ˆki, ˆq, ˆni = ¯ωGW (η, q) " 1 + δGW +  4βq −4 H ϵ + βq ∂ln ¯f(q) ∂ln q  ζL +  3 + βq(η) ∂ln q ∂ln ¯f(q) ∂2 ln ¯f(q) ∂(ln q)2  ζL δGW + + ∂ln Γ ∂η ϵ(η) + ki ∂ln Γ ∂ki + βn(η) ∂ln Γ ∂ln µ  ζL δGW # . (6.39) (6.39) Eq. (6.39) is the basic expression that we need: all quantities at the RHS are evaluated in terms of the original coordinates without the hat. 6.1 Long wavelength modes as coordinate transformations Notice that even in absence of in- trinsic small-scale anisotropies, the GW energy density is modulated by the long mode: a dependence on ζL is indeed still present by setting δGW = 0 in eq (6.39). This is the effect studied by Alba and Maldacena [14]. For example, in pure matted domination, we have βq = H ϵ = 2/5. Setting δGW = 0, eq (6.39) simply becomes ˆωGW  ˆη, ˆki, ˆq, ˆni = ¯ωGW (η, q)  1 + 2 5 ∂ln ¯f(q) ∂ln q ζL  . (6.40) (6.40) In this case, the modulation of ωGW is then controlled by the momentum dependence of the function ¯f(q), associated with the isotropic distribution function of the SGWB energy density [14]. 6.1 Long wavelength modes as coordinate transformations (6.28) (6.28) On the other hand, condition (6.27) gives On the other hand, condition (6.27) gives On the other hand, condition (6.27) gives On the other hand, condition (6.27) gives ˆni = a2(ˆη) a2(η) q ˆq (1 −λ ζL) (1 + ΨL) ni = 1 −ϵ′ ζL −λ ζL + 2ΦL
ni =  1 −2  1 −H ϵ(η) −3 5g(η)  ζL  ni . (6.29) (6.29) These are the results that we need. It is convenient to write more compact expressions as These are the results that we need. It is convenient to write more compact expressions as ˆq = (1 + βq(η) ζL) q , ˆni = (1 + βn(η) ζL) ni , (6.30) ˆni = (1 + βn(η) ζL) ni , (6.30) (6.30) with βq,n functions of time βq(η) = 1 −3 5g(η) , βn(η) = −2  1 −H ϵ(η) −3 5g(η)  . (6.31) (6.31) In matter domination we find βq = 2/5 and βn = 0. In matter domination we find βq = 2/5 and βn = 0. In matter domination we find βq = 2/5 and βn = 0. In matter domination we find βq = 2/5 and βn = 0. 6.2 Coordinate transformations and the GW distribution function We now apply the previous results to the problem at hand. We start by re-writing the GW energy density ωGW η, ki, q, ni
= ¯ωGW (η, q)  1 + δGW η, ki, q, ni
 , (6.32) (6.32) where where where ¯ωGW(η, q) = q4 a4(η) ρcrit ¯f(q) , (6.33) (6.33) and and δGW(η, ⃗k, q, ni) = −∂ln ¯f(q) ∂ln q ΓS  η, ⃗k, q, ni . (6.34) (6.34) We now transform each contribution in the previous formulas under the coordinate transfor- mation discussed in Section 6.1. The background quantities ¯ωGW and ¯f(q) transform as We now transform each contribution in the previous formulas under the coordinate transfor- mation discussed in Section 6.1. The background quantities ¯ωGW and ¯f(q) transform as ¯ωGW(η, q) ⇒¯ωGW(ˆη, ˆq) = ¯ωGW(η, q)  1 + 4 (βq −H ϵ) ζL + βq ∂ln ¯f(q) ∂ln q ζL  , (6.35) (6.35) – 20 – where ∂ln ¯f(q) ∂ln q ⇒∂ln ¯f(ˆq) ∂ln ˆq = ∂ln ¯f(q) ∂ln q + βq(η) ∂2 ln ¯f(q) ∂(ln q)2 ζL . 1Each quantity is evaluated at the same value of η, q, ni, hence we understand such dependence. Here we indicate with ˆδGW the quantity that receives the long-mode modulation. 6.3 The squeezed limit of 3-point correlation functions We now apply the general result of (6.39) to study how correlation functions of small-scale GW anisotropies δGW are influenced by the long mode. We start by studying how two-point correlation functions are modulated by ζL; we continue investigating the squeezed limit of the three-point correlation functions. – 21 – eq (6.39), we find the result1 Using eq (6.39), we find the result1 Using eq (6.39), we find the result1 Using eq (6.39), we find the result1 D ˆδGW(⃗k1) ˆδGW(⃗k2) E′ = (1 + M ζL) D δGW(⃗k1) δGW(⃗k2) E′ , (6.41) (6.41) where the modulating factor M reads where the modulating factor M reads M = 6 + 2βq(η) ∂ln q ∂ln ¯f(q) ∂2 ln ¯f(q) ∂(ln q)2 +ϵ(η) ∂ln D ΓS(⃗k1)ΓS(⃗k2) E′ ∂η + ki 1 ∂ln D ΓS(⃗k1)ΓS(⃗k2) E′ ∂ki 1 + ki 2 ∂ln D ΓS(⃗k1)ΓS(⃗k2) E′ ∂ki 2 +βq(η) ∂ln D ΓS(⃗k1)ΓS(⃗k2) E′ ∂ln q + βn(η) nj ∂ln D ΓS(⃗k1)ΓS(⃗k2) E′ ∂nj . (6.42) (6.42) Notice that the contributions in the first line of eq (6.39) that depend only on the long mode (without being weighted by δGW) do not contribute to M. Therefore they do not modulate the short-mode two point function. Notice that the contributions in the first line of eq (6.39) that depend only on the long mode (without being weighted by δGW) do not contribute to M. Therefore they do not modulate the short-mode two point function. We now apply to the results derived above the definitions of δGW and Γ power spectra, eqs (6.8), (6.9). We find the following expression for the modulation of the power spectrum due to a long mode: PˆδGW(η, k, q, ˆn, ⃗kL) = " 1 + 2 ∂ln Pζ ∂ln k ζ⃗kL + 2 βq(η) ∂ln q ∂ln ¯f(q) ∂2 ln ¯f(q) ∂(ln q)2 ζ(⃗kL) + ϵ(η) ∂ln |TS|2 ∂η + ∂ln |TS|2 ∂ln k + βn(η) ∂ln |TS|2 ∂ln µ ! ζ(⃗kL) # PδGW(η, k, q, ˆn) . (6 43) (6.43) All quantities inside the square parenthesis in the RHS are again evaluated at the same values of η, ˆn, k; hence we understand this dependence. We find that the power spectrum of δGW is modulated by the long mode ζ(⃗kL) through three (physically distinct) effects: All quantities inside the square parenthesis in the RHS are again evaluated at the same values of η, ˆn, k; hence we understand this dependence. 6.3 The squeezed limit of 3-point correlation functions We find that the power spectrum of δGW is modulated by the long mode ζ(⃗kL) through three (physically distinct) effects: 1. A modulation due to the scale dependence of the primordial curvature spectrum, as in Maldacena’s consistency relation. This is contained in the first line of eq (6.43), second term in the RHS. (Notice that the contributions coming from derivatives of the 1/k3 factor cancel out, as expected.) 2. A contribution due to the momentum-dependence of the background distribution ¯f(q). This is contained in the first line of eq (6.43), third term in the RHS. This is a close relative of the effect pointed out by Alba and Maldacena [14], although it is not exactly the same result because we find contributions depending on second derivatives of the function ¯f(q). 3. A contribution due to the time, scale, and direction dependence of the transfer function of scalar modes. This is contained in the second line of Eq. (6.43). 3. A contribution due to the time, scale, and direction dependence of the transfer function of scalar modes. This is contained in the second line of Eq. (6.43). – 22 – In the previous discussion we learned how the long mode modulates the 2-point function. This effect is expected to lead to a non-vanishing squeezed limit for the 3-point function involving the anisotropies δGW. Indeed, expressing a large scale limit of δGW in terms of ζ as In the previous discussion we learned how the long mode modulates the 2-point function. This effect is expected to lead to a non-vanishing squeezed limit for the 3-point function involving the anisotropies δGW. Indeed, expressing a large scale limit of δGW in terms of ζ as ˆδGW(η, ki 3, q, ni) = −∂ln ¯f(q) ∂ln q TS(η, ki 3, µ3) ζ(⃗k3) , (6.44) (6.44) for a small |⃗k3|, we can write the schematic relation (all δGW’s are evaluated at the same values of η, ni, q so we understand their dependence) lim ⃗k3→0 ⟨ˆδGW(⃗k1)ˆδGW(⃗k2)ˆδGW(⃗k3)⟩= −∂ln ¯f(q) ∂ln q TS(η, ki 3, µ3) D ⟨ˆδGW(⃗k1)ˆδGW(⃗k2)⟩ζ(⃗k3) E = −∂ln ¯f(q) ∂ln q TS(η, ki 3, µ3) D ⟨δGW(⃗k1)δGW(⃗k2)⟩(1 + M ζL) ζ(⃗k3) E = −∂ln ¯f(q) ∂ln q TS(η, ki 3, µ3) M D ⟨δGW(⃗k1)δGW(⃗k2)⟩⟨ζLζ(⃗k3) E , (6 45) (6.45) (6.45) where in the second line we used eq (6.41). This non-vanishing result gives the squeezed limit of the three-point function for δGW. 6.3 The squeezed limit of 3-point correlation functions We adopt the following definition 2 for the non-linear parameter fδGW NL : lim ⃗k3→0 D δGW(⃗k1)δGW(⃗k2)δGW(⃗k3) E = fδGW NL  4π4 k3 1 k3 3  PδGW(k1) Pζ(k3) . (6.46) (6.46) our case, using the previous results, we find In our case, using the previous results, we find fδGW NL = −∂ln ¯f(q) ∂ln q TS(η, k3, µ3) h 2 ∂ln Pζ ∂ln k1 + 2βq(η) ∂ln q ∂ln ¯f(q) ∂2 ln ¯f(q) ∂(ln q)2 + +ϵ(η) ∂ln |TS|2 ∂η + ∂ln |TS|2 ∂ln k1 + βn(η) ∂ln |TS|2 ∂ln µ1 i , (6.47) (6.47) and we can apply to this result the very same considerations made after eq (6.43). and we can apply to this result the very same considerations made after eq (6.43). The formula simplifies considerably in the case of pure matter domination. In this case, TS = 3/5, βq = H ϵ = 2/5. Then, fδGW NL = −6 5 ∂ln ¯f(q) ∂ln q ∂ln Pζ ∂ln k1 −12 25 ∂2 ln ¯f(q) ∂(ln q)2 . (6.48) (6.48) Recalling that ¯f(q) is related with the GW isotropic energy density ΩGW by the relation Recalling that ¯f(q) is related with the GW isotropic energy density ΩGW by the relation ∂ln ¯f ∂ln q = ∂ln ΩGW ∂ln q −4 , (6.49) (6.49) the non-linearity parameter fδGW NL can then be enhanced in proximity to large values of second derivatives of ΩGW as a function of the scale q. the non-linearity parameter fδGW NL can then be enhanced in proximity to large values of second derivatives of ΩGW as a function of the scale q. – 23 – 0.5 1.0 1.5 2.0 0 5 10 15 20 25 30 q/(10-5Hz) 1012 ΩGW fNL Figure 2. Representation of the GW spectral density ΩGW and of f δGW NL for the model given in Eqs. (6.50), (6.51), choosing a scale invariant Pζ. Notice that the magnitude of f δGW NL is amplified around the position where the spectral density changes slope. We have chosen the parameters α = 2, β = 5, κ0 = 1/10, Ω0 = 10−12, q⋆= 10−5 Hz−1. 0.5 1.0 1.5 2.0 0 5 10 15 20 25 30 q/(10-5Hz) 1012 ΩGW fNL Figure 2. Representation of the GW spectral density ΩGW and of f δGW NL for the model given in Eqs. (6.50), (6.51), choosing a scale invariant Pζ. 2We use Pζ(k3) instead of PδGW(k3) in the next equation, in order to simplify the overall coefficients in the equations that come next. Recall that the definitions of Pζ and PδGW are related by Eq. (6.9). 6.3 The squeezed limit of 3-point correlation functions Notice that the magnitude of f δGW NL is amplified around the position where the spectral density changes slope. We have chosen the parameters α = 2, β = 5, κ0 = 1/10, Ω0 = 10−12, q⋆= 10−5 Hz−1. As an illustrative toy model which demonstrates this effect, we can consider a GW spectral density with the shape of a broken power law. The following parameterisation for the spectral energy changes slope at a scale q = q⋆: As an illustrative toy model which demonstrates this effect, we can consider a GW spectral density with the shape of a broken power law. The following parameterisation for the spectral energy changes slope at a scale q = q⋆: ΩGW(q) = Ω0 2 ( q q⋆ α  tanh (1 −q/q⋆) κ0  + 1  +  q q⋆ −β  tanh (q/q⋆−1) κ0  + 1 ) (6 50) ) (6.50) ( ) with α, β positive numbers, while the functions inside the square parenthesis represent a regularisation of twice the Heaviside function (that is approached when sending κ0 →0). The function ΩGW has a large second derivative in proximity of the scale q⋆where the change of slope occurs. The value of fδGW NL at q⋆results (for a scale invariant spectrum of ζ) fδGW NL = 3 25 α + β κ0 (4 −(α + β) κ0) . (6.51) (6.51) Hence it can be enhanced taking small values of κ0. See Fig 2 for an illustration of this phenomenon, for a representative choice of parameters. Hence it can be enhanced taking small values of κ0. See Fig 2 for an illustration of this phenomenon, for a representative choice of parameters. 7 Conclusion The amount of information extracted from the detection of GW signals by the LIGO-Virgo collaboration has shown the power of GW to study astrophysical compact object and to give relevant cosmological information on the late time universe. At the same level, the improving angular resolution of future GW detectors will allow one to extract precious in- formation from the detection of the stochastic background of GWs generated both from the superposition of unresolved astrophysical sources and from cosmological sources, like infla- tion, phase transition or topological defects. However, high sensitivity alone will be not – 24 – sufficient for discriminating among different contributions. So it becomes necessary to char- acterize such backgrounds using observables that can give a clear hint about the origin of the signals. As recently studied, a parity-violating SGWB, which represents a smoking gun for some cosmological signals, can be probed using ground and space-based interferometers [63]. Another important tool is the directionality dependence of the SGWB. As shown for astrophysical GW, the distribution of sources implies that the energy density is character- ized by an anisotropic contribution beyond the isotropic one. In the same way we expect that, analogously to CMB photons, also primordial GW are charaterized by anisotropies that can be generated both at the moment of production and during their propagation. In this paper we focused on the stochastic background of cosmological origin and we studied the anisotropies due to the production mechanism (that we encode in an initial condition term) plus those generated from the propagation of GW on the perturbed universe, using a Boltz- mann approach. We solved the Boltzmann equation for the graviton distribution function considering a FLRW metric with both scalar and tensor inhomogeneities. We showed that, contrary to CMB photons, at the moment of production, GWs, which are characterized by a non-thermal spectrum, generically result in angular anisotropies that have an order one dependence on the GW frequency. We provide a criterion to evaluate whether and how much the GW anisotropies depend on frequency. As an example, we evaluate this criterion in the case where an axion inflaton φ sources gauge fields, which in turn generates a large GW background. Additional anisotropies are induced by the GW propagation in the the large-scale scalar and tensor perturbations of the universe. Acknowledgments N.B., D.B. and S.M. acknowledge partial financial support by ASI Grant No. 2016-24- H.0. A.R. is supported by the Swiss National Science Foundation (SNSF), project The Non-Gaussian Universe and Cosmological Symmetries, project number: 200020-178787. The work of G.T. is partially supported by STFC grant ST/P00055X/1. 7 Conclusion We compute the angular power spectrum of the SGWB energy density, and, analogously to CMB photons, also the gravitons distribution function gets mainly affected by the Sachs-Wolfe effect on large scales, while the Integrated Sachs-Wolfe is subdominant. We then focus on a second observable that can be a crucial tool in discriminating an astrophysical from a cosmological background, namely its departure from a Gaussian statistics. While we expect that the astrophysical background is Gaussian, due to central limit theorem, (some) cosmological backgrounds should shown a non-Gaussian statistics. We computed the three-point function (bispectrum) of the SGWB energy density, which is not affected by de-correlation issues, both considering the effects at generation and due to propagation. We have shown that also the SGWB bispectrum carries a memory of the initial condition and that it is proportional to the non-Gaussianity of the scalar perturbations. In this sense, the SGWB can be used as a novel probe (beyond the CMB and the LSS) of the non-Gaussianity of the scalar perturbations. Finally we consider non-linear effects induced by long-wavelength scalar perturbations, which generate a modulation effect on the correlation functions of the short-wavelength modes. We identified the effects of long modes with an appropriate coordinate transforma- tion and we computed the effect of non-linearities in inducing a non-vanishing squeezed limit of the SGWB three-point correlation function. We quantified the dependence of the squeezed bispectrum on the scale-dependence of the spectrum of primordial scalar fluctuations similar to Maldacena consistency relation, on the momentum dependence of the background SGWB distribution function, and on the time, scale, and direction dependence of the scalar transfer function. In summary, in this paper we have approached the possibility to use CMB techniques to describe the cosmological SGWB trying to characterize it using peculiar features that we do not expect to have in the astrophysical background. Of course the detectability with – 25 – interferometers of such effects is one crucial step to address and we plan to work on it on a future paper. At the same time we also plan to analyze several additional physical effects that we have neglected in this first paper, like the effects of neutrinos on the GW amplitude or a possible direct dependence of ΓI on ˆn, which would give distinctive signatures useful for the characterization. A Computation of the tensor sourced term We cope with this by rotating the integrand of the R d2Ωn integration into a basis in which the direction ˆn is decomposed according to Eq. (A.3). We need to evaluate the integral (A.5) for a generic orientation of ⃗k. On the other hand, the explicit expression of the integrand (A.6), holds only when ⃗k is oriented along the z−axis. We cope with this by rotating the integrand of the R d2Ωn integration into a basis in which the direction ˆn is decomposed according to Eq. (A.3). To achieve this, we introduce the rotation matrix S (Ωk) ≡   cos θk cos φk −sin φk sin θk cos φk cos θk sin φk cos φk sin θk sin φk −sin θk 0 cos θk  , (A.7) (A.7) in terms of which in terms of which ˆk = S (Ωk)   0 0 1  ,   sin θn cos φn sin θn sin φn cos φn  = S (Ωk)   sin θk,n cos φk,n sin θk,n sin φk,n cos φk,n  . (A.8) (A.8)  1   cos φn   cos φk,n  Under this rotation Under this rotation Under this rotation Under this rotation Y ∗ ℓm (Ωn) = ℓ X m′=−ℓ D(ℓ) mm′ (S (Ωk)) Y ∗ ℓm′ (Ωk,n) , dΩn = dΩk,n , (A.9) Y ∗ ℓm (Ωn) = ℓ X m′=−ℓ D(ℓ) mm′ (S (Ωk)) Y ∗ ℓm′ (Ωk,n) , dΩn = dΩk,n , (A.9) (A.9) ere the Wigner rotation matrix are given by where the Wigner rotation matrix are given by D(ℓ) ms (S (Ωk)) ≡ r 4π 2ℓ+ 1 (−1)s −sY ∗ ℓm (Ωk) , (A.10) (A.10) in terms of the spin-weighted spherical harmonics in terms of the spin-weighted spherical harmonics −sY ∗ ℓm (Ωk) ≡(−1)m s (ℓ+ m)! (ℓ−m)! (2ℓ+ 1) 4π (ℓ+ s)! (ℓ−s)! sin2ℓ θk 2  × ℓ−s X r=0  ℓ−s r   ℓ+ s r + s −m  (−1)ℓ−r−s eimφk cot2r+s−m θk 2  . A Computation of the tensor sourced term In this appendix we present the steps from Eq. (3.12) to Eq. (3.13) of the main text. The first goal is to obtain an explicit expression for the integrand in Eq. (3.11), when the integration variable ⃗k is oriented along the z−axis. In the {+×} basis, related to the circular basis by eij,λ ≡eij,+ + iλ eij,× √ 2 , eij,λ ≡eij,+ + iλ eij,× √ 2 , this orientation of ⃗k leads to eij,+  ˆkz  = 1 √ 2   1 0 0 0 −1 0 0 0 0  , eij,×  ˆkz  = 1 √ 2   0 1 0 1 0 0 0 0 0  . (A.1) (A.1) so that χ11  ˆkz  = −χ22  ˆkz  = χ (η, k) ξ−2  ⃗k  + ξ2  ⃗k  2 , χ12  ˆkz  = χ21  ˆkz  = χ (η, k) ξ−2  ⃗k  −ξ2  ⃗k  2i . (A.2) (A.2) while the other entries vanish. ˆ while the other entries vanish. We decompose the GW direction ˆn in a basis having ˆk as the z−axis We decompose the GW direction ˆn in a basis having ˆk as the z−axis ˆn = q 1 −µ2 k,n cos φk,n, q 1 −µ2 k,n sin φk,n, µk,n  , (A.3) (A.3) In this basis −ni nj 2 χ′ ij  ⃗k = k ˆkz  = − 1 −µ2 k,n 4 χ′ (η, k) h e2iφk,n ξ2  ⃗k  + e−2iφk,n ξ−2  ⃗k i . (A.4) Our goal is to compute Our goal is to compute Γℓm,T = Z d3k (2π)3 ei⃗k·⃗x0 Z d2Ωn ΓT  η0, ⃗k, Ωn  Y ∗ ℓm (Ωn) , (A.5) Γℓm,T = Z d3k (2π)3 ei⃗k·⃗x0 Z d2Ωn ΓT  η0, ⃗k, Ωn  Y ∗ ℓm (Ωn) , (A.5) (A.5) – 26 – with the knowledge that, when ⃗k is decomposed according to (A.3) (namely, with ˆk directeed along thee z−axis), ΓT  η0, ⃗k, Ωk,n  = − 1 −µ2 k,n 4 X λ=±2 eiλφk,n ξλ  ⃗k  Z η0 ηin dη χ′ (η, k) e−iµk(η0−η)k . (A.6) (A.6) We need to evaluate the integral (A.5) for a generic orientation of ⃗k. On the other hand, the explicit expression of the integrand (A.6), holds only when ⃗k is oriented along the z−axis. A Computation of the tensor sourced term (A.11) (A.11) (A.11) (A.11) ( ) With thi l ti th ti (A 5) b th itt With this relations, the equation (A.5) can be then rewritten as With this relations, the equation (A.5) can be then rewritten as Γℓm,T = Z d3k (2π)3 ei⃗k·⃗x0 ℓ X m′=−ℓ D(ℓ) mm′ (S (Ωk)) Z d2Ωk,n Y ∗ ℓm′ (Ωk,n) ΓT  η0, ⃗k, Ωk,n  . (A.12) where now the innermost integrand is performed in a basis in which the ˆn vector is decom- posed according to (A.3), so that the explicit expression (A.6) can be used. where now the innermost integrand is performed in a basis in which the ˆn vector is decom- posed according to (A.3), so that the explicit expression (A.6) can be used. – 27 – The inner integral evaluates to The inner integral evaluates to The inner integral evaluates to The inner integral evaluates to ,n Y ∗ ℓm′ (Ωk,n) ΓT  η0, ⃗k, Ωk,n  = Z d2Ωk,n s 2ℓ+ 1 4π (ℓ−m′)! (ℓ+ m′)! P m′ ℓ (µk,n) e−im′φk,n Z d2Ωk,n Y ∗ ℓm′ (Ωk,n) ΓT  η0, ⃗k, Ωk,n  = Z d2Ωk,n s 2ℓ+ 1 4π (ℓ−m′)! (ℓ+ m′)! P m′ ℓ (µk,n) e−im′φk,n × (−1) 1 −µ2 k,n 4 X λ=±2 eiλφk,n ξλ  ⃗k  Z η0 ηin dη χ′ (η, k) e−iµk(η0−η)k = − Z η0 ηin dη χ′ (η, k) Z 1 −1 dµk,n 1 −µ2 k,n 4 e−iµk(η0−η)k P 2 ℓ(µk,n) 2π s 2ℓ+ 1 4π (ℓ−2)! (ℓ+ 2)! X λ=±2 δm′λ ξλ  ⃗k  = Z η0 ηin dη χ′ (η, k) (−i)ℓjℓ(k (η0 −η)) k2 (η0 −η)2 p 4π (2ℓ+ 1) s (ℓ+ 2)! (ℓ−2)! 1 4 X λ=±2 δm′λ ξλ  ⃗k  . (A.13) × (−1) 1 −µ2 k,n 4 X λ=±2 eiλφk,n ξλ  ⃗k  Z η0 ηin dη χ′ (η, k) e−iµk(η0−η)k (A.13) Inserting this into Eq. (A.12), and using the relation (A.10) for the Wigner elements we finally arrive to Eq. (3.13) of the main text. Inserting this into Eq. (A.12), and using the relation (A.10) for the Wigner elements we finally arrive to Eq. (3.13) of the main text. B Tensor contribution to the GW bispectrum In this Appendix we present the steps from Eq. (4.14) to Eq. (4.15) of the main text. We start by introducing the quantity Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) from Eq. (2.6) of [60]: * 3 Y i=1 Z dΩkiξλ  ⃗ki  −λY ∗ ℓimi (Ωki) + ≡(2π)3 Fλ ℓ1ℓ2ℓ3 (k1, k2, k3)  ℓ1 ℓ2 ℓ3 m1 m2 m3  , (B.1) (B.1) (where we have also used Eq. (2.6) of [60] at the l.h.s.). This relation is inverted by Eq. (2.7) of [60]: Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) = X m1,m2,m3  ℓ1 ℓ2 ℓ3 m1 m2 m3  Z dΩk1 Z dΩk2 Z dΩk3 −λY ∗ ℓ1m1 (Ωk1)−λ Y ∗ ℓ2m2 (Ωk2)−λ Y ∗ ℓ3m3 (Ωk3) 1 (2π)3 D ξλ  ⃗k1  ξλ  ⃗k2  ξλ  ⃗k3 E . −λY ∗ ℓ1m1 (Ωk1)−λ Y ∗ ℓ2m2 (Ωk2)−λ Y ∗ ℓ3m3 (Ωk3) 1 (2π)3 D ξλ  ⃗k1  ξλ  ⃗k2  ξλ  ⃗k3 E . (B.2) We insert Eq. (B.1) in Eq. (4.14) to obtain We insert Eq. (B.1) in Eq. (4.14) to obtain We insert Eq. (B.1) in Eq. (4.14) to obtain We insert Eq. (B.1) in Eq. (4.14) to obtain B.1) in Eq. (4.14) to obtain * 3 Y i=1 Γℓimi,T + = Gm1m2m3 ℓ1ℓ2ℓ3  ℓ1 ℓ2 ℓ3 0 0 0 −1 s 4π (2ℓ1 + 1) (2ℓ2 + 1) (2ℓ3 + 1) × " 3 Y i=1 4π (−i)ℓi Z k2 i dki (2π)3 T T ℓ,i (ki, η0, ηin) # (2π)3 X λ=±2 Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) . (B 3) (B.3) where the relation (4.12) has also been used. We collect some of the factors in this expression into the combination ˜Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) ≡  ℓ1 ℓ2 ℓ3 0 0 0 −1 s 4π (2ℓ1 + 1) (2ℓ2 + 1) (2ℓ3 + 1) (2π)3 Fλ ℓ1ℓ2ℓ3 (k1, k2, k3) , (B.4) (B.4) – 28 – which then evaluates to the relation (4.16) in the main text. In terms of ˜F we then recover Eq. (4.15) of the main text. C Comparison with the CMB n the CMB case for a temperature T (ˆn) = ¯T + δT (ˆn), we have In the CMB case for a temperature T (ˆn) = ¯T + δT (ˆn), we have ¯f (p) = 1 e p ¯ T −1 f (p, ˆn) = 1 e p T (ˆn) −1 = 1 e p ¯ T −1 + e p ¯ T  e p ¯ T −1 2 p ¯T δT (ˆn) ¯T = ¯f (p) −p ∂¯f (p) ∂p δT (ˆn) ¯T   (C.1) (C.1) from which it follows Γ (ˆn) = δT (ˆn) ¯T , p independent (C.2) from which it follows Γ (ˆn) = δT (ˆn) ¯T , p independent (C.2) Γ (ˆn) = δT (ˆn) ¯T , Γ (ˆn) = δT (ˆn) ¯T , p independent (C.2) (C.2) To connect with the description of the SGWB, we also define wCMB (p, ˆn) = p4 f (p, ˆn) ρcrit , ¯wCMB (p) = p4 ¯f (p) ρcrit (C.3) wCMB (p, ˆn) = p4 f (p, ˆn) ρcrit , ¯wCMB (p) = p4 ¯f (p) ρcrit (C.3) that we have the p−dependent quantity ( ˆ) ( ) p δT (ˆ) (C.3) so that we have the p−dependent quantity so that we have the p−dependent quantity δCMB (p, ˆn) ≡wCMB (p, ˆn) −¯wCMB (p) ¯wCMB (p) = e p ¯ T e p ¯ T −1 p ¯T δT (ˆn) ¯T (C.4) (C.4) as well as the p−dependent quantity as well as the p−dependent quantity 4 −∂ln ¯ωCMB (η0, p) ∂ln p = 4 − ρcrit p3 ¯f (p) 4p3 ¯f (p) ρcrit + p4 ρcrit ∂¯f (p) ∂p  = −p  e p ¯ T −1  −1 ¯T e p ¯ T  e p ¯ T −1 2 = p ¯T e p ¯ T e p ¯ T −1 . (C.5) (C.5) So the ratio δCMB (p, ˆn) 4 −∂ln ¯ωCMB(η0, p) ∂ln p = δT (ˆn) ¯T (C.6) So the ratio So the ratio δCMB (p, ˆn) 4 −∂ln ¯ωCMB(η0, p) ∂ln p = δT (ˆn) ¯T (C.6) So the ratio δCMB (p, ˆn) 4 −∂ln ¯ωCMB(η0, p) ∂ln p = δT (ˆn) ¯T (C.6) δCMB (p, ˆn) 4 −∂ln ¯ωCMB(η0, p) ∂ln p = δT (ˆn) ¯T (C.6) (C.6) is indeed p−independent. is indeed p−independent. References [1] B. P. Abbott et al. [LIGO Scientific and Virgo Collaborations], Phys. Rev. D 100 (2019) no.6, 061101 doi:10.1103/PhysRevD.100.061101 [arXiv:1903.02886 [gr-qc]]. [2] P. Amaro-Seoane et al. [LISA Collaboration], arXiv:1702.00786 [astro-ph.IM]. [3] S K t l Cl Q t G 23 (2006) S125 d i 10 1088/0264 9381/23/8/S17 S. Kawamura et al., Class. Quant. Grav. 23 (2006) S125. doi:10.1088/0264-9381/23/8/S1 [4] B. Sathyaprakash et al., arXiv:1108.1423 [gr-qc]. [4] B. Sathyaprakash et al., arXiv:1108.1423 [gr-qc]. [5] M. Maggiore et al., arXiv:1912.02622 [astro-ph.CO]. [5] M. Maggiore et al., arXiv:1912.02622 [astro-ph.CO]. [5] M. Maggiore et al., arXiv:1912.02622 [astro-ph.CO]. – 29 – [6] B. P. Abbott et al. [LIGO Scientific Collaboration], Class. Quant. Grav. 34, no. 4, 044001 (2017) [arXiv:1607.08697 [astro-ph.IM]]. [7] M. Maggiore, Phys. Rept. 331 (2000) 283 doi:10.1016/S0370-1573(99)00102-7 [gr-qc/9909001]. [8] M C Guzzetti N Bartolo M Liguori and S Matarrese Riv Nuovo Cim 39 no 9 399 [7] M. Maggiore, Phys. Rept. 331 (2000) 283 doi:10.1016/S0370-1573(99)00102-7 [gr-qc/9909001]. [8] M. C. Guzzetti, N. Bartolo, M. Liguori and S. Matarrese, Riv. Nuovo Cim. 39, no. 9, 399 (2016) [arXiv:1605.01615 [astro-ph.CO]]. [9] N. Bartolo et al., JCAP 1612, no. 12, 026 (2016) [arXiv:1610.06481 [astro-ph.CO]]. [10] C. Caprini and D. G. Figueroa, Class. Quant. Grav. 35 (2018) no.16, 163001 doi:10.1088/1361-6382/aac608 [arXiv:1801.04268 [astro-ph.CO]]. [11] T. Regimbau, Res. Astron. Astrophys. 11, 369 (2011) [arXiv:1101.2762 [astro-ph.CO]]. [12] J. D. Romano and N. J. Cornish, Living Rev. Rel. 20, no. 1, 2 (2017) [arXiv:1608.06889 [gr-qc]]. [13] C. Caprini, D. G. Figueroa, R. Flauger, G. Nardini, M. Peloso, M. Pieroni, A. Ricciardone and G. Tasinato, arXiv:1906.09244 [astro-ph.CO]. [14] V. Alba and J. Maldacena, JHEP 1603 (2016) 115 [arXiv:1512.01531 [hep-th]]. [15] C. R. Contaldi, Phys. Lett. B 771 (2017) 9 [arXiv:1609.08168 [astro-ph.CO]]. [16] G. Cusin, C. Pitrou and J. P. Uzan, Phys. Rev. D 96 (2017) no.10, 103019 [arXiv:1704.06184 [astro-ph.CO]]. [17] A. C. Jenkins and M. Sakellariadou, Phys. Rev. D 98 (2018) no.6, 063509 [arXiv:1802.06046 [astro-ph.CO]]. [18] G. Cusin, R. Durrer and P. G. Ferreira, Phys. Rev. D 99 (2019) no.2, 023534 [arXiv:1807.10620 [astro-ph.CO]]. [19] D. Bertacca, A. Ricciardone, N. Bellomo, A. C. Jenkins, S. Matarrese, A. Raccanelli, T. Regimbau and M. Sakellariadou, arXiv:1909.11627 [astro-ph.CO]. [20] N. Barnaby and M. Peloso, Phys. Rev. Lett. 106 (2011) 181301 doi:10.1103/PhysRevLett.106.181301 [arXiv:1011.1500 [hep-ph]]. [20] N. Barnaby and M. Peloso, Phys. Rev. Lett. 106 (2011) 181301 doi:10.1103/PhysRevLett.106.181301 [arXiv:1011.1500 [hep-ph]]. [21] J. L. Cook and L. Sorbo, Phys. Rev. D 85, 023534 (2012) Erratum: [Phys. Rev. References D 86, 069901 (2012)] [arXiv:1109.0022 [astro-ph.CO]]. [22] M. Geller, A. Hook, R. Sundrum and Y. Tsai, Phys. Rev. Lett. 121, no. 20, 201303 (2018) [arXiv:1803.10780 [hep-ph]]. [23] N. Bartolo, V. De Luca, G. Franciolini, A. Lewis, M. Peloso and A. Riotto, Phys. Rev. Lett. 122 (2019) no.21, 211301 [arXiv:1810.12218 [astro-ph.CO]]. [24] N. Bartolo, V. De Luca, G. Franciolini, M. Peloso, D. Racco and A. Riotto, Phys. Rev. D 99, no. 10, 103521 (2019) [arXiv:1810.12224 [astro-ph.CO]]. N. Bartolo et al., arXiv:1909.12619 [astro-ph.CO]. [25] N. Bartolo et al., arXiv:1909.12619 [astro-ph.CO]. [26] C. Powell and G. Tasinato, “Probing a stationary non-Gaussian background of stochastic gravitational waves with pulsar timing arrays,” arXiv:1910.04758 [gr-qc]. [27] S. Weinberg, “Adiabatic modes in cosmology,” Phys. Rev. D 67 (2003) 123504 doi:10.1103/PhysRevD.67.123504 [astro-ph/0302326]. [28] N. Bartolo, D. Bertacca, S. Matarrese, M. Peloso, A. Ricciardone, A. Riotto and G. Tasinato, Phys. Rev. D 100 (2019) no.12, 121501 doi:10.1103/PhysRevD.100.121501 [arXiv:1908.00527 [astro-ph.CO]]. [29] N. Bartolo, A. Hoseinpour, G. Orlando, S. Matarrese and M. Zarei, Phys. Rev. D 98 (2018) no.2, 023518 [arXiv:1804.06298 [gr-qc]]. [30] S. Dodelson, “Modern Cosmology’. [30] S. Dodelson, “Modern Cosmology’. – 30 – [31] N. Bartolo et al., arXiv:1909.12619 [astro-ph.CO]. [32] A. Ricciardone and G. Tasinato, JCAP 1802 (2018) no.02, 011 [arXiv:1711.02635 [astro-ph.CO]]. [33] E. Dimastrogiovanni, M. Fasiello and G. Tasinato, arXiv:1906.07204 [astro-ph.CO]. [34] L. Sorbo, JCAP 1106 (2011) 003 doi:10.1088/1475-7516/2011/06/003 [arXiv:1101.1525 [astro-ph.CO]]. [35] M. M. Anber and L. Sorbo, Phys. Rev. D 85 (2012) 123537 doi:10.1103/PhysRevD.85.123537 [arXiv:1203.5849 [astro-ph.CO]]. [36] M. Shiraishi, A. Ricciardone and S. Saga, JCAP 1311 (2013) 051 doi:10.1088/1475-7516/2013/11/051 [arXiv:1308.6769 [astro-ph.CO]]. [37] D. Bertacca, A. Raccanelli, N. Bartolo and S. Matarrese, Phys. Dark Univ. 20 (2018) 32 [arXiv:1702.01750 [gr-qc]]. [38] A. Renzini and C. Contaldi, arXiv:1907.10329 [gr-qc]. [38] A. Renzini and C. Contaldi, arXiv:1907.10329 [gr-qc]. [39] N. Bartolo, S. Matarrese and A. Riotto, JCAP 0701 (2007) 019 [astro-ph/0610110]. [40] N. Bartolo, S. Matarrese and A. Riotto, JCAP 0606, 024 (2006) [astro-ph/0604416]. [41] J. Chluba, R. Khatri and R. A. Sunyaev, Mon. Not. Roy. Astron. Soc. 425 (2012) 1129 [arXiv:1202.0057 [astro-ph.CO]]. [42] J. Garcia-Bellido, M. Peloso and C. Unal, JCAP 1612, no. 12, 031 (2016) [arXiv:1610.03763 [astro-ph.CO]]. [43] N. Bartolo et al., JCAP 1811, no. 11, 034 (2018) [arXiv:1806.02819 [astro-ph.CO]]. [44] Y. Akrami et al. [Planck Collaboration], arXiv:1905.05697 [astro-ph.CO]. [45] N. Bartolo, E. Komatsu, S. Matarrese and A. Riotto, Phys. Rept. 402, 103 (2004) [astro-ph/0406398]. [46] E. Komatsu and D. N. Spergel, Phys. Rev. References D 63, 063002 (2001) [astro-ph/0005036]. [47] A. Gangui, F. Lucchin, S. Matarrese and S. Mollerach, Astrophys. J. 430, 447 (1994) [astro-ph/9312033]. [48] N. Kaiser, Astrophys. J. 388, 272 (1992). doi:10.1086/171151 [49] J. M. Maldacena, JHEP 0305 (2003) 013 [astro-ph/0210603]. [50] N. Bartolo, S. Matarrese and A. Riotto, JCAP 1202 (2012) 017 [arXiv:1109.2043 [astro-ph.CO]]. [51] P. Creminelli, C. Pitrou and F. Vernizzi, JCAP 1111 (2011) 025 [arXiv:1109.1822 [astro-ph.CO]]. [52] A. Lewis, JCAP 1206 (2012) 023 [arXiv:1204.5018]. [53] S. Weinberg, Phys. Rev. D 69, 023503 (2004) [astro-ph/0306304]. [54] J. M. Maldacena, “Non-Gaussian features of primordial fluctuations in single field inflationary models,” JHEP 0305 (2003) 013 doi:10.1088/1126-6708/2003/05/013 [astro-ph/0210603]. [55] N. Bartolo, S. Matarrese and A. Riotto, “Non-Gaussianity in the Cosmic Microwave Background Anisotropies at Recombination in the Squeezed limit,” JCAP 1202 (2012) 017 doi:10.1088/1475-7516/2012/02/017 [arXiv:1109.2043 [astro-ph.CO]]. [56] P. Creminelli, C. Pitrou and F. Vernizzi, “The CMB bispectrum in the squeezed limit,” JCAP 1111 (2011) 025 doi:10.1088/1475-7516/2011/11/025 [arXiv:1109.1822 [astro-ph.CO]]. [57] A. Lewis, “The full squeezed CMB bispectrum from inflation,” JCAP 1206 (2012) 023 [57] A. Lewis, “The full squeezed CMB bispectrum from inflation,” JCAP 1206 (2012) 023 A. Lewis, “The full squeezed CMB bispectrum from inflation,” JCAP 1206 (2012) 023 – 31 – doi:10.1088/1475-7516/2012/06/023 [arXiv:1204.5018 [astro-ph.CO]]. doi:10.1088/1475-7516/2012/06/023 [arXiv:1204.5018 [astro-ph.CO]]. [58] N. Bartolo, D. Bertacca, S. Matarrese, M. Peloso, A. Ricciardone, A. Riotto and G. Tasinato, Phys. Rev. D 100 (2019) no.12, 121501 doi:10.1103/PhysRevD.100.121501 [arXiv:1908.00527 [astro-ph.CO]]. [59] E. Komatsu, astro-ph/0206039. [60] M. Shiraishi, D. Nitta, S. Yokoyama, K. Ichiki and K. Takahashi, Prog. Theor. Phys. 125 (2011) 795 doi:10.1143/PTP.125.795 [arXiv:1012.1079 [astro-ph.CO]]. [61] O. Lahav, P. B. Lilje, J. R. Primack and M. J. Rees, “Dynamical effects of the cosmological constant,” Mon. Not. Roy. Astron. Soc. 251 (1991) 128. [62] S. M. Carroll, W. H. Press and E. L. Turner, “The Cosmological constant,” Ann. Rev. Astron. Astrophys. 30 (1992) 499. doi:10.1146/annurev.aa.30.090192.002435 [63] V. Domcke, J. Garcia-Bellido, M. Peloso, M. Pieroni, A. Ricciardone, L. Sorbo and G. Tasinato, arXiv:1910.08052 [astro-ph.CO]. – 32 –

https://openalex.org/W4388629077

https://www.frontiersin.org/articles/10.3389/fphar.2023.1277143/pdf?isPublishedV2=False

English

Structure-function and rational design of a spider toxin Ssp1a at human voltage-gated sodium channel subtypes

Frontiers in pharmacology

cc-by

OPEN ACCESS OPEN ACCESS EDITED BY Cassandra M. Modahl, Liverpool School of Tropical Medicine, United Kingdom REVIEWED BY Vladimir Yarov-Yarovoy, University of California, Davis, United States Stephan Kellenberger, Université de Lausanne, Switzerland *CORRESPONDENCE Richard J. Lewis, [email protected] RECEIVED 14 August 2023 ACCEPTED 23 October 2023 PUBLISHED 13 November 2023 CITATION Dongol Y, Wilson DT, Daly NL, Cardoso FC and Lewis RJ (2023), Structure-function and rational design of a spider toxin Ssp1a at human voltage- gated sodium channel subtypes. Front. Pharmacol. 14:1277143. doi: 10.3389/fphar.2023.1277143 EDITED BY Cassandra M. Modahl, Liverpool School of Tropical Medicine, United Kingdom REVIEWED BY Vladimir Yarov-Yarovoy, University of California, Davis, United States Stephan Kellenberger, Université de Lausanne, Switzerland Yashad Dongol1, David T. Wilson2, Norelle L. Daly2, Fernanda C. Cardoso1 and Richard J. Lewis1* 1Centre for Chemistry and Drug Discovery, Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia, 2Australian Institute of Tropical Health and Medicine, James Cook University, Cairns, QLD, Australia The structure-function and optimization studies of NaV-inhibiting spider toxins have focused on developing selective inhibitors for peripheral pain-sensing NaV1.7. With several NaV subtypes emerging as potential therapeutic targets, structure-function analysis of NaV-inhibiting spider toxins at such subtypes is warranted. Using the recently discovered spider toxin Ssp1a, this study extends the structure-function relationships of NaV-inhibiting spider toxins beyond NaV1.7 to include the epilepsy target NaV1.2 and the pain target NaV1.3. Based on these results and docking studies, we designed analogues for improved potency and/or subtype-selectivity, with S7R-E18K-rSsp1a and N14D-P27R-rSsp1a identified as promising leads. S7R-E18K-rSsp1a increased the rSsp1a potency at these three NaV subtypes, especially at NaV1.3 (~10-fold), while N14D-P27R-rSsp1a enhanced NaV1.2/1.7 selectivity over NaV1.3. This study highlights the challenge of developing subtype-selective spider toxin inhibitors across multiple NaV subtypes that might offer a more effective therapeutic approach. The findings of this study provide a basis for further rational design of Ssp1a and related NaSpTx1 homologs targeting NaV1.2, NaV1.3 and/or NaV1.7 as research tools and therapeutic leads. Dongol Y, Wilson DT, Daly NL, Cardoso FC and Lewis RJ (2023), Structure-function and rational design of a spider toxin Ssp1a at human voltage- gated sodium channel subtypes. Front. Pharmacol. 14:1277143. doi: 10.3389/fphar.2023.1277143 COPYRIGHT © 2023 Dongol, Wilson, Daly, Cardoso and Lewis. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). KEYWORDS ICK toxins, rational design, spider toxin, Ssp1a, structure-function, voltage-gated sodium channels TYPE Original Research PUBLISHED 13 November 2023 DOI 10.3389/fphar.2023.1277143 TYPE Original Research PUBLISHED 13 November 2023 DOI 10.3389/fphar.2023.1277143 TYPE Original Research PUBLISHED 13 November 2023 DOI 10.3389/fphar.2023.1277143 OPEN ACCESS The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. frontiersin.org 1 Introduction Voltage-gated sodium (NaV) channels underpin electrical signaling by allowing passive and rapid influx of Na+ ions necessary to control initiation and propagation of action potentials in electrically excitable cells, including neurons and muscles (Ahern et al., 2016). Accordingly, NaV channel dysfunction is associated with various neuronal and neuromuscular disorders, including pain, epilepsy, arrythmia and myopathy (de Lera Ruiz and Kraus, 2015; Cardoso and Lewis, 2018; Dib-Hajj and Waxman, 2019; Cardoso, 2020; Menezes et al., 2020; Goodwin and McMahon, 2021). These channels can be pharmacologically modulated by neurotoxins that bind to different sites on the NaV channel to alter the voltage-dependence of activation, inactivation, and conductance (Stevens et al., 2011; de Lera Ruiz and Kraus, 2015). Peptidic gating modifier toxins, including spider toxins, preferentially target the extracellular binding sites located in the domain II (DII) and domain IV (DIV) of the NaV channels to modulate the channel gating. Frontiers in Pharmacology 01 frontiersin.org 10.3389/fphar.2023.1277143 Dongol et al. FIGURE 1 Structure-function of Ssp1a. (A) Primary structure of native Ssp1a and recombinant Ssp1a (rSsp1a) showing the disulfide bond connectivity. (B) Fold potency difference of rSsp1a alanine mutants vs. wild type (WT) rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7. Data were presented as means ± SEM, with n = 3–12. The W5A, F6A, W24A, K25A, Y26A, W28A, R30A did not complete the dose-response curve at hNaV1.2, hNaV1.3, and hNaV1.7 at the maximum concentration tested. Similarly, Y20A, P27A, Y31A and L33A at hNaV1.3 also demonstrated incomplete dose-response. The IC50 values from the incomplete dose-response for these mutants were used to plot the panel for better comparison of rSsp1a residues key to hNaV1.2, hNaV1.3, and hNaV1.7 activity. The readers are requested to follow Supplementary Table S1 for relative IC50 values and Supplementary Figures S1 for dose-response curves. (C) Active residues in the surface of rSsp1a (PDB: 7SKC) Dongol et al., 2021), GpTx-1 (modeled on engineered GPTX-1, PDB: 6MK5) (Murray et al., 2015), HwTx-IV (PDB: 2M4X) Revell et al., 2013) and m3-HwTx-IV (PDB: 5T3M) (Wisedchaisri et al., 2021) were compared to identify which alanine mutation reduced the toxin potency at hNaV1.7 as indicated by the respective color code (red or pink). All four toxins were aligned at NaSpTx1 signature motif WCK/R (W and K/R bolded). (D) Secondary Hα chemical shift of rSsp1a and the alanine mutants. Frontiers in Pharmacology 2.2 Automated whole-cell patch-clamp electrophysiology NaV channel currents from HEK293 cells stably expressing NaV subtypes and the β1 auxiliary subunit were recorded using an automated whole-cell patch clamp system QPatch 16X (Sophion Bioscience A/S, Ballerup, Denmark). As per the manufacturer’s guidelines, the cells were cultured for 48 h to achieve ~80% confluency, detached using Detachin (Genlantis) and resuspended to 5 × 106 cells/mL in serum free media [CHO-cell SFM (Life Technologies), 25 mM HEPES and 100 U/mL penicillin/ streptomycin]. The extracellular solution comprised (in mM) 1 CaCl2, 1 MgCl2, 5 HEPES, 3 KCl, 140 NaCl and 20 TEA-Cl, with the pH adjusted to 7.3 with NaOH. The intracellular solution comprised (in mM) 140 CsF, 1 EGTA, 5 CsOH, 10 HEPES and 10 NaCl, with the pH adjusted to 7.3 with CsOH. The osmolarity of both solutions were adjusted to 320 mOsm with sucrose. Compounds were prepared in extracellular solution containing 0.1% bovine serum albumin (Sigma-Aldrich). To obtain the dose- response curves, cells were maintained at a holding potential −80 mV and Na+ currents were elicited by 20 m voltage steps to 0 mV from a −120 mV conditioning pulse applied for 200 m. Increasing concentrations of the peptide were incubated with the cells at the holding potential for 2 min before the voltage protocol was applied. In this study, 15 alanine mutants of rSsp1a in the NaV pharmacophore region of NaSpTx1 inhibitors (Li et al., 2004; Klint et al., 2012; Minassian et al., 2013; Revell et al., 2013; Murray et al., 2015; Murray et al., 2016; Shcherbatko et al., 2016) identified the Ssp1a-specific pharmacophore. Using these restraints, docking studies identified specific molecular interactions between rSsp1a and hNaV1.2, hNaV1.3 and hNaV1.7. Through structure- function studies of rSsp1a activity at hNaV1.2, hNaV1.3 and hNaV1.7 and previous optimization studies of NaSpTx1 peptides (Minassian et al., 2013; Revell et al., 2013; Murray et al., 2015; Zhang et al., 2015; Murray et al., 2016; Shcherbatko et al., 2016; Zhang et al., 2018; Neff et al., 2020), we designed rSsp1a analogues with significantly improved potency and subtype-selectivity for hNaV1.3 and hNaV1.2/hNaV1.7, respectively. The findings of this study provide insight into rational design of rSsp1a and NaSpTx1 homologs targeting the hNaV1.2, hNaV1.3, and hNaV1.7 subtypes, with single or multiple subtype-selectivity, with the aim of developing lead molecules that have high value as research tools and/or therapeutic agents. 2.1 Cell culture Human embryonic kidney 293 (HEK293) cells stably expressing recombinant hNaV1.2, hNaV1.3 and hNaV1.7 and the β1 auxiliary subunit (Scottish Biomedical Drug Discovery, Glasgow, UK) were cultured in Minimal Essential medium (MEM) (Sigma-Aldrich, MO, United States) supplemented with 10% v/v FBS-New Zealand origin (Assay Matrix), 2 mM L-glutamine and selection antibiotics as per manufacturer’s recommendation. The HEK293 cells heterologously expressing mNaV1.7 and F813G- mNaV1.7 were generously provided by Prof Irina Vetter and were cultured in Minimal Essential medium (MEM) (Sigma- Aldrich, MO, United States) supplemented with 10% v/v FBS- New Zealand origin (Assay Matrix), 2 mM L-glutamine and hygromycin 100 μg/mL. Ssp1a, a 33-residue inhibitor cystine knot (ICK) peptide (Figure 1A) belonging to the voltage-gated sodium channel modulator spider toxin family 1 (NaSpTx1), is a potent inhibitor of neuronal hNaV subtypes 1.7, 1.6, 1.3, 1.2 and 1.1 (Dongol et al., 2021). The closest homologs with comprehensive structure-function data available are the distantly related GpTx-1 (44% identity) and HwTx-IV (40% identity), with HwTx-IV and recombinant Ssp1a (rSsp1a) showing similar pharmacology at hNaV1.7 (Xiao et al., 2008; Dongol et al., 2021). While the NaSpTx1 toxin studies have focused on the development of hNaV1.7-selective inhibitors (Minassian et al., 2013; Revell et al., 2013; Klint et al., 2015; Murray et al., 2015; Murray et al., 2016; Shcherbatko et al., 2016; Rahnama et al., 2017; Zhang et al., 2018; Neff et al., 2020), the determinants of NaSpTx1 pharmacology at the potential pain target hNaV1.3 (Black et al., 1999; Kim et al., 2001; Hains et al., 2004; Hong et al., 2004; Garry et al., 2005; Lindia et al., 2005; Black et al., 2008; Chen et al., 2014; Tan et al., 2015; Xu et al., 2016) and epilepsy target hNaV1.2 (Menezes et al., 2020) have been largely ignored. Therefore, the characterization of Ssp1a in this study could provide common structure-function information for several closer uncharacterized homologs and at targets hNaV1.2 and hNaV1.3, in addition to hNaV1.7. 1 Introduction The secondary shifts were derived by subtracting the random coil Hα shift from the experimental Hα shifts for the eight alanine substitutions distributed across loop 1, loop 3, loop 4 and the C-terminal. The Hα FIGURE 1 Structure-function of Ssp1a. (A) Primary structure of native Ssp1a and recombinant Ssp1a (rSsp1a) showing the disulfide bond connectivity. (B) Fold potency difference of rSsp1a alanine mutants vs. wild type (WT) rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7. Data were presented as means ± SEM, with n = 3–12. The W5A, F6A, W24A, K25A, Y26A, W28A, R30A did not complete the dose-response curve at hNaV1.2, hNaV1.3, and hNaV1.7 at the maximum concentration tested. Similarly, Y20A, P27A, Y31A and L33A at hNaV1.3 also demonstrated incomplete dose-response. The IC50 values from the incomplete dose-response for these mutants were used to plot the panel for better comparison of rSsp1a residues key to hNaV1.2, hNaV1.3, and hNaV1.7 activity. The readers are requested to follow Supplementary Table S1 for relative IC50 values and Supplementary Figures S1 for dose-response curves. (C) Active residues in the surface of rSsp1a (PDB: 7SKC) Dongol et al., 2021), GpTx-1 (modeled on engineered GPTX-1, PDB: 6MK5) (Murray et al., 2015), HwTx-IV (PDB: 2M4X) Revell et al., 2013) and m3-HwTx-IV (PDB: 5T3M) (Wisedchaisri et al., 2021) were compared to identify which alanine mutation reduced the toxin potency at hNaV1.7 as indicated by the respective color code (red or pink). All four toxins were aligned at NaSpTx1 signature motif WCK/R (W and K/R bolded). (D) Secondary Hα chemical shift of rSsp1a and the alanine mutants. The secondary shifts were derived by subtracting the random coil Hα shift from the experimental Hα shifts for the eight alanine substitutions distributed across loop 1, loop 3, loop 4 and the C-terminal. The Hα for rSsp1a shifted in W5A, F6A and Y20A mutations. The NaV channel is structurally composed of four non- homologous domains (DI–DIV), which collectively forms the functional, pore-forming α-subunit (Shen et al., 2017; Dongol et al., 2019). There are nine known human NaV isoforms (hNaV1.1–1.9), each with distinct tissue localization, channel kinetics and physiological functions (de Lera Ruiz and Kraus, 2015). Modulating a specific isoform is key to avoiding side effects associated with the use of NaV-inhibitor drugs; however, 02 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 2 Materials and methods the high structural homology between NaV subtypes (Vetter et al., 2017) remains a challenge in obtaining subtype-selective inhibitors, which could be addressed by optimizing ligands, including venom peptides. Venoms evolved for prey capture and/or defense are rich in peptide NaV-modulators (Kalia et al., 2015). Research in the 1980s first identified the NaV-modulating effects of spider venom toxins (Fontana and Vital-Brazil, 1985; Adams et al., 1989), many of which are now used as research tools to help define the structure, function and pharmacology of NaV channels (Escoubas et al., 2000; Stevens et al., 2011; Kalia et al., 2015; Wu et al., 2018) and their role in disease (Osteen et al., 2016). More recently, these complex venom peptide libraries have been exploited for potential drug leads (Saez et al., 2010; Pineda et al., 2014; Cardoso and Lewis, 2019; Saez and Herzig, 2019; Cardoso et al., 2022), including CcoTx-1 (Shcherbatko et al., 2016), GpTx-1 (Murray et al., 2015; Murray et al., 2016), ProTx-II (Flinspach et al., 2017) and Tap1a (Hu et al., 2021). the high structural homology between NaV subtypes (Vetter et al., 2017) remains a challenge in obtaining subtype-selective inhibitors, which could be addressed by optimizing ligands, including venom peptides. Venoms evolved for prey capture and/or defense are rich in peptide NaV-modulators (Kalia et al., 2015). Research in the 1980s first identified the NaV-modulating effects of spider venom toxins (Fontana and Vital-Brazil, 1985; Adams et al., 1989), many of which are now used as research tools to help define the structure, function and pharmacology of NaV channels (Escoubas et al., 2000; Stevens et al., 2011; Kalia et al., 2015; Wu et al., 2018) and their role in disease (Osteen et al., 2016). More recently, these complex venom peptide libraries have been exploited for potential drug leads (Saez et al., 2010; Pineda et al., 2014; Cardoso and Lewis, 2019; Saez and Herzig, 2019; Cardoso et al., 2022), including CcoTx-1 (Shcherbatko et al., 2016), GpTx-1 (Murray et al., 2015; Murray et al., 2016), ProTx-II (Flinspach et al., 2017) and Tap1a (Hu et al., 2021). Frontiers in Pharmacology frontiersin.org 2.5 Recombinant production of rSsp1a analogues collected over 4,096 data points in the f2 dimension and 512 increments in the f1 dimension over a spectral width of 12 ppm. Standard Bruker pulse sequences were used with an excitation sculpting scheme for solvent suppression. NMR assignments were made using established protocols (Wüthrich, 1983), and the secondary shifts derived by subtracting the random coil αH shift from the experimental αH shifts (Wishart et al., 1995). Spectra were recorded referenced to external 4,4- dimethyl-4-silapentane-1-sulfonic acid (DSS). rSsp1a analogues were recombinantly expressed in E. coli as a His6 tagged-MBP fused peptide, harvested, and purified as described previously (Dongol et al., 2021). Briefly, the mutant plasmids were transformed into BL21 (λDE3) competent E. coli cells and cultured in LB-Amp medium at 37°C, 120 rpm until the optical density at 600 nm (OD600) reached 0.8–1.0. Peptide expression was induced with 500 μM isopropyl β-D-1-thiogalactopyranoside (IPTG) at 16°C and 120 rpm overnight and pelleted at 6,000 rpm for 10 min at 4°C. The pellet was resuspended in TN buffer (Tris 25 mM, NaCl 150 mM, pH 8.0) and lysed in a constant pressure cell disruptor at 25 kPa at 4°C–8°C. The fusion protein contained in the cell lysate was captured by passing the lysate supernatant through Ni-NTA resin (Hispur NiNTA, Thermo Scientific) and then eluted with TN buffer containing 500 mM imidazole. After desalting, the fusion protein was cleaved with TEV protease in the reducing environment provided by glutathione redox pair. The post-cleavage sample was filtered through a centrifuge filter to isolate the cleaved peptide from tag proteins and loaded onto a reversed-phase C18 column (30Å, 5μm, 4.6 × 250mm, Vydac 218TP, Grace) on an Agilent 1100 series HPLC for purification. The peptide was eluted using the following gradient of solvent B (90% ACN, 0.05% TFA in MilliQ water) in solvent A (0.05% TFA in MilliQ water): 5% solvent B over 0–5 min, 5%–10% solvent B over 5–10 min, 10%–50% solvent B over 10–40 min, 50%–80% solvent B over 40–45 min, wash at 80% solvent B over 45–50 min, 80%–5% solvent B over 50–55 min and a final wash with 5% solvent B over 55–65 min at a flow rate of 1 mL/min. Peak fractions at 214 nm were collected, checked the purity, analysed for the mass using MALDI-TOF/TOF (SCIEX 5800), lyophilized, quantitated using nanodrop, and stored at −20°C until use. 2.3 Nuclear magnetic resonance (NMR) structure determination of rSsp1a alanine mutants Lyophilized peptide (500–1,000 µg) was resuspended in 90% H2O:10%D2O. Two-dimensional 1H-1H TOCSY and 1H-1H NOESY spectra were acquired at 290 K using a 600 MHz AVANCE III NMR spectrometer (Bruker, Karlsruhe, Germany) equipped with a cryogenically cooled probe. All spectra were recorded with an interscan delay of 1 s. NOESY spectra were acquired with mixing times of 200–250 ms and TOCSY spectra were acquired with isotropic mixing periods of 80 ms. Two-dimensional spectra were Frontiers in Pharmacology 03 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 2.4 Alanine scanning and rational design of rSsp1a Based on homology and earlier structure-function studies of NaSpTx1 toxins, 16 residues were selected for alanine substitution. The rSsp1a structure-function data across hNaV1.2, hNaV1.3, and hNaV1.7 were then used along with the homology and structure- function information of other NaSpTx1 toxins to guide rational design of rSsp1a. Unfortunately, the docking model of rSsp1a at hNaV subtypes were not used for rSsp1a rational design as the resting structure of hNaV1.7 DII was published (Wisedchaisri et al., 2021) after the mutants were made. In total, sixteen positions in rSsp1a were mutated to recombinantly produce 38 rSsp1a- analogues, including 30 single mutants, five double mutants, one mutant with an N-terminal extension and two mutants with a C-terminal extension. A N-terminal extended version (GP-Ssp1a) was designed to mimic GP-HwTx-IV (Neff et al., 2020) and GP- ProTx-II (Flinspach et al., 2017), whereas a C-terminal extension (rSsp1a-GK) was designed to mimic the C-terminal amide version of HwTx-IV (Minassian et al., 2013; Neff et al., 2020). Double mutants were designed based on the activity of each single mutation at the three hNaV subtypes. The rSsp1a mutant plasmids were produced using a QuikChange Lightning Site-Directed Mutagenesis Kit (Agilent Technologies) and Ssp1a-pLicC plasmid construct (GeneArt Gene Synthesis, Life Technologies) comprising MalE signal sequence for periplasmic export, His6 affinity tag, maltose binding protein (MBP) tag, a tobacco etch virus (TEV) recognition and cleavage sequence. Briefly, primers were designed using SnapGene software and purchased from Sigma-Aldrich, and the mutant strand was synthesized using polymerase chain reaction (PCR). The amplified PCR product was digested with Dpn I restriction enzyme to eliminate the parental rSsp1a construct and was transformed into Competent E. coli TOP10 cells. The transformed TOP10 cells were plated on Luria-Bertani (LB) agar plates containing 100 μg/mL ampicillin (Amp) and incubated overnight at 37°C. The isolated colonies were sub-cultured overnight at 37°C at 120 rpm, and the plasmid DNA was extracted using a QIAprep Spin Miniprep kit (QIAGEN) following manufacturer’s protocol. The desired mutation in the extracted DNA was confirmed by Sanger sequencing at the Australian Genome Research Facility, Brisbane using Big Dye Terminator (BDT) chemistry version 3.1 (Applied Biosystem). Double mutant plasmids were made using rSsp1a single mutant plasmids as a base construct and a set of primers for the second mutation, and following site-directed mutagenesis, miniprep and Sanger sequencing as described above. Frontiers in Pharmacology frontiersin.org 2.6 Molecular docking The recently solved structure of NaVAb/NaV1.7 VS2A chimera (PDB: 7K48) (Wisedchaisri et al., 2021) provided the resting state structure of the NaV1.7 DII, which was used as a template to build a homology model for hNaV1.2 DII (UniProt: Q99250) and hNaV1.3 DII (UniProt: Q9NY46) in the resting state using SWISS- MODEL (Waterhouse et al., 2018). HADDOCK2.2 webserver was used to perform data-driven docking studies under the Easy interface mode, which only requires the starting structures and the restraint definitions in the form of active and passive residues to drive the docking (de Vries et al., 2010; Van Zundert et al., 2016). Our NMR structure of rSsp1a, cryo-EM structure of the NaVAb/NaV1.7 VS2A chimera, and homology models of DII of hNaV1.2 and hNaV1.3 were uploaded with the structure and restraint definitions to generate putative ligand receptor complex. The rSsp1a active residues were defined based on our activity data of rSsp1a alanine mutants at hNaV1.2, hNaV1.3, and hNaV1.7, whereas the active residues on DII of hNaV subtypes were defined based on the previously published channel mutation data (Xiao et al., 2010; Xiao et al., 2011; Cai et al., 2015; Zeng et al., 2018; Xu et al., 2019). The docking program was allowed to define the passive residues automatically around the active residues. Frontiers in Pharmacology 04 frontiersin.org 10.3389/fphar.2023.1277143 Dongol et al. Dongol et al. means ± standard error of mean (SEM) with number of independent experiments stated and p < 0.05 is considered statistically significant. Statistically insignificant shift in the activity is denoted by “little” while small but statistically significant shift in the activity is denoted by “slight” throughout the manuscript. For the rSsp1a–hNaV1.2 docking, the rSsp1a active residues defined were W5, F6, W24, K25, Y26, W28, and R30. Similarly, the active residues in hNaV1.2 DII S1–S2 loop were E779, Y781, T784, E785, F787, S788 whereas the active residues in the DII S3–S4 loop were E837, E844, and S847. Likewise, the rSsp1a–hNaV1.3 docking was driven by defining the rSsp1a active residues W5, F6, Y20, W24, K25, Y26, P27, W28, R30 and L33, and hNaV1.3 active residues in the DII S1–S2 loop (E780, Y782, T785, E786, F788, and S789) and DII S3–S4 loop (E838, S842, E845, and S848). 3.1 Determining the active surface of rSsp1a Fifteen rSsp1a alanine-analogues were produced via recombinant expression as previously described (Dongol et al., 2021), and their activity tested on hNaV1.2, hNaV1.3 and hNaV1.7 as a function of varying concentration (Figure 1B; Supplementary Figure S1; Supplementary Table S1). The alanine mutations causing ≥10-fold loss in potency at all three hNaV subtypes included W5A, F6A, W24A, K25A, Y26A, W28A and R30A. Y20A was inactive at hNaV1.3 but reduced hNaV1.2 and hNaV1.7 activity by nearly 10-fold. The P27A, Y31A and L33A mutations also preferentially impacted hNaV1.3. Surprisingly, the conserved P11 played little role in rSsp1a activity, and little to slight effects were observed for the N14A mutant. Two mutants, L3A and D32A, slightly improved the rSsp1a potency at hNaV1.2, but only D32A slightly enhanced rSsp1a potency at hNaV1.3. We compared the active surface of rSsp1a with GpTx-1, HwTx- IV, and the m3-HwTx-IV optimized variant, which revealed that the hydrophobic and basic residues that comprise the active surface in these toxins are distributed around a central WCK/R motif, with key residues aligned at corresponding positions in the 3D structure (Figure 1C). All active residues were solvent exposed and on the same face of these toxins, except Y20 in rSsp1a which was buried and K18 in HwTx-IV which was located on the opposite face. The loop 1, loop 4 and C-terminal region act in concert to form the active surface of these NaSpTx1 toxins as previously described for the spider peptide Tap1a (Hu et al., 2021). Next, we determined whether alanine substitutions reducing rSsp1a activity were structural by assessing influences on the Hα secondary chemical shifts of rSsp1a alanine mutants with significant activity loss (Figure 1D). Mutant W5A produced a significant change in Hα chemical shift at Y20 and Y31, while F6A only affected the neighboring W5. The loop 3 mutant Y20A affected both W5 and F6 Hα chemical shifts likely due to their spatial proximity to the buried Y20, suggesting loop 3 residues play a structural role in these ICK peptides. The loop 4 mutations W24A, K25A and Y26A and C-terminal mutations R30A and Y31A had only minor local effects on the Hα chemical shifts, indicating that the activity losses attributed to these mutations are functional rather than structural. Unfortunately, W28A-rSsp1a analogue was obtained in insufficient yield for NMR analysis. 2.6 Molecular docking Finally, to dock rSsp1a at hNaV1.7, the active residues defined in rSsp1a were W5, F6, W24, K25, Y26, W28 and R30, while the active residues defined in the hNaV1.7 DII S1–S2 loop were E753, E759, E760, F761 and K762, and active residues defined in the DII S3–S4 loop were E811, F813, D816 and E818. The docking results were displayed as a cluster of water-refined models, which were then downloaded and visualized using Pymol 2.4.1 (Schrodinger, 2018). Generally, the top 10 clusters were listed in the order of their HADDOCK score—the top position in the list occupied by the cluster with the lowest HADDOCK score. Further, each cluster contains the top four best scoring structures. A z-score was also determined for each cluster, indicating the number of standard deviations by which the HADDOCK score of a particular cluster differed from the mean score of all clusters. The top clusters in each list providing the lower z-score values are considered more reliable (de Vries et al., 2010; Van Zundert et al., 2016). Additionally, each of the top 10 generated models were evaluated to identify the docking pose that best supported the pharmacology for interactions. The 3D structure of rSsp1a analogues used for docking studies were obtained by introducing the mutation in the rSsp1a structure using Pymol. The docking of rSsp1a analogues were driven by defining the active residues at rSsp1a analogues and hNaV subtypes as described above for rSsp1a docking at each hNaV subtypes. The presence of key molecular interactions was determined by the distance separating the two specified interacting atoms participating in interactions between the toxin and the channel. Electrostatic interactions were categorized as hydrogen bonds (H-bonds) when an electronegative O-atom engaged with a H-atom covalently bonded either to an electronegative O- or N-atom within 2 Å. Conversely, electrostatic interactions were designated as salt bridges between positively charged N-atom (amino group) and the negatively charged O-atom (carboxyl group) within 3.3 Å. Similarly, hydrophobic interactions were defined interactions between hydrophobic residues of the toxin and channel (<5 Å), with the van der Waals radii represented as dots in the structural images. Dots representations were generated using the default parameters of the Pymol program for atom van der Waals radii, dot radius, width, and density. Additionally, aromatic amino acids from the toxin and the channel, located within a 5 Å distance, were considered to form π-π interactions. Frontiers in Pharmacology FIGURE 2 Docking pose of m3-HwTx-IV and rSsp1a. (A) Docking orientation of m3-HwTx-IV at NaVAb/NaV1.7 VS2A captured in cryo-EM and (B–D) docking orientation of rSsp1a at hNaV1.2, hNaV1.3, and hNaV1.7 obtained using HADDOCK are shown with side and top views. The red surface indicates the activity surface of the corresponding toxin, with dark red indicating the Trp residue from the WCK/R motif. The cyan colour indicates non-identical residues among hNaV1.2, hNaV1.3, and hNaV1.7. The recently published NaVAb/NaV1.7 VS2A structure (Wisedchaisri et al., 2021) provided the resting structure of DII that served as the hNaV1.7 DII used to model the hNaV1.2 and hNaV1.3 DII resting state. structure of NaV1.7-DII trapped by m3-HwTx-IV was recently solved (Wisedchaisri et al., 2021), allowing predictive docking of rSsp1a using HADDOCK (Van Zundert et al., 2016). We validated the HADDOCK docking results by comparing the docking orientation and molecular interaction of m3-HwTx-IV at NaVAb/ NaV1.7 VS2A between the HADDOCK-generated in silico structure and the cryo-EM structure captured by Wisedchaisri et al. (Wisedchaisri et al., 2021) (Supplementary Figure S2). The docking orientation of rSsp1a at hNaV1.2 revealed molecular interactions, including salt bridges, H-bonds and a series of hydrophobic interactions within the aqueous cleft (Figure 3A). Specifically, K25–E844 (2.8 Å, -NH3 +••••−OOC-) and R30–E837 (2.6 Å, =NH2 +••••−OOC-) salt bridges, and a H-bond between Y26 and E844 (1.8 Å, -OH••••−OOC-) were observed. Additionally, W5 projects into the hydrophobic pocket formed by the LGLA residues in the DII S3–S4 loop and Y31 interacted with W5–LGLA complex, while F6 and W28 occupied the aqueous cleft and interacted with the neighboring hydrophobic residues in the DII S2 and S3–S4 loop. Further, an interaction was observed between W24, located on the edge of the hydrophobic patch in the rSsp1a, and Y781 from the DII S1–S2 loop. The docking results of rSsp1a to hNaV1.2-, hNaV1.3- and hNaV1.7-DII illustrated that rSsp1a bound in the aqueous cleft formed between the S1–S2 and S3–S4 loops of each channel subtype, primarily targeting the S3–S4 loop as observed for m3- HwTx-IV (Figures 2A–D). rSsp1a docked at hNaV1.7 similar to m3- HwTx-IV, with the key residue W28 (W30 in m3-HwTx-IV, Supplementary Figures S3A) positioned towards the S3–S4 loop in the aqueous cleft (Figures 2A, B) and interacting with hydrophobic residues in the S3–S4 loop to trap the S4 segment, as shown for m3-HwTx-IV in Supplementary Figures S3B. 3.2 Molecular interaction of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 The experimental data were analysed using QPatch Assay software v5.6.4 and GraphPad Prism 7.0 using a four-parameter Hill equation [Y = Bottom + (Top—Bottom)/(1 + 10^(Log IC50—X)* Hillslope)] to fit concentration response curves by non-linear regression analysis and Student’s t-test. Data are presented as Our previous work indicated that rSsp1a traps DII of hNaV1.7 in the resting conformation (Dongol et al., 2021). The resting state 05 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 2 Docking pose of m3-HwTx-IV and rSsp1a. (A) Docking orientation of m3-HwTx-IV at NaVAb/NaV1.7 VS2A captured in cryo-EM and (B–D) docking orientation of rSsp1a at hNaV1.2, hNaV1.3, and hNaV1.7 obtained using HADDOCK are shown with side and top views. The red surface indicates the activity surface of the corresponding toxin, with dark red indicating the Trp residue from the WCK/R motif. The cyan colour indicates non-identical residues among hNaV1.2, hNaV1.3, and hNaV1.7. The recently published NaVAb/NaV1.7 VS2A structure (Wisedchaisri et al., 2021) provided the resting structure of DII that served as the hNaV1.7 DII used to model the hNaV1.2 and hNaV1.3 DII resting state. Frontiers in Pharmacology frontiersin.org FIGURE 2 When compared to hNaV1.7, the docking orientation of rSsp1a at hNaV1.2 and hNaV1.3 twisted, which allowed rSsp1a to occupy more space in the DII aqueous cleft with W28 occupying the center of the cleft (Figures 2C, D). Similar to the rSsp1a–hNaV1.2 results, the rSsp1a–hNaV1.3 docking orientation also showed K25 and R30 forming a salt bridge with E845 (3.0 Å, -NH3 +••••−OOC-) and E838 (2.6 Å, =NH2 +••••−OOC-), respectively, and Y26 forming a H-bond with S848 (1.7 Å, -O••••HO-) (Figure 3B). The bulky hydrophobic residue W5 fit within the hydrophobic pocket formed by LGL in the DII S3–S4 loop, with Y31and L33 positioned above the W5–LGL complex to hinder the upward transition of the S4, while F6 and V794 from S2 segment formed a hydrophobic 06 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 3 Molecular docking of rSsp1a. Molecular docking of rSsp1a at (A) hNaV1.2 DII (B) hNaV1.3 DII and (C) hNaV1.3 DII illustrating electrostatic and hydrophobic interactions. The resting conformation of hNaV1.2 and hNaV1.3 DII was modelled on the resting state structure of NaVAb/NaV1.7-VS2A chimera (PDB: 7K48) (Wisedchaisri et al., 2021). HADDOCK2.2 Easy interface (Van Zundert et al., 2016) was used for docking and the results were visualized using Pymol 2.4.1 (Schrodinger, 2018). The green and orange dots around the residues corresponds to the van der Waals radii. (D) Dose- response of rSsp1a at wild type (WT) and F823G-mNaV1.7. Data presented as means ± SEM, with number of experiments (n) indicated in the legend. Frontiers in Pharmacology FIGURE 3 Data presented as means ± SEM, with number of experiments (n) indicated in the legend. R30–D816 (2.7 Å, =NH2 +••••−OOC-) were observed, with additional electrostatic interactions between W24–E759 (1.8 Å, >NH••••−OOC- ), W28–D816 (1.8 Å, >NH••••−OOC-) and Y26–E810 (1.8 Å, -OH••••−OOC-) (Figure 3C). The hydrophobic residues LFLA in the S3–S4 loop project into the hydrophobic groove in rSsp1a formed by W5, F6, Y20, Y31, and W28. This binding mode shifted W28 away from the aqueous cleft center towards the S3–S4 loop, with K25 expected to play a key role in restricting the upward movement of S4 upon depolarization to trap DII in the resting state. We also assessed the role of the non-conserved residue F813 from hNaV1.7 DII S3–S4 in rSsp1a binding by using a F823G-mNaV1.7 variant. The F823 in mNaV1.7 DII corresponds to the F813 in the hNaV1.7 DII. The F823G mutation slightly decreased the mNaV1.7 sensitivity to rSsp1a (~2.5-fold) but was not critical for rSsp1a interaction (Figure 3D). Channel residues predicted by docking studies to interact with rSsp1a were highlighted in Supplementary Figure S4. introduction of an acidic residue Asp at the adjacent residue N14 slightly improved rSsp1a potency at hNaV1.2, with little reduction of potency at hNaV1.3 and >20-fold reduction of potency at hNaV1.7. This provided N14D-rSsp1a variant hNaV1.2- selectivity by >24-fold against hNaV1.7 and by 7-fold against hNaV1.3 (Figure 4; Supplementary Table S2). Reflecting the slight improvement observed at hNaV1.2, the docking orientation of N14D-rSsp1a at hNaV1.2 (Figure 5A) showed a small shift, which removed the Y26–E844 H-bond but favored a comparatively stronger π-stacking between W24 and Y781. However, in the docking orientation of N14D-rSsp1a at hNaV1.7 (Supplementary Figures S7A), the major interaction D816–R30–E818 necessary to trap the DII S4 segment in its resting state was lost, contributing to the weaker toxin-channel interaction observed. Of the five substitutions made in loop 3 (E18G, E18K, E18F, E18Y and Y20L), only the E18K substitution provided little or slight improvement in rSsp1a potency at all three hNaV subtypes (Figure 4; Supplementary Figures S5B; Supplementary Table S2), with the largest effect observed at hNaV1.2 (>3-fold). E18 is spatially distant from the rSsp1a interaction face (Supplementary Figures S6D), suggesting that E18K-rSsp1a might allow longer-range dipole interactions with the lipid bilayer that enhance potency. Y20 is buried within the rSsp1a hydrophobic patch and contributes to rSsp1a activity (Figures 1B, D). FIGURE 3 FIGURE 3 Molecular docking of rSsp1a. Molecular docking of rSsp1a at (A) hNaV1.2 DII (B) hNaV1.3 DII and (C) hNaV1.3 DII illustrating electrostatic and hydrophobic interactions. The resting conformation of hNaV1.2 and hNaV1.3 DII was modelled on the resting state structure of NaVAb/NaV1.7-VS2A chimera (PDB: 7K48) (Wisedchaisri et al., 2021). HADDOCK2.2 Easy interface (Van Zundert et al., 2016) was used for docking and the results were visualized using Pymol 2.4.1 (Schrodinger, 2018). The green and orange dots around the residues corresponds to the van der Waals radii. (D) Dose- response of rSsp1a at wild type (WT) and F823G-mNaV1.7. Data presented as means ± SEM, with number of experiments (n) indicated in the legend. FIGURE 3 Molecular docking of rSsp1a. Molecular docking of rSsp1a at (A) hNaV1.2 DII (B) hNaV1.3 DII and (C) hNaV1.3 DII illustrating electrostatic and hydrophobic interactions. The resting conformation of hNaV1.2 and hNaV1.3 DII was modelled on the resting state structure of NaVAb/NaV1.7-VS2A chimera (PDB: 7K48) (Wisedchaisri et al., 2021). HADDOCK2.2 Easy interface (Van Zundert et al., 2016) was used for docking and the results were visualized using Pymol 2.4.1 (Schrodinger, 2018). The green and orange dots around the residues corresponds to the van der Waals radii. (D) Dose- response of rSsp1a at wild type (WT) and F823G-mNaV1.7. Data presented as means ± SEM, with number of experiments (n) indicated in the legend. interaction. The toxin penetrated deep into the channel where the bulky hydrophobic W28 occupied the aqueous cleft and interacted with the neighboring hydrophobic residues from DII S1 and S2 segments. Consistent with the hNaV1.2 docking, the uniquely positioned W24 showed a hydrophobic interaction with Y782 in the DII S1–S2 loop to further stabilize rSsp1a interactions at the S3–S4 loop. rSsp1a docked to hNaV1.7 DII in a slightly altered orientation compared to hNaV1.2 and hNaV1.3. Salt bridges between K25–E753 (2.7 Å, -NH3 +••••−OOC-), R30–E818 (2.6 Å, =NH2 +••••−OOC-) and 07 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 4 Dose-response of rSsp1a mutations. (A–C) Dose-response curve of selected rSsp1a analogues compared with rSsp1a at hNaV1.2, hNaV1.3, and hNaV1.7 obtained using the whole-cell automated patch clamp electrophysiology (QPatch 16X) platform. Data presented as means ± SEM, with number of experiments (n) indicated in the legend. FIGURE 4 Dose-response of rSsp1a mutations. (A–C) Dose-response curve of selected rSsp1a analogues compared with rSsp1a at hNaV1.2, hNaV1.3, and hNaV1.7 obtained using the whole-cell automated patch clamp electrophysiology (QPatch 16X) platform. Frontiers in Pharmacology frontiersin.org 3.3.1 rSsp1a optimization through single residue substitution Based on the active surface defined for rSsp1a through the alanine substitutions and evidence from earlier optimization studies of closely related homologs, we designed and tested several single residue substitutions for rSsp1a (Supplementary Figures S5A–E; Supplementary Figures S6A–F). A charge-reversing substitution of D1K, located away from the predicted interaction face (Supplementary Figures S6A), improved rSsp1a-activity at hNaV1.2 and hNaV1.3, but not at hNaV1.7 (Figure 4; Supplementary Table S2). This suggests that the increased affinity of D1K-rSsp1a may be associated with lipid bilayer interactions that could facilitate interaction with hNaV1.2 and hNaV1.3. Substitution of the loop 1 hydrophobic residue W5 with a less bulky Phe was tolerated at hNaV1.2 but not at hNaV1.7 and hNaV1.3 (Figure 4; Supplementary Table S2). Other substitutions at loop 1 had smaller effects on rSsp1a activity (Supplementary Table S2; Supplementary Figures S5A) and they surround the rSsp1a interaction face (Supplementary Figures S6B). Th f d l 2 i t N13G S 1 (S l t Loop 4 residues (residues 23–28) comprise the active surface of rSsp1a (Supplementary Figures S6E). To examine the role of loop 4, nine analogues mutating five residues from the loop 4 of rSsp1a were generated. Mutations of H23 caused a little increase in rSsp1a activity only for H23S at hNaV1.2 (Supplementary Figures S5C; Supplementary Table S2). The W24R mutation partially inhibited hNaV1.2 (~65%), and incompletely inhibited hNaV1.7 (~50%) and hNaV1.3 (~75%) at 3 µM (Supplementary Figures S8). In contrast, the Y26H mutation did not improve rSsp1a activity, while mutations reinstating conserved basic residues at position 27 (P27R and P27K) had opposing affinity at hNaV1.3 and hNaV1.2, providing >10-fold hNaV1.2 selectivity (Figures 4A–C; Supplementary Figures S5C; Supplementary Table S2). However, the P27R and P27K analogues retained rSsp1a activity at hNaV1.7 and the fold-potency differences of these two analogues between hNaV1.2 and hNaV1.7 is <2-fold. We then investigated the molecular basis for the improved selectivity of P27R-rSsp1a between hNaV1.2 (Figure 5B) and hNaV1.3 The surface exposed loop 2 variant N13G-rSsp1a (Supplementary Figures S6C) only slightly weakened activity at hNaV1.7 (Supplementary Table S2; Supplementary Figures S5B). In contrast, 08 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 5 Molecular docking of rSsp1a single mutants. Visualization of molecular interactions of (A) N14D-rSsp1a, (B) P27R-rSsp1a and (C) D32Y-rSsp1a compared to rSsp1a at hNaV1.2. Frontiers in Pharmacology 3.3.1 rSsp1a optimization through single residue substitution Consistent with the activity data, rSsp1a and N14D-rSsp1a docked similarly at hNaV1.2 while P27R-rSsp1a formed additional ionic interactions through the substituted R27 and E785 and Y781 (main chain). In addition, Y26–E779 engaged the S1–S2 loop to further strengthen the toxin–channel complex. Compared to rSsp1a, D32Y-rSsp1a formed a stronger interaction with hNaV1.2, engaging both the S1–S2 and S3–S4 loops with ionic interactions, and consistent with the activity data. FIGURE 5 FIGURE 5 Molecular docking of rSsp1a single mutants. Visualization of molecular interactions of (A) N14D-rSsp1a, (B) P27R-rSsp1a and (C) D32Y-rSsp1a compared to rSsp1a at hNaV1.2. Consistent with the activity data, rSsp1a and N14D-rSsp1a docked similarly at hNaV1.2 while P27R-rSsp1a formed additional ionic interactions through the substituted R27 and E785 and Y781 (main chain). In addition, Y26–E779 engaged the S1–S2 loop to further strengthen the toxin–channel complex. Compared to rSsp1a, D32Y-rSsp1a formed a stronger interaction with hNaV1.2, engaging both the S1–S2 and S3–S4 loops with ionic interactions, and consistent with the activity data. At the C-terminal, we mutated Y31 and D32 to generate six analogues. These two residues are close to the functional residues on the surface of rSsp1a, with Y31 forming the hydrophobic patch (Supplementary Figures S6F). Substitution of Y31 with bulky Trp significantly improved rSsp1a potency, while the Y31T mutation removed the activity (Supplementary Figures S5D; Supplementary Table S2). Substitutions made at D32, preferentially improved rSsp1a potency at hNaV1.2 (up to 3-fold) and hNaV1.3 (up to 6-fold). The D32F mutant significantly improved rSsp1a potency at hNaV1.3 but not selectivity, while the D32Ymutant significantly improved rSsp1a potency (3-fold) and selectivity (~6-fold) for hNaV1.2 against hNaV1.3. The docking orientation of D32Y-rSsp1a at hNaV1.2 (Figure 5C) differed compared to rSsp1a, allowing engagement of the S1–S2 loop with the ionic bond K25–E779. The DII S3–S4 loop key interactions, including R30–E837 and Y26–E837, were maintained while a new interaction (W24–E785) was formed. In addition, Y32 and Y781 showed the potential to engage through a π–π interaction based on their side chain flexibility; however, the orientation to support a π–π interaction was not evident in the molecular docking studies. (Supplementary Figures S7B). P27R-rSsp1a docked at hNaV1.2 in a similar way to rSsp1a but included additional interactions with the S1–S2 loop. Specifically, in addition to the R30–E837 and K25–E844 ionic interactions in the S3–S4 loop, the substituted R27 formed a salt bridge with E785 and a H-bond with Y781 in the main chain, while Y26 interacted with E779 in the S1–S2 loop. These additional interactions at the S1–S2 loop stabilized the toxin-channel complex and improved the potency at hNaV1.2. For the P27R-rSsp1a–hNaV1.3 complex, the docking orientation most similar to the rSsp1a docking orientation was selected (Supplementary Figures S7B). Frontiers in Pharmacology FIGURE 5 (A) Potency fold-difference of selected rSsp1a analogues presented against the potency of wild type S 1 hN 1 2 hN 1 3 d hN 1 7 D d SEM ( 3 12 hN 1 2 f E18F 2) (B D) D f FIGURE 6 Potency and selectivity optimization of rSsp1a. (A) Potency fold-difference of selected rSsp1a analogues presented against the potency of wild type rSsp1a at hNaV1.2, hNaV1.3, and hNaV1.7. Data were presented as means ± SEM (n = 3–12, except at hNaV1.2 for E18F n = 2). (B–D) Dose-response curve of rSsp1a double mutants compared with rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7, obtained using the whole-cell automated patch clamp electrophysiology (QPatch 16X) platform. Data presented as means ± SEM, with number of experiments (n) indicated in the legend. TABLE 1 Double mutant designs for improved subtype-selectivity. Mutant Intended subtype-selectivity Outcome D1G-D32Y hNaV1.2 Enhanced affinity at hNaV1.3 and hNaV1.7, generating a non-selective inhibitor with nanomolar potency at the three hNaV subtypes tested W5F-N14D hNaV1.2 Predicted cumulative effects of single mutants that enhanced hNaV1.2 selectivity were not realised S7R-E18K hNaV1.2 and/or hNaV1.3 Enhanced affinity at hNaV1.2, hNaV1.3 and hNaV1.7, generating a non-selective inhibitor with nanomolar potency at the three hNaV subtypes S7K-W28F hNaV1.7 Reduced potency at hNaV1.7 N14D-P27R hNaV1.2 Reverted the potency loss by N14D at hNaV1.7 and provided hNaV1.2/ hNaV1.7 selectivity against hNaV1.3 TABLE 1 Double mutant designs for improved subtype-selectivity. Mutant Intended subtype-selectivity Outcome D1G-D32Y hNaV1.2 Enhanced affinity at hNaV1.3 and hNaV1.7, generating a non-selective inhibitor with nanomolar potency at the three hNaV subtypes tested W5F-N14D hNaV1.2 Predicted cumulative effects of single mutants that enhanced hNaV1.2 selectivity were not realised S7R-E18K hNaV1.2 and/or hNaV1.3 Enhanced affinity at hNaV1.2, hNaV1.3 and hNaV1.7, generating a non-selective inhibitor with nanomolar potency at the three hNaV subtypes S7K-W28F hNaV1.7 Reduced potency at hNaV1.7 N14D-P27R hNaV1.2 Reverted the potency loss by N14D at hNaV1.7 and provided hNaV1.2/ hNaV1.7 selectivity against hNaV1.3 FIGURE 5 These results revealed a spatial shift in the P27R-rSsp1a functional residues K25, Y26 and R30 that removed interactions between the active residues K25 and Y26 in rSsp1a and E838 and S848 in hNaV1.3. These results are consistent with the reduced activity observed for P27R-rSsp1a at hNaV1.3. The W28 residue of the WCK/R motif mutated to generate W28F and W28K analogues had a deleterious effect on hNaV activity; however, the decrease in activity by W28F at hNaV1.7 was only ~5- fold (Supplementary Figures S5C; Supplementary Table S2). (Supplementary Figures S7B). P27R-rSsp1a docked at hNaV1.2 in a similar way to rSsp1a but included additional interactions with the S1–S2 loop. Specifically, in addition to the R30–E837 and K25–E844 ionic interactions in the S3–S4 loop, the substituted R27 formed a salt bridge with E785 and a H-bond with Y781 in the main chain, while Y26 interacted with E779 in the S1–S2 loop. These additional interactions at the S1–S2 loop stabilized the toxin-channel complex and improved the potency at hNaV1.2. For the P27R-rSsp1a–hNaV1.3 complex, the docking orientation most similar to the rSsp1a docking orientation was selected (Supplementary Figures S7B). These results revealed a spatial shift in the P27R-rSsp1a functional residues K25, Y26 and R30 that removed interactions between the active residues K25 and Y26 in rSsp1a and E838 and S848 in hNaV1.3. These results are consistent with the reduced activity observed for P27R-rSsp1a at hNaV1.3. The W28 residue of the WCK/R motif mutated to generate W28F and W28K analogues had a deleterious effect on hNaV activity; however, the decrease in activity by W28F at hNaV1.7 was only ~5- fold (Supplementary Figures S5C; Supplementary Table S2). Frontiers in Pharmacology 09 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 6 Potency and selectivity optimization of rSsp1a. (A) Potency fold-difference of selected rSsp1a analogues presented against the potency of wild type rSsp1a at hNaV1.2, hNaV1.3, and hNaV1.7. Data were presented as means ± SEM (n = 3–12, except at hNaV1.2 for E18F n = 2). (B–D) Dose-response curve of rSsp1a double mutants compared with rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7, obtained using the whole-cell automated patch clamp electrophysiology (QPatch 16X) platform. Data presented as means ± SEM, with number of experiments (n) indicated in the legend. FIGURE 6 Potency and selectivity optimization of rSsp1a. frontiersin.org 3.3.2 rSsp1a optimization through combinatorial mutation against hNaV1.3 for the D32Y mutation, the D1G-D32Y mutation had significantly improved potency at hNaV1.3 and hNaV1.7 (Figure 6A). The S7R-E18K-rSsp1a analogue showed the greatest improvement in potency of the rSsp1a analogues tested, with hNaV1.2 potency enhanced 6-fold, hNaV1.3 enhanced ~10-fold, and hNaV1.7 enhanced >3-fold. The N14D-P27R analogue Guided by the potency and subtype-selectivity results of the single mutants, the double mutants D1G-D32Y, S7R-E18K, N14D- P27R, W5F-N14D and S7K-W28F were synthesised and tested (Figures 6A–D; Table 1). In contrast to the selectivity observed 10 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 7 Molecular docking of rSsp1a double mutants. (A) D1G-D32Y-rSsp1a docked differently at hNaV1.2 compared to rSsp1a (Supplementary Figures S9A) to form stronger ionic bonds, engaging both the S1–S2 and S3–S4 loops, as well as a possible π–π interaction between Y32–Y781, which was consistent with the activity data. (B) D1G-D32Y-rSsp1a docked at hNaV1.3 similar to docking to hNaV1.2, but in a different orientation compared to rSsp1a (Supplementary Figures S9B), which enabled stronger interactions that engaged both the S1–S2 and S3–S4 loops and a possible Y32–Y782 π–π interaction. (C) At hNaV1.7, both rSsp1a (Supplementary Figures S9C) and D1G-D32Y-rSsp1a docked with a similar orientation, forming new interactions including W5–D816 and Y32–H755. (D) At hNaV1.2, S7R-E18K-rSsp1a docked similar to rSsp1a (Supplementary Figures S9D) with R7 in S7R-E18K-rSsp1a (Continued) FIGURE 7 FIGURE 7 Molecular docking of rSsp1a double mutants. (A) D1G-D32Y-rSsp1a docked differently at hNaV1.2 compared to rSsp1a (Supplementary Figures S9A) to form stronger ionic bonds, engaging both the S1–S2 and S3–S4 loops, as well as a possible π–π interaction between Y32–Y781, which was consistent with the activity data. (B) D1G-D32Y-rSsp1a docked at hNaV1.3 similar to docking to hNaV1.2, but in a different orientation compared to rSsp1a (Supplementary Figures S9B), which enabled stronger interactions that engaged both the S1–S2 and S3–S4 loops and a possible Y32–Y782 π–π interaction. (C) At hNaV1.7, both rSsp1a (Supplementary Figures S9C) and D1G-D32Y-rSsp1a docked with a similar orientation, forming new interactions including W5–D816 and Y32–H755. (D) At hNaV1.2, S7R-E18K-rSsp1a docked similar to rSsp1a (Supplementary Figures S9D) with R7 in S7R-E18K-rSsp1a (Continued) 11 11 Frontiers in Pharmacology frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 FIGURE 7 (Continued) forming an ionic bond with E785 from the S1–S2 loop. In addition, a π–π interaction between W24–Y781 in a parallel displaced geometry was noted. These extra interactions engaging the S1–S2 loop are predicted to form a tighter toxin–channel complex, thus improving the toxin potency. (E) S7R- E18K-rSsp1a docked at hNaV1.3 similar to docking to hNaV1.2 and in an orientation similar to rSsp1a (Supplementary Figures S9E). An additional R7–E786 salt bridge engaging the S1–S2 loop is predicted to form a tighter toxin–channel complex. (F) At hNaV1.7, S7R-E18K-rSsp1a docked differently to rSsp1a (Supplementary Figures S9F), but contained salt bridges K25–E818, R30–E811 and R7–E759 that tightly engaged the S1–S2 and S3–S4 loop. In addition to the extra R7–E759 salt bridge, a possible W24–H755 π–π interaction was observed. (G) At hNaV1.2, N14D-P27R-rSsp1a docked similar to rSsp1a (Supplementary Figures S9G) with the R27 in N14D-P27R-rSsp1a forming an ionic bond with E785 from the S1–S2 loop. (H) The docking orientation for N14D-P27R-rSsp1a at hNaV1.3 was chosen to resemble rSsp1a docking. The lack of activity of N14D-P27R-rSsp1a at hNaV1.3 was consistent with the absence of major interactions. (I) At hNaV1.7, N14D-P27R-rSsp1a docked similar to rSsp1a (Supplementary Figures S9I) with R27 interacting with the L812 carbonyl group. The substituted residues are highlighted in red. docking orientation, with the loss of a major salt bridge formed by K25 and the substituted R27 did not engage with hNaV1.3 DII residues (Figure 7H; Supplementary Figures S9H), contributing to its observed weaker binding at hNaV1.3 (Figure 6). 4 Discussion In recent decades, NaV-modulating spider ICK toxins have generated significant interest in drug discovery (Dongol et al., 2019), with potential to target NaV subtypes of therapeutic interest (Xie et al., 2015; Su et al., 2017; Salvatierra et al., 2018; Cardoso, 2020; Menezes et al., 2020). Despite the need to discover and develop subtype-selective NaV-modulators, few comprehensive structure- function studies of NaSpTxs have been reported, with most focusing on hNaV1.7 (Minassian et al., 2013; Revell et al., 2013; Murray et al., 2015; Hu et al., 2021). These findings have helped guide analogue development of NaSpTxs aimed at improving potency and subtype-selectivity for hNaV1.7 (Revell et al., 2013; Klint et al., 2015; Murray et al., 2015; Murray et al., 2016; Shcherbatko et al., 2016; Agwa et al., 2017; Flinspach et al., 2017; Rahnama et al., 2017; Zhang et al., 2018; Mueller et al., 2020; Neff et al., 2020; Rupasinghe et al., 2020; Hu et al., 2021), while NaSpTx optimization at other subtypes have been overlooked. This study addresses the gap by investigating the structure-function relationships and optimization of Ssp1a at hNaV1.2, hNaV1.3 and hNaV1.7. g pp y g S7R-E18K-rSsp1a docked at hNaV1.2 (Figure 7D; Supplementary Figures S9D) and hNaV1.3 (Figure 7E; Supplementary Figures S9E) in a similar way to rSsp1a, maintaining the major S3–S4 loop interactions. However, a new salt bridge was predicted to be formed between the substituted R7 and Glu from the S1–S2 loop in both hNaV subtypes. In addition, the π–π interaction between W24 and Y781 (in hNaV1.2) or Y782 (in hNaV1.3) potentially stabilized the toxin–channel complex. At hNaV1.7, the S7R-E18K-rSsp1a mutant docked in a different orientation to rSsp1a but retained the major R30–E811 and K25–E818 salt bridges, in addition to an extra R7–E759 salt bridge (Figure 7F; Supplementary Figures S9F). The closely oriented W24 and H755 were predicted to form a π–π interaction instead of the W24–E759 electrostatic interaction observed for rSsp1a. The docking orientation of N14D-P27R-rSsp1a at hNaV1.2 and hNaV1.7 (Figures 7G, I; Supplementary Figures S9G, I) showed all major salt bridges were retained and the substituted R27 interacted with E785 from the S1–S2 loop in hNaV1.2 and the L812 main chain carbonyl oxygen from the S3–S4 loop in hNaV1.7. The docking of N14D-P27R-rSsp1a at hNaV1.3 closely overlapped the rSsp1a FIGURE 7 decreased the rSsp1a potency at hNaV1.3 (~40% block at 3µM) and reversed the potency reduction at hNaV1.7 caused by the N14D mutation. The results reveal that the potency and subtype-selectivity at hNaV1.2 conferred by the N14D and P27R mutations was not additive in the double mutant. The W5F-N14D and S7K-W28F mutants designed for hNaV1.2- and hNaV1.7-selectivity were unsuccessful. 3.3.3 rSsp1a optimization through N-terminal and C-terminal extension Further, docking studies were performed to visualize the molecular basis of potency and subtype-selectivity on the D1G- D32Y-rSsp1a, S7R-E18K-rSsp1a and N14D-P27R-rSsp1a double mutants at the three hNaV subtypes (Figures 7A–I; Supplementary Figures S9A–I). At hNaV1.2, D1G-D32Y-rSsp1a docked differently to rSsp1a to form stronger ionic bonds that engaged both the S1–S2 and S3–S4 loops (Figure 7A; Supplementary Figures S9A). The R30–E837 and Y26–E837 interactions engaged the hNaV1.2 DII S3–S4 loop, while the K25–E779 and W24–E785 interactions engaged with the S1–S2 loop. The proximity of Y32 to Y781 suggested π–π interactions were probable. Similar interactions were also observed for D1G-D32Y-rSsp1a docking with hNaV1.3 (Figure 7B; Supplementary Figures S9B). When docked at hNaV1.7, such π–π interaction with Y32 was not formed with the equivalent H755 in the hNaV1.7 DII S1–S2 loop. However, interactions between Y32–H755 and W5–D816, in addition to R30–E818, Y26–E811, K25–E753 and W24–E759, predicted enhanced D1G-D32Y-rSsp1a potency compared to rSsp1a at hNaV1.7 (Figure 7C; Supplementary Figures S9C). Finally, we extended the rSsp1a N-terminal with GP- to mimic GP-HwTx-IV (Neff et al., 2020) and GP-ProTx-II (Flinspach et al., 2017), while the C-terminal was extended with -GK to mimic the C-terminal amides as previously designed for HwTx-IV (Minassian et al., 2013; Neff et al., 2020). Both GP-rSsp1a and rSsp1a-GK (Supplementary Table S3) did not improve rSsp1a activity or selectivity (Supplementary Figures S5E). Further, the Y26H- rSsp1a mutant with an extra two Leu at the C-terminal (Y26H- rSsp1a-LL) was made to assess the effect of the C-terminal extension with small, linear hydrophobic residues, however, the analogue did not improve the activity or selectivity. Frontiers in Pharmacology FIGURE 7 (Continued) forming an ionic bond with E785 from the S1–S2 loop. In addition, a π–π interaction between W24–Y781 in a parallel displaced geometry was noted. These extra interactions engaging the S1–S2 loop are predicted to form a tighter toxin–channel complex, thus improving the toxin potency. (E) S7R- E18K-rSsp1a docked at hNaV1.3 similar to docking to hNaV1.2 and in an orientation similar to rSsp1a (Supplementary Figures S9E). An additional R7–E786 salt bridge engaging the S1–S2 loop is predicted to form a tighter toxin–channel complex. (F) At hNaV1.7, S7R-E18K-rSsp1a docked differently to rSsp1a (Supplementary Figures S9F), but contained salt bridges K25–E818, R30–E811 and R7–E759 that tightly engaged the S1–S2 and S3–S4 loop. In addition to the extra R7–E759 salt bridge, a possible W24–H755 π–π interaction was observed. (G) At hNaV1.2, N14D-P27R-rSsp1a docked similar to rSsp1a (Supplementary Figures S9G) with the R27 in N14D-P27R-rSsp1a forming an ionic bond with E785 from the S1–S2 loop. (H) The docking orientation for N14D-P27R-rSsp1a at hNaV1.3 was chosen to resemble rSsp1a docking. The lack of activity of N14D-P27R-rSsp1a at hNaV1.3 was consistent with the absence of major interactions. (I) At hNaV1.7, N14D-P27R-rSsp1a docked similar to rSsp1a (Supplementary Figures S9I) with R27 interacting with the L812 carbonyl group. The substituted residues are highlighted in red. 4.1 Structure-function of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 For example, P27A-rSsp1a only lost activity at hNaV1.3, suggesting that the non-conserved Pro at this position likely played less of a structural role in rSsp1a and could be exploited to make designs non-selective to hNaV1.3. This is supported by P27R mutation, which significantly improved rSsp1a activity at hNaV1.2 while losing activity at hNaV1.3 (Figure 4; Supplementary Table S2). residues and the channel subtype’s acidic residues in DII S3–S4 loop (Figures 3A–C), including E753 in the DII S1–S2 loop in hNaV1.7 that might allow the rational design of subtype-selective inhibitors. At hNaV1.7, additional interactions between indole nitrogen (W24 and W28) and carboxylate (E759 and D816) were observed, possibly strengthening the rSsp1a binding. In contrast, hydrophobic interactions between rSsp1a residue W24 and a Tyr residue in DII S1–S2 loop of hNaV1.2 and hNaV1.3 were observed. These hydrophobic interactions represent another key molecular mechanism involved in trapping DII in the resting state. For example, prior research (Wisedchaisri et al., 2021) predicted that the hydrophobic interaction between the m3-HwTx-IV hydrophobic patch (I5, F6, W30, and W33) and the LFLA stretch in DII S3–S4 loop (Supplementary Figures S3B) hinders the upward movement of the S4 segment, effectively trapping DII in the resting state. A similar mechanism was observed in the docking of rSsp1a with hNaV1.7, although distinct hydrophobic interactions were observed in the case of hNaV1.2 and hNaV1.3 (Figures 3A–C). Despite the high sequence homology across NaV subtypes, there is an important variation in the DII S3–S4 hydrophobic stretch in NaV1.7 where a bulky hydrophobic Phe replaces the small, non- polar Gly prevalent across other NaV subtypes except NaV1.6 and NaV1.9. Using the F823G-mNaV1.7 variant (equivalent to F813 in hNaV1.7) revealed that the Phe unique to hNaV1.7 was not critical for rSsp1a interaction (Figure 3D), similarly to HwTx-IV (Xiao et al., 2010). In contrast, this mutation desensitized mNaV1.7 binding to Pn3a from NaSpTx2 family by 28-fold (Mueller et al., 2020) and the binding to ProTx-II from NaSpTx3 family by 9–100 fold (Schmalhofer et al., 2008; Xiao et al., 2010). This pharmacology was verified by the structure of ProTx-II–DII-NaVAb complex, where two hydrophobic residues W5 and M6 from ProTx-II flanked F813 to stabilize the DII S3–S4 helix (Xu et al., 2019). 4.1 Structure-function of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 However, the docking orientation of rSsp1a (Figure 3C) and m3- HwTx-IV (Supplementary Figures S3B) (Wisedchaisri et al., 2021) at hNaV1.7 DII revealed that these two toxins positioned themselves on top of the LFLA motif rather than surrounding F813, agreeing to small reduction of F813G-hNaV1.7 or F823G-mNaV1.7 affinity to HwTx-IV (Xiao et al., 2010) or rSsp1a, respectively. pp y Based on the biophysical studies of rSsp1a (Dongol et al., 2021), we predicted that rSsp1a interacted with hNaV subtypes similarly to HwTx-IV (Xiao et al., 2008) to trap the hNaV1.7 DII in the resting state conformation. The available 3D structure of rSsp1a, activity data of rSsp1a alanine mutants, recently published cryo-EM structure of the resting state hNaV1.7 DII (Wisedchaisri et al., 2021), previously published channel mutation data (Xiao et al., 2010; Xiao et al., 2011; Cai et al., 2015; Zeng et al., 2018; Xu et al., 2019), and the high sequence homology at the predicted interacting face encouraged data- driven docking of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7. Data- driven docking can often accurately predict the molecular interaction at the binding interface in the absence of high-resolution atomic structures to illustrate the molecular interactions (Rodrigues and Bonvin, 2014). The rSsp1a docking across hNaV subtypes revealed the molecular mechanism of interaction is comparable to m3-HwTx- IV–NaV1.7 (Wisedchaisri et al., 2021), while highlighting the subtle differences in rSsp1a binding mode across hNaV subtypes (Figures 2, 3), as illustrated by the twisted docking orientation of rSsp1a at hNaV1.7 compared to hNaV1.2 and hNaV1.3. The more hydrophobic LFLA stretch in hNaV1.7 DII S3–S4, as compared to LGLA (hNaV1.2) and LGLS (hNaV1.3), potentially contributed to an observed twist and a shift in the docking of rSsp1a towards the S3–S4 loop in hNaV1.7, and to a lesser extent in hNaV1.2 and hNaV1.3. This shift and twist likely facilitated stronger subtype-specific hydrophobic interactions in hNaV1.7, with rSsp1a′s hydrophobic pocket locking the channel’s LFLA stretch, as opposed to hNaV1.2 and hNaV1.3, where the channel’s hydrophobic patch accommodated the hydrophobic residues of rSsp1a (Figures 3A–C). Additionally, the distinct binding pose of rSsp1a at hNaV1.7 also enabled electrostatic interactions with DII S1–S2 loop, which are not observed in hNaV1.2 and hNaV1.3. Such differences in binding mode across hNaV subtypes provide novel opportunities to optimize toxin- channel interactions across the hNaV subtypes. 4.1 Structure-function of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 The active surface defined for rSsp1a and close homologs GpTx- 1 (Murray et al., 2015), HwTx-IV (Revell et al., 2013), and m3- 12 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 10.3389/fphar.2023.1277143 residues and the channel subtype’s acidic residues in DII S3–S4 loop (Figures 3A–C), including E753 in the DII S1–S2 loop in hNaV1.7 that might allow the rational design of subtype-selective inhibitors. At hNaV1.7, additional interactions between indole nitrogen (W24 and W28) and carboxylate (E759 and D816) were observed, possibly strengthening the rSsp1a binding. In contrast, hydrophobic interactions between rSsp1a residue W24 and a Tyr residue in DII S1–S2 loop of hNaV1.2 and hNaV1.3 were observed. These hydrophobic interactions represent another key molecular mechanism involved in trapping DII in the resting state. For example, prior research (Wisedchaisri et al., 2021) predicted that the hydrophobic interaction between the m3-HwTx-IV hydrophobic patch (I5, F6, W30, and W33) and the LFLA stretch in DII S3–S4 loop (Supplementary Figures S3B) hinders the upward movement of the S4 segment, effectively trapping DII in the resting state. A similar mechanism was observed in the docking of rSsp1a with hNaV1.7, although distinct hydrophobic interactions were observed in the case of hNaV1.2 and hNaV1.3 (Figures 3A–C). HwTx-IV (Wisedchaisri et al., 2021) are similar, consistent with the concept of conservation in the interaction face of the peptides adopting the same fold with >30% sequence identity (Russell et al., 2004; Tuncbag et al., 2011). These comparisons reveal that the hydrophobic patch of rSsp1a forms the major binding face, with basic residues K25 (loop 4) and R30 (C-terminal) and the hydrophobic Y26 (loop 4) also expected to contribute to binding (Figure 1B; Figures 3A–C). Specifically, the hydrophobic residues W5, F6, W24 and W28 were predicted to be key to the rSsp1a activity. This is consistent with the conservation of Trp, Arg and Tyr in the binding site of peptides/proteins (Moreira et al., 2007), including the NaSpTx1 toxins (Klint et al., 2012; Minassian et al., 2013; Revell et al., 2013; Murray et al., 2015; Wisedchaisri et al., 2021). Although the structure-function of close homologs can be predicted, variations in the selectivity of analogs across NaV subtypes provides an opportunity to optimize peptides to improve potency and selectivity towards a specific target. Frontiers in Pharmacology frontiersin.org 4.2 Optimization of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 Two mutations in loop 2, N13G and N14D, were anticipated to provide subtype-selectivity and improved potency at hNaV1.7, respectively (Minassian et al., 2013; Neff et al., 2020). In contrast to N13G-HwTx-IV (Neff et al., 2020), N13G-rSsp1a lost the activity at hNaV1.7. But surprisingly, N14D-rSsp1a significantly lost the hNaV1.7 activity with slight improvement in hNaV1.2 activity and little reduction in hNaV1.3 activity, providing N14D-rSsp1a hNaV1.2-selectivity (Figure 6A) and could be considered for combination with other mutations to achieve hNaV1.2-selectivity. Two mutations in loop 2, N13G and N14D, were anticipated to provide subtype-selectivity and improved potency at hNaV1.7, respectively (Minassian et al., 2013; Neff et al., 2020). In contrast to N13G-HwTx-IV (Neff et al., 2020), N13G-rSsp1a lost the activity at hNaV1.7. But surprisingly, N14D-rSsp1a significantly lost the hNaV1.7 activity with slight improvement in hNaV1.2 activity and little reduction in hNaV1.3 activity, providing N14D-rSsp1a hNaV1.2-selectivity (Figure 6A) and could be considered for combination with other mutations to achieve hNaV1.2-selectivity. Acidic residues in loop 3 are less frequent in NaSpTx1 family toxins (Klint et al., 2012). Therefore, E18 in rSsp1a was substituted to remove the negative charge, reverse the charge, or to introduce hydrophobicity. The charge reversal mutation E18K improved potency at all three hNaV subtypes and its location away from the rSsp1a interaction face, like D1K mutation, suggested its indirect role in potency improvement, as discussed above. The potency and selectivity data from single-point mutations guided us to design five double mutants, from which only three designs improved rSsp1a activity (Figures 6A–D). The D1G-D32Y- rSsp1a designed for hNaV1.2-selectivity instead improved potency at all three subtypes, showing that the effects of combining single mutations can be unpredictable. Of the 38 optimized rSsp1a analogues, the S7R-E18K-rSsp1a showed the most enhanced potency (up to 10-fold), with the effect of each single mutation being additive when combined and improved potency at all three hNaV subtypes, thus minimising subtype-selectivity (Figure 6A). In contrast to the double mutant S7R-E18K-rSsp1a, the N14D-P27R- rSsp1a double mutant showed additive effects at hNaV1.3 but not at hNaV1.7, while the neutral effect at hNaV1.2 helped it to achieve dual selectivity for hNaV1.2 and hNav1.7. While the use of hNaV1.2/ hNav1.7-selective drugs is typically limited due to potential side effects (Eaton et al., 2021; Zhang et al., 2021; Echevarria-Cooper et al., 2022), this double mutant provides a new starting point for the design of hNaV1.2-selective and hNav1.7-selective leads. 4.2 Optimization of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 The P27R that reinstated the more frequent basic residue in the position improved rSsp1a potency specifically at hNaV1.2, suggesting incorporating it in a combined mutation to generate hNaV1.2-selective analogues. The comprehensive substitution of HwTx-IV residues suggested acidic residues at the C-terminal are not preferred to improve activity at hNaV1.2 and hNaV1.7 (Neff et al., 2020). Thus, D32, which also neighbors the rSsp1a interaction face, was substituted with several functionalities, including polar uncharged, basic, and hydrophobic entities, to evaluate their role in rSsp1a potency and selectivity (Figure 6A; Supplementary Table S2). D32S improved the rSsp1a potency at hNaV1.2 (2.5-fold) and hNaV1.3 (4-fold) compared to hNaV1.7 (1.3-fold) but did not provide subtype-selectivity. In contrast, D32F improved rSsp1a potency only at hNaV1.3 (~3.5-fold) but did not provide hNaV1.3-selectivity, given the lower potency of rSsp1a at hNaV1.3 compared to remaining two hNaV subtypes. Similarly, the lack of hNaV1.3-selectivity was also observed for D32K mutation despite improved hNaV1.3 inhibition by 6-fold. Interestingly, D32Y provided hNaV1.2-selectivity (5.5-fold) against hNaV1.3 with improved potency (3-fold). This contrasts with hNaV1.3-active CcoTx-2 which is a natural variant of hNaV1.3-inactive CcoTx-1 with D32Y mutation (Bosmans et al., 2006). Thus, D32Y could be combined to generate hNaV1.2-selective analogues. Optimization studies of CcoTx-1 (Shcherbatko et al., 2016), GpTx-1 (Murray et al., 2016) and HwTx-IV (Revell et al., 2013; Neff et al., 2020) suggested mutating Y31 in rSsp1a could improve potency and selectivity. Surprisingly, Y31T-rSsp1a did not contribute to hNaV1.7-selectivity (Figure 6A), in contrast to the 10-fold hNaV1.7-selectivity gained against hNaV1.2 by equivalent Y33T- HwTx-IV (Neff et al., 2020). Instead, Y31W-rSsp1a improved the hNaV activity between 1.5 and 3.5-fold, suggesting a preference of bulky hydrophobic in this position. In rSsp1a loop 1, reinstating cationic residue at position 7 had little effect on rSsp1a potency (Supplementary Figures S5A; Supplementary Table S2) contrasting to the key role played by R7 in GpTx-I activity (Murray et al., 2015; Murray et al., 2016) and K7 in HwTx-IV activity (Minassian et al., 2013) on hNaV1.7. Interestingly, reinstating the predominant Lys did not improve rSsp1a potency while restoring the less predominant Arg showed little improvement in the rSsp1a potency, making it a prospective single mutation to combine with other mutations. Further, substituting key residue W5 with less hydrophobic Phe provided rSsp1a hNaV1.2-selectivity with <3-fold reduction in potency. 4.2 Optimization of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 Several optimization studies on NaSpTx toxins have indicated that removing the negative charge at the N-terminal by introducing pyroglutamate, 1-Nal (1-naphthylalanine), Gly, or GP- or simply substituting the acidic residue with Ala or Gly might contribute to toxin binding, potency, and/or selectivity (Minassian et al., 2013; Revell et al., 2013; Rong et al., 2013; Shcherbatko et al., 2016). Extra N-terminal Gly and GP- residues are remnants of TEV and HRV3C digestion of recombinantly expressed fusion proteins, respectively (Klint et al., 2013; Minassian et al., 2013; Flinspach et al., 2017). These non-native residues at the N-terminal are an advantage of the recombinant expression system, as replacing N-terminal acidic residue with Gly in HwTx-IV and HnTx-I has improved the toxin potency (Revell et al., 2013; Rahnama et al., 2017; Zhang et al., 2018). rSsp1a inclusion of a non-native Gly (G0) resulting from TEV digestion and the D1G mutation removed the N-terminal The key molecular interactions revealed by the docking studies involve strong salt bridges (≤3 Å) formed between rSsp1a basic 13 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 10.3389/fphar.2023.1277143 negative charge. In contrast to HwTx-IV (Revell et al., 2013), the D1G-rSsp1a mutant reduced activity at hNaV1.7, while no significant changes were observed at hNaV1.2 and hNaV1.3 (Supplementary Table S2). Instead, a charge reversal mutation (D1K) slightly improved hNaV1.2 and hNaV1.3 activity with little reduction in hNaV1.7 activity, suggesting it could be used when combining mutations to achieve hNaV1.2- and/or hNaV1.3- selectivity. The D1K mutation is located outside the predicted interacting face of rSsp1a, suggesting this position plays an indirect role in binding, likely through longer-range charge effects. Previous studies have identified that the increased cationicity, as well as the presence of both native and substituted cationic residues (primarily Lys) contributes to the enhanced electrostatic interaction with anionic moieties within the lipid bilayer, including anionic lipid head groups, that can facilitate initial toxin-ion channel interactions (Henriques et al., 2016; Agwa et al., 2017; Agwa et al., 2018; Lawrence et al., 2019). Again, such toxin-membrane lipid binding can be subtype- specific, with D1K-rSsp1a showing reduced potency at hNaV1.7 but not at hNaV1.2 and hNaV1.3. (Zhang et al., 2018). However, the design neither improved potency nor selectivity of rSsp1a (Supplementary Figures S5C; Supplementary Table S2), confirming the significance of Y26 as an active residue. Frontiers in Pharmacology frontiersin.org Cai, T., Luo, J., Meng, E., Ding, J., Liang, S., Wang, S., et al. (2015). Mapping the interaction site for the tarantula toxin Hainantoxin-IV (β-TRTX-Hn2a) in the voltage sensor module of domain II of voltage-gated sodium channels. Peptides 68, 148–156. doi:10.1016/j.peptides.2014.09.005 Funding selectivity across hNaV1.2, 1.3 and 1.7. Supporting these experimental findings, in silico docking studies of optimized rSsp1a analogues revealed key molecular interactions underpinning the improved potency and selectivity observed, including new or altered electrostatic, hydrophobic, π-π interactions, and interactions with the S1–S2 loop of hNaV channels. The authors declare financial support was received for the research, authorship, and/or publication of this article. This work was funded by the Australian National Health and Medical Research Council (NHMRC) Program Grant APP1072113 and Principal Research Fellowship APP1119056 (RJL); an NHMRC Ideas Grant APP1188959 (RJL, FCC); University of Queensland Research Training Scholarship (YD); and Australian Research Council LIEF Grants [LE120100015, LE160100218, (NLD)]. In conclusion, we examined the pharmacology of 54 rSsp1a analogues to understand how this NaSpTx1 toxin can be modified to differentially alter interactions at hNaV1.2, hNaV1.3 and hNaV1.7. The inclusion of hNaV1.3 interactions provides the first view of NaSpTx1 pharmacology at this subtype. Given Ssp1a is distantly related to HwTx-IV (40% identity), which has comprehensive structure-function and optimization data and shares related pharmacology, the structure-function and optimization data of rSsp1a can be used to enrich the rational design of NaSpTx1 family toxins more broadly. Finally, this study reveals the complexities of moving from single to dual and triple mutations to develop improved research tools and/or potential therapeutic leads. Acknowledgments We thank Prof Irina Vetter and Jennifer Deuis from Vetter lab for providing HEK293 cells stably expressing mNaV1.7 and F823G- mNaV1.7. Data availability statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material. The authors declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision. Author contributions YD: Conceptualization, Formal Analysis, Investigation, Methodology, Visualization, Writing–original draft, Writing–review and editing. DTW: Methodology, Writing–review and editing. NLD: Formal Analysis, Writing–review and editing, Funding acquisition. FCC: Supervision, Writing–review and editing, Funding acquisition. RJL: Conceptualization, Funding acquisition, Resources, Supervision, Visualization, Writing–review and editing, Project administration. Publisher’s note Ethical approval was not required for the studies on animals in accordance with the local legislation and institutional requirements because only commercially available established cell lines were used. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. 4.2 Optimization of rSsp1a at hNaV1.2, hNaV1.3 and hNaV1.7 Overall, rSsp1a optimization by combining two single-point mutations provided two major designs, S7R-E18K with improved potency and N14D-P27R with improved selectivity. Both S7R-E18K and N14D-P27R mutants represent promising starting points for further analogue studies to improve potency and/or subtype- Acidic residues in loop 3 are less frequent in NaSpTx1 family toxins (Klint et al., 2012). Therefore, E18 in rSsp1a was substituted to remove the negative charge, reverse the charge, or to introduce hydrophobicity. The charge reversal mutation E18K improved potency at all three hNaV subtypes and its location away from the rSsp1a interaction face, like D1K mutation, suggested its indirect role in potency improvement, as discussed above. Loop 4 in rSsp1a constitutes active residues, including P27 that was critical only for hNaV1.3 activity (Figure 1B). Substituting the rare W24 with more conserved Arg or Ser did not improve rSsp1a activity at hNaV subtypes. W24 was important for rSsp1a activity, contrasting with the equivalent Arg in HwTx-IV and GpTx-1, which was important for HwTx-IV activity at hNaV1.2 (Minassian et al., 2013), and GpTx-1 (Murray et al., 2015) and HwTx-IV (Minassian et al., 2013) activity at hNaV1.7. Y26H was designed to reinstate more common His, as the equivalent H27A-GpTx-1 decreased in potency at hNaV1.7 by >10-fold (Murray et al., 2015; Murray et al., 2016), while D26H-HnTx-I improved hNaV1.7 activity by >7-fold 14 frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 10.3389/fphar.2023.1277143 Supplementary material The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2023.1277143/ full#supplementary-material Black, J. A., Nikolajsen, L., Kroner, K., Jensen, T. S., and Waxman, S. G. (2008). Multiple sodium channel isoforms and mitogen-activated protein kinases are present in painful human neuromas. Ann. Neurol. 64, 644–653. doi:10.1002/ana.21527 Bosmans, F., Rash, L., Zhu, S., Diochot, S., Lazdunski, M., Escoubas, P., et al. (2006). Four novel tarantula toxins as selective modulators of voltage-gated sodium channel subtypes. Mol. Pharmacol. 69, 419–429. doi:10.1124/mol.105.015941 Black, J. A., Cummins, T. R., Plumpton, C., Chen, Y. H., Hormuzdiar, W., Clare, J. J., et al. (1999). Upregulation of a silent sodium channel after peripheral, but not central, nerve injury in DRG neurons. J. Neurophysiol. 82, 2776–2785. doi:10.1152/jn.1999.82.5.2776 Frontiers in Pharmacology Ahern, C. A., Payandeh, J., Bosmans, F., and Chanda, B. (2016). The hitchhiker’s guide to the voltage-gated sodium channel galaxy. J. Gen. Physiol. 147, 1–24. doi:10. 1085/jgp.201511492 References Selective spider toxins reveal a role for the NaV1.1 channel in mechanical pain. Nature 534, 494–499. doi:10.1038/nature17976 Garry, E. M., Delaney, A., Anderson, H. A., Sirinathsinghji, E. C., Clapp, R. H., Martin, W. J., et al. (2005). Varicella zoster virus induces neuropathic changes in rat dorsal root ganglia and behavioral reflex sensitisation that is attenuated by gabapentin or sodium channel blocking drugs. Pain 118, 97–111. doi:10.1016/j.pain.2005.08.003 Pineda, S. S., Undheim, E. A., Rupasinghe, D. B., Ikonomopoulou, M. P., and King, G. F. (2014). Spider venomics: implications for drug discovery. Future Med. Chem. 6, 1699–1714. doi:10.4155/fmc.14.103 Goodwin, G., and McMahon, S. B. (2021). The physiological function of different voltage-gated sodium channels in pain. Nat. Rev. Neurosci. 22, 263–274. doi:10.1038/ s41583-021-00444-w Rahnama, S., Deuis, J. R., Cardoso, F. C., Ramanujam, V., Lewis, R. J., Rash, L. D., et al. (2017). The structure, dynamics and selectivity profile of a NaV1.7 potency-optimised Huwentoxin-IV variant. PLoS One 12, e0173551. doi:10.1371/journal.pone.0173551 Hains, B. C., Saab, C. Y., Klein, J. P., Craner, M. J., and Waxman, S. G. (2004). Altered sodium channel expression in second-order spinal sensory neurons contributes to pain after peripheral nerve injury. J. Neurosci. 24, 4832–4839. doi:10.1523/jneurosci.0300-04.2004 Revell, J. D., Lund, P.-E., Linley, J. E., Metcalfe, J., Burmeister, N., Sridharan, S., et al. (2013). Potency optimization of Huwentoxin-IV on hNaV1.7: a neurotoxin TTX-S sodium-channel antagonist from the venom of the Chinese bird-eating spider Selenocosmia huwena. Peptides 44, 40–46. doi:10.1016/j.peptides.2013.03.011 Henriques, S. T., Deplazes, E., Lawrence, N., Cheneval, O., Chaousis, S., Inserra, M., et al. (2016). Interaction of tarantula venom peptide ProTx-II with lipid membranes is a prerequisite for its inhibition of human voltage-gated sodium channel NaV1.7. J. Biol. Chem. 291, 17049–17065. doi:10.1074/jbc.M116.729095 Rodrigues, J. P., and Bonvin, A. M. (2014). Integrative computational modeling of protein interactions. FEBS J. 281, 1988–2003. doi:10.1111/febs.12771 Hong, S., Morrow, T. J., Paulson, P. E., Isom, L. L., and Wiley, J. W. (2004). Early painful diabetic neuropathy is associated with differential changes in tetrodotoxin- sensitive and -resistant sodium channels in dorsal root ganglion neurons in the rat. J. Biol. Chem. 279, 29341–29350. doi:10.1074/jbc.M404167200 Rong, M., Duan, Z., Chen, J., Li, J., Xiao, Y., and Liang, S. (2013). Native pyroglutamation of Huwentoxin-IV: a post-translational modification that increases the trapping ability to the sodium channel. PLoS One 8, e65984. doi:10.1371/journal. pone.0065984 Hu, H., Mawlawi, S. E., Zhao, T., Deuis, J. R., Jami, S., Vetter, I., et al. (2021). References Adams, M. E., Herold, E. E., and Venema, V. J. (1989). Two classes of channel-specific toxins from funnel web spider venom. J. Comp. Physiol. A 164, 333–342. doi:10.1007/BF00612993 Agwa, A. J., Lawrence, N., Deplazes, E., Cheneval, O., Chen, R. M., Craik, D. J., et al. (2017). Spider peptide toxin HwTx-IV engineered to bind to lipid membranes has an increased inhibitory potency at human voltage-gated sodium channel hNaV1.7. Biochim. Biophys. Acta 1859, 835–844. doi:10.1016/j.bbamem.2017.01.020 Agwa, A. J., Lawrence, N., Deplazes, E., Cheneval, O., Chen, R. M., Craik, D. J., et al. (2017). Spider peptide toxin HwTx-IV engineered to bind to lipid membranes has an increased inhibitory potency at human voltage-gated sodium channel hNaV1.7. Biochim. Biophys. Acta 1859, 835–844. doi:10.1016/j.bbamem.2017.01.020 Agwa, A. J., Peigneur, S., Chow, C. Y., Lawrence, N., Craik, D. J., Tytgat, J., et al. (2018). Gating modifier toxins isolated from spider venom: modulation of voltage-gated sodium channels and the role of lipid membranes. J. Biol. Chem. 293, 9041–9052. doi:10. 1074/jbc.RA118.002553 Ahern, C. A., Payandeh, J., Bosmans, F., and Chanda, B. (2016). The hitchhiker’s guide to the voltage-gated sodium channel galaxy. J. Gen. Physiol. 147, 1–24. doi:10. 1085/jgp.201511492 Ahern, C. A., Payandeh, J., Bosmans, F., and Chanda, B. (2016). The hitchhiker’s guide to the voltage-gated sodium channel galaxy. J. Gen. Physiol. 147, 1–24. doi:10. 1085/jgp.201511492 15 Frontiers in Pharmacology frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 Dongol et al. Cardoso, F. C. (2020). Multi-targeting sodium and calcium channels using venom peptides for the treatment of complex ion channels-related diseases. Biochem. Pharmacol. 181, 114107. doi:10.1016/j.bcp.2020.114107 Klint, J. K., Chin, Y. K., and Mobli, M. (2015). Rational engineering defines a molecular switch that is essential for activity of spider-venom peptides against the analgesics target NaV1.7. Mol. Pharmacol. 88, 1002–1010. doi:10.1124/mol.115.100784 Cardoso, F. C., and Lewis, R. J. (2018). Sodium channels and pain: from toxins to therapies. Br. J. Pharmacol. 175, 2138–2157. doi:10.1111/bph.13962 Klint, J. K., Senff, S., Rupasinghe, D. B., Er, S. Y., Herzig, V., Nicholson, G. M., et al. (2012). Spider-venom peptides that target voltage-gated sodium channels: pharmacological tools and potential therapeutic leads. Toxicon 60, 478–491. doi:10. 1016/j.toxicon.2012.04.337 Cardoso, F. C., and Lewis, R. J. (2019). Structure-function and therapeutic potential of spider venom-derived cysteine knot peptides targeting sodium channels. Front. Pharmacol. 10, 366. doi:10.3389/fphar.2019.00366 Klint, J. K., Senff, S., Saez, N. J., Seshadri, R., Lau, H. Y., Bende, N. S., et al. (2013). References Production of recombinant disulfide-rich venom peptides for structural and functional analysis via expression in the periplasm of E. coli. PLoS One 8, e63865. doi:10.1371/ journal.pone.0063865 Cardoso, F. C., Servent, D., and de Lima, M. E. (2022). Editorial: venom peptides: a rich combinatorial library for drug development. Front. Mol. Biosci. 9, 924023. doi:10. 3389/fmolb.2022.924023 Lawrence, N., Wu, B., Ligutti, J., Cheneval, O., Agwa, A. J., Benfield, A. H., et al. (2019). Peptide-membrane interactions affect the inhibitory potency and selectivity of spider toxins ProTx-II and GpTx-1. ACS Chem. Biol. 14, 118–130. doi:10.1021/ acschembio.8b00989 Chen, H. P., Zhou, W., Kang, L. M., Yan, H., Zhang, L., Xu, B. H., et al. (2014). Intrathecal miR-96 inhibits NaV1.3 expression and alleviates neuropathic pain in rat following chronic constriction injury. Neurochem. Res. 39, 76–83. doi:10.1007/s11064- 013-1192-z Li, D., Xiao, Y., Xu, X., Xiong, X., Lu, S., Liu, Z., et al. (2004). Structure-activity relationships of Hainantoxin-IV and structure determination of active and inactive sodium channel blockers. J. Biol. Chem. 279, 37734–37740. doi:10.1074/jbc. M405765200 de Lera Ruiz, M., and Kraus, R. L. (2015). Voltage-gated sodium channels: structure, function, pharmacology, and clinical indications. J. Med. Chem. 58, 7093–7118. doi:10. 1021/jm501981g de Vries, S. J., Van Dijk, M., and Bonvin, A. M. (2010). The HADDOCK web server for data-driven biomolecular docking. Nat. Protoc. 5, 883–897. doi:10.1038/nprot. 2010.32 Lindia, J. A., Kohler, M. G., Martin, W. J., and Abbadie, C. (2005). Relationship between sodium channel NaV1.3 expression and neuropathic pain behavior in rats. Pain 117, 145–153. doi:10.1016/j.pain.2005.05.027 Dib-Hajj, S. D., and Waxman, S. G. (2019). Sodium channels in human pain disorders: genetics and pharmacogenomics. Annu. Rev. Neurosci. 42, 87–106. doi:10. 1146/annurev-neuro-070918-050144 Menezes, L. F. S., Sabiá Júnior, E. F., Tibery, D. V., Carneiro, L. d. A., and Schwartz, E. F. (2020). Epilepsy-related voltage-gated sodium channelopathies: a review. Front. Pharmacol. 11, 1276. doi:10.3389/fphar.2020.01276 Dongol, Y., Cardoso, F. C., and Lewis, R. J. (2019). Spider knottin pharmacology at voltage-gated sodium channels and their potential to modulate pain pathways. Toxins (Basel) 11, 626. doi:10.3390/toxins11110626 Minassian, N. A., Gibbs, A., Shih, A. Y., Liu, Y., Neff, R. A., Sutton, S. W., et al. (2013). Analysis of the structural and molecular basis of voltage-sensitive sodium channel inhibition by the spider toxin Huwentoxin-IV (μ-TRTX-Hh2a). J. Biol. Chem. 288, 22707–22720. doi:10.1074/jbc.M113.461392 Dongol, Y., Choi, P. M., Wilson, D. T., Daly, N. L., Cardoso, F. C., and Lewis, R. J. (2021). References Voltage-gated sodium channel modulation by a new spider toxin Ssp1a isolated from an Australian theraphosid. Front. Pharmacol. 12, 795455. doi:10.3389/fphar.2021. 795455 Moreira, I. S., Fernandes, P. A., and Ramos, M. J. (2007). Hot spots—a review of the protein–protein interface determinant amino-acid residues. Proteins Struct. Funct. bioinform. 68, 803–812. doi:10.1002/prot.21396 Eaton, M., Zhang, J., Ma, Z., Park, A. C., Lietzke, E., Romero, C. M., et al. (2021). Generation and basic characterization of a gene-trap knockout mouse model of Scn2a with a substantial reduction of voltage-gated sodium channel NaV1. 2 expression. Genes, Brain Behav 20, e12725. doi:10.1111/gbb.12725 Mueller, A., Dekan, Z., Kaas, Q., Agwa, A. J., Starobova, H., Alewood, P. F., et al. (2020). Mapping the molecular surface of the analgesic NaV1.7-selective peptide Pn3a reveals residues essential for membrane and channel interactions. ACS Pharmacol. Transl. Sci. 3, 535–546. doi:10.1021/acsptsci.0c00002 Echevarria-Cooper, D. M., Hawkins, N. A., Misra, S. N., Huffman, A. M., Thaxton, T., Thompson, C. H., et al. (2022). Cellular and behavioral effects of altered NaV1.2 sodium channel ion permeability in Scn2aK1422E mice. Hum. Mol. Genet. 31, 2964–2988. doi:10. 1093/hmg/ddac087 Murray, J. K., Ligutti, J., Liu, D., Zou, A., Poppe, L., Li, H., et al. (2015). Engineering potent and selective analogues of GpTx-1, a tarantula venom peptide antagonist of the NaV1.7 sodium channel. J. Med. Chem. 58, 2299–2314. doi:10.1021/jm501765v Murray, J. K., Long, J., Zou, A., Ligutti, J., Andrews, K. L., Poppe, L., et al. (2016). Single residue substitutions that confer voltage-gated sodium ion channel subtype selectivity in the NaV1.7 inhibitory peptide GpTx-1. J. Med. Chem. 59, 2704–2717. doi:10.1021/acs.jmedchem.5b01947 Escoubas, P., Diochot, S., and Corzo, G. (2000). Structure and pharmacology of spider venom neurotoxins. Biochimie 82, 893–907. doi:10.1016/S0300-9084(00)01166-4 Flinspach, M., Xu, Q., Piekarz, A., Fellows, R., Hagan, R., Gibbs, A., et al. (2017). Insensitivity to pain induced by a potent selective closed-state NaV1.7 inhibitor. Sci. Rep. 7, 39662. doi:10.1038/srep39662 Neff, R. A., Flinspach, M., Gibbs, A., Shih, A. Y., Minassian, N. A., Liu, Y., et al. (2020). Comprehensive engineering of the tarantula venom peptide Huwentoxin-IV to inhibit the human voltage-gated sodium channel hNaV1.7. J. Biol. Chem. 295, 1315–1327. doi:10.1074/jbc.RA119.011318 Fontana, M., Vital-Brazil, O., and Vital-Brasil, O. (1985). Mode of action of Phoneutria nigriventer spider venom at the isolated phrenic nerve-diaphragm of the rat. Braz. J. Med. Biol. Res. 18, 557–565. Available at:. https://pubmed.ncbi.nlm.nih.gov/2425876/. Osteen, J. D., Herzig, V., Gilchrist, J., Emrick, J. J., Zhang, C., Wang, X., et al. (2016). Frontiers in Pharmacology References Engineering of a spider peptide via conserved structure-function traits optimizes sodium channel inhibition in vitro and anti-nociception in vivo. Front. Mol. Biosci. 8, 742457. doi:10.3389/fmolb.2021.742457 Rupasinghe, D. B., Herzig, V., Vetter, I., Dekan, Z., Gilchrist, J., Bosmans, F., et al. (2020). Mutational analysis of ProTx-I and the novel venom peptide Pe1b provide insight into residues responsible for selective inhibition of the analgesic drug target NaV1.7. Biochem. Pharmacol. 181, 114080. doi:10.1016/j.bcp.2020.114080 Kalia, J., Milescu, M., Salvatierra, J., Wagner, J., Klint, J. K., King, G. F., et al. (2015). From foe to friend: using animal toxins to investigate ion channel function. J. Mol. Biol. 427, 158–175. doi:10.1016/j.jmb.2014.07.027 Russell, R. B., Alber, F., Aloy, P., Davis, F. P., Korkin, D., Pichaud, M., et al. (2004). A structural perspective on protein–protein interactions. Curr. Opin. Struct. Biol. 14, 313–324. doi:10.1016/j.sbi.2004.04.006 Kim, C. H., Oh, Y., Chung, J. M., and Chung, K. (2001). The changes in expression of three subtypes of TTX sensitive sodium channels in sensory neurons after spinal nerve ligation. Mol. Brain Res. 95, 153–161. doi:10.1016/s0169-328x(01)00226-1 Saez, N. J., and Herzig, V. (2019). Versatile spider venom peptides and their medical and agricultural applications. Toxicon 158, 109–126. doi:10.1016/j.toxicon.2018.11.298 16 Frontiers in Pharmacology frontiersin.org Dongol et al. 10.3389/fphar.2023.1277143 Dongol et al. Investigations of nearest-neighbor effects. J. Biomol. NMR 5, 67–81. doi:10.1007/ BF00227471 Saez, N. J., Senff, S., Jensen, J. E., Er, S. Y., Herzig, V., Rash, L. D., et al. (2010). Spider- venom peptides as therapeutics. Toxins (Basel) 2, 2851–2871. doi:10.3390/ toxins2122851 Wu, Y., Ma, H., Zhang, F., Zhang, C., Zou, X., and Cao, Z. (2018). Selective voltage- gated sodium channel peptide toxins from animal venom: pharmacological probes and analgesic drug development. ACS Chem. Neurosci. 9, 187–197. doi:10.1021/ acschemneuro.7b00406 Salvatierra, J., Castro, J., Erickson, A., Li, Q., Braz, J., Gilchrist, J., et al. (2018). NaV1.1 inhibition can reduce visceral hypersensitivity. JCI Insight 3, e121000. doi:10. 1172/jci.insight.121000 Wüthrich, K. (1983). Sequential individual resonance assignments in the 1H-NMR spectra of polypeptides and proteins. Biopolymers 22, 131–138. doi:10.1002/bip. 360220121 Schmalhofer, W., Calhoun, J., Burrows, R., Bailey, T., Kohler, M. G., Weinglass, A. B., et al. (2008). ProTx-II, a selective inhibitor of NaV1. 7 sodium channels, blocks action potential propagation in nociceptors. Mol. Pharmacol. 74, 1476–1484. doi:10.1124/mol. 108.047670 Xiao, Y., Bingham, J.-P., Zhu, W., Moczydlowski, E., Liang, S., and Cummins, T. R. (2008). References Tarantula Huwentoxin-IV inhibits neuronal sodium channels by binding to receptor site 4 and trapping the domain II voltage sensor in the closed configuration. J. Biol. Chem. 283, 27300–27313. doi:10.1074/jbc.M708447200 Schrodinger (2018). The PyMOL molecular graphics system. Shcherbatko, A., Rossi, A., Foletti, D., Zhu, G., Bogin, O., Galindo Casas, M., et al. (2016). Engineering highly potent and selective microproteins against NaV1.7 sodium channel for treatment of pain. J. Biol. Chem. 291, 13974–13986. doi:10.1074/jbc.M116.725978 Xiao, Y., Blumenthal, K. M., Jackson, J. O., Liang, S., and Cummins, T. R. (2010). The tarantula toxins ProTx-II and HwTx-IV differentially interact with human NaV1.7 voltage-sensors to inhibit channel activation and inactivation. Mol. Pharmacol. 78, 1124–1134. doi:10.1124/mol.110.066332 Shen, H., Zhou, Q., Pan, X., Li, Z., Wu, J., and Yan, N. (2017). Structure of a eukaryotic voltage-gated sodium channel at near-atomic resolution. Science 355, eaal4326. doi:10. 1126/science.aal4326 Xiao, Y., Jackson, J. O., Liang, S., and Cummins, T. R. (2011). Common molecular determinants of tarantula Huwentoxin-IV inhibition of Na+ channel voltage-sensors in domains II and IV. J. Biol. Chem. 286, 27301–27310. doi:10.1074/jbc.M111.246876 Stevens, M., Peigneur, S., and Tytgat, J. (2011). Neurotoxins and their binding areas on voltage-gated sodium channels. Front. Pharmacol. 2, 71. doi:10.3389/fphar.2011. 00071 Xie, W., Strong, J. A., and Zhang, J. M. (2015). Local knockdown of the NaV1.6 sodium channel reduces pain behaviors, sensory neuron excitability, and sympathetic sprouting in rat models of neuropathic pain. Neuroscience 291, 317–330. doi:10.1016/j.neuroscience.2015.02.010 Su, S., Shao, J., Zhao, Q., Ren, X., Cai, W., Li, L., et al. (2017). MiR-30b attenuates neuropathic pain by regulating voltage-gated sodium channel Nav1.3 in rats. Front. Mol. Neurosci. 10, 126. doi:10.3389/fnmol.2017.00126 Tan, A. M., Samad, O. A., Dib-Hajj, S. D., and Waxman, S. G. (2015). Virus-mediated knockdown of NaV1.3 in dorsal root ganglia of STZ-induced diabetic rats alleviates tactile allodynia. Mol. Med. 21, 544–552. doi:10.2119/molmed.2015.00063 Xu, H., Li, T., Rohou, A., Arthur, C. P., Tzakoniati, F., Wong, E., et al. (2019). Structural basis of NaV1.7 inhibition by a gating-modifier spider toxin. Cell 176, 702–715. doi:10.1016/j.cell.2018.12.018 Tuncbag, N., Gursoy, A., and Keskin, O. (2011). Prediction of protein-protein interactions: unifying evolution and structure at protein interfaces. Phys. Biol. 8, 035006. doi:10.1088/1478-3975/8/3/035006 Xu, W., Zhang, J., Wang, Y., Wang, L., and Wang, X. (2016). Changes in the expression of voltage-gated sodium channels NaV1.3, NaV1.7, NaV1.8, and NaV1.9 in rat trigeminal ganglia following chronic constriction injury. Neuroreport 27, 929–934. References doi:10.1097/wnr.0000000000000632 Van Zundert, G., Rodrigues, J., Trellet, M., Schmitz, C., Kastritis, P., Karaca, E., et al. (2016). The HADDOCK2.2 web server: user-friendly integrative modeling of biomolecular complexes. J. Mol. Biol. 428, 720–725. doi:10.1016/j.jmb.2015.09.014 Zeng, X., Li, P., Chen, B., Huang, J., Lai, R., Liu, J., et al. (2018). Selective closed-state NaV1.7 blocker JzTx-34 exhibits analgesic effects against pain. Toxins (Basel) 10, E64. doi:10.3390/toxins10020064 Vetter, I., Deuis, J. R., Mueller, A., Israel, M. R., Starobova, H., Zhang, A., et al. (2017). NaV1.7 as a pain target–from gene to pharmacology. Pharmacol. Ther. 172, 73–100. doi:10.1016/j.pharmthera.2016.11.015 Zhang, F., Liu, Y., Zhang, C., Li, J., Yang, Z., Gong, X., et al. (2015). Natural mutations change the affinity of μ-theraphotoxin-Hhn2a to voltage-gated sodium channels. Toxicon 93, 24–30. doi:10.1016/j.toxicon.2014.11.220 Waterhouse, A., Bertoni, M., Bienert, S., Studer, G., Tauriello, G., Gumienny, R., et al. (2018). SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 46, W296–W303. doi:10.1093/nar/gky427 Zhang, J., Chen, X., Eaton, M., Wu, J., Ma, Z., Lai, S., et al. (2021). Severe deficiency of the voltage-gated sodium channel NaV1.2 elevates neuronal excitability in adult mice. Cell Rep. 36, 109495. doi:10.1016/j.celrep.2021.109495 Wisedchaisri, G., Tonggu, L., Gamal El-Din, T. M., McCord, E., Zheng, N., and Catterall, W. A. (2021). Structural basis for high-affinity trapping of the NaV1.7 channel in its resting state by tarantula toxin. Mol. Cell 81, 38–48.e4. doi:10.1016/j.molcel.2020.10.039 Zhang, Y., Yang, Q., Zhang, Q., Peng, D., Chen, M., Liang, S., et al. (2018). Engineering gain-of-function analogues of the spider venom peptide HnTx-I, a potent blocker of the hNaV1.7 sodium channel. Toxins (Basel) 10, 358. doi:10.3390/ toxins10090358 Wishart, D. S., Bigam, C. G., Holm, A., Hodges, R. S., and Sykes, B. D. (1995). 1H, 13C and 15N random coil NMR chemical shifts of the common amino acids. I. 17 17 Frontiers in Pharmacology Frontiers in Pharmacology frontiersin.org

https://openalex.org/W3175124630

https://acp.copernicus.org/articles/21/11723/2021/acp-21-11723-2021.pdf

English

Response to both referee reports

cc-by

Aitken mode particles as CCN in aerosol- and updraft-sensitive regimes of cloud droplet formation Mira L. Pöhlker1, Minghui Zhang2, Ramon Campos Braga1, Ovid O. Krüger1, Ulrich Pöschl1, and Barbara Ervens2 1Multiphase Chemistry Department, Max Planck Institute for Chemistry, 55128 Mainz, Germany 2Université Clermont Auvergne, CNRS, SIGMA Clermont, Institut de Chimie de Clermont-Ferrand, 63000 Clermont-Ferrand, France orrespondence: Mira L. Pöhlker ([email protected]) and Barbara Ervens ([email protected] Received: 12 March 2021 – Discussion started: 15 March 2021 Revised: 27 June 2021 – Accepted: 30 June 2021 – Published: 6 August 2021 Received: 12 March 2021 – Discussion started: 15 March 2021 Revised: 27 June 2021 – Accepted: 30 June 2021 – Published: 6 August 2021 Abstract. The high variability of aerosol particle concentra- tions, sizes and chemical composition makes their descrip- tion challenging in atmospheric models. Aerosol–cloud in- teraction studies are usually focused on the activation of accumulation mode particles as cloud condensation nuclei (CCN). However, under specific conditions Aitken mode par- ticles can also contribute to the number concentration of cloud droplets (Nd), leading to large uncertainties in pre- dicted cloud properties on a global scale. We perform sen- sitivity studies with an adiabatic cloud parcel model to con- strain conditions under which Aitken mode particles con- tribute to Nd. The simulations cover wide ranges of aerosol properties, such as total particle number concentration, hy- groscopicity (κ) and mode diameters for accumulation and Aitken mode particles. Building upon the previously sug- gested concept of updraft (w)- and aerosol-limited regimes of cloud droplet formation, we show that activation of Aitken mode particles does not occur in w-limited regimes of ac- cumulation mode particles. The transitional range between the regimes is broadened when Aitken mode particles con- tribute to Nd, as aerosol limitation requires much higher w than for aerosol size distributions with accumulation mode particles only. In the transitional regime, Nd is similarly de- pendent on w and κ. Therefore, we analyze the sensitivity of Nd to κ, ξ(κ), as a function of w to identify the value combinations above which Aitken mode particles can af- fect Nd. As ξ(κ) shows a minimum when the smallest ac- tivated particle size is in the range of the “Hoppel mini- mum” (0.06 µm ≤Dmin ≤0.08 µm), the corresponding (w– κ) pairs can be considered a threshold level above which Aitken mode particles have significant impact on Nd. This threshold is largely determined by the number concentration of accumulation mode particles and by the Aitken mode di- ameter. Aitken mode particles as CCN in aerosol- and updraft-sensitive regimes of cloud droplet formation Our analysis of these thresholds results in a simple parametric framework and criterion to identify aerosol and updraft conditions under which Aitken mode particles are ex- pected to affect aerosol–cloud interactions. Our results con- firm that Aitken mode particles likely do not contribute to Nd in polluted air masses (urban, biomass burning) at mod- erate updraft velocities (w ≤3 m s−1) but may be important in deep convective clouds. Under clean conditions, such as in the Amazon, the Arctic and remote ocean regions, hygro- scopic Aitken mode particles can act as CCN at updrafts of w < 1 m s−1. 1 Introduction The representation of aerosol–cloud interactions in atmo- spheric models is challenging due to the high variability of aerosol particle loading, properties and processes on small temporal and spatial scales. Aerosol–cloud interactions in- clude both the effects of aerosol particles on clouds by acting as cloud condensation nuclei (CCN) and the modification of aerosol due to chemical and physical cloud processing. The interaction of aerosol particles with water vapor is described by the Köhler theory (Köhler, 1936). It combines the curvature (Kelvin) effect that describes the enhancement of the water vapor pressure above a curved particle surface and the solute (Raoult) effect that accounts for water uptake by hygroscopic particle mass, which is often parameterized by the hygroscopicity parameter κ (Petters and Kreidenweis, Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 © Author(s) 2021. This work is distributed under the Creative Commons Attribution 4.0 License. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Thus, conclusions based on equilibrium conditions as implied by Köhler theory often represent overestimates of the effect of aerosol properties on clouds, e.g., Ervens et al. (2005). p p g The relative importance of aerosol parameters (e.g., chem- ical composition (κ), dry particle diameter (Ds), shape of aerosol size distribution (ASD), particle number concen- tration (Na)) and updraft velocity (w) on cloud properties was explored in previous sensitivity studies. Feingold (2003) showed that Na has the largest influence on the effective ra- dius of a cloud droplet size population, followed by the ge- ometric mean mode diameters and standard deviations of ASDs (Dg and σg). A similar ranking was discussed by Hernández Pardo et al. (2019), who showed that conclusions regarding the relative importance of the aerosol properties and w hold true for both the effective droplet radius and num- ber concentration (Nd) with lower sensitivity of the effective radius than that of Nd. Other sensitivity studies also identi- fied w and Na, followed by κ and other chemical composi- tion effects, as the most important parameters determining Nd (Ervens et al., 2005; McFiggans et al., 2006; Anttila and Kerminen, 2007; Reutter et al., 2009; Ward et al., 2010) or the supersaturation in clouds (Hammer et al., 2015). A simi- lar relative importance of Na and w in the shape of the cloud droplet size distribution (CDSD) was shown (Cecchini et al., 2017). Based on an intercomparison of 16 global models, it was concluded that Aitken mode particles do not significantly contribute to CCN in clouds with maximum supersaturations Smax =0.2 % (Fanourgakis et al., 2019). Based on another global model study, Lee et al. (2013) compared the influence of 28 parameters characterizing aerosol emissions, processes and size distributions on the CCN number concentration at S =0.3 %. They identified the width of the Aitken mode as the second most important parameter after the dry deposition of accumulation mode particles. Chang et al. (2017) found that on a global scale, the fraction of Aitken mode particles to total CCN is negligible at S =0.2 %, while it can be sig- nificant at S =0.4 % above the continental Northern Hemi- sphere. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11724 (Komppula et al., 2005). Similarly low aerosol loading was encountered above the tropical ocean, where ≤40 % of Aitken mode particles were predicted to act as CCN (Roelofs et al., 2006). Also in Arctic clouds, high contributions of Aitken mode particles to Nd were predicted (Korhonen et al., 2008; Jung et al., 2018; Bulatovic et al., 2021). In marine stratocumuli off the Californian coast, CCN and droplet clo- sure could only be achieved when contributions of Aitken mode particles to Nd were taken into account in clouds with w ≥0.6 m s−1 (Schulze et al., 2020). In deep convective clouds above the Amazon (w ≤12 m s−1), it was predicted that the formation of cloud droplets on Aitken mode particles (D ≥0.02 µm) might even impact the thermodynamic cloud structure by amplifying the convective invigoration and af- fecting precipitation rates (Fan et al., 2018). 2007). The maximum of the Köhler curve represents the crit- ical supersaturation (Scrit) above which a particle of a given composition and dry size (critical diameter, Dcrit) is activated and efficiently grows to a cloud droplet. (Komppula et al., 2005). Similarly low aerosol loading was encountered above the tropical ocean, where ≤40 % of Aitken mode particles were predicted to act as CCN (Roelofs et al., 2006). Also in Arctic clouds, high contributions of Aitken mode particles to Nd were predicted (Korhonen et al., 2008; Jung et al., 2018; Bulatovic et al., 2021). In marine stratocumuli off the Californian coast, CCN and droplet clo- sure could only be achieved when contributions of Aitken mode particles to Nd were taken into account in clouds with w ≥0.6 m s−1 (Schulze et al., 2020). In deep convective clouds above the Amazon (w ≤12 m s−1), it was predicted that the formation of cloud droplets on Aitken mode particles (D ≥0.02 µm) might even impact the thermodynamic cloud structure by amplifying the convective invigoration and af- fecting precipitation rates (Fan et al., 2018). Clouds are dynamic systems where the supersaturation is continuously altered due to increasing water vapor concen- tration in cooling air parcels and other processes and water vapor condensation onto particles. As the supply of water vapor and the growth timescales in clouds are limited, parti- cles and droplets may not reach their equilibrium sizes. Published by Copernicus Publications on behalf of the European Geosciences Union. Published by Copernicus Publications on behalf of the European Geosciences Union. 2.1 Model description We use an adiabatic parcel model to examine droplet for- mation on a population of aerosol particles (Feingold and Heymsfield, 1992; Ervens et al., 2005). The evolution of par- ticle and droplet sizes is described on a moving size grid. The calculation of the equilibrium saturation (seq) is based on Köhler theory, including the hygroscopicity parameter κ (Petters and Kreidenweis, 2007; Rose et al., 2008). pp In sensitivity tests, Dg,Ait and Dg,acc are shifted to 0.05, 0.06 and 0.07 µm and to 0.13 and 0.17 µm, respectively (Sect. 3.3.2). We define 15 model ASDs that differ (i) in the relative contributions of Aitken and accumulation mode particles (Na,Ait, Na,acc) to total Na (columns I–V in Fig. 1) and (ii) in the total number concentration Na (rows a–c in Fig. 1). ASDs I and V are monomodal with an accumulation or Aitken mode only; ASDs II, III and IV are bimodal with Na,Ait corresponding to 50 % (ASD II), 100 % (ASD III) and 150 % (ASD IV) of Na,acc. We distinguish the two modes by the diameter of the Hoppel minimum (∼0.07 µm). While strictly both modes have “tails” beyond the Hoppel minimum (Aitken mode particles being larger and accumulation mode particles being smaller than the Hoppel minimum), we do not consider this in our model. This simplification seems jus- tified since the mode classification in measured ASDs is usu- ally ascribed based on particle size and not on composition, which may differ because of different sources of particles in the two modes. The particle hygroscopicity is assumed to be equal in both modes; 27 values are used in the simulations to cover a range of 0.02 ≤κ ≤1. Additional tests are performed for fixed κ values in the accumulation mode (κacc = 0.1 and 0.5; Sect. 3.3.1). A total of 30 values for the updraft veloc- ities are applied (0.1 m s−1 ≤w ≤3 m s−1), resulting in 810 simulations (27κ × 30w) for each of the 15 ASDs. seq =  1 + κ Vs Vw −1 exp  4σsolMw RT ρwDwet  , (1) (1) where Dwet is the wet particle diameter, σsol the surface ten- sion of the wet particle (72 mN m−1), ρs the density of the dry particle, ρw the density of pure water, Mw the molecu- lar weight of water, R the constant for ideal gases and T the absolute temperature. 2.1 Model description We note that we do not consider additional composition effects (such as surface tension suppression) as κ represents the effective hygroscopicity as derived from experimental data. The model includes the standard thermodynamic equa- tions for particle and droplet growth and the derivatives to time of temperature, saturation and pressure (Pruppacher and Klett, 2003). These differential equations are iteratively solved within each model time step. The times steps are func- tion of w and chosen such that they cover a vertical change of the air parcel of 0.1 m. The change in the saturation is cal- culated as ds dt = 91w |{z} updraft term −922π ρw ρs G Z DwetNa(Ds)(s −seq)dDwet | {z } condensation term , (2) ds dt = 91w |{z} updraft term −922π ρw ρs G Z DwetNa(Ds)(s −seq)dDwet | {z } condensation term , (2) 2 Adiabatic parcel model 0.0028 µm < Ds < 1.4 µm, in lognormal distributions with geometric mean mode diameters Dg,Ait = 0.037 µm and Dg,acc = 0.145 µm with standard deviations of σg = 0.5 (cor- responding to 1.4 in commonly used Heisenberg fits). Note that different versions of lognormal fit functions are used in the literature (Pöhlker et al., 2021); a standard lognormal fit function was applied here (Pöhlker et al., 2018). M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets In their later global model study, cloud supersatura- tion in each grid cell was calculated based on the mean ver- tical velocity, and Nd was derived as the number of particles whose Scrit was approximated and compared for three cloud activation schemes (Chang et al., 2021). While the schemes mostly agreed in the Nd prediction from accumulation mode particles, large discrepancies were found in the predictions of the contribution of Aitken mode particles to Nd. CCN can be modified in clouds by mass addition due to chemical reactions in cloud droplets and by collision– coalescence processes (e.g., Ervens, 2015). These processes are suggested to lead to a size separation of cloud-processed and interstitial particles, resulting in a gap between the Aitken and accumulation modes (“Hoppel minimum”) (Hop- pel et al., 1986; Cantrell et al., 1999; Feingold and Kreiden- weis, 2000). It is traditionally assumed that only accumula- tion mode particles (D >∼0.07 µm) undergo cloud process- ing, leading possibly to a broadening of this mode. These prior studies provide strong evidence that Aitken mode particles can cause large uncertainties in predicted aerosol–cloud interactions under conditions of low Na, small fractions of accumulation mode particles to total Na and/or high w. In the current study, we perform simulations with an adiabatic parcel model to systematically explore the pa- rameter ranges of aerosol properties (Na, Na,Ait/Na,acc, κ, Dg,Ait, Dg,acc) and of w to identify aerosol and cloud con- ditions under which Aitken mode particles contribute to Nd. (All parameters are defined Table A1 in the Appendix.) Our analysis results in a framework that can be used to assess un- der which aerosol and cloud conditions detailed information on Aitken mode particles is needed to describe their potential role in aerosol–cloud interactions. Model and observational studies challenge this assump- tion, as ambient conditions were identified under which su- persaturation in clouds is sufficiently high to form cloud droplets on Aitken mode particles. For example, at a con- tinental remote background site in France, Aitken mode par- ticles with D ≥∼0.025 µm were shown to contribute the ma- jor fraction to Nd due to the absence of a significant ac- cumulation mode (Gérémy et al., 2000). About 30 % of all Aitken mode particles were observed to form cloud droplets at a background site in Finland at very low Na (∼150 cm−3) https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 11725 2.2.1 Initialization ξ(κ) = ∂lnNd ∂lnκ . (4) ξ(κ) = ∂lnNd ∂lnκ . (4) The model is initialized below cloud at RH = 98 %, T = 290 K and p = 829 mbar. The initial ASDs consist of 545 particle size classes in a diameter range of For comparison to conclusions on parameter regimes as discussed in previous model sensitivity studies, we also in- 2.2.2 Analysis of sensitivities ξ(κ) and ξ(Na) (2) Droplets are defined as particles with Dwet ≥3 µm. This def- inition does not strictly follow Köhler theory that defines droplets as particles whose Scrit is exceeded. The reasoning for our definition is the comparability of model results to observational studies that commonly report data from cloud probes detecting particles above a fixed size threshold, e.g., Campos Braga et al. (2021). As our study is intended to give guidance to future field and model studies, we use the size- based (3 µm) rather than the Köhler-based (Dcrit) droplet def- inition. We define the smallest dry particle size on which droplets grow to ≥3 µm as Dmin. The droplet number con- centration Nd is calculated as the cumulative particle number concentration between Dmin and the largest Ds (1.4 µm). We describe the predicted change in Nd as a function of particle hygroscopicity (κ) as the sensitivity ξ(κ): condensation term where 91 and 92 are functions of temperature and saturation (s). The updraft term describes the increase of s due to cool- ing of a rising air parcel in an adiabatic environment; the con- densation term accounts for the condensation of water vapor on aerosol particles and droplets. Particle and droplet growth are driven by the gradient between s and the particle-specific seq: seq: dDwet dt ∝(s −seq) Dwet . (3) dDwet dt ∝(s −seq) Dwet . (3) Atmos. Chem. Phys., 21, 11723–11740, 2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Figure 1. Schematic of model input aerosol size distributions (ASDs, designated I–V, a–c) with different number concentrations of Aitken mode particles (Na,Ait, red) and accumulation mode particles (Na,acc, blue). Particle number concentrations are given in units of cubic centimeters (cm−3). The modal geometric mean diameters are Dg,Ait = 0.037 µm and Dg,acc = 0.145µm. Figure 1. Schematic of model input aerosol size distributions (ASDs, designated I–V, a–c) with different number concentrations of Aitken mode particles (Na,Ait, red) and accumulation mode particles (Na,acc, blue). Particle number concentrations are given in units of cubic centimeters (cm−3). The modal geometric mean diameters are Dg,Ait = 0.037 µm and Dg,acc = 0.145µm. vestigate the sensitivity of Nd to Na: vestigate the sensitivity of Nd to Na: particles form cloud droplets (Nd ∼Na,acc = 500 cm−3) and Nd cannot further increase for ASD I.b. In the presence of an Aitken mode, a significant fraction of highly hygro- scopic Aitken mode particles grow to cloud droplets (Nd ∼ 620 cm−3, Fig. 2b) as Dmin is significantly smaller than the size range of the Hoppel minimum (Ds ∼0.07 µm). ξ(Na) = ∂lnNd ∂lnNa . (5) ξ(Na) = ∂lnNd ∂lnNa . (5) These definitions follow the same approach as in previ- ous model studies that investigated the sensitivity of Nd to κ and Na for monomodal accumulation mode ASDs (e.g., Mc- Figgans et al., 2006; Reutter et al., 2009; Ward et al., 2010; Hernández Pardo et al., 2019). g pp ( s µ ) The ξ(κ) evolution for ASD I.b (Fig. 2e) repeats the trends of Dmin and mirrors those of Nd; i.e., ξ(κ) is lowest for the highest κ and w. For ASD I.b, ξ(κ) reaches the lowest values for the highest w when nearly all particles are grown to cloud droplets (Nd ∼500 cm−3), and a decrease in Dmin does not further increase Nd. The difference in ξ(κ) for ASD I.b and III.b is significant (Fig. 2e, f) as ξ(κ) for the bimodal ASD III.b is predicted to increase for κ >∼0.5 above the level of Smax. This inversion of ξ(κ) occurs at the height at which Aitken mode particles start contributing to Nd (Fig. 2).While for the monomodal ASD I.b ξ(κ) is predicted to continuously decrease with height, the increasing contribution of Aitken mode particles to Nd leads to the opposite trend, i.e., to the highest ξ(κ) values for particles with the highest κ. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Since Nd is predicted to increase above cloud base, we perform most of our sensitivity analyses at 20 m above the height level of maximum supersaturation (Smax). In a recent Nd closure study, we found not only the best agreement of measured and predicted Nd at this height, but also of the liq- uid water content, independently of the pollution level of the air mass (Campos Braga et al., 2021). https://doi.org/10.5194/acp-21-11723-2021 https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 11726 3.1 Vertical profiles of Nd, Dmin and ξ(κ) Generally, the differences in the vertical profiles for ASD I.b and III.b are smaller with lower w (Figs. S1 and S2). At w = 1 m s−1, Nd is only slightly lower for ASD I.b than for ASD III.b. The two modes overlap at the Hoppel minimum (Fig. 1), and the small concentration of Aitken mode parti- cles at this size explains the somewhat higher Nd in Fig. S1a as compared to Fig. S1b. At w = 0.2 m s−1, only large ac- cumulation mode particles are activated (activated fraction, Fact,acc <∼0.5 and Dmin ≥0.12 µm), and ξ(κ) remains gen- erally higher than for larger w. A high sensitivity implies that Dmin is in a size range in which the ASD exhibits a steep slope, where a small change in Dmin translates into a rela- tively large change in Nd. Figure 2 shows the vertical profiles of Nd, Dmin and ξ(κ) as a function of κ (color-coding) for simulations with ASD I.b and ASD III.b at an updraft velocity of w = 2.9 m s−1; com- plementary results for w = 1.0 and 0.2 m s−1 are shown in Figs. S1 and S2. The black lines denote the height at which the supersaturation reaches its maximum value (Fig. S3). At high w and κ, Nd shows significantly smaller Nd for the monomodal ASD I.b than for the bimodal ASD III.b (Fig. 2a and b). The Dmin values are nearly identical for the two ASDs for a given w (comparison of left and right columns in Figs. 2, S1 and S2), and they are inversely correlated with Nd. Under these conditions, Dmin reaches minimum val- ues of ∼0.05 µm, which means that all accumulation mode https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets In line with the results in Figs. 2, ξ(κ) and Dmin are high- est for small κ and w and decrease with increasing w (Fig. 3a and b). Analogous trends are shown in Fig. 3b for the ξ(κ) values as a function of Dmin along the horizontal lines (con- stant w) in Fig. 3a. The ξ(κ) lines for the three κ or three w values, respectively, overlap as ξ(κ) depends on the slope of the ASD at Dmin. Different combinations of κ and w can re- sult in identical Dmin values and thus yield the same ξ(κ) val- ues (contour lines of identical color in Fig. 3a). When Dmin is near the Hoppel minimum, a change in Dmin by either of the parameters does not lead to a significant change in Nd leading to low ξ(κ) values. Accordingly, ξ(κ) increases when Dmin reaches sizes smaller than the Hoppel minimum. If cloud droplets were defined based on Dcrit, Nd would be computed at the level of Smax (black lines in Figs. 2, S1, and S2) and remain constant above this height. This would result in different Nd as large haze particles may not be counted if their Scrit were not reached in cloud, and small particles whose Scrit is exceeded might also not be counted using our size-based droplet definition. Since Nd would not change above Smax, ξ(κ) would also be constant above this level. A detailed comparison of predicted cloud properties apply- ing the two droplet definitions (Dcrit versus Dwet ≥2 µm) was performed in a previous sensitivity study (Loftus, 2018). There it was shown that predicted Nd based on the two def- initions shows largest discrepancies at the lowest w and/or high Na and that the droplet size distributions are generally narrower if droplets are defined based on Dcrit. Also sensi- tivities would be overestimated if they are computed in the unstable bottom layer of the cloud at the level of Smax. It can be concluded that sensitivity studies in which a droplet def- inition based on Dcrit is applied will not only lead to higher absolute values of ξ(κ) but, at the same time, also to an un- derestimate of the sensitivity to κ as the predicted differences of ξ(κ) for different κ values are much smaller (e.g., Reutter et al., 2009). M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11728 hygroscopic particles, Dmin is near the Hoppel minimum and thus ξ(κ) is smaller than for high κ (Fig. 3). above the levels of Smax. This height corresponds to ∼35 m above the level where RH = 100 % in Fig. 2e and f. Figure 3 depicts the ξ(κ) values resulting from the 810 simulations for ASD III.b as a function of κ and w. The contour lines are color-coded by 0 ≤ξ(κ) ≤1. Parallel to the axes, six lines are marked for three κ values (vertical lines at κ = 0.7 (or- ange), 0.3 (blue) and 0.04 (red)), and three updraft velocities (horizontal lines at w = 2.9 m s−1 (orange), 1.0 m s−1 (blue) and 0.2 m s−1 (red)). In all simulations, Nd is predicted to increase while Dmin decreases until the air parcel reaches up to several tens of meters above the height level of Smax (Figs. 2, S1, and S2), as particles continue to grow and eventually reach the size threshold of 3 µm. The size distributions of haze particles and droplets are shown in Fig. S4 at four heights for the three w and κ = 0.04, 0.3 and 0.7; the red vertical lines in- dicate the size threshold (3 µm) for droplets. It can be seen that the separation of haze particles and droplets occurs at different heights, depending on κ and w. Hygroscopic Aitken mode particles successively grow at high w to similar droplet sizes as accumulation mode particles (orange box at the bot- tom of the figure), coinciding with the height at which ξ(κ) increases. This ξ(κ) trend is opposite to that for ASD I.b and also as found in previous studies of monomodal ASDs (Hernández Pardo et al., 2019; Cecchini et al., 2017) that showed decreasing sensitivities to aerosol properties with height. Each ξ(κ) can be related to a Dmin (Fig. 3). This rela- tionship is shown in Fig. 3b, where the ξ(κ) values along the vertical lines are overlaid by the aerosol size distribution. Figure 3c shows the Dmin range that is covered by the sim- ulations for the three constant κ values. The end points of the Dmin ranges in Fig. 3b and c are connected by vertical dashed lines. In the same way, Fig. 3d shows Dmin for ξ(κ), i.e., along the horizontal lines in Fig. 3a. 3.2 Sensitivity to aerosol properties: ξ(κ) and ξ(Na) While for Na = 200 cm−3 and 1000 cm−3 the minima in the ξ(κ) curves coincide with the Hoppel minimum of the ASD, the minimum of ξ(κ) is shifted to somewhat larger sizes for Na = 5000 cm−3 (Fig. S6). At such high Na, the su- persaturation is very low (Fig. S7). Under these conditions, an increase in w or κ might result in only small changes in Nd because of buffering effects; i.e., the growth of additional M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Such studies might thus lead to biased conclu- sions if they are applied to ambient Nd measurements that are based on fixed size thresholds to discriminate cloud droplets. Overall, ξ(κ) as a function of Dmin traces the shape of the ASD (black line in Fig. 3a and c). The combinations of w and κ for which an increase in ξ(κ) with a decrease in Dmin is predicted are those in the upper right corner of Fig. 3a. The Dmin ranges, covered by a variation of w or κ by the same factor (∼30) (colored bars in Fig. 3c), are simi- lar. This suggests that an equal change in either of the pa- rameters affects Dmin to the same extent. However, this re- lationship cannot be generalized. Figures S5 and S6 show equivalent results to those in Fig. 3 but for Na = 200 cm−3 and Na = 5000 cm−3, respectively. Smaller Na implies a smaller condensation term (Eq. 2), resulting in a higher su- persaturation, which allows also smaller particles to grow to cloud droplets. Thus, for Na = 200 cm−3, the Dmin ranges are shifted to smaller values, resulting in higher activated frac- tions (Fig. S5b and c). Accordingly, the Dmin ranges move to larger values for Na = 5000 cm−3 (Fig. S6b and c). M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11727 p re 2. Vertical profiles of (a, b) cloud droplet number concentration (Nd), (c, d) dry size of smallest particles that contribute to Nd (Dmin), aid by the corresponding ASD (grey lines) and (e, f) sensitivity of Nd to κ (ξ(κ)) for an updraft velocity w = 2.9 m s−1. Left and right mns show results for ASD I.b and III.b, respectively. The black lines in all panels mark the height of Smax above the level of saturation = 100 %). Figure 2. Vertical profiles of (a, b) cloud droplet number concentration (Nd), (c, d) dry size of smallest particles that contribute to Nd (Dmin), overlaid by the corresponding ASD (grey lines) and (e, f) sensitivity of Nd to κ (ξ(κ)) for an updraft velocity w = 2.9 m s−1. Left and right columns show results for ASD I.b and III.b, respectively. The black lines in all panels mark the height of Smax above the level of saturation (RH = 100 %). Similar trends of ξ(κ) with w were discussed in previous studies in which it was generalized that ξ(κ) is highest at low w as only a small but significant fraction of the accumulation mode particles is activated (e.g., Moore et al., 2013; Ervens et al., 2005). Reutter et al. (2009) showed low sensitivities of Nd at low w for very high Na(∼10000 cm−3). In this case, the supersaturation is efficiently suppressed as the conden- sation term (Eq. 2) is dominated by Na. As only very large particles grow to droplet sizes, Nd would only include parti- cles with Ds >∼0.3 µm , i.e., only in a flat part of the ASD, where a small change in Dmin does not lead to a significant change in Nd. The increased ξ(κ) at high w for the bimodal ASD III.b (Fig. 2f) follows the same reasoning as Dmin is lo- cated at a size range where the Aitken mode exhibits approx- imately the same slope as at Dmin for w = 0.2 m s−1. For less Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 3.2.1 Dependence of ξ(κ) on Dmin (ASD III.b) In the following, we investigate more generally the parame- ter ranges at which Aitken mode particles affect sensitivities and Nd in clouds. Our discussion will be limited to cloud conditions at a height of 20 m above Smax, i.e., when Nd has reached a constant value. Figure 3 shows results at 20 m Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11729 Figure 3. (a) Sensitivity of cloud droplet number concentration to aerosol hygroscopicity (ξ(κ)) as a function of κ and w for ASD III.b at 20 m above Smax. (b) ξ(κ) (right axis) as a function of Dmin for κ = 0.04 (red), κ = 0.3 (blue) and κ = 0.7 (orange). (c) Ranges of Dmin for the simulations in panel (b) and (d). (b) ξ(κ) (right axis) as a function of Dmin for w = 0.2 m s−1 (red), w = 1.0 m s−1 (blue) and w = 2.9 m s−1 (orange). Figure 3. (a) Sensitivity of cloud droplet number concentration to aerosol hygroscopicity (ξ(κ)) as a function of κ and w for ASD III.b at 20 m above Smax. (b) ξ(κ) (right axis) as a function of Dmin for κ = 0.04 (red), κ = 0.3 (blue) and κ = 0.7 (orange). (c) Ranges of Dmin for the simulations in panel (b) and (d). (b) ξ(κ) (right axis) as a function of Dmin for w = 0.2 m s−1 (red), w = 1.0 m s−1 (blue) and w = 2.9 m s−1 (orange). droplets suppresses the supersaturation and prevents further activation. limited regime is characterized by high activated fractions, i.e., when an increase in Nd can only be caused by an in- crease in Na and Nd depends linearly only on Na; the w- limited regime occurs for small activated fractions where an increase in w leads to sufficient decrease in Dmin to in- crease Nd. Our analysis in Sect. 3.2.1 suggests that for wide parameter ranges, Nd is similarly sensitive to κ and w. To discuss these results in the context of Na- and w-limited regimes, we explore the sensitivity of Nd to Na, ξ(Na), as a function of w for κ = 0.7 for ASD I, III and V with 500 cm−3 ≤Na ≤5000 cm−3. 3.2.1 Dependence of ξ(κ) on Dmin (ASD III.b) Figure 4a shows Nd for ASD I as a function of Na and w and confirms the thresholds be- tween the regimes as suggested by Reutter et al. (2009), i.e., Na limitation above w/Na > ∼10−3 m s−1 cm3 and w lim- itation below w/Na <∼10−4 m s−1 cm3, with a transitional regime in between. Accordingly, ξ(Na) approaches unity in the Na-limited regime when nearly all particles are activated (upper left corner of Fig. 4d) and even exceeds this value at high Na and low w (bottom right corner). The comparison of Figs. 3, S5 and S6 reveals that not only the ranges of Dmin values are shifted as a function of Na but also that their widths differ depending on Na. The Dmin ranges are widest for Na = 200 cm−3, which implies that for these conditions Nd is most sensitive to κ and w, as a change in these parameters causes a significant change in Dmin and Nd. Correspondingly, for Na = 5000 cm−3, a change in w or κ only leads to a small change in Dmin. While such a shift in Dmin only leads to small change in the activated fraction, it translates into a relatively large difference in Nd, resulting in high ξ(κ). This analysis demonstrates that the similarity in the Dmin ranges in Fig. 3b and c resulting from a change in κ or w by the same factor is coincidental and should not be general- ized to all conditions as the relative sensitivities to κ and w depend on Na. However, it also shows that conditions exist under which Nd is similarly sensitive to w and κ, and both parameters need to be taken into account to accurately pre- dict Nd. The ξ(Na) pattern in Fig. 4d exhibits a minimum when the activated fraction is ∼0.7. This fraction corresponds to the size range at which the cumulative ASD exhibits the largest slope. If more particles are activated, the relative change in Nd and therefore in ξ(Na) becomes smaller. 3.2.2 Sensitivity regimes of ξ(Na) for mono- and bimodal ASDs In the presence of an Aitken mode, in addition to an accu- mulation mode (ASD III; Fig. 4b), Nd is not limited by Na,acc under the same Na and w conditions as for ASD I since only a small fraction of the Aitken mode is activated. Figure 4e Previously, sensitivity of Nd to Na and to w was presented in terms of aerosol (Na)- and updraft (w)-limited regimes (Reutter et al., 2009, 2014; Chang et al., 2015). The Na- M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Accordingly, ξ(Na) ≥1, similar to the ξ(Na) values reached for much higher Na,acc and lower w for the monomodal accumulation mode ASD I. The ξ(Na) pattern in Fig. 4b exhibits a minimum when the activated fraction is ∼0.7. This fraction corresponds to the size range at which the cumulative ASD has the largest slope. If more particles are activated, the relative change in Nd and therefore in ξ(Na) becomes smaller. In summary, based on these trends of ξ(Na) and those of ξ(κ) in Sect. 3.2.1, the following conclusions can be drawn: At the highest Na, the ξ(κ) patterns do not show any sig- nificant difference (I.c to IV.c in Fig. 5). As under these con- ditions for κ, w and Na only accumulation mode particles are activated, the presence of the Aitken mode does not affect ξ(κ) within the κ and w ranges considered here. If our scales were extended to updraft velocities of several meters per sec- ond as encountered in deep convective clouds, Aitken mode particle particles may activate even if Na,acc ≥2500 cm−3, resulting in similar contour patterns as for lower Na and the w ranges considered here. However, in polluted air masses eventually a saturation effect in terms of droplet formation is reached above which Nd and effective radii do not signifi- cantly change, as shown for convective clouds in the Amazon region (Polonik et al., 2020). – The Na- and w-limited regimes are dependent on par- ticle size, and thus the w/Na limits are different for monomodal Aitken vs. accumulation mode ASDs and also for monomodal vs. bimodal ASDs. – The Na- and w-limited regimes are dependent on par- ticle size, and thus the w/Na limits are different for monomodal Aitken vs. accumulation mode ASDs and also for monomodal vs. bimodal ASDs. – The sensitivities of Nd to ASD parameters (Na, Dg, κ) and to w depend on their value combinations. – Under most w and Na conditions as considered here for bimodal ASDs, the Na-limited regime is not reached, as they cover the transitional regime. At first sight, the ξ(κ) patterns for ASD V are very dif- ferent to those of the other ASDs (columns I–IV vs. V in Fig. 5). These apparently different trends can be reconciled based on the discussion in Sect. 3.2.1. and 3.2.2: the higher ξ(κ) for highly hygroscopic particles (red area at the bottom right corner in Fig. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11730 shows the same ξ(Na) patterns as the part of Figure 4d for Na ≤2500 cm−3 under conditions where only accumulation mode particles are activated and thus contribute to Nd and to ξ(Na). Thus, the transitional regime is extended to a broader parameter space as ξ(Na) does not show a constant value of unity as it should in an Na-limited regime. As at high Na and low w, only accumulation mode particles contribute to Nd, the contours for the w-limited regime do not differ between Fig. 4d and e. The ξ(κ) values for the monomodal accumulation mode ASD I are shown in the first column of Fig. 5 (a: Na = 100 cm−3, b: 500 cm−3 and c: 2500 cm−3). The white regions in the upper right part of the figures mark the parame- ter spaces above which ξ(κ) ∼0, i.e., the Na-limited regime. At high Na, this space is shifted to higher κ and w, in agree- ment with the trends of larger Dmin at higher Na (Sect. 3.1.1). The trends in ξ(κ) along columns II, III and IV show the ef- fect on ξ(κ) due to increasing Na,Ait/Na,acc (0.5, 1.0, 1.5) for a fixed Na,acc in each row. The ξ(κ) contours in the param- eter space at which only accumulation mode particles con- tribute to Nd do not significantly change for a given Na,acc (bottom left corners of panels ASD I–IV). However, with increasing Na,Ait/Na,acc, the absolute ξ(κ) values increase, i.e., ∼0.15≲ξ(κ)≲0.25 for the major part of panels IV.a and IV.b whereas 0 ≤ξ(κ)≲0.25 in panels II.a and II.b. A higher Na,Ait/Na,acc causes the range to narrow at which ξ(κ) shows a minimum, while the w–κ combinations for minimum ξ(κ) are not significantly shifted (ASD I–IV within each row in Fig. 5). This trend in ξ(κ) is caused by the higher Na,Ait near the Hoppel minimum with increasing Na,Ait/Na,acc and thus higher Nd when Dmin ∼0.07 µm. g For the monomodal ASD V (Aitken mode), the acti- vated fraction reaches at most ∼0.6 (Fig. 4c, f); thus, the aerosol-limited regime is not reached. However, for Na >∼ 1000 cm−3 and w < ∼1.0 m s−1, Nd is linearly dependent on w and independent of Na, which implies a w-limited regime for w/Na <∼10−3 m s−1 cm3, i.e., shifted by an order of magnitude as compared to ASD I. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 5 V.a–c) implies that at low w, a large number of particles with high κ form droplets, whereas Nd is smaller for less hygroscopic particles. This suggests that Dmin for high κ is located in a “steep” part of the ASD with relatively high Na,Ait, and a decrease in κ increases Dmin, such that it is in the flat part of the ASD (Ds ∼0.1 µm). – The equivalency of a change in w and κ to affect Dmin (Fig. 3) implies that a κ-sensitive regime could be equally defined and taken into account as a w-sensitive regime when exploring sensitivities of Nd. https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11731 Figure 4. (a)–(c) Nd as a function of w and Na for ASD and κ = 0.7. (a) I, (b) III and (c). Color scale: Nd,acc, black/white scale: Nd,Ait. (d)–(f) Corresponding ξ(Na) for ASD. (d) I, (e) III and (f) V. Solid black lines show contours of activated fraction of the accumulation mode (Fact,acc) and dashed grey lines show contours of activated fraction of the Aitken mode (Fact,Ait). Figure 4. (a)–(c) Nd as a function of w and Na for ASD and κ = 0.7. (a) I, (b) III and (c). Color scale: Nd,acc, black/white scale: Nd,Ait. (d)–(f) Corresponding ξ(Na) for ASD. (d) I, (e) III and (f) V. Solid black lines show contours of activated fraction of the accumulation mode (Fact,acc) and dashed grey lines show contours of activated fraction of the Aitken mode (Fact,Ait). As discussed in Sect. 3.2.2., the conditions for ASD V can be described as a w- and/or κ-limited regime that it is char- acterized by a relatively low w/Na (or κ/Na) ratio. Using ASD I, similar ξ(κ) patterns as in Fig. 5V.a could be ob- tained using ASD I.a for smaller κ and/or w or higher Na than considered here. This can be seen in the upper row of Fig. 5 where panel V.a is apparently a continuation of panel I.a to smaller κ values. However, the match of the corre- sponding panels (V.b to I.b or V.c to I.c, respectively) is not as perfect which suggests different sensitivities of ξ(κ) to Dg and κ, depending on Na. Thus, not only are the limits of the w-limited regime depending on Dg (i.e., accumulation or Aitken mode) but also on the parameter ranges that charac- terize the regimes. tainties in cloud properties predicted in global model studies when Aitken mode particles significantly influence Nd and other cloud properties (e.g., Lee et al., 2013; Chang et al., 2021). 3.2.3 Sensitivities ξ(κ) for different aerosol size distributions The parameter ranges of w, κ and Na considered in our sim- ulations in the presence of an Aitken mode constrain tran- sitional or w-limited regimes in which Nd can be equally influenced by w and κ. Therefore, we explore in detail the w–κ combinations above which Aitken mode particles sig- nificantly affect Nd and ξ(κ). Figure 5 shows ξ(κ) contour plots for all cases as defined in Fig. 1; the middle column (ASD III) repeats Figs. 3a, S5a and S6a. As the supersatu- ration and activated fractions are closely related to ξ(κ), the corresponding figures for the 15 cases are shown in Figs. S7 and S8. For ASD I, such situations are not encountered for the cho- sen parameter ranges. Even for the lowest κ and w, a signif- icant fraction of Na,acc is activated, and any change in Dmin due to a change in w or κ shifts it along the steep part of the cumulative ASD. If we performed simulations for higher Na,acc, i.e., for conditions typical for a w-limited regime, Dmin would be shifted to even larger sizes than shown in Fig. S6b and c, resulting in very small Nd and ξ(κ). https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11732 Figure 5. Sensitivity of Nd to κ (ξ(κ)) as a function of w and κ for ASDs I.a–V.c (Fig. 1). Contour plots (0 ≤ξ(κ) ≤1, color code in panel (a)) are based on 810 model simulations assuming 30 different values of w and 27 different values of κ for each ASD. Column ASD III repeats panel (a) in Figs. 3, S5 and S6. Figure 5. Sensitivity of Nd to κ (ξ(κ)) as a function of w and κ for ASDs I.a–V.c (Fig. 1). Contour plots (0 ≤ξ(κ) ≤1, color code in panel (a)) are based on 810 model simulations assuming 30 different values of w and 27 different values of κ for each ASD. Column ASD III repeats panel (a) in Figs. 3, S5 and S6. during the wet season in the Amazon, with κAit ∼0.1 and κacc ∼0.2 (Zhou et al., 2002; Gunthe et al., 2009). In a global model study, fairly large differences were predicted above oceans (κAit ∼0.5, κacc ∼1) and more similar values for both modes above continents (∼0.3 ≤κ≲0.8; Chang et al., 2017). supersaturation and prevent efficient growth of accumulation mode particles. The large overlap of the activated fractions from both modes in Fig. S8a (κacc = 0.1) demonstrates that at w≳1 m s−1, large Aitken mode particles with κAit≳0.3 may grow to droplet sizes, even though only ∼70 % of accumula- tion mode particles are activated, whereas the smallest 30 % of the accumulation mode particles have not been activated yet. To explore situations with κacc ̸= κAit, we repeat the simu- lations for ASD III.b but use a single value of κacc, whereas the full range of 0.02 ≤κAit ≤1 is applied (Fig. 6a). For low w and κ, ξ(κAit) is zero (white space in Fig. 6c and d) be- cause in this parameter range droplets only form on accumu- lation mode particles and a change in κAit does not affect Nd. When κAit > κacc, small accumulation mode particles may not become activated, whereas more hygroscopic (but smaller) Aitken mode particles sufficiently grow and con- tribute to Nd. Thus, the total Nd is the sum based on two separate Dmin values for Aitken and accumulation modes, respectively. The activated fractions for the two simulations are compared in Fig. 3.3.1 Influence of constant κacc on ξ(κ) While we have assumed so far that both Aitken and accu- mulation modes have the same κ, such conditions are rarely encountered in the atmosphere. They might occur, for ex- ample, when sea salt contributes significantly to both modes (Wex et al., 2016). However, more frequently the accumula- tion mode consists of material of higher hygroscopicity as it accumulates sulfate and other compounds during cloud pro- cessing and other aging processes. Continental accumulation modes typically exhibit values in a range of 0.1 < κacc≲0.5, with an average value of κacc ∼0.3 (Andreae and Rosenfeld, 2008). The Aitken mode is comprised of fresher, less hygro- scopic particles with 0 < κ≲0.05 in urban and continental air masses, corresponding to hygroscopic growth factors ≤1.1 at RH = 90 %. Aged Aitken mode particles are more hygro- scopic, with κ ∼0.3 in the free troposphere and κ ∼0.6 in remote marine air (McFiggans et al., 2006). Similar trends in the hygroscopicity of the two modes were also observed The comparison of the influence on ξ(κ) due to increas- ing Na,Ait (V.a, b, c in Fig. 5) and on ξ(Na) (Fig. 4f) shows that ξ(κ) is increasing, whereas ξ(Na) is decreasing. Therefore, when high concentrations of Aitken mode parti- cles dominate the ASD, Nd is highly sensitive to κ and w and less to Na,Ait. While in the presence of accumulation- mode-dominated ASDs such w- and κ-sensitive conditions might only be encountered in highly polluted air masses (e.g., biomass burning), they can occur for Aitken mode- dominated ASDs at much lower aerosol concentrations. This shift in sensitivities might partially explain the large uncer- https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets S8; Fact,acc appears as horizontal lines as it is independent of κAit; only at high κAit values are there small deviations from this behavior as very hygro- scopic Aitken mode particles may sufficiently suppress the The similar ξ(κ(Ait)) values in Fig. 6b–d show their weak dependence on κacc. The supersaturation is largely controlled by Na,acc, resulting in very similar values over the full κ range, independent of the presence of an Aitken mode (Fig- ure S7 I.b–IV.b). When κacc = 0.1, droplet formation on Aitken mode particles occurs for slightly lower values of κ and w as compared to the case with κacc = 0.5 (Fig. 6c and d). More hygroscopic accumulation mode particles ef- ficiently suppress the supersaturation and prevent smaller (Aitken mode) particles from efficient growth. Thus, for the same w–κ combinations, Fact,Ait is smaller when κacc = 0.5 compared to κacc = 0.1 (Fig. S8). Similar feedbacks of the two modes on cloud properties were described in previous model sensitivity studies that Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11733 Pöhlker et al.: Aitken mode contributions to cloud droplets 6. Sensitivity of Nd to κ (ξ(κ)) as a function of w and κ for variable ASDs and aerosol properties as outlined in the top atic ASD with constant κacc; (b) identical to Fig. 5 III.b; (c) ξ(κAit), assuming κacc = 0.1; (d) ξ (κAit), assuming κacc = 0.5 fferent Dg,Ait in addition to ASD III; (f) identical to Fig. 5 III.b; (g) ξ(κ) for Dg,Ait = 0.05 µm; (h) ξ(κ) for Dg,Ait = 0.07 µm fferent Dg,acc in addition to ASD III; (j) ξ(κ) for Dg,acc = 0.13 µm; (k) identical to Fig. 5 III.b; and (l) ξ(κ) for Dg,acc = 0 Figure 6. Sensitivity of Nd to κ (ξ(κ)) as a function of w and κ for variable ASDs and aerosol properties as outlined in the top panels. (a) Schematic ASD with constant κacc; (b) identical to Fig. 5 III.b; (c) ξ(κAit), assuming κacc = 0.1; (d) ξ (κAit), assuming κacc = 0.5; (e) ASDs with different Dg,Ait in addition to ASD III; (f) identical to Fig. 5 III.b; (g) ξ(κ) for Dg,Ait = 0.05 µm; (h) ξ(κ) for Dg,Ait = 0.07 µm; (i) ASDs with different Dg,acc in addition to ASD III; (j) ξ(κ) for Dg,acc = 0.13 µm; (k) identical to Fig. 5 III.b; and (l) ξ(κ) for Dg,acc = 0.17 µm. ble (hygroscopic) Aitken than in small, less soluble accu- mulation mode particles. A significant contribution of Aitken mode particles to Nd (> 50 %) was observed at a background site in northern Finland, with average activated fractions of Fact,acc ∼87 % and Fact,Ait ∼30 % (Komppula et al., 2005). While such observations could be equally explained by exter- showed more Aitken mode particles to activate in the pres- ence of less hygroscopic accumulation mode particles (Kul- mala et al., 1996). In the latter study, soluble and insoluble mass fractions were used as proxies of particle composition, and it was shown that CCN activation can be parameterized by the soluble mass, which may be higher in large, solu- 3.3.2 Influence of Dg,Ait and Dg,acc on ξ(κ) 3.3.2 Influence of Dg,Ait and Dg,acc on ξ(κ) The parameters commonly used to characterize lognormal ASDs, Dg and σg, were identified as the most important aerosol parameters in affecting cloud properties (e.g., Fein- gold, 2003; Ervens et al., 2005; Reutter et al., 2009; Ward et al., 2010; Anttila et al., 2012). To compare their impor- tance for accumulation and Aitken modes, we vary Dg,Ait and Dg,acc within the range of observed values. Our base case values (Dg,Ait = 0.037 µm, Dg,acc = 0.145 µm, Fig. 1) are typical for oceanic aerosol (Wex et al., 2016). In addi- tion, we apply Dg,acc = 0.13 µm and 0.17 µm for continental aerosol (Pöhlker et al., 2016, 2018). The size of Aitken mode particles strongly depends on their aging state; it is larger for continental aerosol than above the ocean (e.g., Birmili et al., 2001; Heintzenberg et al., 2004; Pöhlker et al., 2016, 2018). In sensitivity tests, we assume Dg,Ait = 0.05 µm and 0.07 µm while keeping Dg,acc = 0.145 µm. To provide a general framework of our model results, we extract from the sensitivity simulations the w–κ com- binations where Fact,Ait = 0.05 as a threshold above which Aitken mode particles may significantly contribute to the to- tal droplet number concentration. Figure 7a and c present a selection of our model ASDs, together with additional ASDs (Na,Ait/Na,acc ∼10 and ∼0.1) to further map out the param- eter space. In the bottom panels, the w–κ lines are summa- rized from each simulation that correspond to Fact,Ait = 0.05 (Fig. 7b, d). The black lines in Fig. 7 show two ASDs with equal contri- butions of Aitken and accumulation mode to Na (1000 cm−3, ASD III.b). For ASDs of these Dg and Na, any combination along the lines yields Fact,Ait ≥0.05, such as w ≥1.5 m s−1 and κAit ∼1 or κ ≥0.3 and w ∼3 m s−1, respectively. The red lines in Fig. 7b mark the shift of the w–κ as a function of Na and Na,acc/Na,Ait: decreasing Na of both modes by a factor 5 moves the line to much lower w and κ values, such that Aitken mode particles with κ ∼0.4 may be activated at w ≳1 m s−1 and more hygroscopic Aitken mode particles at even lower w (dotted red line). 3.3.2 Influence of Dg,Ait and Dg,acc on ξ(κ) When the ASD is domi- nated by an Aitken mode (Na,Ait ∼10Na,acc), Aitken mode particles with κAit ≥0.1 will be activated at w ≥1 m s−1 (dashed red line). Conversely, when Na,acc ∼10Na,Ait with Na,Ait = 55 cm−3, the w and κ values are nearly outside the ranges of our κ and w scales (solid red line). Results for even higher Na (2500–6750 cm−3) are consequently not included in the figure because under these conditions, the efficient sup- pression of the supersaturation by the high Na,acc prevents Aitken mode particles from being activated within the con- sidered parameter ranges of w and κ. Obviously, for wider w ranges as relevant for pyrocumuli or other highly convective cloud systems, corresponding thresholds and w–κ combina- tions could be derived on extended axes. p g g,acc µ We use ASD III.b as the reference case; it is shown to- gether with the Dg-shifted ASDs in Fig. 6e and i. The pan- els below the ASDs show the effect on ξ(κ) of increasing Dg,Ait (f, g, h) and Dg,acc (j, k, l). With increasing Dg,Ait, the parameter space at which ξ(κ) exhibits minimum val- ues is shifted to lower w and κ values. The lowest values of ξ(κ) (∼0) are predicted for the smallest Dg,Ait, whereas ξ(κ) ∼0.2 for most of the w–κ space above which Aitken mode particles are activated. This is in agreement with our interpre- tation of Fig. 3 that the dependence of ξ(κ) on Dmin traces the ASD shape. With Dg,Ait = 0.07 µm, there is no w–κ space in which ξ(κ) shows a distinct minimum as both modes largely overlap (dotted green line in Fig. 6e). As Dmin is determined by the supersaturation, which, in turn, is largely controlled by the accumulation mode properties (Na,acc, Dg,acc, κacc), a shift of Dg,Ait to larger sizes moves Dmin to a different part of the ASD. For example, while Dmin ∼0.06 µm only leads to a small activated fraction of Aitken mode particles (Fact,Ait < 0.1) if Dg,Ait = 0.037 µm, it would be > 0.5 with Dg,Ait = 0.07 µm. The purple lines in Fig. 7b correspond to results for con- stant Na,acc = 500 cm−3 and for Na,Ait, being 0.5 and 1.5 times that of the accumulation mode (ASD II.b, IV.b). M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11734 as compared to Dg,Ait for the w–κ parameter space above which Aitken mode particles contribute to Nd. as compared to Dg,Ait for the w–κ parameter space above which Aitken mode particles contribute to Nd. nally mixed aerosol, this would result in the same effects dur- ing cloud processing: efficient formation of mass (e.g., sul- fate) in droplets formed on Aitken mode particles that led to a narrowing of the Hoppel minimum rather than to a widening. In a global model study, it was demonstrated that efficient sulfate formation in such droplets could contribute several percent (≳5 %) to the global sulfate budget (Roelofs et al., 2006). 4 Updraft and hygroscopicity regimes of Aitken mode CCN activation Our sensitivity studies have shown that for bimodal (Aitken and accumulation mode) ASDs, the w–κ combinations re- sulting in ξ(κ) minimum values can be used as a criterion of conditions under which Aitken mode particles contribute to Nd. An increase in w or κ decreases Dmin to sizes smaller than the region of the Hoppel minimum. https://doi.org/10.5194/acp-21-11723-2021 https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 5 Summary and conclusions Previous field and model studies suggested that not only ac- cumulation, but also – under specific conditions – Aitken mode particles increase cloud droplet number concentrations and are involved in aerosol–cloud interactions. However, the conditions under which the Aitken mode significantly con- tributes to cloud droplet number concentration (Nd) had not been fully constrained. y Using an adiabatic parcel model, we systematically inves- tigated the conditions under which Aitken mode particles contribute to Nd for wide ranges of aerosol size distribution (ASD) parameters (particle number concentrations of accu- mulation and Aitken modes, Na,acc, Na,Ait, mode diameters, Dg,acc, Dg,Ait, and hygroscopicities κacc, κAit) and of the up- draft velocity w for Nd. In previous model sensitivity studies of monomodal ASDs, aerosol- and updraft-limited regimes were defined in which Nd depends linearly on Na or w (Reut- ter et al., 2009). Using this concept, we show that Aitken mode particles are not activated if updraft-limited conditions prevail in the presence of a dominant accumulation mode (high Na,acc). Also, aerosol-limited conditions do not occur as by far not all Aitken mode particles are activated (for w ≤3 m s−1, i.e., updraft velocities of abundant cloud types), and thus the transitional regime (between Na and w limita- tions) exists over wider parameter spaces than in the pres- ence of monomodal accumulation mode ASDs. When ASDs are dominated by an Aitken mode, we find that Nd is highly sensitive to w even at low Na,Ait which implies that the w/Na regime limits as identified previously for accumulation mode ASDs are not always applicable, but they depend on param- eter value combinations of Dg, Na, κ and w. Figure 7. Updraft and hygroscopicity regimes where Aitken mode particles are relevant as CCN. (a) Aerosol size distributions with different Na and Na,Ait/Na,acc. (b) Lines indicate the w–κ combi- nations for which Fact,Ait = 0.05 using the corresponding ASDs in (a) (same color). (c) Aerosol size distributions with different Dg,Ait and Dg,acc. (d) Lines indicate the w–κ combinations for which Fact,Ait = 0.05 using the corresponding ASDs in (c) (same color). g Exceeding the previous framework that was restricted to w and Na limitations, we show that the sensitivities of Nd in the transitional and w-limited regimes equally depend on w and κ. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11735 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Figure 7. Updraft and hygroscopicity regimes where Aitken mode particles are relevant as CCN. (a) Aerosol size distributions with different Na and Na,Ait/Na,acc. (b) Lines indicate the w–κ combi- nations for which Fact,Ait = 0.05 using the corresponding ASDs in (a) (same color). (c) Aerosol size distributions with different Dg,Ait and Dg,acc. (d) Lines indicate the w–κ combinations for which Fact,Ait = 0.05 using the corresponding ASDs in (c) (same color). (Na, Na,Ait/Na,acc, Dg,acc, Dg,Ait), the ranking of their rela- tive importance in determining the position of the w–κ line is expected to hold generally true. 3.3.2 Influence of Dg,Ait and Dg,acc on ξ(κ) The fact that they do not show any noticeable difference to results using ASD III.b demonstrates that the ratio Na,Ait/Na,acc – if Na,acc is approximately constant – does not significantly im- pact the position of the w–κ line. Accordingly, the trends in ξ(κ) for a change in Dg,acc (Fig. 6j, k, l) can be explained: the Hoppel minimum is widest for the ASD with Dg,acc = 0.17 µm, which is reflected by the large space in which ξ(κ) ∼0 (Fig. 6l). However, un- like the shift of the range in which ξ(κ) shows a minimum to lower w–κ values for increasing Dg,Ait, the w–κ space only becomes broader for larger Dg,acc but barely changes its po- sition. It can be concluded that Dg,acc is of minor importance Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 5 Summary and conclusions Therefore, we explored in detail the sensitivity of Nd to κ, ξ(κ), as a function of w for ASDs that differ in the number of modes (mono- or bimodal), Na,Ait/Na,acc and total Na. Based on the patterns of ξ(κ) as a function of w and κ, we analyze the dependence of the w–κ range above which Aitken mode particles contribute to Nd on the ASD parameters. We show that ξ(κ) exhibits minimum values for w–κ combinations for which the smallest activated parti- cle size (Dmin) is near the Hoppel minimum and increases when smaller Aitken mode particles are activated. Defining lines near these w–κ combinations as the minimum thresh- old, it can be estimated under which aerosol (Na, κ, Dg) and w conditions Aitken mode particles start contributing to Nd. We conclude that the most important requirements are a low number concentration of total and accumulation mode In Fig. 7d, we show the effects of Dg,Ait and Dg,acc on the w–κ space. A change in Dg,acc from 0.145 to 0.13 µm or to 0.17 µm has a negligible effect (solid and dotted blue lines). The low sensitivity of the w–κ line position to Dg,acc suggests that it is applicable to continental and marine air masses, largely independent of Dg,acc. However, an increase in Dg,Ait from 0.037 µm (ASD III.b) to 0.05, 0.06 and 0.07 µm (green lines) significantly reduces the w and κ val- ues that are required to activate Aitken mode particles. Thus, aging processes might efficiently increase Dg,Ait to those of CCN. Figure 7b and d represent a simple scheme that can be ap- plied to estimate whether Aitken mode particles contribute to Nd under ambient conditions in various air masses and cloud types. While we did not explore all relevant combinations of the parameters that are commonly used to characterize ASDs Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11736 particles (Na, Na,acc) and/or a large mode diameter of the Aitken mode (Dg,Ait). While this ranking repeats previous findings for sensitivities to monomodal ASDs (e.g., Ervens et al., 2005; Reutter et al., 2009; Ward et al., 2010; Cecchini et al., 2017; Hernández Pardo et al., 2019), our analysis ex- ceeds these studies as it evaluates the relative importance of these parameters of accumulation and Aitken modes for the activation of Aitken mode particles. Global model studies have identified large uncertainties in CCN number concentration and Nd predictions due to the as- sumptions associated with Aitken mode particle properties, specifically in the Southeast US, Europe and to a small extent in the Amazon region (Lee et al., 2013; Chang et al., 2021). Our framework, together with global maps of κAit and κacc (e.g., Chang et al., 2017), will help to reduce these uncer- tainties and constrain aerosol–cloud interactions in regions where Aitken mode particles affect cloud properties. particles (Na, Na,acc) and/or a large mode diameter of the Aitken mode (Dg,Ait). While this ranking repeats previous findings for sensitivities to monomodal ASDs (e.g., Ervens et al., 2005; Reutter et al., 2009; Ward et al., 2010; Cecchini et al., 2017; Hernández Pardo et al., 2019), our analysis ex- ceeds these studies as it evaluates the relative importance of these parameters of accumulation and Aitken modes for the activation of Aitken mode particles. Applying this framework to typical ambient aerosol condi- tions, it seems likely that above the ocean where aerosol load- ing is usually low and ASDs often exhibit bimodal shapes with very hygroscopic particles (Wex et al., 2016; Campos Braga et al., 2021), Aitken mode particles are activated to cloud droplets. This confirms findings in marine stratocu- mulus clouds with moderate w ≥0.5 m s−1 (Schulze et al., 2020). Given that marine stratocumuli comprise a large frac- tion of global cloud coverage, the contribution of Aitken mode particles to Nd above the ocean should thus be included in global estimates of aerosol–cloud interactions. Similarly, our concept is consistent with the large observed fractions of activated Aitken mode particles at Arctic sites (Komppula et al., 2005). Conversely, it implies that in highly polluted re- gions even high Na,Ait (e.g., in megacities; Mönkkönen et al., 2005) are not relevant in stratocumulus and shallow cumulus clouds as droplets will only form on accumulation mode par- ticles. Atmos. Chem. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 11737 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets https://doi.org/10.5194/acp-21-11723-2021 Competing interests. The authors declare that they have no conflict of interest. Competing interests. The authors declare that they have no conflict of interest. p g Cecchini, M. A., Machado, L. A. T., Andreae, M. O., Martin, S. T., Albrecht, R. I., Artaxo, P., Barbosa, H. M. J., Borrmann, S., Fütterer, D., Jurkat, T., Mahnke, C., Minikin, A., Molleker, S., Pöhlker, M. L., Pöschl, U., Rosenfeld, D., Voigt, C., Weinzierl, B., and Wendisch, M.: Sensitivities of Amazonian clouds to aerosols and updraft speed, Atmos. Chem. Phys., 17, 10037– 10050, https://doi.org/10.5194/acp-17-10037-2017, 2017. Disclaimer. Publisher’s note: Copernicus Publications remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Financial support. Barbara Ervens and Minghui Zhang were financially supported by the French National Research Agency (ANR) (grant no. ANR-17-MPGA-0013). Chang, D., Cheng, Y., Reutter, P., Trentmann, J., Burrows, S. M., Spichtinger, P., Nordmann, S., Andreae, M. O., Pöschl, U., and Su, H.: Comprehensive mapping and characteristic regimes of aerosol effects on the formation and evolution of pyro-convective clouds, Atmos. Chem. Phys., 15, 10325–10348, https://doi.org/10.5194/acp-15-10325-2015, 2015. The article processing charges for this open-access publication were covered by the Max Planck Society. Chang, D. Y., Lelieveld, J., Tost, H., Steil, B., Pozzer, A., and Yoon, J.: Aerosol physicochemical effects on CCN activation simulated with the chemistry-climate model EMAC, Atmos. Environ., 162, 127–140, https://doi.org/10.1016/j.atmosenv.2017.03.036, 2017. Review statement. This paper was edited by Hailong Wang and re- viewed by Jingyi Chen and one anonymous referee. Chang, D. Y., Lelieveld, J., Steil, B., Yoon, J., Yum, S. S., and Kim, A.-H.: Variability of aerosol-cloud interactions induced by differ- ent cloud droplet nucleation schemes, Atmos. Res., 250, 105367, https://doi.org/10.1016/j.atmosres.2020.105367, 2021. Appendix A Table A1. Definition of parameters that are used in the discussion. Parameter Description Dg Geometric mean mode diameter Dg,Ait Mean Aitken mode diameter Dg,acc Mean Aitken mode diameter Dmin D of smallest particle that forms a cloud droplet Ds Diameter of dry particle Fact,Ait Activated fraction of Aitken mode particles Nd,Ait/Na,Ait Fact,acc Activated fraction of Aitken mode particles Nd,acc/Na,acc κ Hygroscopicity parameter κAit Hygroscopicity parameter for Aitken mode particles κacc Hygroscopicity parameter for accumulation mode particles Na Particle number concentration Na,acc Particle number concentration of Aitken mode particles Na,Ait Particle number concentration of accumulation mode particles Nd Predicted droplet number concentration Nd,Ait Number concentration of droplets formed on Aitken mode particles Nd,acc Number concentration of droplets formed on accumulation mode particles s Saturation seq Equilibrium saturation based on Köhler theory Smax Supersaturation σg Geometric standard deviation for mode w Updraft velocity ξ(κ) Sensitivity of Nd to κ (Eq. 4) ξ(Na) Sensitivity of Nd to Na (Eq. 5) Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11738 Code and data availability. Details on the model codes and further model results can be obtained from the corresponding authors upon request. Campos Braga, R., Ervens, B., Rosenfeld, D., Andreae, M. O., Förster, J.-D., Fütterer, D., Hernández Pardo, L., Holanda, B. A., Jurkat, T., Krüger, O. O., Lauer, O., Machado, L. A. T., Pöhlker, C., Sauer, D., Voigt, C., Walser, A., Wendisch, M., Pöschl, U., and Pöhlker, M. L.: Cloud droplet number clo- sure for tropical convective clouds during the ACRIDICON– CHUVA campaign, Atmos. Chem. Phys. Discuss. [preprint], https://doi.org/10.5194/acp-2021-80, in review, 2021. Supplement. The supplement related to this article is available on- line at: https://doi.org/10.5194/acp-21-11723-2021-supplement. Bulatovic, I., Igel, A. L., Leck, C., Heintzenberg, J., Riipinen, I., and Ekman, A. M. L.: The importance of Aitken mode aerosol particles for cloud sustenance in the summertime high Arc- tic – a simulation study supported by observational data, At- mos. Chem. Phys., 21, 3871–3897, https://doi.org/10.5194/acp- 21-3871-2021, 2021. Author contributions. MLP and BE led the study and wrote the manuscript with input from all coauthors. MZ and BE performed the model simulations. MZ, RCB, OOK and UP commented on the manuscript. Cantrell, W., Shaw, G., and Benner, R.: Cloud proper- ties inferred from bimodal aerosol number distribu- tions, J. Geophys. Res.-Atmos., 104, 27615–27624, https://doi.org/10.1029/1999JD900252, 1999. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11739 Korhonen, H., Carslaw, K. S., Spracklen, D. V., Ridley, D. A., and Ström, J.: A global model study of pro- cesses controlling aerosol size distributions in the Arctic spring and summer, J. Geophys. Res.-Atmos., 113, D08211, https://doi.org/10.1029/2007JD009114, 2008. M., Daskalakis, N., Decesari, S., Gysel-Beer, M., Kalivitis, N., Liu, X., Mahowald, N. M., Myriokefalitakis, S., Schrödner, R., M., Daskalakis, N., Decesari, S., Gysel-Beer, M., Kalivitis, N., Liu, X., Mahowald, N. M., Myriokefalitakis, S., Schrödner, R., Sfakianaki, M., Tsimpidi, A. P., Wu, M., and Yu, F.: Evaluation of global simulations of aerosol particle and cloud condensation nuclei number, with implications for cloud droplet formation, At- mos. Chem. Phys., 19, 8591–8617, https://doi.org/10.5194/acp- 19-8591-2019, 2019. Kulmala, M., Korhonen, P., Vesala, T., Hansson, H.-C., Noone, K., and Svenningsson, B.: The effect of hygroscopic- ity on cloud droplet formation, Tellus B, 48, 347–360, https://doi.org/10.3402/tellusb.v48i3.15903, 1996. Feingold, G.: Modeling of the first indirect effect: Analysis of measurement requirements, Geophys. Res. Lett., 30, 1997, https://doi.org/10.1029/2003GL017967, 2003. Lee, L. A., Pringle, K. J., Reddington, C. L., Mann, G. W., Stier, P., Spracklen, D. V., Pierce, J. R., and Carslaw, K. S.: The mag- nitude and causes of uncertainty in global model simulations of cloud condensation nuclei, Atmos. Chem. Phys., 13, 8879–8914, https://doi.org/10.5194/acp-13-8879-2013, 2013. Feingold, G. and Heymsfield, A. J.: Parameterizations of conden- sational growth of droplets for use in general circulation mod- els, J. Atmos. Sci., 49, 2325–2342, https://doi.org/10.1175/1520- 0469(1992)049<2325:POCGOD>2.0.CO;2, 1992. Loftus, A. M.: Towards an enhanced droplet activation scheme for multi-moment bulk microphysics schemes, Atmos. Res., 214, 442–449, https://doi.org/10.1016/j.atmosres.2018.08.025, 2018. Feingold, G. and Kreidenweis, S.: Does cloud processing of aerosol enhance droplet concentrations?, J. Geophys. Res., 105, 24351– 24361, https://doi.org/10.1029/2000JD900369, 2000. Gérémy, G., Wobrock, W., Flossmann, A. I., Schwarzenböck, A., and Mertes, S.: A modelling study on the activation of small Aitken-mode aerosol particles during CIME 97, Tellus B, 52, 959–979, https://doi.org/10.3402/tellusb.v52i3.17078, 2000. McFiggans, G., Artaxo, P., Baltensperger, U., Coe, H., Facchini, M. C., Feingold, G., Fuzzi, S., Gysel, M., Laaksonen, A., Lohmann, U., Mentel, T. F., Murphy, D. M., O’Dowd, C. D., Snider, J. R., and Weingartner, E.: The effect of physical and chemical aerosol properties on warm cloud droplet activation, Atmos. Chem. Phys., 6, 2593–2649, https://doi.org/10.5194/acp-6-2593- 2006, 2006. Gunthe, S. S., King, S. M., Rose, D., Chen, Q., Roldin, P., Farmer, D. K., Jimenez, J. L., Artaxo, P., Andreae, M. O., Martin, S. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets T., and Pöschl, U.: Cloud condensation nuclei in pristine tropi- cal rainforest air of Amazonia: size-resolved measurements and modeling of atmospheric aerosol composition and CCN activity, Atmos. Chem. Phys., 9, 7551–7575, https://doi.org/10.5194/acp- 9-7551-2009, 2009. Mönkkönen, P., Koponen, I. K., Lehtinen, K. E. J., Hämeri, K., Uma, R., and Kulmala, M.: Measurements in a highly polluted Asian mega city: observations of aerosol number size distribu- tion, modal parameters and nucleation events, Atmos. Chem. Phys., 5, 57–66, https://doi.org/10.5194/acp-5-57-2005, 2005. Hammer, E., Bukowiecki, N., Luo, B. P., Lohmann, U., Mar- colli, C., Weingartner, E., Baltensperger, U., and Hoyle, C. R.: Sensitivity estimations for cloud droplet formation in the vicinity of the high-alpine research station Jungfrau- joch (3580 m a.s.l.), Atmos. Chem. Phys., 15, 10309–10323, https://doi.org/10.5194/acp-15-10309-2015, 2015. Moore, R. H., Karydis, V. A., Capps, S. L., Lathem, T. L., and Nenes, A.: Droplet number uncertainties associated with CCN: an assessment using observations and a global model adjoint, At- mos. Chem. Phys., 13, 4235–4251, https://doi.org/10.5194/acp- 13-4235-2013, 2013. Petters, M. D. and Kreidenweis, S. M.: A single parameter representation of hygroscopic growth and cloud condensa- tion nucleus activity, Atmos. Chem. Phys., 7, 1961–1971, https://doi.org/10.5194/acp-7-1961-2007, 2007. Heintzenberg, J., Birmili, W., Wiedensohler, A., Nowak, A., and Tuch, T.: Structure, variability and persistence of the submicrometre marine aerosol, Tellus B, 56, 357–367, https://doi.org/10.3402/tellusb.v56i4.16450, 2004. Pöhlker, M. L., Pöhlker, C., Ditas, F., Klimach, T., Hrabe de Ange- lis, I., Araújo, A., Brito, J., Carbone, S., Cheng, Y., Chi, X., Ditz, R., Gunthe, S. S., Kesselmeier, J., Könemann, T., Lavriˇc, J. V., Martin, S. T., Mikhailov, E., Moran-Zuloaga, D., Rose, D., Sat- urno, J., Su, H., Thalman, R., Walter, D., Wang, J., Wolff, S., Bar- bosa, H. M. J., Artaxo, P., Andreae, M. O., and Pöschl, U.: Long- term observations of cloud condensation nuclei in the Amazon rain forest – Part 1: Aerosol size distribution, hygroscopicity, and new model parametrizations for CCN prediction, Atmos. Chem. Phys., 16, 15709–15740, https://doi.org/10.5194/acp-16-15709- 2016, 2016. Hernández Pardo, L., Toledo Machado, L. A., Amore Cecchini, M., and Sánchez Gácita, M.: Quantifying the aerosol effect on droplet size distribution at cloud top, Atmos. Chem. Phys., 19, 7839–7857, https://doi.org/10.5194/acp-19-7839-2019, 2019. Hoppel, W. A., Frick, G. M., and Larson, R. E.: Effect of non- precipitating clouds on the aerosol size distribution in the marine boundary layer, Geophys. Res. Lett., 13, 125–128, https://doi.org/10.1029/GL013i002p00125, 1986. Jung, C. H., Yoon, Y. J., Kang, H. J., Gim, Y., Lee, B. References Ervens, B.: Modeling the Processing of Aerosol and Trace Gases in Clouds and Fogs, Chem. Rev., 115, 4157–4198, https://doi.org/10.1021/cr5005887, 2015. Andreae, M. O. and Rosenfeld, D.: Aerosol-cloud-precipitation interactions. Part 1: The nature and sources of cloud-active aerosols, Earth-Sci. Rev., 89, 13–41, 2008. Ervens, B., Feingold, G., and Kreidenweis, S. M.: Influ- ence of water-soluble organic carbon on cloud drop num- ber concentration, J. Geophys. Res.-Atmos., 110, D18211, https://doi.org/10.1029/2004JD005634, 2005. Anttila, T. and Kerminen, V.-M.: On the contribution of Aitken mode particles to cloud droplet populations at continental back- ground areas – a parametric sensitivity study, Atmos. Chem. Phys., 7, 4625–4637, https://doi.org/10.5194/acp-7-4625-2007, 2007. Fan, J., Rosenfeld, D., Zhang, Y., Giangrande, S. E., Li, Z., Machado, L. A. T., Martin, S. T., Yang, Y., Wang, J., Artaxo, P., Barbosa, H. M. J., Braga, R. C., Comstock, J. M., Feng, Z., Gao, W., Gomes, H. B., Mei, F., Pöhlker, C., Pöhlker, M. L., Pöschl, U., and de Souza, R. A. F.: Substantial convection and precipi- tation enhancements by ultrafine aerosol particles, Science, 359, 411–418, https://doi.org/10.1126/science.aan8461, 2018. Anttila, T., Brus, D., Jaatinen, A., Hyvärinen, A.-P., Kivekäs, N., Romakkaniemi, S., Komppula, M., and Lihavainen, H.: Relationships between particles, cloud condensation nu- clei and cloud droplet activation during the third Pallas Cloud Experiment, Atmos. Chem. Phys., 12, 11435–11450, https://doi.org/10.5194/acp-12-11435-2012, 2012. Fanourgakis, G. S., Kanakidou, M., Nenes, A., Bauer, S. E., Bergman, T., Carslaw, K. S., Grini, A., Hamilton, D. S., John- son, J. S., Karydis, V. A., Kirkevåg, A., Kodros, J. K., Lohmann, U., Luo, G., Makkonen, R., Matsui, H., Neubauer, D., Pierce, Fanourgakis, G. S., Kanakidou, M., Nenes, A., Bauer, S. E., Bergman, T., Carslaw, K. S., Grini, A., Hamilton, D. S., John- son, J. S., Karydis, V. A., Kirkevåg, A., Kodros, J. K., Lohmann, U., Luo, G., Makkonen, R., Matsui, H., Neubauer, D., Pierce, J. R., Schmale, J., Stier, P., Tsigaridis, K., van Noije, T., Wang, H., Watson-Parris, D., Westervelt, D. M., Yang, Y., Yoshioka, Birmili, W., Wiedensohler, A., Heintzenberg, J., and Lehmann, K.: Atmospheric particle number size distribution in cen- tral Europe: Statistical relations to air masses and me- teorology, J. Geophys. Res.-Atmos., 106, 32005–32018, https://doi.org/10.1029/2000JD000220, 2001. https://doi.org/10.5194/acp-21-11723-2021 Atmos. Chem. Phys., 21, 11723–11740, 2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets 11740 rain forest aerosols, Atmos. Chem. Phys., 18, 10289–10331, https://doi.org/10.5194/acp-18-10289-2018, 2018. Rose, D., Gunthe, S. S., Mikhailov, E., Frank, G. P., Dusek, U., Andreae, M. O., and Pöschl, U.: Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment, Atmos. Chem. Phys., 8, 1153–1179, https://doi.org/10.5194/acp- 8-1153-2008, 2008. Pöhlker, M. L., Krüger, O. O., Förster, J.-D., Elbert, W., Fröhlich- Nowoisky, J., Pöschl, U., Pöhlker, C., Bagheri, G., Bodenschatz, E., Huffman, J. A., Scheithauer, S., and Mikhailov, E.: Respira- tory aerosols and droplets in the transmission of infectious dis- eases, arXiv [preprint], arXiv:2103.01188, 1 March 2021. Schulze, B. C., Charan, S. M., Kenseth, C. M., Kong, W., Bates, K. H., Williams, W., Metcalf, A. R., Jonsson, H. H., Woods, R., Sorooshian, A., Flagan, R. C., and Seinfeld, J. H.: Charac- terization of Aerosol Hygroscopicity Over the Northeast Pacific Ocean: Impacts on Prediction of CCN and Stratocumulus Cloud Droplet Number Concentrations, Earth and Space Science, 7, e2020EA001098, https://doi.org/10.1029/2020EA001098, 2020. Polonik, P., Knote, C., Zinner, T., Ewald, F., Kölling, T., Mayer, B., Andreae, M. O., Jurkat-Witschas, T., Klimach, T., Mahnke, C., Molleker, S., Pöhlker, C., Pöhlker, M. L., Pöschl, U., Rosenfeld, D., Voigt, C., Weigel, R., and Wendisch, M.: The challenge of simulating the sensitivity of the Amazonian cloud microstruc- ture to cloud condensation nuclei number concentrations, At- mos. Chem. Phys., 20, 1591–1605, https://doi.org/10.5194/acp- 20-1591-2020, 2020. D., Voigt, C., Weigel, R., and Wendisch, M.: The challenge of simulating the sensitivity of the Amazonian cloud microstruc- ture to cloud condensation nuclei number concentrations, At- mos. Chem. Phys., 20, 1591–1605, https://doi.org/10.5194/acp- 20-1591-2020, 2020. Ward, D. S., Eidhammer, T., Cotton, W. R., and Kreidenweis, S. M.: The role of the particle size distribution in assessing aerosol composition effects on simulated droplet activation, At- mos. Chem. Phys., 10, 5435–5447, https://doi.org/10.5194/acp- 10-5435-2010, 2010. Pruppacher, H. R. and Klett, J. D.: Microphysics of clouds and pre- cipitation, 2nd edn., Kluwer Academics Publisher, Dordrecht, the Netherlands, 2003. Reutter, P., Su, H., Trentmann, J., Simmel, M., Rose, D., Gunthe, S. S., Wernli, H., Andreae, M. O., and Pöschl, U.: Aerosol- and updraft-limited regimes of cloud droplet formation: influence of particle number, size and hygroscopicity on the activation of cloud condensation nuclei (CCN), Atmos. Chem. Phys., 9, 7067– 7080, https://doi.org/10.5194/acp-9-7067-2009, 2009. Wex, H., Dieckmann, K., Roberts, G. C., Conrath, T., Izaguirre, M. M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets Y., Ström, J., Krejci, R., and Tunved, P.: The sea- sonal characteristics of cloud condensation nuclei (CCN) in the arctic lower troposphere, Tellus B, 70, 1–13, https://doi.org/10.1080/16000889.2018.1513291, 2018. Pöhlker, M. L., Ditas, F., Saturno, J., Klimach, T., Hrabˇe de Ange- lis, I., Araùjo, A. C., Brito, J., Carbone, S., Cheng, Y., Chi, X., Ditz, R., Gunthe, S. S., Holanda, B. A., Kandler, K., Kesselmeier, J., Könemann, T., Krüger, O. O., Lavriˇc, J. V., Martin, S. T., Mikhailov, E., Moran-Zuloaga, D., Rizzo, L. V., Rose, D., Su, H., Thalman, R., Walter, D., Wang, J., Wolff, S., Barbosa, H. M. J., Artaxo, P., Andreae, M. O., Pöschl, U., and Pöh- lker, C.: Long-term observations of cloud condensation nuclei over the Amazon rain forest – Part 2: Variability and charac- teristics of biomass burning, long-range transport, and pristine Köhler, H.: The nucleus in and the growth of hygroscopic droplets, Transact. Faraday Soc., 32, 1152–1161, 1936. Komppula, M., Lihavainen, H., Kerminen, V. M., Kulmala, M., and Viisanen, Y.: Measurements of cloud droplet activation of aerosol particles at a clean subarctic background site, J. Geophys. Res.- Atmos., 110, D06204, https://doi.org/10.1029/2004JD005200, 2005. Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021 M. L. Pöhlker et al.: Aitken mode contributions to cloud droplets A., Hartmann, S., Herenz, P., Schäfer, M., Ditas, F., Schmeiss- ner, T., Henning, S., Wehner, B., Siebert, H., and Stratmann, F.: Aerosol arriving on the Caribbean island of Barbados: physical properties and origin, Atmos. Chem. Phys., 16, 14107–14130, https://doi.org/10.5194/acp-16-14107-2016, 2016. Reutter, P., Trentmann, J., Seifert, A., Neis, P., Su, H., Chang, D., Herzog, M., Wernli, H., Andreae, M. O., and Pöschl, U.: 3- D model simulations of dynamical and microphysical interac- tions in pyroconvective clouds under idealized conditions, At- mos. Chem. Phys., 14, 7573–7583, https://doi.org/10.5194/acp- 14-7573-2014, 2014. Zhou, J., Swietlicki, E., Hansson, H. C., and Artaxo, P.: Sub- micrometer aerosol particle size distribution and hygroscopic growth measured in the Amazon rain forest during the wet season, J. Geophys. Res.-Atmos., 107, LBA 22-1–LBA 22-10, https://doi.org/10.1029/2000JD000203, 2002. Roelofs, G. J., Stier, P., Feichter, J., Vignati, E., and Wilson, J.: Aerosol activation and cloud processing in the global aerosol- climate model ECHAM5-HAM, Atmos. Chem. Phys., 6, 2389– 2399, https://doi.org/10.5194/acp-6-2389-2006, 2006. Atmos. Chem. Phys., 21, 11723–11740, 2021 https://doi.org/10.5194/acp-21-11723-2021

https://openalex.org/W4289745418

https://digitalcommons.wustl.edu/cgi/viewcontent.cgi?article=1292&context=oa_4

English

Enhancing interprofessional collaboration and interprofessional education in women’s health

Medical education online

cc-by

Follow this and additional works at: https://digitalcommons.wustl.edu/oa_4 Part of the Medicine and Health Sciences Commons Please let us know how this document benefits you. Follow this and additional works at: https://digitalcommons.wustl.edu/oa_4 Part of the Medicine and Health Sciences Commons Please let us know how this document benefits you. Follow this and additional works at: https://digitalcommons.wustl.edu/oa_4 Washington University School of Medicine Washington University School of Medicine Digital Commons@Becker Digital Commons@Becker 2020-Current year OA Pubs Open Access Publications 8-4-2022 Enhancing interprofessional collaboration and interprofessional Enhancing interprofessional collaboration and interprofessional education in women's health education in women's health Laura Baecher-Lind Tufts University Angela C Fleming Beaumont Hospital Rashmi Bhargava University of Saskatchewan Susan M Cox Dell Medical School Elise N Everett University of Vermont See next page for additional authors Open Access Publications Open Access Publications This Open Access Publication is brought to you for free and open access by the Open Access Publications at Digital Commons@Becker. It has been accepted for inclusion in 2020-Current year OA Pubs by an authorized administrator of Digital Commons@Becker. For more information, please contact [email protected]. Enhancing interprofessional collaboration and interprofessional Enhancing interprofessional collaboration and interprofessional education in women's health education in women's health Laura Baecher-Lind Tufts University Angela C Fleming Beaumont Hospital Rashmi Bhargava University of Saskatchewan Susan M Cox Dell Medical School Elise N Everett University of Vermont See next page for additional authors Authors Authors Laura Baecher-Lind, Angela C Fleming, Rashmi Bhargava, Susan M Cox, Elise N Everett, David A Forstein, Shireen Madani Sims, Helen K Morgan, Christopher M Morosky, Celeste S Royce, Tammy S Sonn, Jill M Sutton, and Scott C Graziano Recommended Citation Recommended Citation Baecher-Lind, Laura; Fleming, Angela C; Bhargava, Rashmi; Cox, Susan M; Everett, Elise N; Forstein, David A; Madani Sims, Shireen; Morgan, Helen K; Morosky, Christopher M; Royce, Celeste S; Sonn, Tammy S; Sutton, Jill M; and Graziano, Scott C, "Enhancing interprofessional collaboration and interprofessional education in women's health." Medical Education Online. 27, 1. 2107419 (2022). https://digitalcommons.wustl.edu/oa_4/296 This Open Access Publication is brought to you for free and open access by the Open Access Publications at Digital Commons@Becker. It has been accepted for inclusion in 2020-Current year OA Pubs by an authorized administrator of Digital Commons@Becker. For more information, please contact [email protected]. This open access publication is available at Digital Commons@Becker: https://digitalcommons.wustl.edu/oa_4/296 This open access publication is available at Digital Commons@Becker: https://digitalcommons.wust Medical Education Online Medical Education Online ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/zmeo20 Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=zmeo20 MEDICAL EDUCATION ONLINE 2022, VOL. 27, 2107419 https://doi.org/10.1080/10872981.2022.2107419 MEDICAL EDUCATION ONLINE 2022, VOL. 27, 2107419 https://doi.org/10.1080/10872981.2022.2107419 MEDICAL EDUCATION ONLINE 2022, VOL. 27, 2107419 https://doi.org/10.1080/10872981.2022.2107419 ABSTRACT Without the complemen­ tary yet distinct skills and expertise of each of these disciplines, more profound morbidity and mortality may have occurred [1]. Put simply, the health-care challenges that we face cannot be surmounted by physicians alone. The complexity of health and health care in the 21st century demands increasingly coor­ dinated and agile collaboration among varied health- care professionals [2]. Medical educators must ensure future physicians are trained experts at engaging with non-physician health-care professionals in a respectful, productive, and collaborative manner in order to manage day-to-day health care as well as to solve the challenges ahead. KEYWORDS Accreditation; clinical competence; communication skills; competency-based education; curriculum development; health-care professionals; interdisciplinary; medical education Department of ObGyn, 800 Washington Street, Box 022, Boston, MA This article has been corrected with minor changes. These changes do not impact the academic content of the article. © 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Enhancing interprofessional collaboration and interprofessional education in women’s health Laura Baecher-Linda, Angela C. Flemingb, Rashmi Bhargavac, Susan M. Coxd, Elise N. Everette, David A. Forsteinf, Shireen Madani Simsg, Helen K. Morganh, Christopher M. Moroskyi, Celeste S. Roycej,k, Tammy S. Sonnl, Jill M. Suttonm and Scott C. Grazianon aDepartment of Obstetrics and Gynecology, Tufts University School of Medicine, Boston, MA, USA; bDepartment of Obstetrics and Gynecology, Beaumont Hospital, Novi, MI, USA; cDepartment of Obstetrics and Gynecology, University of Saskatchewan College of Medicine, Regina, SK, Canada; dDepartment of Obstetrics and Gynecology, Dell Medical School, Austin, TX, USA; eDepartment of Obstetrics and Gynecology, Larner College of Medicine at the University of Vermont, Burlington, VT, Canada; fDepartment of Obstetrics and Gynecology, Rocky Vista College of Osteopathic Medicine, Parker, CO, USA; gDepartment of Obstetrics and Gynecology, University of Florida College of Medicine, Gainesville, FL, USA; hDepartment of Obstetrics and Gynecology, University of Michigan Medical School, Ann Arbor, MI, USA; iDepartment of Obstetrics and Gynecology, University of Connecticut School of Medicine, Farmington, CT, USA; jDepartment of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center, Boston, MA, USA; kDepartment of Obstetrics, Gynecology, and Reproductive Biology, Harvard Medical School, Boston, MA, USA; lDepartment of Obstetrics and Gynecology, Washington University School of Medicine, St. Louis, MO, USA; mDepartment of Obstetrics and Gynecology, Brody School of Medicine at East Carolina University, Greenville, NC, USA; nDepartment of Obstetrics and Gynecology, Loyola University Medical Center, Hines, IL, USA ABSTRACT ARTICLE HISTORY Received 21 November 2021 Revised 5 May 2022 Accepted 26 July 2022 ARTICLE HISTORY Received 21 November 2021 Revised 5 May 2022 Accepted 26 July 2022 ABSTRACT This article is from the ‘To The Point’ series from the Association of Professors of Gynecology and Obstetrics Undergraduate Medical Education Committee. The purpose of this review is to provide an understanding of the differing yet complementary nature of interprofessional collaboration and interprofessional education as well as their importance to the specialty of Obstetrics and Gynecology. We provide a historical perspective of how interprofessional collaboration and interprofessional education have become key aspects of clinical and educa­ tional programs, enhancing both patient care and learner development. Opportunities to incorporate interprofessional education within women’s health educational programs across organizations are suggested. This is a resource for medical educators, learners, and practicing clinicians from any field of medicine or any health-care profession. KEYWORDS Accreditation; clinical competence; communication skills; competency-based education; curriculum development; health-care professionals; interdisciplinary; medical education Health-care teams have faced enormous challenges over the past several years, navigating a pandemic while also continuing to provide routine and preven­ tive health care, striving to mitigate disparities and racism, training our next generation of health-care providers, and combatting chronic public health crises such as violence and mental health in the USA. Each of these tasks are Herculean and require collective ingenuity and efforts across specialties. Taken alone, the COVID-19 pandemic has required collaboration between physicians, public health experts, respiratory technicians, nurses, environmen­ tal services, among others. Without the complemen­ tary yet distinct skills and expertise of each of these disciplines, more profound morbidity and mortality may have occurred [1]. Put simply, the health-care challenges that we face cannot be surmounted by physicians alone. The complexity of health and health care in the 21st century demands increasingly coor­ dinated and agile collaboration among varied health- Health-care teams have faced enormous challenges over the past several years, navigating a pandemic while also continuing to provide routine and preven­ tive health care, striving to mitigate disparities and racism, training our next generation of health-care providers, and combatting chronic public health crises such as violence and mental health in the USA. Each of these tasks are Herculean and require collective ingenuity and efforts across specialties. Taken alone, the COVID-19 pandemic has required collaboration between physicians, public health experts, respiratory technicians, nurses, environmen­ tal services, among others. [email protected] ARTICLE HISTORY Received 21 November 2021 Revised 5 May 2022 Accepted 26 July 2022 KEYWORDS Accreditation; clinical competence; communication skills; competency-based education; curriculum development; health-care professionals; interdisciplinary; medical education Enhancing interprofessional collaboration and interprofessional education in women’s health Laura Baecher-Lind, Angela C. Fleming, Rashmi Bhargava, Susan M. Cox, Elise N. Everett, David A. Forstein, Shireen Madani Sims, Helen K. Morgan, Christopher M. Morosky, Celeste S. Royce, Tammy S. Sonn, Jill M. Sutton & Scott C. Graziano To cite this article: Laura Baecher-Lind, Angela C. Fleming, Rashmi Bhargava, Susan M. Cox, Elise N. Everett, David A. Forstein, Shireen Madani Sims, Helen K. Morgan, Christopher M. Morosky, Celeste S. Royce, Tammy S. Sonn, Jill M. Sutton & Scott C. Graziano (2022) Enhancing interprofessional collaboration and interprofessional education in women’s health, Medical Education Online, 27:1, 2107419, DOI: 10.1080/10872981.2022.2107419 To link to this article: https://doi.org/10.1080/10872981.2022.2107419 © 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. Published online: 04 Aug 2022. Submit your article to this journal Article views: 620 View related articles View Crossmark data © 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. Published online: 04 Aug 2022. Submit your article to this journal Article views: 620 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=zmeo20 2 L. BAECHER-LIND ET AL. 2 L. BAECHER-LIND ET AL. encourages IPE should produce health-care profes­ sionals that are more agile in involving and respect­ ing one another’s contributions. Conversely, an educational environment where learners remain siloed, surrounded only by students from their own discipline and taught exclusively by faculty from their own domain, might be less likely to understand, involve, and respect the contributions of other pro­ fessionals involved in patient care. Figure 1. Relationship between interprofessional collabora­ tion and interprofessional education with examples. p Both IPC and IPE have been increasing in promi­ nence in health-care and medical education over the past several decades. Prior to 1999, IPE and IPC were considered fringe concepts in higher medical educa­ tion and health-care settings. In 1999, To Err is Human was published by the then Institute of Medicine which revealed the frequency of medical errors and recom­ mended improved collaboration between health-care professionals – called ‘caregiving microsystems’ in the report – to improve patient safety [6]. These caregiving microsystems are what we would call IPC today. With the increasing emphasis on IPC in health-care work­ places, educators developed the concept of IPE to train students to be able to interact in an increasingly colla­ borative health-care environment. In 2010, the World Health Organization published a framework for IPE for global medical education [7]. This was followed in the US by the release of recommendations from a collaboration of the six main educational organiza­ tions in health-care education – specifically, American Association of Colleges of Nursing, American Association of Colleges of Osteopathic Medicine, American Association of Colleges of Pharmacy, American Dental Education Association, Association of American Medical Colleges, and Association of Schools of Public Health – which hallmarked the wide­ spread introduction of IPE into medical education curricula at medical schools nationwide [8]. Figure 1. Relationship between interprofessional collabora­ tion and interprofessional education with examples. professional societies and includes four core competen­ cies related to the value of collaboration, roles and responsibilities, communication, and teamwork [4]. Medical students often work alongside physician assis­ tants, advanced practice nurses, and other students dur­ ing their clinical education. Learners from different schools typically have distinct learning objectives and educational goals reflective of their future discipline and learn not only the patient care-related material alongside one another but also how the roles and perspectives of the other discipline complement patient care to achieve the core interprofessional education competencies. Collaboration versus Education The terms Interprofessional Collaboration (IPC) and Interprofessional Education (IPE) are often used inter­ changeably; however, they refer to distinct yet overlap­ ping entities (Figure 1) [3]. Interprofessional education (IPE) refers to the co-education of learners from different disciplines, such as pharmacy and medicine, learning alongside one another as well as about each other’s pro­ fessions in a classroom or workplace learning environ­ ment. A common set of learning objectives for all students engaging in interprofessional education – regardless of discipline – has been endorsed by over 60 CONTACT Laura Baecher-Lind 02111, USA 2 L. BAECHER-LIND ET AL. In comparison, interprofessional collaboration (IPC) refers more specifically to the workplace rather than the educational environment and focuses on health-care pro­ fessionals from different disciplines working coopera­ tively to optimize patient care. IPC occurs when clinicians from different backgrounds – such as physical therapy, physicians, and audiologists – work together using their distinct yet complementary skills and exper­ tise to provide the highest quality of care to patients, families, and communities [5]. Substantial overlap between IPE and IPC exists given the inclusion of stu­ dents into multidisciplinary teams on a regular basis in the clinical setting. As such, learners can achieve similar educational goals through IPC experiences. The commu­ nication, perspective enhancing, and collaborative skills that are the objectives of IPE may be more readily achieved through IPC opportunities already existent in health-care settings [4]. For example, arranging for case- based or patient simulation learning experiences that are mutually beneficial for students from differing voca­ tions – such as a birthing simulation experience for medical and midwifery students – are often time- consuming and challenging to implement [5]. Instead, having a medical student spend a Labor and Delivery shift with a midwife rather than the obstetrics team, and having a midwifery student work with the obstetrics team can be much simpler to implement and potentially more effective to achieving educational goals. As of 2013, the Liaison Committee on Medical Education (LCME) required medical schools to demonstrate that students are being prepared to function collaboratively on health-care teams that include other professionals. This requirement remains a part of LCME accreditation currently represented as Standard 6.7, which requires that medical students ‘have opportunities to learn in academic environments that permit interaction with students enrolled in other health professions, graduate, and professional degree programs, and in clinical environments that provide opportu­ nities for interaction with physicians in graduate medical education programs and in continuing medical education programs’, and Standard 7.9, which requires that medical schools ensure that students are prepared to ‘function collaboratively on health-care teams that include health profes­ sionals from other disciplines as they provide coordinated services to patients’ [9]. An emphasis on IPE should lead to improved IPC. That is, an educational environment that respects and MEDICAL EDUCATION ONLINE 3 3 IPE and IPC in obstetrics and Gynecology all residency training programs [23,24]. IPE in ObGyn clinical learning environments As pioneers in IPC, Obstetrician-Gynecologists are well positioned to be leaders in IPE. Opportunities for direct education of medical students by non-physician health- care professionals, and for collaborative education for learners from various disciplines, may be found through­ out the Obstetrics and Gynecology Clerkship. Table 1 provides examples of interprofessional collaboration in Obstetrics and Gynecology subspecialties and potential opportunities for associated IPE that can be incorporated into clinical education curricula. Gynecologic subspecialties also commonly rely on non-physician health-care professionals to deliver care and improve patient outcomes and safety. In Gynecologic Oncology, nurse navigators improve coor­ dination of care and are associated with increased patient satisfaction and reduced anxiety [15]. Reproductive Endocrinology regularly incorporates psychological assessment and support for patients undergoing evalua­ tion and treatment for infertility which is associated with reduced anxiety and improved success with fertility treat­ ments [16,17]. Urogynecologists routinely incorporate pelvic floor physical therapy into treatment planning for incontinence and prolapse which improves successful treatment and patient satisfaction[18]. Medical students often share clinical learning envir­ onments with students from physician assistant, nursing, nurse practitioner, and genetic counseling programs, among others. To encourage IPE, educators may con­ sider placing students with complementary yet differing educators rather than adhering to siloed educational spheres [5]. For example, a genetic counseling student may benefit from spending time with sonographers; a sonography student may benefit from spending time in a Maternal-Fetal-Medicine clinic session; and a medical student may benefit from spending time with a genetic counselor. Such an approach is an example of IPC itself, as this would require coordination across and among various educational leaders. This approach may not only complement educational goals but may also increase collaboration that translates across other domains such as patient care, clinical research, or quality and safety initiatives. Medical education leaders should create opportunities for students from other disciplines to learn from physicians and concurrently develop relation­ ships to allow opportunities for medical students to learn from non-physician professionals in return. Other non-physician health-care professionals have substantial roles in improving outcomes for Obstetrics and Gynecology patients across subspecialties. Involving pharmacists in bedside rounds reduces the risk of med­ ication errors by two-thirds [19]. When physical thera­ pists regularly participate in discharge planning, readmission rates are decreased more than twofold [20]. Routinely involving genetic counselors across disciplines is associated with increased patient satisfaction, a greater sense of control, and positive health behaviors [21]. 2 L. BAECHER-LIND ET AL. Evidence indicates that IPE activities improve learners’ atti­ tudes towards other members of interdisciplinary teams and enhance communication and shared pro­ blem-solving among interdisciplinary team members [25]. Given the clear patient care and population health benefits that interdisciplinary teams offer, the evidence supporting the efficacy of IPE on achieving a physician workforce more proficient in interdisci­ plinary teamwork and problem-solving, and the chal­ lenges inherent in health care in the 21st century, educators should strive to incorporate IPE into med­ ical education at every opportunity. Obstetrics and Gynecology has long been a collaborative specialty. Even well into the 20th century, women sought care from traditional birth attendants for expertise in pregnancy and childbirth rather than from a physician. In the 1940s, nurse midwifery was promoted by public health nurses, social reformers, and obstetricians in order to reduce maternal morbidity and mortality [10]. Expertise in pregnancy and childbirth is now shared between obstetricians and other health-care providers including nurse midwives, family medicine physicians, women’s health nurse practitioners, physician assistants, and doulas. Nearly 13% of women in the USA choose a midwife rather than an obstetrician for their care [11]. Women receiving care with midwives experience fewer interventions in labor and have reduced risks of cesarean section or operative vaginal delivery [12]. It is recognized that increasing access to and learning best practices from nurse midwifery may be a primary strategy to continuing to reduce maternal morbidity and mortality in the USA [11–13]. Team-training, a form of interprofessional edu­ cation, has been shown to reduce rates of adverse obste­ tric events including return to the operating room and birth injury [14]. IPE in ObGyn clinical learning environments Patients receiving pastoral care report greater sense of peace and reduced anxiety about their prognoses [22]. Similarly, both IPC and IPE may be enhanced when students participate in interdisciplinary activities already embedded into clinical settings. For instance, interdisciplinary team rounding on antepartum patients may involve a Maternal Fetal Medicine physi­ cian, a pharmacist, and a nurse. Involving a student from each discipline to accompany each clinician – a medical student, pharmacy student, and nursing Interprofessional education is paramount to ensur­ ing a future workforce that is receptive to the knowl­ edge, perspectives, and expertise of other disciplines. Working effectively in interdisciplinary teams has been identified as the single-most important skill for incoming residents and is a required component of 4 L. BAECHER-LIND ET AL. L. BAECHER-LIND ET AL. Table 1. Opportunities for interprofessional collaboration and education within Obstetrics and Gynecology learning environments. Clinical learning environment Interprofessional collaboration (IPC) Opportunity for interprofessional education (IPE) Obstetrics – outpatient Social workers (SW) Spend ½ day with perinatal SW Ultrasound technologists Spend ½ day with OB US technician Genetic counselors Spend ½ day with GC; create family tree Lactation counselors (LC) Participate in education session with LC Nutrition Sit in on nutrition counseling session Pelvic Floor Physical Therapist (PT) Spend ½ day with Pelvic Floor PT Obstetrics – inpatient Certified nurse midwifes (CNM) Participate in labor of 1+ CNM patients Labor and delivery nurses Shadow L&D nurse for one shift Physician assistants Postpartum round with PA Gynecology (all) Pharmacists Invite pharmacists to round daily Interpreter services Shadow interpreter services for ½ day Gynecologic Oncology Physical therapists Shadow PT on 1+ shared patients Occupational therapists (OT) Shadow OT on 1+ shared patients Respiratory therapists (RT) Shadow RT on 1+ shared patients Social workers Spend ½ day with social worker Nutrition Sit in on nutrition counseling session Pastoral care Attend family meeting; shadow on rounds Genetic counselors (GC) Spend ½ day with GC; create family tree Chemotherapy nurses Shadow chemo nurse for one shift Palliative care Participate in palliative care consult Urogynecology Pelvic floor physical therapists Designated session with pelvic floor PT Nurse practitioners (NP) Designated session with NP Reproductive Endocrinology Psychologists Sit in on infertility-related counseling Embryologists Shadow during infertility care student, in this example – can complement both disci­ pline-specific learning objectives alongside shared interprofessionalism competencies[4]. Funding The author(s) reported there is no funding associated with the work featured in this article. Discussion Working effectively, productively, and respectfully with diverse team members is crucial for health-care delivery across settings and in both routine as well as pandemic- related care. Medical educators and leaders must ensure our learners – the workforce of our future – are capable and comfortable collaborating with other health-care professionals in order to prepare our students for success, both personally and for the tremendous challenges that lie ahead in health care and population health. Educational leaders should be familiar with where these opportunities already exist in their organization, where they might be developed, and strive to incorporate learners in these activities wherever possible. The oppor­ tunities listed in Table 1 are not exhaustive; organizations may have additional or unique IPE opportunities depending upon the patient populations and services offered within that organization. Similarly, every situa­ tion of IPC does not need to be utilized as an educational opportunity. Educational leaders hoping to increase IPE within their curriculum should identify opportunities that are most easily and reliably integrated into their program in order to be self-sustaining and available to all learners. To this end, securing buy-in with key stake­ holders from the collaborating profession is crucial. For example, if an educational leader hopes to incorporate a medical student working with a Labor and Delivery nurse for a shift, obtaining support from perinatal nur­ sing and nursing leadership will be paramount. Educational leaders should monitor the impact of the IPE experience regularly with both learners and colla­ borators, to ensure that the experience meets educational objectives and clinical goals on an ongoing basis. The impact of IPE experiences typically involves assessment Disclosure statement No potential conflict of interest was reported by the author(s). IPE in ObGyn clinical learning environments Opportunities for similar IPE exist in any situation where physicians regularly engage with non-physician health-care profes­ sionals, including subspecialty interdisciplinary rounds, consult particularly with non-physician professionals embedded in the workplace, in management of depres­ sion or other mental health concerns, or discharge planning rounds. of learners’ attitudes, perspectives, and communication skills. These may be self-assessed by the learner and/or by the non-physician health-care professionals or learners within a particular IPE experience [26,27]. References [1] Xyrichis A, Williams U. Strengthening health systems response to COVID-19: interprofessional science ris­ ing to the challenge. J Interprof Care. 2020;34 (5):577–579. [2] Rosen MA, Diaz Granados D, Dietz AS, et al. Teamwork in healthcare: key discoveries enabling safer, high-quality care. Am Psychol. 2018;73(4):433–450. MEDICAL EDUCATION ONLINE 5 5 [3] Bridges DR, Davidson RA, Odegard PS, et al. Interprofessional collaboration: three best practice models of interprofessional education. Med Ed Online. 2011;16(1):6035. [15] Gordils-Perez J, Schneider SM, Gabel M, et al. Oncology nurse navigation: development and imple­ mentation of a program at a comprehensive cancer center. Clin J Oncol Nurs. 2017;21(5):581–588. [16] Boivin J, Gameiro S. Evolution of psychology and counseling in infertility. Fertil Steril. 2015;104 (2):251–259. [4] Interprofessional Education Collaborative. Core com­ petencies for interprofessional collaborative practice: 2016 update. Washington DC: Interprofessional Education Collaborative; 2016. Accessed September 1, 2021. Available from: https://ipec.memberclicks. net/assets/2016-Update.pdf [17] Rooney KL, Domar AD. The impact of stress on fertility treatment. Curr Opin Obstet Gynecol. 2016;28(3):198–201. [18] Wallace SL, Miller LD, Mishra K. Pelvic floor physical therapy in the treatment of pelvic floor dysfunction in women. Curr Opin Obstet Gynecol. 2019;31(6):485–493. [5] Meijer LJ, deGroot E, Blaauw-Westerlaken M, et al. Intraprofessional collaboration and learning between specialists and general practitioners during postgrad­ uate training: a qualitative study. BMC Health Serv Res. 2016;16(a):376. p y [19] Leape LL, Cullen DJ, Clapp MD, et al. Pharmacist participation on physician rounds and adverse drug events in intensive care units. JAMA. 1999;282 (3):267–270. [6] Institute of Medicine. To Err is Human: building a Safer Health System. Washington DC: The National Academies Press; 2000. [20] Kadivar Z, English A, Marx BD. Understanding the relationship between physical therapist participation in interdisciplinary rounds and hospital readmission rates: preliminary study. Phys Ther. 2016;96 (11):1705–1713. [7] World Health Organization. The Health Professions Network Nursing and Midwifery Office within the Department of HumanResources for Health. Framework for action on interprofessional education and collabora­ tive practice. WHO; 2010. Accessed September 1, 2021. https://apps.who.int/iris/handle/10665/70185 [21] Madlensky L, Trepanier AM, Cragun D, et al. A rapid systematic review of outcome studies in genetic counseling. J Genet Couns. 2017;26(3):361–378. p pp [8] Interprofessional Education Collaborative. Core com­ petencies for interprofessional collaborative practice. Washington DC: Interprofessional Education Collaborative; 2011. Accessed September 1, 2021. Available from: https://ipec.memberclicks.net/assets/ 2011-Original.pdf [22] Lobb EA, Schmidt S, Jerzmanowska N, et al. Patient reported outcomes of pastoral care in a hospital setting. J Health Care Chaplain. 2019;25(4):131–146. References [23] Angus S, Vu R, Halvorsen AJ, et al. What skills should new internal medicine residents have in July? A national survey of internal medicine residency pro­ gram directors. Acad Med. 2014;89(3):432–435. g [9] LCME Functions and Structure of a Medical School. Published Mar 2021. g [24] Accreditation Council for Graduate Medical Education. Common program requirements (residency). Editorial revision effective Jul 2021. https://www.acgme.org/globalassets/PFAssets/ ProgramRequirements/CPRResidency2021.pdf [10] Dawley K. Origins of nurse-midwifery in the USA and its expansion in the 1940s. J Midwifery Womens Health. 2003;48(2):86–95. [11] Weisband YL, Gallo MF, Klebanoff M, et al. Who uses a midwife for prenatal care and for birth in the USA? A secondary analysis of Listening to Mothers III. Womens Health Issues. 2018;29(1):89–96. [25] Dyess AL, Brown JS, Brown ND, et al. Impact of interprofessional education on students of the health professions: a systematic review. J Educ Eval Health Prof. 2019;16:33. [12] Souter V, Nethery E, Kopas ML, et al. Comparison of midwifery and obstetric care in low-risk hospital births. Obstet Gynecol. 2019;134(5):1056–1065. [26] Brock D, Abu-Rish E, Chiu C, et al. Interprofessional education in team communication: working together to improve patient safety. BMJ Qual Saf. 2013;22 (5):414–423. y [13] Styles C, Kearney L, George K. Implementation and upscaling of midwifery continuity of care: the experi­ ence of midwifes and obstetricians. Women Birth. 2020;33(4):343–351. [27] Weinstein AR, Dolce MC, Koster M, et al. Integration of systematic clinical interprofessional training in a student-faculty collaborative primary care practice. J Interprof Care. 2018;32(1):104–107. [14] Wagner B, Meirowitz N, Shah J, et al. Comprehensive perinatal safety initiative to reduce adverse obstetric events. J Healthc Qual. 2012;34(1):6–15.

https://openalex.org/W4242860153

https://www.nature.com/articles/s41467-020-17112-9.pdf

English

Identifying Domains of Applicability of Machine Learning Models for Materials Science

cc-by

ARTICLE ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 R aiming to minimize the expected error (also called prediction risk) eðf Þ ¼ Z X ´ R lðf ðxÞ; yÞdPðx; yÞ ð2Þ ð2Þ measured by some non-negative loss function l that quantifies the cost incurred by predicting the actual property value y with f(x). Examples for loss functions are the squared error (lðy0; yÞ ¼ ðy0  yÞ2), the absolute error (lðy0; yÞ ¼ jy0  yj), and, for non-zero properties, the relative error (lðy0; yÞ ¼ jy0  yj=jyj). Here P denotes some fixed probability distribution that captures how candidate materials are assumed to be sampled from the materials class (this concept, while commonly assumed in ML, is an unnecessary restriction for high-throughput screening as we discuss in more detail below). Since the true prediction risk is impossible to compute directly without perfect knowledge of the investigated materials class, models are evaluated by the test error (or empirical risk) p y p y y Here we present an informed diagnostic tool based on subgroup discovery (SGD)22–24 that detects domains of applicability (DA) of ML models within a materials class. These domains are given as a combination of simple conditions on the unit-cell structure (e.g., on the lattice vectors, lattice angles, and bond distances) under which the model error is substantially lower than its global average in the complete materials class. Thus, in contrast to methods that provide uncertainty estimates for individual data points (such as probabilistic models or ensemble methods), the presented approach provides logical descriptions of contiguous regions with an overall low estimated uncertainty. These descriptions allow (a) to understand and subsequently address systematic shortcomings of the investigated ML model and (b) to focus sampling of can- didate materials on regions of low expected model uncertainty. We demonstrate this procedure by analyzing the three state-of-the-art ML models for the above mentioned TCO challenge. Despite having a globally indistinguishable and unsatisfactory average error, the models have domains of applicability (DAs) with notably improved performance and distinctive features. That is, they all perform well for different characteristics of the unit cell. In terms of error improvement, the MBTR-based model stands out with a ca. twofold reduction in the average error and ca. 7.5-fold reduc- tion in the fraction of errors above the required accuracy to identify the ground state polymorph (i.e., from 12.8 to 1.7%). NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 G iven sufficient predictive accuracy, machine learning (ML) can accelerate the discovery of novel materials by allowing to rapidly screen compounds at orders of magnitude lower computational cost than first-principles electronic-structure approaches1–7. In practice, however, the accuracy of ML models is often insufficient to draw reliable conclusions about materials for specific applications7. Therefore, different ML representations for materials are actively developed to provide accurate predictions over diverse materials classes and properties8–20. A critical obstacle for this effort is that the complex choices involved in designing ML models are currently made based on the overly simplistic metric of the average model test error with respect to the entire materials class. This treatment of models as a black box that produces a single error statistic can render different models indistinguishable although their performance actually differs substantially across materials. Moreover, models may appear generally insufficient for certain screening tasks while they actually predict the target property accurately in specific subdomains. For example, for a large public ML challenge for predicting the for- mation energies of transparent conducting oxides (TCOs)21, three approaches have a nearly indistinguishable performance: the competition winning model adapted from natural language pro- cessing (n-gram method)21, smooth overlap of atomic positions (SOAP)13,14, and the many-body tensor representation (MBTR)12. Importantly, as shown below, they all appear unsatisfactory for screening applications as they fail to reliably identify the ground state polymorph structure for many of the examined systems. G according to some chosen representation. The more complex state-of-the-art representations evaluated in this work are defined further below. A first simple example is to use features ϕ1, …, ϕn of the isolated atoms that constitute the material (e.g., ϕi(Z) may be the “electronegativity of the species with atomic number Z” (see Supplementary Table 4) and then to lift these to repre- sentation coordinates xi for compounds ðZj; μjÞk j¼1 defined as xi ¼ X k j¼1 μjϕiðZjÞ ð1Þ ð1Þ where μj corresponds to the mixture coefficient for atomic number Zj. Moreover, let y be a numeric material property according to which screening should be performed (in this work, we focus on formation energy, which is relevant for performing a ground state search). A predictive ML model is then a function f : X ! NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Thus, we demonstrate that the MBTR-based model is in fact feasible for screening materials that lie within its DA while it is highly unre- liable outside of it. This illustrates how the proposed method can be used to guide the development of ML representations through the identification of their systematic strengths and weaknesses. We expect this form of analysis to advance ML methods for materials as well as ML methods for science more broadly. ^eðf Þ ¼ X m i¼1 eiðf Þ=m ð3Þ ð3Þ defined as the average of the individual errors (losses) ei(f) = l(f (xi), yi) on some test set of m reference data points ðxi; yiÞm i¼1. The samples in this test set are drawn independently and identically distributed according to P and are also independent of the model— which means in practice that it is a random subset of all available reference data that has been withheld from the ML algorithm. In order to reduce the variance of this estimate, a common strategy is cross-validation, where this process is repeated multiple times based on partitioning the data into a number of non-overlapping “folds” and then to use each of these folds as test sets and the remaining data as a training set to fit the model. This test error properly estimates the model performance globally over the whole representation space X (weighted by the distribution P used to generate the test points). This is an appropriate evaluation metric for selecting a model that is required to work well on average for arbitrary new input materials that are sampled according to the same distribution P. This is, however, not the condition of high-throughput screening. Here, rather than being presented with random inputs, we can decide which candidate materials to screen next. This observation leads to the central idea enabled by the DA analysis proposed in this work: if the employed model is particularly applicable in a specific subdomain of the materials class, and if that subdomain has a simple and interpretable shape that permits to generate new materials from it, then we can directly focus the screening there. Identifying domains of applicability of machine learning models for materials science Christopher Sutton 1,7✉, Mario Boley 2,7✉, Luca M. Ghiringhelli 1✉, Matthias Rupp 1,3,6, Jilles Vreeken 4 & Matthias Scheffler1,5 Although machine learning (ML) models promise to substantially accelerate the discovery of novel materials, their performance is often still insufficient to draw reliable conclusions. Improved ML models are therefore actively researched, but their design is currently guided mainly by monitoring the average model test error. This can render different models indis- tinguishable although their performance differs substantially across materials, or it can make a model appear generally insufficient while it actually works well in specific sub-domains. Here, we present a method, based on subgroup discovery, for detecting domains of applic- ability (DA) of models within a materials class. The utility of this approach is demonstrated by analyzing three state-of-the-art ML models for predicting the formation energy of transparent conducting oxides. We find that, despite having a mutually indistinguishable and unsatisfactory average error, the models have DAs with distinctive features and notably improved performance. 1 NOMAD Laboratory, Fritz Haber Institute of the Max Planck Society, Berlin, Germany. 2 Faculty of IT, Monash University, Clayton VIC 3800, Australia. 3 Citrine Informatics, Redwood City, CA 94063, USA. 4 CISPA Helmholtz Center for Information Security, Saarbrücken, Germany. 5 Physics Department, IRIS Adlershof Humboldt-Universität, Berlin, Germany. 6Present address: Department of Computer and Information Science, University of Konstanz, Konstanz, Germany. 7These authors contributed equally: Christopher Sutton, Mario Boley. ✉email: [email protected]; [email protected]; [email protected] 1 NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Boolean functions (σ: X →{true, false}) of the form σðxÞ  π1ðxÞ ^ π2ðxÞ ^ ¼ ^ πpðxÞ Boolean functions (σ: X →{true, false}) of the form: σðxÞ  π1ðxÞ ^ π2ðxÞ ^ ¼ ^ πpðxÞ Boolean functions (σ: X →{true, false}) of the form: σðxÞ  π1ðxÞ ^ π2ðxÞ ^ ¼ ^ πpðxÞ for high x2 values. Discovering applicable domains reveals how different models cope differently with this setting even if they have a comparable average error. To show this, let us examine the error distributions obtained from three different kernelized regression models of the form σðxÞ  π1ðxÞ ^ π2ðxÞ ^ ¼ ^ πpðxÞ σðxÞ  π1ðxÞ ^ π2ðxÞ ^ ¼ ^ πpðxÞ where “∧” denotes the “and” operation and each proposition πi is a simple inequality constraint on one of the coordinates, i.e., πi(x) ≡xj ≤v for some constant v. Thus, these selectors describe intersections of axis-parallel half-spaces resulting in simple convex regions ({x ∈X: σ(x) = true}) in X. This allows to systematically reason about the described subdomains (e.g., it is easy to determine their differences and overlap) and also to sample novel points from them. To specifically obtain regions where a given model has a decreased error, standard SGD algorithms25,26 can be configured to yield a selector with maximum impact on the model error. The impact is defined as the product of selector coverage, i.e., the probability of the event σ(x) = true, and the selector effect on the model error, i.e., the model error minus the model error given that the features satisfy the selector. f ðÞ ¼ X n i¼1 νikðxF i ; Þ with parameter vector ν that are fitted around a training, or fitting (F), set ðxF i ; yF i Þn i¼1 with three different choices for the kernel function k. We observe: ● When using the linear (lin) kernel (kðx; x0Þ ¼ hx; x0i), the resulting linear model is globally incapable to trace the variation of the third-order polynomial except for a small stripe on the x1-axis where it can be approximated well by a linear function. Consequently, there is a very high error globally that is substantially reduced in the DA described by σlin(x1, x2) ≡−0.3 ≤x1 ≤0.3. An illustrative example. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Before describing the details of DA identification and its integration into the ML process, let us illustrate the concept and its utility via a synthetic example (see Fig. 1). We consider a simple two-dimensional representation consisting of independent features x1 and x2 that are each dis- tributed according to a normal distribution with mean 0 and variance 2 (N(0, 2)) and a target property y that is a third-degree polynomial in x1 with an additive noise component that scales exponentially in x2: ● When using the Gaussian kernel kðx; x0Þ ¼ exp k x  x0k2= 2ϵ2), the resulting radial basis function (rbf) model is able to represent the target property well locally unless (a) the noise component is too large and (b) the variation of the target property is too high relative to the number of training points. The second restriction is because the rbfs have non-negligible values only within a small region around the training examples. Consequently, the discovered DA is not only restricted in x2-direction but also excludes high absolute x1- values: σrbf ≡−3.3 ≤x11 ≤3.1 ∧x2 ≤0.1. ● When using the Gaussian kernel kðx; x0Þ ¼ exp k x  x0k2= 2ϵ2), the resulting radial basis function (rbf) model is able to represent the target property well locally unless (a) the noise component is too large and (b) the variation of the target property is too high relative to the number of training points. The second restriction is because the rbfs have non-negligible values only within a small region around the training examples. Consequently, the discovered DA is not only restricted in x2-direction but also excludes high absolute x1- values: σrbf ≡−3.3 ≤x11 ≤3.1 ∧x2 ≤0.1. y  x3 1  x1 þ Nð0; expðx2=2ÞÞ: y  x3 1  x1 þ Nð0; expðx2=2ÞÞ: That is, the y values are almost determined by the third-degree polynomial for low x2 values but are almost completely random In contrast, when using the non-local third-degree poly- nomial (poly) kernel kðx; x0Þ ¼ ðhx; x0i þ 1Þ3, data sparsity a c b Fig. 1 Domains of applicability of three 2d-models of a noisy third-degree polynomial. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Three different models, linear (top), radial basis function (rbf, center), and polynomial (poly, bottom), are shown approximating the same distribution of two independent features x1 ~ N(0, 2) and x2 ~ N(0, 2), and the target property y  x3 1  x1 þ Nð0; expðx2=2ÞÞ, where N(μ, ϵ2) denotes a normal distribution with mean μ and standard deviation ϵ. Test points are plotted in 3d plots against the prediction surface of the models (color corresponds to absolute error) where the DA is highlighted in gray. The distributions of individual errors for the DA (gray) and globally (black) are shown in the 2d plots of each panel with the mean error (solid) and the 95th percentile (95 perc./dashed) marked by vertical lines. Note that the global error distribution of the linear model has a considerably long tail, which is capped in the image. a b a b c c Fig. 1 Domains of applicability of three 2d-models of a noisy third-degree polynomial. Three different models, linear (top), radial basis function (rbf, center), and polynomial (poly, bottom), are shown approximating the same distribution of two independent features x1 ~ N(0, 2) and x2 ~ N(0, 2), and the target property y  x3 1  x1 þ Nð0; expðx2=2ÞÞ, where N(μ, ϵ2) denotes a normal distribution with mean μ and standard deviation ϵ. Test points are plotted in 3d plots against the prediction surface of the models (color corresponds to absolute error) where the DA is highlighted in gray. The distributions of individual errors for the DA (gray) and globally (black) are shown in the 2d plots of each panel with the mean error (solid) and the 95th percentile (95 perc./dashed) marked by vertical lines. Note that the global error distribution of the linear model has a considerably long tail, which is capped in the image. Fig. 1 Domains of applicability of three 2d-models of a noisy third-degree polynomial. Three different models, linear (top), radial basis function (rbf, center), and polynomial (poly, bottom), are shown approximating the same distribution of two independent features x1 ~ N(0, 2) and x2 ~ N(0, 2), and the target property y  x3 1  x1 þ Nð0; expðx2=2ÞÞ, where N(μ, ϵ2) denotes a normal distribution with mean μ and standard deviation ϵ. Results Domain of applicability identification via subgroup discovery. To formally introduce the method for DA identification, we recall some notions of ML for materials. In order to apply smooth function approximation techniques like ridge regression, the materials of interest are represented as vectors in a vector space X Such simply described DA can be identified by the descriptive data mining technique of subgroup discovery (SGD)22–24. This technique finds selectors in the form of logical conjunctions, i.e., 2 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 2 Workflow for domain of applicability (DA) identification and Fig. 2 Workflow for domain of applicability (DA) identification and validation for an ML model. The DA is described by a selector (σf) that is comprised of logical conjunctions of a representation space (here symbolized by a single dimension x for simplicity but may be multidimensional). The selector is identified by applying subgroup discovery (SGD) to the individual ML-model errors for subset of test set (DA identification set). An unbiased estimate of the model performance within the DA is obtained on the remaining samples of the test set that were left out of the DA identification (DA validation set). DA representation and objective function. In the illustrative example above, all evaluated models share the same simple representation. However, in practice different models are typically fitted with different and more complicated representations. For instance, for the study on formation energies of transparent oxides below, we compare models based on the n-gram, SOAP, and MBTR representations. These representations use different descriptions of the local atomic geometry, leading to high- dimensional non-linear transforms of the material configurations (e.g., 1400, 681, and 472 dimensions for MBTR, SOAP, and n- gram representations). A DA described directly in terms of these complex representations cannot easily be mapped back to intui- tive conditions on the unit cell of a given material. This not only hinders interpreting the DA but also to construct novel materials from it. Finally, using different representations to describe DAs of different models makes it impossible to assess their overlap and differences. Therefore, we define a single representation com- prised of features that are specifically intended for the description of insightful subdomains. A first natural group of features per- tains directly to the shape of the unit cell such as the sorted lattice vectors and angles, the number of atoms in the unit cell, and the unit-cell volume. In addition, when we are interested in a fixed compositional space, we can add features describing the com- position (e.g., “percentage of Al cations”) as well as structural features describing the bonding environments (e.g., “average nearest-neighbor distance between Al and O”, which we define using the effective coordination number27). The description of DAs in these simple terms of the unit-cell structure and com- position allows to easily interpret, compare, and sample from them (e.g., for focused screening). NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 However, we note that the representation space inputted into subgroup discovery can be adapted for various purposes depending on the focus of the investigation. See Table 1 for a summary of all features used. DA validation set for obtaining an unbiased performance estimate of the identified DA (see Fig. 2 for an illustration of the overall workflow). Technically, the data points withheld in the DA validation set mimic novel independent sample points that can be used to evaluate both the coverage of the DA, as well as, the reduction in model error. As an extension of this, one can also repeat the DA optimization/validation on several splits (cross- validation) to reduce the variance of the coverage and model error estimates and, moreover, to assess the stability of the DA selector elements. DA validation set for obtaining an unbiased performance estimate of the identified DA (see Fig. 2 for an illustration of the overall workflow). Technically, the data points withheld in the DA validation set mimic novel independent sample points that can be used to evaluate both the coverage of the DA, as well as, the reduction in model error. As an extension of this, one can also repeat the DA optimization/validation on several splits (cross- validation) to reduce the variance of the coverage and model error estimates and, moreover, to assess the stability of the DA selector elements. For ease of notation we assume the DA identification set consists of the first k points of the test set. We end up with the following objective function for the SGD algorithm: impactðσÞ ¼ s k   |{z} coverage 1 k X k i¼1 liðf Þ  1 s X i2IðσÞ liðf Þ 0 @ 1 A |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} effect on test error ð4Þ ð4Þ {z effect on test error where s denotes the number of points in the DA identification set selected by σ and I(σ) = {i: 1 ≤i ≤k, σ(xi) = true} denotes the set of selected indices itself. Here, we focus on DA identification based on the relative error lðy0; yÞ ¼ jy0  yj=jyj, as it is less correlated with the target values than the absolute error. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Test points are plotted in 3d plots against the prediction surface of the models (color corresponds to absolute error) where the DA is highlighted in gray. The distributions of individual errors for the DA (gray) and globally (black) are shown in the 2d plots of each panel with the mean error (solid) and the 95th percentile (95 perc./dashed) marked by vertical lines. Note that the global error distribution of the linear model has a considerably long tail, which is capped in the image. 3 3 ARTICLE ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Table 1 Features used for discovery of domain of applicability (DA) selectors. Type Label Definition Unit a, b, c Lattice-vector lengths sorted from largest (a) to smallest (c) Å α Angle between b and c ∘ Unit cell β Angle between a and c ∘ γ Angle between a and b ∘ V/N Volume of unit cell divided by number of atoms Å3 N Number of atoms – Composition %Al, %Ga, %In Number of cations divided by total number of cations % Structural R{Al,Ga,In}-{Al,Ga,In,O} Average nearest-neighbor distance between Al, Ga, In, and O Å able 1 Features used for discovery of domain of applicability (DA) selectors. does not prevent an accurate modeling of the target property along the x1-axis. However, this non-locality is counter- productive along the x2-axis where overfitting of the noise component has a global influence that results in higher prediction errors for the almost deterministic data points with low x2-values. This is reflected in the identified DA σpoly(x1, x2) ≡−3.5 ≤x2 ≤0.1, which contains no restriction in x1-direction, but excludes both high and low x2-values. This highlights an important structural difference between the rbf and the polynomial model that is not reflected in their similar average errors. Fig. 2 Workflow for domain of applicability (DA) identification and validation for an ML model. The DA is described by a selector (σf) that is comprised of logical conjunctions of a representation space (here symbolized by a single dimension x for simplicity but may be multidimensional). The selector is identified by applying subgroup discovery (SGD) to the individual ML-model errors for subset of test set (DA identification set). An unbiased estimate of the model performance within the DA is obtained on the remaining samples of the test set that were left out of the DA identification (DA validation set). Fig. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 In addition, promoting large, i.e., general, DAs through the coverage term is important as those have a higher chance to (a) contain data points of interest and (b) to have an accurate effect estimate, i.e., the empirical error reduction measured by the effect term is likely to generalize to other points in the DA that are not contained in the DA identification set. Thus, the coverage term has a similar role as a regularization term in common objective functions for model fitting. With the above objective function, we reduce the bi- criterial coverage/effect optimization problem to a uni-criterial impact optimization problem where both individual criteria are equally weighted and non-compensatory, i.e., due to the multi- plicative combination, very low values of one criterion cannot be compensated by very high values in the other. The relative weight of both criteria can be re-calibrated by introducing a simple exponential weight parameter (see the Supplementary Methods section on Coverage/Effect Trade-off for a detailed discussion). MBTR, SOAP, and n-gram all display a similar test error (using the absolute error as the loss function l (see Eq. (3)); the resulting quantity we refer to as the mean absolute error, MAE) of 14.2, 14.1, and 14.7 meV/cation, respectively. This confirms previously reported virtually indistinguishable accuracies for MBTR and SOAP in the prediction of formation energies of alloys40. However, using the proposed method, key differences can be observed in the MAEs of their respective DAs (see Table 2 and Fig. 3 for a summary of all model performances). More specifically, the ML models built from MBTR, SOAP, and n- gram have an MAE (standard deviation) over the relevant DA validation sets of 7.6 (±1.5), 11.7 (±1.8), 10.2 (±0.9) meV/cation, respectively. All identified DAs for the models utilizing MBTR, SOAP, and n-gram have a large coverage (i.e., percent of samples within the DA) with an average (standard deviation) subpopula- tion contained within the DA validation set of 44% (±6%), 78% (±3%), and 52% (±5%), respectively. Optimizing the impact function over all conjunctive selectors that can be formed from a given set of base propositions is an NP-hard problem. This implies that there is no solver for it with worst-case polynomial time complexity (unless P = NP). NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Thus, this choice promotes DAs that contain a representative distribu- tion of target values and, by extension, more distinct and thus more characteristic DAs for the different models (see Supple- mentary Note 2 for a discussion of the DAs resulting from using the absolute error). g y The DA optimization and validation can be performed as a by- product from the labels and ML predictions of the test set. However, just as for the ML-model fitting itself, we can only estimate these quantities based on empirical data. For that purpose, it is sensible to also split the test data into two parts: a DA identification set for optimizing the empirical impact and a The effect term of the objective function ensures that the model is estimated to be more accurate in the described region than in NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications 4 4 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 used reliably for high-throughput screening. Specifically, we focus on the state-of-the-art representations of MBTR, SOAP, and the n-gram representation (all described in the Methods section). As an additional benchmark, we also perform DA identification for a simple representation containing just atomic properties averaged by the compositions (this corresponds to the simplistic choice of a representation given in Eq. (1); see Supplementary Table 4 for a list of atomic properties used in this representation). Since this representation is oblivious to configurational disorder (i.e., many distinct structures that are possible at a given composition), it is expected to perform poorly across all space groups and concentrations. Formally, there is no unique y-value associated with each x but rather a distribution P(y∣x). Thus, even the optimal prediction at each composition of the test set (the median energy) to predict the test set energies results in a mean absolute error of 32.6 meV/cation, which is the highest accuracy that can be obtained using just composition-based properties. Therefore, it is a candidate for a representation that does not have any useful DA when compared with its full domain. the global representation space. Thus, selectors with a large effect value describe domains of increased applicability as desired (see also Supplementary Note 4). NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 However, there is a practically efficient branch-and-bound algorithm that turns out to be very fast in practice if the dimensionality of the DA representation is not too high—in particular, substantially faster than the model training process (see Methods and Supplementary Methods). Domains of applicability for TCO models. Equipped with the DA concept, we can now examine the ML models for the pre- diction of stable alloys with potential application as transparent conducting oxides (TCOs). Materials that are both transparent to visible light and electrically conductive are important for a variety of technological devices such as photovoltaic cells, light-emitting diodes for flat-panel displays, transistors, sensors, touch screens, and lasers28–38. However, only a small number of TCOs have been realized because typically the properties that maximize transparency are detrimental to conductivity and vice versa. Because of their promise for technologically relevant applications, a public data-analytics competition was organized by the Novel Materials Discovery Center of Excellence (NOMAD39) and hos- ted by the on-line platform Kaggle using a dataset of 3000 ðAlxGayInzÞ2O3 sesquioxides, spanning six different space groups. The target property in this examination is the formation energy, which is a measure of the energetic stability of the specific elements in a local environment that is defined by the specific lattice structure. In contrast, the atomic model is not only the worst model globally with a test error of 65.5 meV/cation, but, as anticipated, the DA error is virtually indistinguishable from the global model error (MAE = 60.2 meV/cation). This model performs worse than the MAE = 32.6 meV/cation that can be obtained by using the median energy at each composition of the test set to predict the test set energies. Therefore, this result illustrates the case of a weak representation for which no DA with substantial error reduction can be identified. Although the reduction of the mean error for the three state- of-the-art representations is notable, the difference between the whole materials space and the DAs is even more pronounced when comparing the tails of the error distributions using the 95th percentile. For the global models, the average 95th percentile across all relevant splits is reduced by a factor of 2.9, 1.4, and 1.6 for the DA compared with the global error for MBTR, SOAP, and n-gram. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 To put these error values into context, we consider the reference value of 24.9 meV/cation corresponding to half of the mean energy difference between the minimum energy and the second-to-minimum energy polymorph for all concentrations. The fraction of data points with these errors from the MBTR model above this reference value is reduced by a factor of 7.5 from 12.8% in the entire test set to 1.7% (averaged over each relevant split) within the DA. A smaller reduction in the fraction of errors is observed for the SOAP model (13.3 to 9.0%) and n- gram model (16.2 to 10.8%). For the MBTR model, the 95th percentile of the DA errors (18.8 meV/cation) lies below the reference value. We close this results section by an investigation of the effect of the individual DA selector elements of the SOAP-based model (details for MBTR and n-gram based models are provided in Supplementary Figs. 1 and 2). The inclusion of the attributes γ < 90.35∘and β ≥88.68∘excludes 18.3% and 1.8% samples that have irregular unit cells based on the relatively large γ and small β values compared with the rest of the data points (see Fig. 4 for the distribution of the selected a/c, γ, and β values for the SOAP- based model). The inclusion of the term a/c ≤3.87, which describes 86.2% of the test set, is attributed to the fact that SOAP employs a real-space radial cut-off value rcut = 10 Å in construct- ing the local atomic density for all samples (see above for a description of this representation). The algorithm threshold choice of a/c ≤3.87 separates two modes of a relatively dense region of points (see Fig. 4 left panel); however, for structures with asymmetric unit cell, the spherical radius could lead to inaccurate depiction of the local atomic environment, therefore, we repeat the procedure for two additional rcut values of 20 and 30 Å. Compared with the selector identified for rcut = 10 Å, a largely consistent selector is observed when the cut-off value is changed to a value of rcut = 20 Å: Since the restriction of features values in the DAs generally affects the distribution of target values, the observed MAE improvements might simply be a function of reduced variation in the property values. This would mean that the DAs are not actually characteristic for the models as those reduced variations would be independent of them. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Fig. 3 DA versus global ML-model performance. Comparison of absolute error distributions for entire test set (global/gray) and for DA validation sets (DA/blue) using violin plots that extend to the 98th percentile. Boxplots inside violin plots indicate 25th, 50th, and 75th percentiles (boxes) of the absolute errors, as well as 5th and 95th percentiles (whiskers). Horizontal line indicates reference error level of half of the mean energy difference between the minimum energy and the second-to- minimum energy polymorph (mean over all considered concentrations). The selector of the n-gram model includes both features describing the unit-cell shape [medium lattice vector (b) and angle (γ)] and structural motifs [interatomic bond distances between Al–O (RAl–O) and Ga–O (RGa–O) within the first coordination shell]: σn-gram  b ≥5:59 Å ^ γ < 90:35^ RAlO ≤2:06 Å ^ RGaO ≤2:07 Å It is worth noting that applying these DA selectors to the training set results in a similar reduction in error between the global and local populations and sample coverages (i.e., local population size) to what was observed for the test set: The training MAEs are reduced by factors of 1.87, 1.18, and 1.43 and the training DA coverages are 44%, 76%, and 54% for MBTR, SOAP, and n-gram models, respectively. Fig. 3 DA versus global ML-model performance. Comparison of absolute error distributions for entire test set (global/gray) and for DA validation sets (DA/blue) using violin plots that extend to the 98th percentile. Boxplots inside violin plots indicate 25th, 50th, and 75th percentiles (boxes) of the absolute errors, as well as 5th and 95th percentiles (whiskers). Horizontal line indicates reference error level of half of the mean energy difference between the minimum energy and the second-to- minimum energy polymorph (mean over all considered concentrations). The qualitative differences observed in the DA selectors for these three models can be quantified by examining the over- lapping samples in the DAs using the Jaccard similarity, which is the ratio of the number of overlapping samples over the total number of samples in both DAs. We find Jaccard similarities of 0.61 for n-gram vs. SOAP, 0.66 for n-gram vs. MBTR, 0.57 for SOAP vs. MBTR (computed over the whole test set). In other words, the discovered DA selectors are not only syntactically different, but, despite some overlap, they do indeed describe substantially different sub-populations of the investigated materials class. σSOAP;rcut¼30Å  c ≥4:05 Å ^ γ ≤90:35 The absence of the a/c term for the SOAP representation utilizing a rcut = 30 Å indicates that the choice of a cut-off value less than the length of the unit cell directly impacts the model performance for the larger unit cells within this dataset, and thus, directly affects the selector chosen by SGD. σMBTR  N ≥50 atoms ^ γ ≤98:83 ^ RAlO ≤2:06 Å: For SOAP, selectors include features exclusively based on the unit-cell shape such as the ratio of the longest (a) and shortest (c) lattice vectors, and lattice-vector angles (β and γ): NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Our aim is to demonstrate the ability of the proposed DA analysis to (i) differentiate the performance of models based on different representations of the local atomic information of each structure and (ii) to identify subdomains in which they can be Table 2 Summary statistics for DAs for all investigated models. Global (test set) DA (validation set) DA (identification set) MAE 95AE R cov MAE 95AE R cov MAE 95AE R MBTR 14.2 54.1 0.83 44 (6) 7.6 (1.5) 18.8 (2.9) 0.88 (0.03) 44 (1) 7.6 (0.3) 20.7 (0.2) 0.89 (0.01) SOAP 14.1 51.0 0.84 78 (3) 11.7 (1.8) 36.6 (10.8) 0.85 (0.01) 76 (1) 11.9 (0.4) 37.8 (2.0) 0.85 (0.00) n-gram 14.7 51.1 0.83 52 (5) 10.2 (0.9) 32.6 (2.6) 0.86 (0.02) 54 (1) 10.3 (0.2) 35.5 (1.0) 0.86 (0.00) Atomic 65.5 154.5 0.24 85 (1) 60.2 (7.8) 141.6 (28.5) 0.25 (0.09) 85 (0) 63.3 (1.5) 153.9 (5.5) 0.25 (0.02) Coverage (cov), mean absolute error (MAE), 95th-percentile absolute error (95AE), and coefficient of determination based on absolute error (R) are all estimated via the mean value of the relevant DA validation sets and DA identification sets. Standard deviations are in parentheses. Global values are computed over whole test set. MAE and 95AE are in units of meV/cation, cov values are in percentages. Table 2 Summary statistics for DAs for all investigated models. NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 However, comparing the coefficient of determination (R) of the models globally and within their DA reveals that this is not the case. The R values are increased from 0.83 to 0.88 (MBTR), 0.84 to 0.85 (SOAP), and 0.83 to 0.86 (n-gram). This means, while there is a reduction in target dispersion, there is a disproportionate reduction in MAE in the model-specific predictions. Note that, matching our interest in absolute error performance, we consider here the R-value defined as one minus the sum of absolute errors over dispersion measured as the sum of absolute deviations from the median41. σSOAP;rcut¼20Å  a c ≤3:89 ^ γ ≤90:35 The identified selectors are mostly stable, i.e., appearing in four out of six splits for MBTR and SOAP and five of six for n-gram. Interestingly, the variables that comprise the selectors of the DA are qualitatively different for each of these models. Selectors for MBTR include the number of atoms (N), the angle between the two longest lattice vectors in the unit cell (γ), and the average bond distance between aluminum and oxygen within the first coordination shell (that is defined by the effective coordination number), RAl–O: However, increasing rcut to a value of 30 Å—which exceeds the largest unit-cell vector length (a) of ca. 24 Å in the structures contained within this dataset—results in the selector: σSOAP;rcut¼30Å  c ≥4:05 Å ^ γ ≤90:35 Methods MBTR Th MBTR. The MBTR representation space X can vary depending on the employed many-body order (e.g., interatomic distances for a two-body model, and/or angles for a two- and/or three-body model, and/or torsions for up to four-body models12). The results reported herein are calculated using a representation consisting of broadened histograms of element counts (one-body terms) and pairwise inverse interatomic distances (two-body terms), one for each unique pair of elements in the structure (i.e., for this dataset: Al–Al, Al–Ga, Al–In, Al–O, Ga–Ga, Ga–In, Ga–O, In–In, In–O, and O–O). These are generated according to: gið1=rÞ ¼ 1 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2πϵ2 atom p X j exp  ð1=r  1=jri  rjjÞ2 2ϵ2 atom ! wMBTRði; jÞ where a normal distribution function is centered at each inverse distance between pairs of atoms (1/ri,j) to ensure smoothness of the representation. The function wMBTRði; jÞ dampens contributions from atoms separated by large distances and is defined as wMBTRði; jÞ ¼ ð1=ri;jÞ2. The MBTR representation was generated using QMMLpack12,44. p g A further potential application of the proposed approach is to form ensembles of different local models, each of which is only contributing to the overall predictions for data points inside its DA. The general approach of forming ensembles of local models is an emergent idea to cope with heterogeneous materials classes42,43. So far these efforts have focused on a priori parti- tioning of the training set into sub-populations identified by automated clustering methods or prior knowledge, followed by fitting local models using the same regression technique for all subpopulations. In contrast, the DA approach can be used to incorporate the relative model advantages and disadvantages into the partitioning of the materials class. SOAP. The SOAP representation space is constructed by transforming pairwise atomic distances as overlapping densities of neighboring atoms and expanding the resulting density in terms of radial and spherical harmonics basis functions. The local density is modeled through a sum of gaussian distributions on each of the atomic neighbors j of atom i: ρiðrÞ ¼ X j exp  ðr  ri;jÞ2 2ϵ2 b ! Methods MBTR Th wSOAPðrÞ where j ranges over neighbors within a specific cut-off radius (rcut) relative to i, where the cut-off function wSOAP is defined as: where j ranges over neighbors within a specific cut-off radius (rcut) relative to i, where the cut-off function wSOAP is defined as: wSOAPðrÞ ¼ 1 ; for r ≤rcut  d cos π rrcutþd d   þ 1   =2 ; for rcut  d < r ≤rcut 0 ; otherwise 8 > < > : ; for r ≤rcut  d g In this context, it is an intuitive expectation that an improved model can be obtained by fitting only training data from within the discovered DA. However, this is not true in general: points outside of the DA, while having a higher error on average, can still contribute positively to the prediction inside the DA. For instance, refitting to a training set trimmed according to the DA selectors of the three model types investigated here leads to a change in test MAE of −1.5 (MBTR), −1.0 (SOAP), and +0.1 (n- gram) meV/cation. That is, we see an improvement for the MBTR and SOAP models but a slight decline in model performance for the n-gram model. Note that, only the DA validation set can be used to obtain an unbiased error estimate of a refitted model because it contains the only data that is independent of the overall fitting and refitting process. The statistical considerations related to model refitting are an interesting subject for future investiga- tions and a better understanding could lead to an iterative The density ρi(r) is then expanded in terms of spherical harmonics Yk,m(r/∣r∣) and orthogonal radial functions gn(∣r∣): The density ρi(r) is then expanded in terms of spherical harmonics Yk,m(r/∣r∣) and orthogonal radial functions gn(∣r∣): ρiðrÞ ¼ X n;k;m cn;k;mgnðjrjÞYk;m r jrj   : The number of coefficients cn,k,m is given by the choice of basis set expansion values. Rotationally invariant features are then computed from the coefficients of the expansion and averaged to create a single per-structure representation, forming the input space X. A real-space radial cutoff of rcut = 10 Å and ϵb = 0.5 Å are used in this work. The SOAP representation was computed with the QUIPPY package available at https://libatoms.github.io/QUIP/index.html. n-gram. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 ARTICLE a b c Fig. 4 Effect of individual DA selector conditions for SOAP-based model. The distributions of the three features referenced in the selector σSOAP ≡a/c ≤ 3.87 ∧γ < 90.35∘∧β ≥88.68∘, i.e., a/c (a), γ (b), and β (c), are shown with subpopulation selected by condition in blue, subpopulation deselected in gray, and threshold by red line. b a Fig. 4 Effect of individual DA selector conditions for SOAP-based model. The distributions of the three features referenced in the selector σSOAP ≡a/c ≤ 3.87 ∧γ < 90.35∘∧β ≥88.68∘, i.e., a/c (a), γ (b), and β (c), are shown with subpopulation selected by condition in blue, subpopulation deselected in gray, and threshold by red line. coverage of the underlying materials class (44–78%). In parti- cular, the MBTR model displays a subdomain with a 95th- percentile error that is about a factor of 3 smaller than its global 95th-percentile error. Besides these quantitative assessments, the discovered DAs enable a qualitative comparison of the three investigated representations of materials by analyzing their defining logical formulas. These show notable differences that can be attributed to variation in the physics captured by the models. For example, the appearance of the number of atoms in the selector for MBTR indicates preferred fitting of specific unit-cell sizes due to the dependence on an unnormalized histogram in the representation. For SOAP, the selectors include features exclu- sively based on the unit-cell shape, which is attributed to the choice of a cut-off radius in the construction of the local atomic environment. In order to be applicable to a wider domain, improved versions of these representations need to address those systematic shortcomings—a conclusion which is illustrative of how the method of DA identification can guide the improvement of material representations and ML methods in general. refitting scheme where DAs are refined until convergence. Such a scheme could also contain an active learning component where additional data points are sampled from within the identified subdomains. Discussion σSOAP  a c ≤3:87 ^ γ < 90:35 ^ β ≥88:68 The presented approach identified DAs for each investigated model with notably improved predictive accuracy and a large 6 NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications 6 NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 References Big Data-Driven Materials Science and its FAIR Data Infrastructure, 1–25 (Springer International Publishing, Cham, 2019). Domain of applicability analysis. Each single run of DA identification and eva- luation was performed by applying SGD using 500 random data points of the TCO test set (DA identification set) and then determining DA performance on the remaining 100 data points of the test set (DA validation set). For each ML model, a single DA selector was then determined by (i) performing six runs using sixfold cross-validation (i.e., the test set was randomly partitioned into 6 non-overlapping folds of 100 data points, and each fold was used as DA validation set once) and (ii) choosing the selector that resulted from a majority of runs. For step (ii), selectors were considered equivalent if they only differed in threshold values in some inequality conditions and these differences did not result in a different selection of data points on the whole test set. The selectors were identified using a DA iden- tification set, but the DA performance was then assessed as the mean value of the performance measures on the relevant DA validation sets. For an increased robustness of results, the above process was carried out three times for each model and the median performance over those repetitions was reported. The DA majority DA selectors themselves did not vary across those repetitions. 8. Rupp, M., Tkatchenko, A., Müller, K.-R. & von Lilienfeld, O. A. Fast and accurate modeling of molecular atomization energies with machine learning. Phys. Rev. Lett. 108, 058301 (2012). 9. Montavon, G. et al. Machine learning of molecular electronic properties in chemical compound space. N. J. Phys. 15, 095003 (2013). 10. Hirn, M., Poilvert, N. & Mallat, S. Wavelet scattering regression of quantum chemical energies. Multiscale Model. Sim. 15, 827–863 (2017). 11. Hansen, K. et al. Machine learning predictions of molecular properties: accurate many-body potentials and nonlocality in chemical space. J. Phys. Chem. Lett. 6, 2326–2331 (2015). 12. Huo, H. & Rupp, M. Unified Representation of Molecules and Crystals for Machine Learning. Preprint at https://arxiv.org/abs/1704.06439 (2017). 13. Bartók, A. P., Payne, M. C., Kondor, R. & Csányi, G. Gaussian approximation potentials: The accuracy of quantum mechanics, without the electrons. Phys. Rev. Lett. 104, 136403 (2010). 14. Bartók, A. P., Kondor, R. & Csányi, G. On representing chemical environments. Phys. Rev. B 87, 184115 (2013). Subgroup discovery. References 1. Meredig, B. et al. Combinatorial screening for new materials in unconstrained composition space with machine learning. Phys. Rev. B 89, 094104 (2014). Machine learning models. All representations were combined with kernel ridge regression using the rbf kernel as implemented in scikit-learn, version 0.21.3 (https://pypi.org/project/scikit-learn/0.23.1/). That is, ML models f νðxÞ ¼ Pn i¼1 νi expðk xF  xF i k2=ð2ϵ2ÞÞ with parameter vector ν are found by mini- mizing the objective 2. Isayev, O. et al. Materials cartography: representing and mining materials space using structural and electronic fingerprints. Chem. Mater. 27, 735–743 (2015). 3. Oliynyk, A. O. et al. High-throughput machine-learning-driven synthesis of full-Heusler compounds. Chem. Mater. 28, 7324–7331 (2016). X n i¼1 ðfνðxF i Þ  yF i Þ2 þ λνTKν 4. Schmidt, J. et al. Predicting the thermodynamic stability of solids combining density functional theory and machine learning. Chem. Mater. 29, 5090–5103 (2017). using a training set ðxF i ; yF i Þn i¼1 of n = 2400 points. Here, K refers to the n ´ n kernel matrix with entries corresponding to the application of the rbf kernel to all pairs of training points, i.e., Kij ¼ expðk xF  xF i k2=ð2ϵ2ÞÞ for 1 ≤i; j ≤n. The values for the two hyperparameters ϵ and λ are determined through a grid search within an inner loop of fivefold cross-validation where, in each iteration, the training set is split into a parameter tuning and a parameter validation set. using a training set ðxF i ; yF i Þn i¼1 of n = 2400 points. Here, K refers to the n ´ n kernel matrix with entries corresponding to the application of the rbf kernel to all pairs of training points, i.e., Kij ¼ expðk xF  xF i k2=ð2ϵ2ÞÞ for 1 ≤i; j ≤n. The values for the two hyperparameters ϵ and λ are determined through a grid search within an inner loop of fivefold cross-validation where, in each iteration, the training set is split into a parameter tuning and a parameter validation set. 5. Pilania, G. et al. Machine learning bandgaps of double perovskites. Sci. Rep. 6, 19375 (2016). 6. Lee, J., Seko, A., Shitara, K., Nakayama, K. & Tanaka, I. Prediction model of band gap for inorganic compounds by combination of density functional theory calculations and machine learning techniques. Phys. Rev. B 93, 115104 (2016). 7. Draxl, C. & Scheffler, M. References 5, 111 (2019). Data availability 22. Atzmueller, M. Subgroup discovery. Wiley Interdiscip. Rev.: Data Min. Knowl. Discov. 5, 35–49 (2015). All datasets involved in this research are available via the first author’s GitHub account at https://github.com/csutton7/ML_domain_of_applicability. The dataset used in this analysis was originally produced for the NOMAD 2018 Kaggle competition (https://www. kaggle.com/c/nomad2018-predict-transparent-conductors). The output of all materials computations are available in the NOMAD Repository (https://doi.org/10.17172/ NOMAD/2019.06.14-1). The curated competition dataset including the test data is also available at https://github.com/csutton7/nomad_2018_kaggle_dataset. A web-app allowing to reproduce the results of the Kaggle competition can be found at https:// analytics-toolkit.nomad-coe.eu/tutorial-kaggle-competition. 23. Wrobel, S. in European Symposium on Principles of Data Mining and Knowledge Discovery, 78–87 (Springer, 1997). 24. Friedman, J. H. & Fisher, N. I. Bump hunting in high-dimensional data. Stat. Comput. 9, 123–143 (1999). 25. Lemmerich, F., Atzmueller, M. & Puppe, F. Fast exhaustive subgroup discovery with numerical target concepts. Data Min. Knowl. Discov. 30, 711–762 (2016). 26. Boley, M., Goldsmith, B. R., Ghiringhelli, L. M. & Vreeken, J. Identifying consistent statements about numerical data with dispersion-corrected subgroup discovery. Data Min. Knowl. Discov. 31, 1391–1418 (2017). References The subgroup discovery target variable for DA identification was the relative (main text results) and absolute (results in Supplementary Infor- mation) model residuals, respectively. The basic propositions for the potential subgroup selectors were formed by first applying the feature map from Table 1 and then finding thresholds for inequality constraints via 5-means clustering45. Con- junctive selectors were optimized with respect to the impact objective function as given in Eq. (4) using branch-and-bound search with equivalence-aware non- redundant branching46 and tight (selection-unaware) bounding25,26. That is, selectors are found using (1) a branching operator that generates exactly one selector for any set of selectors describing the same subset of data points, and (2) a bounding function that provides an upper bound to the best attainable speciali- zation of a given selector by assuming that all subsets of its extension can be exactly described by some refined selector. All computations were run using the subgroup discovery implementation in realKD 0.7.2. More details on selector optimi- zation and a refined objective function are discussed in the Supplementary Information. 15. Seko, A., Hayashi, H., Nakayama, K., Takahashi, A. & Tanaka, I. Representation of compounds for machine-learning prediction of physical properties. Phys. Rev. B 95, 144110 (2017). y 16. Schütt, K. T. et al. How to represent crystal structures for machine learning: towards fast prediction of electronic properties. Phys. Rev. B 89, 205118 (2014). 17. Faber, F., Lindmaa, A., von Lilienfeld, O. A. & Armiento, R. Crystal structure representations for machine learning models of formation energies. Int. J. Quantum Chem. 115, 1094–1101 (2015). 18. Behler, J. Atom-centered symmetry functions for constructing high- dimensional neural network potentials. J. Chem. Phys. 134, 074106 (2011). dimensional neural network potentials. J. Chem. Phys. 134, 074106 (2011). 19. Behler, J. Neural network potential-energy surfaces in chemistry: a tool for large scale simulations Phys Chem Chem Phys 13 17930 17955 (2011) p y 19. Behler, J. Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations. Phys. Chem. Chem. Phys. 13, 17930–17955 (2011). for large-scale simulations. Phys. Chem. Chem. Phys. 13, 17930–17955 (2011). 20. Shapeev, A. V. Moment tensor potentials: a class of systematically improvable interatomic potentials Multiscale Modeling Simul 14 1153 1173 (2016) 20. Shapeev, A. V. Moment tensor potentials: a class of systematically improvable interatomic potentials. Multiscale Modeling, Simul. 14, 1153–1173 (2016). 21. Sutton, C. et al. Crowd-sourcing materials-science challenges with the NOMAD 2018 Kaggle competition. npj Comput. Mater. ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 computations involved in this research are available via the first authors GitHub account at https://github.com/csutton7/ML_domain_of_applicability. All are provided under the MIT open source software license. See http://analytics-toolkit.nomad-coe.eu/ ML_domain_of_applicability for an interactive tutorial that allows to reproduce the results of the DA analysis. occurs if the interatomic distance in the 3D crystal is less than a pre-specified cut- off distance (rcut) that is proportional to the sum of the ionic radii of the two species. The number of edges of a given node i corresponds to its coordination environment (CNi): CNi ¼ X j ri;j1ðri;j < rcutÞ: Received: 29 August 2019; Accepted: 22 May 2020; The parameter rcut was taken to be lattice dependent as used in ref. 21. Here, only the cation coordination environment is considered, which is defined entirely by the number of oxygen atoms in the first coordination shell. The n-gram representation utilizes contiguous sequences of up to four nodes (see ref. 21 for a detailed The parameter rcut was taken to be lattice dependent as used in ref. 21. Here, only the cation coordination environment is considered, which is defined entirely by the number of oxygen atoms in the first coordination shell. The n-gram representation utilizes contiguous sequences of up to four nodes (see ref. 21 for a detailed description of this approach). An implementation of the n-gram model used here is available at https://analytics-toolkit.nomad-coe.eu/home/. g q p description of this approach). An implementation of the n-gram model used here is available at https://analytics-toolkit.nomad-coe.eu/home/. Methods MBTR Th The n-gram features are generated using a histogram of contiguous sequences of nodes (i.e., atoms) that are connected by edges (i.e., bonds) in a crystalline graph representation. An edge between nodes in the crystalline graph NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications 7 ARTICLE ARTICLE Author contributions M.B. conceived the idea. C.S. and M.B. designed the initial method and research, ana- lyzed the initial results together with L.M.G., M.R., J.V., and M.S., and wrote the initial paper. M.B., C.S, L.M.G., and M.S. designed revised versions of the method and the research. All authors reviewed the paper and contributed to its final revision. 31. Nakamura, S., Mukai, T. & Senoh, M. Candela-class high-brightness InGaN/ AlGaN double-heterostructure blue-light-emitting diodes. Appl. Phys. Lett. 64, 1687–1689 (1994). 32. Arulkumaran, S. et al. Improved dc characteristics of AlGaN/GaN high- electron-mobility transistors on AlN/sapphire templates. Appl. Phys. Lett. 81, 1131–1133 (2002). ARTICLE ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17112-9 Acknowledgements 28. Kinoshita, A., Hirayama, H., Ainoya, M., Aoyagi, Y. & Hirata, A. Room- temperature operation at 333 nm of Al0.03Ga0.97N/Al0.25Ga0.75N quantum-well light-emitting diodes with Mg-doped superlattice layers. Appl. Phys. Lett. 77, 175–177 (2000). This work received funding from the European Union’s Horizon 2020 Research and Innovation Programe (grant agreement No. 676580), the NOMAD laboratory CoE, and ERC:TEC1P (No. 740233). C.S. gratefully acknowledges funding by the Alexander von Humboldt Foundation. 29. Ohta, H. et al. Current injection emission from a transparent p-n junction composed of p-SrCu2O2/n-ZnO. Appl. Phys. Lett. 77, 475–477 (2000). p p pp y 30. Tsukazaki, A. et al. Repeated temperature modulation epitaxy for p-type doping and light-emitting diode based on ZnO. Nat. Mater. 4, 42 (2005). The authors declare no competing interests. 34. Hoffman, R., Norris, B. J. & Wager, J. ZnO-based transparent thin-film transistors. Appl. Phys. Lett. 82, 733–735 (2003). Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 020-17112-9. Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 020-17112-9. pp y 35. Nishii, J. et al. High mobility thin film transistors with transparent ZnO channels. Jpn. J. Appl. Phys. 42, L347 (2003). 36. Nomura, K. et al. Thin-film transistor fabricated in single-crystalline transparent oxide semiconductor. Science 300, 1269–1272 (2003). Correspondence and requests for materials should be addressed to C.S., M.B. or L.M.G. Correspondence and requests for materials should be addressed to C.S., M.B. or L.M.G. 37. Nomura, K. et al. Room-temperature fabrication of transparent flexible thin-film transistors using amorphous oxide semiconductors. Nature 432, 488 (2004). Peer review information Nature Communications thanks Patrick Riley and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. Peer review information Nature Communications thanks Patrick Riley and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. g p 38. Dehuff, N. et al. Transparent thin-film transistors with zinc indium oxide channel layer. J. Appl. Phys. 97, 064505 (2005). y pp y 39. Draxl, C. & Scheffler, M. NOMAD: the FAIR concept for big data-driven materials science. MRS Bull. 43, 676–682 (2018). Competing interests 33. Kubovic, M. et al. Microwave performance evaluation of diamond surface channel FETs. Diam. Relat. Mater. 13, 802–807 (2004). Reprints and permission information is available at http://www.nature.com/reprints Reprints and permission information is available at http://www.nature.com/reprints Reprints and permission information is available at http://www.nature.com/reprints 40. Nyshadham, C. et al. Machine-learned multi-system surrogate models for materials prediction. npj Comput. Mater. 5, 51 (2019). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Code availability 27. Hoppe, R. Effective coordination numbers (ECoN) and mean fictive ionic radii (MEFIR). Z. für. Kristallographie - Crystalline Mater. 150, 23 (1979). The SGD implementation is publicly available through the open source Java library realKD 0.7.2 available at https://bitbucket.org/realKD/realkd/. Python scripts for a NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunicati 8 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. pj p 41. Cade, B. S. & Richards, J. D. Permutation tests for least absolute deviation regression. Biometrics 52, 886–902 (1996). 42. Kailkhura, B., Gallagher, B., Kim, S., Hiszpanski, A. & Han, T. Y.-J. Reliable and explainable machine-learning methods for accelerated material discovery. npj Comput. Mater. 5, 1–9 (2019). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. pj p 43. Haghighatlari, M., Shih, C.-Y. & Hachmann, J. Thinking globally, acting locally: on the issue of training set imbalance and the case for local machine learning models in chemistry. Preprint at https://doi.org/10.26434/ chemrxiv.8796947.v2 (2019). 44. Rupp, M. Machine learning for quantum mechanics in a nutshell. Int. J. Quantum Chem. 115, 1058–1073 (2015). 45. Boley, M. & Kariryaa, A. On the intuitiveness of common discretization methods. In KDD 2016 Workshop on Interactive Data Exploration and Analytics, 22–29 (Association for Computing Machinery (ACM), 2016). 46. Boley, M. & Grosskreutz, H. Non-redundant subgroup discovery using a closure system. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 179–194 (Springer, 2009). NATURE COMMUNICATIONS | (2020) 11:4428 | https://doi.org/10.1038/s41467-020-17112-9 | www.nature.com/naturecommunications

W2990206712.txt

https://www.intechopen.com/citation-pdf-url/69311

en

Concepts and Trends for Extraction and Application of Microalgae Carbohydrates

IntechOpen eBooks

cc-by

We are IntechOpen, the world’s leading publisher of Open Access books Built by scientists, for scientists 6,900 186,000 200M Open access books available International authors and editors Downloads Our authors are among the 154 TOP 1% 12.2% Countries delivered to most cited scientists Contributors from top 500 universities Selection of our books indexed in the Book Citation Index in Web of Science™ Core Collection (BKCI) Interested in publishing with us? Contact [email protected] Numbers displayed above are based on latest data collected. For more information visit www.intechopen.com Chapter Concepts and Trends for Extraction and Application of Microalgae Carbohydrates Maiara Priscilla de Souza, Andrea Sanchez-Barrios, Tiele Medianeira Rizzetti, Lisianne Brittes Benitez, Michele Hoeltz, Rosana de Cassia de Souza Schneider and Fábio de Farias Neves Abstract The proposed chapter aims to provide a more in-depth explanation of the composition of carbohydrates in microalgae biomass, focusing on separation methods, chemistry, molecular characterization, as well as their application in several areas. The purpose of this review chapter is to show that biological products from microalgae have potential in health, food, and industry applications (materials and biofuel production). Steps for extraction and purification will be discussed, as well as the relationship between the type of microalgae and its composition, as a way of optimizing protocol selection and product making, without breaking down the cell to begin with (total carbohydrate extraction present in the cell). An overall overview of the current and prospective trends and methodologies for the use of microalgae carbohydrate will be included as starting points to shed light on some of the possible issues that currently do not allow the development and feasibility of microalgae biorefineries. Keywords: microalgae, carbohydrates, analytical chemistry, biorefinery, trends 1. Introduction The constant necessity of looking for new alternatives to produce sustainable versions of products has led to the discovery and development of new technique and biological models. Organisms with the potential of producing molecules that can be used for the development of bioproducts in different areas (food, beauty, health, and biodiesel, among others) have led to the discovery, study, and use of new organisms. Algae and microalgae have become potential and promising model organisms to be used for carbohydrate production and use, but besides that, it is also deeply studied due to its lipids, proteins, and photosynthetic activity, making them a prospective source of bioenergy production [1, 2]. The problems associated to the first- and second-generation feedstock for biofuels have seen to become more and more complicated to solve considering the food crisis and complex conversion of lignocellulosic materials [3] researchers, has led to the exploration of a third-generation feedstock, mostly represented by photosynthetic organisms, primarily algae/microalgae [4]. 1 Microalgae - From Physiology to Application Microalgae are considered to have the potential to produce third-generation biodiesel (due to its capability of fixing carbon dioxide (CO2), which is eventually converted to biomass and other products), which can also be referred as thirdgeneration feedstock, providing mainly lipids, proteins, and carbohydrates. The utilization of these molecules in a sequential way allows the treatment of biomass in biorefineries, including its use in fermentative production of a range of platform biochemical [5]. Through this process, carbohydrates are used as a fundamental piece for the production of certain products. Besides the cell wall and reserves of photosynthetic reaction, the carbohydrates can be excreted by the cell. The exopolysaccharides (EPS) are complex carbohydrates produced from some microalgae, which are long chains composed of sugar derivative structures, mucilaginous and with reactive functional groups, such as sulfate, hydroxyl, or carboxylic [6]. The major components of EPS include mainly the polysaccharides and others as proteins, nucleic acid, and lipids [7]. Addition of these molecules is considered to be of extreme importance for enriching the nutritional value of food items [8]. Although these characteristics are beneficial, the extraction of these compounds from microalgae becomes a real challenge. In this context, several treatments can be performed for the disruption of microalgal cells, including chemical modifications and mechanical, thermal, or ultra-sonication processes [9]. Although promising, it is still hard to manage the cost and work that developing new technologies have for investors (in the industry for applied approaches and academia for basic development and standardization), which presents some limitations for the advance of research in this area. On the long run, an implementation of the use of microalgae as a substitute for many of the other crop options still used will have an important impact on the economics, environment, and more sustainable practices. Due to the vast diversity of species of microalgae, we present an analysis of the current trends and importance and potential of the use of carbohydrates present in Figure 1. Diagram of the main topics involved in the harnessing of microalgae carbohydrates. 2 Concepts and Trends for Extraction and Application of Microalgae Carbohydrates DOI: http://dx.doi.org/10.5772/intechopen.89323 microalgae. The aim of highlighting these points is to bring awareness and present new alternatives of methods that will allow the use of carbohydrates in microalgae without the breakdown of these carbohydrates, as a way of improving and lowering cost and making the use of these organisms a more feasible way of developing new products and technologies [10]. The main topics that will be discussed in this chapter can be seen in Figure 1. 2. Carbohydrate importance and composition Carbohydrates from microalgae are considered to have a great application in industry, which has led to the development of new techniques and studies but that has exposed new challenges to industry [11, 12]. The diversity of microalgae, the composition, and cell organization are some of the few trials that many scientists are currently facing. Carbohydrates are poly- or oligosaccharides that can be present in vacuoles and cell walls or that could also be excreted as exopolysaccharides (EPS) [13]. Microalgae come to be an interesting key organism to study due to the high content Figure 2. Glucose polymers found in microalgae. 3 Microalgae - From Physiology to Application of carbohydrates that some of them have. Some examples are Desmodesmus spp. (41%) from wastewater and landfill leachate treatment [14], Chlorella stigmatophora (~55%) [5], or Chlorella vulgaris (60% [15] or >52% [16]) cultivated under conditions of nitrogen depletion. Similarly, indigenous microalgae species have been bioprospected in nitrogendepleted environments, such as Desmodesmus sp. (57%), an unidentified one with more carbohydrate content (70%) [17], and Arthrospira platensis that accumulated up to 74% of carbohydrates [18]. Despite their vast potential, using different species imposes a few challenges when trying to establish a consistent methodology for extraction and use. Based on the differences in metabolism that each species presents, a possible and feasible approach is to consider that they all have reserves made out of polymers of glucose (glucans), such as chrysolaminarin (1:11), laminarin (3:1) (β-1,3 and β-1,6 branches), paramylon(β-1,3), glycogen-type, cyanophycean, floridean (semiamylopectin: α-1,4 and α-1,6 branches), and amylose-type starch (α-1, 4) or both (Figure 2). The external covering of cell with polysaccharides could be peptidoglycan matrices, cellulosic wall, and galactose polymer matrix, and others [18–20]. The potential of using microalgae for the development of various products, using carbohydrates as the main source, exposes some of the challenges that using these organisms can have. The main of it is the process of extraction of the diverse saccharides present in each species, but in comparison with plant-derived products, it is actually a much easier path, since they do not present lignification of the cell wall [11, 21]. In the same way, having various species together as a main feedstock source may require adaptations to a proper and more efficient method of extraction. 3. Trends for microalgae use and production, using carbohydrates as the main molecule of interest Microalgae use for development of new products that could be beneficial to multiple industries has raised many questions in the quest of finding what will be the next technology that will be developed, what products will change our industries, and how this could benefit populations. As we know, the use of microalgae carbohydrates could be a solution to many current limitations that agriculture, pharmacy, nutrition, and other areas are facing while trying to develop better solutions to fulfill the needs of people [22]. One of the main reasons for the development of new technologies and the arise of trends is to be able to satisfy the demands of the public and produce the income that the market will like to have in return for investing in these areas. Unfortunately, it is still a difficult task since there is still a long way to go for developments of products outside of the food industry. Carbohydrates/starch is considered to be a positive/beneficial step for the chemical development of products, and when produced in high quantities, its use for fermentation is considered a better option. To date, the most common way to induce the production of specific molecules (in this case, carbohydrates and starch) is by creating stress conditions in the environment where the alga is growing, mainly by altering nutrient concentrations or by changing light, temperature, and other parameters [23, 24]. Photosynthetic electron transport approach seems to be the most researched area [25] since genetic modification of algae becomes challenging due to the great variety of species. Cyanobacteria are being used more for genetic engineering, but no complete success has come out it yet. Besides the interest cyanobacteria, some algae species have been explored as potential species for biofuel production through 4 Concepts and Trends for Extraction and Application of Microalgae Carbohydrates DOI: http://dx.doi.org/10.5772/intechopen.89323 modification of metabolic pathways that could increase the production of carbohydrates, lipids, and other compounds, but the progress in this area has seen to be slow, which although discouraging is still considered one of the main focus for the future of biotechnologies in microalgae [26]. Microbiome studies of microalgae populations seem to focus more on the ecology aspects and the importance of their use in multiple biological processes. Toxicity of waters seems to attract the attention of most research groups. Its importance lies in not only the understanding of structure communities and safety of fresh water resources (which is the main source of drinking water for the world) but also the discovery of new species that may be hard to culture through regular isolation practices. Similarly, human microbiome studies seem to be interested in the potential use of carbohydrates (including exopolysaccharides) derived from microalgae, due to its prebiotic potential, improving the health of adults and infants that cannot consume breast milk in their early stages, strengthening their immune responses [12, 27]. 4. Carbohydrate extraction and analysis Some microalgae species are capable of accumulating a large amount of intracellular starch and have structural polysaccharides in their cell walls [28]. However, the development of standard analytical procedures for the characterization of microalgae biomass has been difficult because of the existence of several microalgae species with different cellular structures and chemical compositions [28]. Even a specific microalgae species may present variability in their chemical composition because they can strongly be affected by factors of cultivation conditions such as temperature, salinity, and nutrient availability [9]. Thereby, sample preparation for the determination of carbohydrates in microalgae is very complex. The analytical method applied depends on the intended usage of the algal carbohydrates. Some require the qualitative or quantitative composition of the monosaccharides and others the total carbohydrate analysis [29]. In algal biomass, the carbohydrate profile analysis requires preliminary acid hydrolysis. Acid hydrolysis allows depolymerizing the intracellular starch and structural polysaccharides into their monomers, which are then further quantified [28]. The optimal hydrolysis conditions (acid, temperature, and time) should ensure complete hydrolysis of the polysaccharides and at the same time avoid excessive degradation of monosaccharides [29]. In this context, the selective cleaving of algal polysaccharides by enzymatic hydrolysis is another promising approach once the formation of degradation products like furfural or hydroxymethylfurfural is avoided [29]. However, for some microalgae species, the composition of the cell wall is complicated and unknown; also, some enzymes are very expensive [30]. The cell wall present in the microalgae limits extraction yields of high-value products or results in a low bioavailability of intracellular components [9, 30]. The methodology widely employed in acid hydrolysis is the procedure published by the National Renewable Energy Laboratory (NREL) based on two-step hydrolysis using H2SO4 [31, 32]. Despite being a reliable method, it is a multistep procedure, increasing the chance of experimental errors, and also presents a high time of sample preparation. Northcote et al. [33] proposed another method based on one-step acid hydrolysis with dilute H2SO4 and use smaller biomass samples. Other methods use chemical extraction method like alkaline pretreatment [34] and physical method, such as hot-water treatment, microwave-assisted extraction, and ultrasonic-assisted extraction [30, 35]. The choice of extraction method is that the 5 Microalgae - From Physiology to Application pretreatment is effective qualitatively and quantitatively, and the technology is simple to operate and economical for scale-up [30]. Zhao’s research team [30] investigated three methods of conventional solvent extraction (CSE), fluidized bed extraction (FBE), and ultrasonic-assisted extraction (UAE) to obtain an effective extraction method of carbohydrates/glucose. The CSE employed lyophilized microalgae extracted with distilled water and agitation in a vortex. For FBE, the Chlorella sp. culture was harvested and washed with distilled water and then diluted using distilled water and added into a fluidized reactor with air aeration. To UAE, the algal cells were harvested and washed with distilled water, diluted, and taken to the ultrasonic processor. The ultrasonic-assisted extraction was more effective than the other methods. Information in the literature related to the amount of cell wall microalgae polysaccharides is scarce. Usually, the quantification of polysaccharides in microalgal is made by analyzing the total carbohydrate, thus including storage polysaccharides (SPS) and cell wall-related polysaccharides, which exhibit different functions in the microalgal cell [9]. According to Bernaerts et al. [9], the insight into the composition of cell wall-related polysaccharides, such as the monosaccharide profile or the degree of sulfation, is not only desired in terms of process optimization but also as a potential for several biotechnological. Thereby, the authors investigated to apply a universal procedure for extraction of the total cell wall-related polysaccharides, including cell wall polysaccharides (CWPS) and extracellular polymeric substances (EPMS), of 10 commercially available microalgae species followed by a characterization of the monosaccharide profile, uronic acid content, and sulfate [9]. Initially, the procedure consisted of dry biomass suspended in saline solution incubated for 16 h at 25°C, followed by a two-step centrifugation. Afterward, the supernatant was submitted for extraction of EPMS and the residual biomass (pellet) for extraction of CWPS. Ethanol was added to the supernatant precipitating EPMS; the solution was vacuum filtered, and the insoluble residue was dialyzed against demineralized water for 48 h, and finally, the extracts were lyophilized. In the extraction of CWPS, the pellets were suspended in MOPS buffer, and the cells were disrupted using UHPH after cold ethanol was added to the suspensions and for pellet recovery. Lipids were removed by addition of hexane/isopropanol to the pellet, mixed and centrifuged to remove the upper solvent layer. Afterward, SPS and protein were enzymatically removed using endo-β-1,3-glucanase or a combination of α-amylase/amyloglucosidase and Subtilisin A protease, respectively. The mixtures were incubated and after addition of cold ethanol they were centrifuged. The pellet was finally washed in acetone, vacuum filtered, and dried overnight at 40°C, and this residue was considered as CWPS. Monosaccharide and uronic acid composition of CWPS and EPMS were hydrolyzed according to De Ruiter et al. [36] using methanolysis combined with trifluoroacetic acid (TFA) hydrolysis. After sample preparation, carbohydrate analysis is a very complex field. Usually, after microalgal acid hydrolysis, the total carbohydrate content of the hydrolysate can be determined using colorimetric procedures like the phenol-sulfuric acid [37–39] or anthrone-based [40–42]. These methods are available, giving excellent and robust results with low effort in a very short time. Nevertheless, detailed information about the monosaccharide composition cannot be generated [29]. Qualitative investigations can be performed using TLC methods with silica-based separation materials making the separation of most monosaccharides possible. However, quantification with the TLC methods is not possible [29], and for quantification of monosaccharides, analytical methods such as high performance liquid chromatography (HPLC) are often used. The HPLC equipped with a refractive index detector (RID) [32] and HPLC combined with pulsed amperometric 6 Concepts and Trends for Extraction and Application of Microalgae Carbohydrates DOI: http://dx.doi.org/10.5772/intechopen.89323 detection (HPAEC-PAD) [9, 28] and liquid chromatography mass spectrometer (LC–MS) [29] were related. The trends for carbohydrate analysis are the exploration of methods that study both cell wall polysaccharides and extracellular polymeric substances. The developed methods always aim to use small amounts of sample, reagents, and shorter analysis. Liquid chromatography has been highlighted in the carbohydrate determination since it presented a good separation and quantification of these compounds. Although it is a more expensive analytical technique, it provides data on the composition of individual monomeric sugars that make it of interest for this type of analysis. 5. Applications Microalgae have several types of polysaccharides in their composition, such as phycocolloids, agar, alginate, carrageenan, fucoidan, ulvana, and cellulose, among others. phycocolloids can be formed by different monomers such as glucose, galactose, mannuronic acid, guluronic acid, mannitol, and laminarin. These carbohydrates can be inserted into functional beverages and food products such as functional bread, ready to serve soups, functional snack foods and a variety of sauces, creams, bakery products, and additional food products [43, 44]. Due to the high carbohydrate content, poultry and aquaculture feed is one of the main study targets for the use of microalgae biomass. In 2007, around 30% of the world’s current algae production was sold for animal feed application [8]. Microalgae are also a suitable alternative for growing fish, larvae, and zooplankton. Chlorella is one of the main examples of microalgae that can play a key role in food and feed due to the properties of its biomass, which can simultaneously provide high concentrations of carbohydrates, vitamins, and proteins [45]. Besides Chlorella, other species used in aquaculture can be highlighted: Tetraselmis, Isochrysis, Pavlova, Phaeodactylum, Chaetoceros, Nannochloropsis, Skeletonema, and Thalassiosira. Spirulina and Chlorella microalgae can be applied in the feeding of cats, dogs, aquarium fish, ornamental birds, horses, birds, cows, and breeding bulls. The most common genera of larval microalgae include Chaetoceros, Thalassiosira, Tetraselmis, Isochrysis, and Nannochloropsis [46]. 1,3-β-glucan is an important carbohydrate present in microalgae composition due to its applications in the food industry as a thickener, and health applications, especially in the protection against infections and also to inhibit cancer cell growth in vivo [44, 47]. According to [48], the global β-glucan market was valued at USD 307.8 million in 2016, and it is predicted that in 2022, the global carbohydrate market could reach up to USD 476.5 million, which indicates the huge potential for development in many different types of applications. According to Koller et al. [49], sulfated polysaccharides produced by microalgae can be applied in therapies against bacterial infections. Carrageenan polysaccharide, also known as food additive E407, can be used in pharmaceutical applications. Marine carbohydrates have been widely used in the cosmetics industries due to their chemical and physical properties. Brown algal fucoidans/alginates, green algal ulcers, and red algal carrageenans/agar are used as gelling, thickening, and stabilizing agents. In addition, marine carbohydrates have potential skin benefits, and biological activities are linked to their structure as determined by molecular weights or the presence of sulfate groups and other sugars [50]. Red algae, such as Chondrus sp., Gigartina sp. Eucheuma sp., Hypnea sp., and Furcellaran sp., are widely used for the production of carrageenan. This compound can be used in food and pharmaceutical industries for applications in fruit gel, fruit juices, sweets, and jellies, among others. Another carbohydrate group 7 Microalgae - From Physiology to Application molecule is fucoidan, which is associated with brown algal cell wall components (Phaeophyceae). Among the bioactivities derived from this molecule, the anticoagulant, antitumor, antivirus, and antioxidant properties stand out, making it attractive for pharmaceutical applications [51]. Besides these applications, the remaining biomass of microalgae presents carbohydrate-rich molecules, which have been widely used in the production of bioplastics, agar, sugars, and other high-added value chemicals. However, despite being a growing area, the biorefinery stage must be studied in order to extend its applicability on an industrial scale [51]. According to Mihranyan [52], the rheological behavior of cellulose found in Cladophora algae is similar to micro fibrillated cellulose. Because this cellulose is very robust and not susceptible to chemical reactions, the properties of cellulose found in these algae provide excellent rheological properties making this material interesting in food, pharmaceuticals, paints, dressings, and biodegradable plastic applications. The high carbohydrate content and low-ash values make microalgae more suitable for conversion to biofuels [43]. The production of bioethanol from microalgae gained importance due to their high biomass productivity, diversity, variable chemical composition, and high photosynthetic rates of these organisms [53]. Due to the large amount of carbohydrates/polysaccharides and cellulose walls, these microorganisms become favorable for the production of this biofuel [54, 55]. In many countries, ethanol is produced on a large scale from crops containing sugars and starches in its composition through fermentation. The biomass is ground, and the starch is converted into sugars by different methods. Polysaccharide starch is also accessible as a storage material for various algal species and can be anaerobically converted into bioethanol [49]. 6. Conclusions Microalgae biomass conversion technologies involve carbohydrates as the main source in the production of biofuels and other compounds of high commercial value. Changes in metabolic pathways aiming at increased carbohydrate production are seen as a potential for enhancing microalgae biotechnology. Extraction methods and trends in analytical methodologies focus on microalgae cell wall polysaccharides and the polymers excreted by these microorganisms. The high carbohydrate content makes microalgae excellent candidates for the production of numerous biocomposites, especially beta-glucan, which is on the international market, indicating its strong potential for its use in different biotechnological applications. 8 Concepts and Trends for Extraction and Application of Microalgae Carbohydrates DOI: http://dx.doi.org/10.5772/intechopen.89323 Author details Maiara Priscilla de Souza1, Andrea Sanchez-Barrios1, Tiele Medianeira Rizzetti1, Lisianne Brittes Benitez1, Michele Hoeltz1, Rosana de Cassia de Souza Schneider1 and Fábio de Farias Neves2* 1 University of Santa Cruz do Sul, Santa Cruz do Sul, Brazil 2 Santa Catarina State University, Laguna, Brazil *Address all correspondence to: [email protected] © 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 9 Microalgae - From Physiology to Application References [1] Das P, Aziz SS, Obbard JP. Two [10] Fasaei F et al. Techno-economic phase microalgae growth in the open system for enhanced lipid productivity. Renewable Energy. 2011;36(9):2524-2528 evaluation of microalgae harvesting and dewatering systems. Algal Research. 2018;31:347-362 [11] Wells ML et al. Algae as nutritional [2] Markou G, Nerantzis E. Microalgae for high-value compounds and biofuels production: A review with focus on cultivation under stress conditions. Biotechnology Advances. 2013;31(8):1532-1542 [3] Lavoie J-M, Marie-Rose S, Lynch D. Non-homogeneous residual feedstocks to biofuels and chemicals via the methanol route. Biomass Conversion and Biorefinery. 2013;3(1):39-44 [4] Huang G et al. Biodiesel production by microalgal biotechnology. Applied Energy. 2010;87(1):38-46 and functional food sources: Revisiting our understanding. Journal of Applied Phycology. 2017;29(2):949-982 [12] de Jesus Raposo MF, de Morais AMMB, de Morais RMSC. Emergent sources of prebiotics: Seaweeds and microalgae. Marine Drugs. 2016;14(2):27 [13] Larronde-Larretche M, Jin X. Microalgal biomass dewatering using forward osmosis membrane: Influence of microalgae species and carbohydrates composition. Algal Research. 2017;23:12-19 [14] Hernández-García A et al. [5] Mayers JJ et al. Identifying a marine microalgae with high carbohydrate productivities under stress and potential for efficient flocculation. Algal Research. 2018;31:430-442 Wastewater-leachate treatment by microalgae: Biomass, carbohydrate and lipid production. Ecotoxicology and Environmental Safety. 2019;174:435-444 [15] Uyaguari-Diaz MI et al. [6] Cunha C et al. Marine vs freshwater microalgae exopolymers as biosolutions to microplastics pollution. Environmental Pollution. 2019;249:372-380 A comprehensive method for amplicon-based and metagenomic characterization of viruses, bacteria, and eukaryotes in freshwater samples. Microbiome. 2016;4(1):20 [7] Xiao R, Zheng Y. Overview of [16] Chen C-Y, Chang H-Y, Chang J-S. microalgal extracellular polymeric substances (EPS) and their applications. Biotechnology Advances. 2016;34(7):1225-1244 Producing carbohydrate-rich microalgal biomass grown under mixotrophic conditions as feedstock for biohydrogen production. International Journal of Hydrogen Energy. 2016;41(7):4413-4420 [8] Becker EW. Micro-algae as a source of protein. Biotechnology Advances. 2007;25(2):207-210 [9] Bernaerts TMM et al. Comparison of microalgal biomasses as functional food ingredients: Focus on the composition of cell wall related polysaccharides. Algal Research. 2018;32:150-161 10 [17] Sanchez Rizza L et al. Bioprospecting for native microalgae as an alternative source of sugars for the production of bioethanol. Algal Research. 2017;22:140-147 [18] Depraetere O et al. Harvesting carbohydrate-rich Arthrospira platensis Concepts and Trends for Extraction and Application of Microalgae Carbohydrates DOI: http://dx.doi.org/10.5772/intechopen.89323 by spontaneous settling. Bioresource Technology. 2015;180:16-21 within the intestinal environment. Frontiers in Microbiology. 2018;9:1-15 [19] Gaignard C et al. New horizons [28] Souza MFD et al. Neutral sugars in culture and valorization of red microalgae. Biotechnology Advances. 2019;37(1):193-222 determination in chlorella: Use of a onestep dilute sulfuric acid hydrolysis with reduced sample size followed by HPAEC analysis. Algal Research. 2017;24:130-137 [20] Sheath RG, Wehr JD. Introduction to the freshwater algae. In: Wehr JD, Sheath RG, Kociolek JP, editors. Freshwater Algae of North America. 2nd ed. Boston: Academic Press; 2015. pp. 1-11 [21] Mišurcová L et al. Amino acid composition of algal products and its contribution to RDI. Food Chemistry. 2014;151:120-125 [22] Jutur PP, Nesamma AA, Shaikh KM. Algae-derived marine oligosaccharides and their biological applications. Frontiers in Marine Science. 2016;3(83):1-5 [23] Jerez CG et al. Effect of nutrient starvation under high irradiance on lipid and starch accumulation in Chlorella fusca (Chlorophyta). Marine Biotechnology. 2016;18(1):24-36 [24] Sun X et al. Effect of nitrogen- starvation, light intensity and iron on triacylglyceride/carbohydrate production and fatty acid profile of Neochloris oleoabundans HK-129 by a two-stage process. Bioresource Technology. 2014;155:204-212 [29] Schulze C et al. Carbohydrates in microalgae: Comparative determination by TLC, LC-MS without derivatization, and the photometric thymol-sulfuric acid method. Algal Research. 2017;25:372-380 [30] Zhao G et al. Ultrasound assisted extraction of carbohydrates from microalgae as feedstock for yeast fermentation. Bioresource Technology. 2013;128:337-344 [31] Van Wychen S, Laurens LML. Determination of Total Carbohydrates in Algal Biomass: Laboratory Analytical Procedure (LAP). United States: NREL/ TP-5100-60957, National Renewable Energy Laboratory (NREL); 2016 [32] Hanifzadeh M, Garcia EC, Viamajala S. Production of lipid and carbohydrate from microalgae without compromising biomass productivities: Role of Ca and Mg. Renewable Energy. 2018;127:989-997 [33] Northcote DH, Goulding KJ, [25] Rockwell NC, Lagarias JC. A brief Horne RW. The chemical composition and structure of the cell wall of Chlorella pyrenoidosa. The Biochemical Journal. 1958;70(3):391-397 historyof phytochromes.ChemPhysChem. 2010;11(6):1172-1180 [34] Correia PR, Beirão-da-Costa ML. [26] Rabinovitch-Deere CA et al. Synthetic biology and metabolic engineering approaches to produce biofuels. Chemical Reviews. 2013;113(7):4611-4632 [27] Castro-Bravo N et al. Interactions of surface exopolysaccharides from bifidobacterium and lactobacillus 11 Starch isolation from chestnut and acorn flours through alkaline and enzymatic methods. Food and Bioproducts Processing. 2012;90(2):309-316 [35] Velmurugan R, Muthukumar K. Ultrasound-assisted alkaline pretreatment of sugarcane bagasse for fermentable sugar production: Optimization through response surface Microalgae - From Physiology to Application methodology. Bioresource Technology. 2012;112:293-299 feed and aquaculture. Journal of Biological Research (Thessaloniki). 2014;21(1):6 [36] de Ruiter GA et al. Carbohydrate [45] Guccione A et al. Chlorella for analysis of water-soluble uronic acid-containing polysaccharides with high-performance anion-exchange chromatography using methanolysis combined with TFA hydrolysis is superior to four other methods. Analytical Biochemistry. 1992;207(1):176-185 protein and biofuels: From strain selection to outdoor cultivation in a Green Wall panel photobioreactor. Biotechnology for Biofuels. 2014;7(1):84 [37] DuBois M et al. Colorimetric method for determination of sugars and related substances. Analytical Chemistry. 1956;28(3):350-356 [46] Priyadarshani I, Rath B. Commercial and industrial applications of micro algae–A review. Journal of Algal Biomass Utilization. 2012;3(4):89-100 [47] Carballo C et al. Yeast β-glucans to stimulate high carbohydrate content in Spirulina biomass. Bioresource Technology. 2018;269:221-226 and microalgal extracts modulate the immune response and gut microbiome in Senegalese sole (Solea senegalensis). Fish and Shellfish Immunology. 2019;92:31-39 [39] Cea-Barcia G et al. A cost-effective [48] Bozbulut R, Sanlier N. Promising strategy for the bio-prospecting of mixed microalgae with high carbohydrate content: Diversity fluctuations in different growth media. Bioresource Technology. 2014;163:370-373 effects of β-glucans on glyceamic control in diabetes. Trends in Food Science and Technology. 2019;83:159-166 [38] Braga VDS et al. Cultivation strategy [40] Lupatini AL et al. Protein and carbohydrate extraction from S. platensis biomass by ultrasound and mechanical agitation. Food Research International. 2017;99:1028-1035 [41] Batista AP et al. Comparison of microalgal biomass profiles as novel functional ingredient for food products. Algal Research. 2013;2(2):164-173 [42] Osborne DRVP. In: Acribia E, editor. Analisis de nutrientes de los alimentos. Zaragoza: Editorial Acribia; 1986 [49] Koller M, Muhr A, Braunegg G. Microalgae as versatile cellular factories for valued products. Algal Research. 2014;6:52-63 [50] Kim J et al. Beneficial effects of marine algae-derived carbohydrates for skin health. Marine Drugs. 2018;16(11):459 [51] Khanra S et al. Downstream processing of microalgae for pigments, protein and carbohydrate in industrial application: A review. Food and Bioproducts Processing. 2018;110:60-84 [52] Mihranyan A. Cellulose from [43] Sudhakar MP et al. A review on bioenergy and bioactive compounds from microalgae and macroalgaesustainable energy perspective. Journal of Cleaner Production. 2019;228:1320-1333 [44] Yaakob Z et al. An overview: Biomolecules from microalgae for animal 12 cladophorales green algae: From environmental problem to high-tech composite materials. Journal of Applied Polymer Science. 2011;119(4):2449-2460 [53] Sharma P, Sharma N. Industrial and biotechnological applications of algae: A review. Journal of Advances in Plant Biology. 2017;1(1):01 Concepts and Trends for Extraction and Application of Microalgae Carbohydrates DOI: http://dx.doi.org/10.5772/intechopen.89323 [54] Abdulla R et al. Microalgae chlorella as a sustainable feedstock for bioethanol production. In: Green Engineering for Campus Sustainability. Singapore: Springer; 2020. pp. 81-103 [55] Smachetti MES et al. Microalgal biomass as an alternative source of sugars for the production of bioethanol. In: Principles and Applications of Fermentation Technology. United States: Scrivener Publishing LLC; 2018. pp. 351-386 13 

https://openalex.org/W2090553538

https://europepmc.org/articles/pmc3302021?pdf=render

English

<i>In Silico</i> Experimental Modeling of Cancer Treatment

ISRN Oncology

cc-by

1. Introduction Our contention in this paper, a view shared by many researchers in the closely related fields of computational, theoretical and mathematical biology, is that in silico experi- ments can be used as precursors to, or in combination with, preclinical experimental studies to provide guidance for the development of more refined hypotheses and experimental studies. In silico and mathematical modeling lends itself to the determination of preliminary information such as toxi- city, pharmacokinetics, and efficacy, which can then be used to guide preclinical and clinical studies. Traditional laboratory-based cancer research involves expen- sive trial and error experimental strategies applied to hu- mans, animals, and their harvested tissues. “In silico experi- mentation,” the coupling of current computing technologies with mathematical or theoretical characterizations of cancer cell biology, provides a novel approach to guiding the early stages of hypothesis development and experimental design that has the potential to create subsequent efficiencies and cost savings in the laboratory. This computational approach is advantageous because it allows vast numbers of experi- ments to be carried out that are easily observed at any desired level of detail and can be repeated and controlled at will. g p In silico experimentation involves the combination of biological data and expert opinion with mathematical and computer-based representations to construct models of bio- logy. Computer-based experiments can then be carried out using these models rather than, or in combination with, lab- oratory research. Using parameter distributions based on current expert opinion (“fuzzy” inputs) or actual biological data (random variables) as inputs into the in silico models, it is possible to create what are effectively “computational patients” upon which to experiment. It is of course also pos- sible to consider smaller-scale experiments and even mul- tiscale experiments, conducted on molecular, cellular, and tissue/organ levels. Appropriate use of in silico models in- volves making predictions based on experimental data and expert information and allows the models to be effectively It seems difficult to argue that preclinical studies in can- cer biology are expensive. Such studies involving in vitro and in vivo animal experiments involve hypothesis generation and testing to determine whether further trials are warranted and are extremely costly both in terms of researchers’ time and the associated financial investment. Trisilowati1 and D. G. Mallet1, 2 1Mathematical Sciences Discipline, Queensland University of Technology, P.O. Box 2434, Brisbane, QLD 4001, Australia 2Institute of Health and Biomedical Innovation, Queensland University of Technology, P.O. Box 2434, Brisbane, QLD 4001, Australia 1Mathematical Sciences Discipline, Queensland University of Technology, P.O. Box 2434, Brisbane, QLD 4001, Australia 2Institute of Health and Biomedical Innovation, Queensland University of Technology, P.O. Box 2434, Brisbane, QLD 4001, Australia Correspondence should be addressed to D. G. Mallet, [email protected] Received 21 September 2011; Accepted 28 October 2011 Academic Editors: A. M. Garcia-Lora, F. Kuhnel, and M. Stracke Copyright © 2012 Trisilowati and D. G. Mallet. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. In silico experimental modeling of cancer involves combining findings from biological literature with computer-based models of biological systems in order to conduct investigations of hypotheses entirely in the computer laboratory. In this paper, we discuss the use of in silico modeling as a precursor to traditional clinical and laboratory research, allowing researchers to refine their experimental programs with an aim to reducing costs and increasing research efficiency. We explain the methodology of in silico experimental trials before providing an example of in silico modeling from the biomathematical literature with a view to promoting more widespread use and understanding of this research strategy. International Scholarly Research Network ISRN Oncology Volume 2012, Article ID 828701, 8 pages doi:10.5402/2012/828701 International Scholarly Research Network ISRN Oncology Volume 2012, Article ID 828701, 8 pages doi:10.5402/2012/828701 1. Introduction Costs, such as labo- ratory setup, equipment and space, time spent by academics training others, and the time, equipment, and materials costs involved in repetitive, hands-on experimental work, all con- tribute to the expense of laboratory-based experimental re- search. 2 2 ISRN Oncology ISRN Oncology used to inform clinical trials with a view to reducing costs and increasing efficiency. Mallet and de Pillis [8] and later de Pillis et al. [9] ex- plored a particular type of in silico model known as a hybrid cellular automata-partial differential equation (CA-PDE) model to describe the interactions between a growing tumor and the host immune response. A hybrid CA-PDE model combines the traditional continuum methods of applied mathematics, such as macroscale reaction-diffusion equa- tions describing chemical concentrations, with more mod- ern, individual, or grid-based automaton methods, which are used for describing individual cell-level phenomena. The hybrid CA-PDE modeling approach has been successfully used in the past to model tumor growth, chemotherapeutic treatment, and the effects of vascularization on a growing tumor [7, 10, 11, 14]. In Section 3 we discuss this model in some detail, explaining how the model is constructed as well as typical outputs of an in silico model of this type. To provide an example, consider the study of cell transfer therapy for metastatic melanoma patients of Rosenberg et al. [1]. The authors commented on the difficulty of deriv- ing meaningful results from human experiments because of the variations in cell types, tumor types, immune states, and more fundamentally the human subjects themselves. While Rosenberg et al. suggest a solution to such a problem is to treat the same patient in differing ways over a period of time, another more ethical and flexible, and less hazardous method is through the use of in silico models and experimentation. This approach was used in the model discussed in Section 3. pp There is a rich history of theoretical studies involving mathematical and computational approaches to studying cancer. Burton and Greenspan pioneered the mathematical modeling of tumor growth with models of growth dynamics explained as a problem of diffusion [2–5]. Since that time, theoretical studies of most aspects of tumor growth and re- lated processes have been investigated at least to some extent, using various different methodologies including differential equations, stochastic models, and cellular automata. 2. Methods—In Silico Trials While in vitro and in vivo models use actual biological ma- terials and/or actual animals to investigate hypotheses and, for example, predict effectiveness of treatment strategies, in silico models use specifically designed computer programs to mimic these “real” experimental environments and to con- duct computational experiments. There exist a number of different types of in silico model including differential equa- tion models that track changes in quantities over time and/or space, network models that trace lines of probabilistic causa- tion and/or correlation, discrete cellular automata- or indi- vidual-based models, and hybrids of all of these models. Rather than providing models of real biological phenomena and structures that have a basis in some sort of extracted tissue or a somehow related animal species, these in silico models are comprised of mathematical and computational representations such as formulae, equations, and/or com- puter programs. A key feature of such models is that they can be “parameterized” so that quantities or rates not known in the real world or which are specific to different experiments can be investigated via computational experiments, or as we dub them “in silico trials.” The concept of the in silico trial can be thought of as akin to clinical trials. Just as each patient in a clinical trial has their own set of characteristics such as height, age, and status with regard to smoking and alcohol consumption so too we can run the program of an in silico model multiple times with varied parameters to produce “computational patients” in an in silico trial. It is also becoming more common to find mathematical studies appearing in the cancer literature. Utley et al., for ex- ample, discuss improvement in survival rates resulting from postoperative chemotherapy for lung cancer patients [12]. They note that the marginal (5%) survival rate improvement due to chemotherapy may be outweighed for some patients by the morbidity due to the treatment and that further trials do not actually improve information provided to patients, but rather improve the certainty of that prediction. Utley et al. propose the use of a mathematical model, utilizing pa- tient-specific pathological cancer stage data combined with existing techniques, to arrive at better evidence for informing patients regarding their postoperative treatment choices. The development of in silico model is often a process of cross-disciplinary collaboration between cancer biologists and mathematicians or modelers. 1. Introduction Araujo and McElwain provide an excellent review of the mathemati- cal modeling work carried out up to middle of the last decade [6]. More recently, Alarc´on et al. [7], Mallet and coworkers [8, 9], and Ferreira et al. [10, 11] have used a new paradigm— that of spatiotemporal, stochastic models using hybrid cel- lular automata techniques—to represent “computational pa- tients” or “in silico experiments” in a new direction for cancer research. This experimental paradigm extends the traditional mathematical modeling of cancer to incorporate computa- tional simulations that are parameterized in such a way to represent different patients or different experiments. 2. Methods—In Silico Trials Generally, the initial stages involve the model builder obtaining an understanding of the tumor biology required for developing the in silico model. This will be a period of intense collaborative work involving discussions between all investigators and a review of the the- oretical and experimental literature. The next stage involves abstraction of biological information into a mathematical or computational form, that is, building the update rules. This requires the creation of mathematical representations of relevant micro level biological phenomena and mecha- nisms (such as rates and results of cell division, methods for In a study more at the preclinical stage of research, de Pillis et al. describe a differential equation-based model for the interactions between a growing tumor, natural killer cells, and CD8+ T cells of the host immune system [13]. With a view to understanding how the immune system assists in rejecting growing tumors, de Pillis et al. present mathemati- cal descriptions of key mechanisms in the immune response before fitting the model to data from published mouse and human studies. A parameter sensitivity analysis reveals the key role of a patient-specific variable and that the model may in fact provide a means to predict positive response of par- ticular patients to treatment. 3 ISRN Oncology representing distributions of chemical molecules, and inter- actions between antigen and antigen presenting cells) and the compilation of these into a macrolevel description of the real experimental situation. The type of cellular automata model considered here is executed as follows. The system is first initialized so that the computational representation presented in the cellular auto- mata grid matches some initial condition for the ensuing computational experiment. Next, a sequence of “time steps” is carried out such that the model time is incremented by a small amount at each step. Within each time step, every spatial location or element in the CA grid is investigated to identify its contents. Depending on the contents, an appro- priate update rule is applied which may involve the states of the neighboring elements. Updates are made throughout the grid, time is incremented, and the process continues. p Following the development of the update rules, the algo- rithm for the entire process is computerized usually employ- ing generic programming languages such as C++ or with mathematical software such as MATLAB. 2. Methods—In Silico Trials This algorithm allows for the solution of the in silico model and facilitates easy simulation of large numbers of experiments, that is, re- peated simulation of the model using many different param- eter sets in order to mimic running slightly different exper- iments in the laboratory. This could reflect, for example, an investigation of the effect of different quantities of gold nano- particles on effectiveness of radiotherapy or the effect of dif- ferent concentrations of chemotherapeutic treatments. To extend this model to allow in silico trials, the computer program for the algorithm described in the above paragraph is wrapped in a further program. This involves providing a collection of two or more (depending on the number of ex- periments or trials required) parameter value sets to the algo- rithm and running the algorithm once with each set. The output data, for example, cell counts over time, for each trial is exported to memory at the completion of each trial. While largely automated via the computer program, the simulation of the in silico model requires careful and contin- uous monitoring to ensure that computations converge (i.e., solutions are obtained rather than computational errors) and to make adjustments to investigations when results of interest are observed. 2.2. Development of Rules. As mentioned earlier, with regard to developing an accurate description of the biological proc- ess of interest, the specification of the update rules for a cel- lular automata-based in silico model is the most important part of the modelling process. To demonstrate this, consider the seemingly simple case of the movement of one cell to a neighboring location and the following increasingly complex but increasingly accurate rules. Following simulation of the in silico model, the results of the computational experiments are analyzed and inter- preted. This generally involves the use of custom-designed visualization of the resulting data. The investigators use the outputs of the model to determine what results are already useful for informing any associated experimental studies as well as what parts of the in silico model are deficient and re- quire refinement along with a follow-up round of in silico experiments. The whole process can be repeated, with refine- ment, as often as new information is required, and in general the costs of follow-up in silico experimentation decrease as the fundamental computational framework has already been developed. 2. Methods—In Silico Trials In the remainder of this section, we present an oversimplified and generic model along with the computa- tional algorithm to further illuminate this concept. Rule 1. If there is one or more empty CA elements surround- ing a cell, move to a randomly chosen empty element, other- wise, do not move. Rule 2. If there is one or more empty CA elements surround- ing a cell and moving to one would increase the cell’s satis- faction in some way, move to a randomly chosen element of this type, otherwise, do not move. 2.1. The CA Approach. A cellular automaton (CA) is a type of mathematical model, discrete in both space and time. Here we consider a two-dimensional CA, such as that which could be used to model the surface of the skin or possibly a petri dish, but note that three-dimensional models are simple, if computationally expensive, extensions of the same concepts. A two-dimensional CA consists of a lattice or grid of CA elements covering a region of space (see Figure 1). Applied in the biological context, each element is allowed to house one or more biological cells and, depending on the experimental situation being modeled, may also hold other matter such as molecules, debris, fluid, or bacteria. The cells in the CA elements are allowed to interact with one another via update rules. The set of update rules defines how the state of each element changes in response to its current state and the cur- rent state of its neighbors—the definition of these rules is the fundamental modelling stage in the development of the in silico model (see Figure 2). The accuracy of the model is heavily dependent on designing rules that adequately reflect the real interactions between cells. Rule 3. If there is one or more empty CA elements surround- ing a cell, consider moving to one of these locations with a probability that depends on factors such as cell adhesion levels, nutrient supply, and chemoattractants, otherwise, do not move. Each of these rules could be implemented in an in silico model as the determining factor regarding whether or not a cell moves. Clearly moving from Rule 1 to Rule 3, the amount of realism increases, but, simultaneously, the amount of information required to design the rule also increases. 2. Methods—In Silico Trials Rule 1 does not require any information about the cells of interest— the cell simply moves if it can, and the location it moves to is randomly chosen. On the other hand, Rule 3 requires that the modeler has some preexisting or obtainable understanding regarding how cells respond to chemoattractants, how cell adhesion affects motility, and what impact nutrient levels have on the decision of a cell to move from location to loca- tion. Thus we note that with more information about the bi- ological process, the modeler can construct more realistic ISRN Oncology ISRN Oncology ISRN Oncology Figure 1: A two-dimensional grid is imposed on a region of space of interest with cells of different types, molecules, debris, fluid, and/or bacteria housed in each element of the grid. Figure 1: A two-dimensional grid is imposed on a region of space of interest with cells of different types, molecules, debris, fluid, and/or bacteria housed in each element of the grid. Current state Update function Next state State of neighbours Figure 2: The transition from the current state to the next state for each element of the CA grid is determined only by its current state, that of its neighbors and the update rule. Figure 2: The transition from the current state to the next state for each element of the CA grid is determined only by its current state, that of its neighbors and the update rule. update rules, but at the same time, a lack of information by no means rules out in silico modeling. In fact, in silico models can yield rich information when they are used from the very early stages as part of hypothesis generation and testing when there is a dearth of biological information. using the Mallet and de Pillis model agreed qualitatively with the experimental results demonstrated by Zhang et al. [15], Schmollinger et al. [16], and Soiffer et al. [17]. While a laboratory model is usually designed to focus on a particular stage of a process or a specific event, in silico models can be designed to focus on arbitrarily small or large- scale phenomena. 2. Methods—In Silico Trials Mallet and de Pillis chose to focus on the early stages of tumor growth during which the tumor is ad- jacent to, but not yet infiltrated by, nutrient supplying vas- culature in order to allow for an investigation of the initial interactions between the immune system and the emerging tumor. The simple model incorporated a simplified immune system comprised of two cell types, namely, the natural killer (NK) cells of the innate immune system and the cytotoxic T lymphocytes (CTLs) of the specific immune system. A hybrid cellular automata and partial differential equation model was constructed with an aim to demonstrate the combined effects of the innate and specific immune systems on the growth of a two-dimensional representation of a growing tu- mor. This was accomplished by constructing a model with computerized cell behaviors built from descriptions in the experimental literature and findings of dynamic models of tumor—immune system interactions developed in the theo- retical literature such as the work of Kuznetsov and Knott [18] and de Pillis and Radunskaya [19, 20]. 3. An Example in Cancer Biology Mallet and de Pillis [8] presented a so-called “hybrid cellular automata model” of the interactions between the cells of a growing tumor and those of the host immune system. Mallet and de Pillis successfully designed a computational method for investigating the interactions between an idealized host immune system and a growing tumor. The simulated tumor growth experiments were found to be in qualitative agree- ment with both the experimental and theoretical literature. It was found that even with quite simple mathematical des- criptions of the biological processes and with an overly sim- plified description of the host immune system, the compu- tational model had the potential to produce the behavior observed in laboratory experiments including spherical and papillary tumor growth geometries, stable and oscillatory tumor growth dynamics, and the infiltration of the tumor by immune cells. It was also possible to show the dependence of these different morphologies on key model parameters re- lated to the immune response. Numerical solutions produced 5 ISRN Oncology ISRN Oncology 6 5 4 3 2 1 0 Figure 4: An example of two-dimensional tumor growth after 800 cell cycles, simulated using the Mallet and de Pillis in silico model. Red intensity indicates tumor cell density. The domain shown is approximately 10–20 mm square, and growth is over a time period of at least a year. 0 100 200 300 400 500 600 700 800 0 1 2 3 4 5 6 7 Tumor cell cycles Tumor cell count (×105) Figure 3: An example of growth curve produced by the Mallet and de Pillis in silico model showing total number of tumor cells over time for a tumor growing in the absence of immune response. 0 100 200 300 400 500 600 700 800 0 1 2 3 4 5 6 7 Tumor cell cycles Tumor cell count (×105) Tumor cell count (×105) Figure 4: An example of two-dimensional tumor growth after 800 cell cycles, simulated using the Mallet and de Pillis in silico model. Red intensity indicates tumor cell density. The domain shown is approximately 10–20 mm square, and growth is over a time period of at least a year. Figure 3: An example of growth curve produced by the Mallet and de Pillis in silico model showing total number of tumor cells over time for a tumor growing in the absence of immune response. 3. An Example in Cancer Biology that depended on some combination of nutrient levels, local immune response, and crowding due to the presence of other tumor cells. Natural killer cells were maintained at or near a “normal” level by recruitment from outside the domain of interest whenever the local density dropped too far below its equilibrium level. Both natural killer cells and cytotoxic T cells were able to lyse tumor cells, although CTLs could do so more than once and were able to attract other CTLs to the local area. CTLs were also subject to removal from the local region with a probability depending on the local tumor cell density. Mallet and de Pillis’ hybrid cellular automata model em- ployed a coupled deterministic-stochastic approach that had the benefit of being conceptually accessible as well as compu- tationally straightforward to implement. The authors used reaction-diffusion equations, to describe chemical species such as growth nutrients, and a cellular automata strategy to track the tumor cells and two distinct immune cell species. Together, these elements simulated the growth of the tumor and the interactions of the immune cells with the tumor growth. The rules used to represent these phenomena are devel- oped as approximations of reality and involve considering in- dividual events, such as an interaction between a cell on the periphery of a tumor and a natural killer cell, and attempting to quantify what happens as a result of this interaction. This act of quantifying is guided by accepted results in the exper- imental and theoretical literature, expert elicitation, and simple physical arguments. As mentioned in the previous section, the development of these rules is the most important step in model development. g The model tracked cells both through time and through space—a clear advantage over dynamic models that assume a spatially well-mixed population of cells, which is not often the case in reality. Unlike continuum-based spatiotemporal models, which are generally comprised entirely of partial dif- ferential equations, the hybrid cellular automata approach allows for the consideration of individual cell behavior and associated randomness, rather than applying a general rule to a collection of cells, as is the case with continuum models. The cellular automata approach is also very flexible in terms of its computational implementation. 3. An Example in Cancer Biology While the Mallet and de Pillis model considered only four cell species with an over- ly simplistic view of the immune system, it is easily modified to cater for the inclusion of more cell types or new chemical species. While the design and statement of all the CA rules are presented in the original paper, here we expand on the design of one of the rules to elucidate how such objects are con- structed. Take, for example, the individual cell level event of cell division. This process is extremely complex and involves countless subprocesses each with many participants. Just as an experimentalist in the laboratory does not consider each of these explicitly, we do not attempt to represent each of them in the computational model either. Instead, we distil what information is available in the literature and from col- laborators to arrive at a model representation of the chance that the event occurs given certain conditions. This distilled model representation is the cellular automata rule. The evolution of the cell species considered in the Mallet and de Pillis model proceeds according to a combination of probabilistic and deterministic rules, developed in an at- tempt to describe the phenomena considered important in the theoretical model. In particular, Mallet and de Pillis imposed a simplifying assumption to the host cells such that, other than their consumption of nutrients, they allow tumor cells to freely divide and migrate and were more or less passive bystanders to tumor growth. Tumor cells on the other hand were able to move, divide, die due to nutrient levels and die because of the immune response, each with a probability For the case of cell division, Mallet and de Pillis consider that given a tumor cell, the probability of division increases with the ratio of nutrient concentration to the number of tu- mor cells already present in the local region. Note that there 6 ISRN Oncology 6 4 5 3 2 1 0 −1 (a) 25 20 15 10 5 0 (b) Figure 5: Two-dimensional snapshots of a tumor exhibiting high levels of necrosis (a) and populations of immune cells that have infiltrated the tumor mass causing cell death (b). 5 (b) (b) Figure 5: Two-dimensional snapshots of a tumor exhibiting high levels of necrosis (a) and populations of immune cells that have infiltrated the tumor mass causing cell death (b). 3. An Example in Cancer Biology Mallet and de Pillis also presented a particularly inter- esting application of their model that produced qualitatively similar simulated tumors to the results of some recent ex- perimental studies of immune response to tumor growth. The experimental studies of Schmollinger et al. [16], Soiffer et al. [17], and Kuznetsov and Knott [18] discussed the rela- tionship between increased survival rates of cancer patients, tumor necrosis, and fibrosis, and the presence of intratu- moral T cells or infiltrated T lymphocytes. In Figures 5(a) and 5(b), immune cells are shown to have infiltrated a grow- ing tumor. In particular, the darker regions in Figure 5(a) are evidence of tumor necrosis while lighter regions of Figure 5(b) are indicative of high immune cell populations. These solution plots are similar to experimental results shown by Schmollinger et al. [16], Soiffer et al. [17], and Kuznetsov and Knott [18] where strings of immune cells are moving into the tumor, surrounding individual cells, and causing tumor cell necrosis. is no mention of subcellular signal processing and neither is there any consideration of macrolevel pressure fields. Instead, the chance of the occurrence of a cell division is con- densed into a consideration of whether or not there are suf- ficient nutrients nearby and whether or not the region is al- ready crowded with tumor cells. This rule is interesting because it also incorporates a sec- ond subrule—that of the placement of the daughter cell. The model dictates that the grid location upon which the daughter cell is placed depends upon the cells occupying the neighborhood of the mother cell. For example, a dividing cell with at least one host cell or necrotic space surrounding it will place its daughter cell randomly in one of those noncan- cerous locations and either destroy the host cell or simply replace the necrotic material. On the other hand, if all ele- ments around the dividing cell are filled with tumor cells, the daughter cell will be placed in the neighboring element con- taining the fewest tumor cells. The authors viewed this as one approach to modeling tumor cell crowding. The simulation results showed employee parameters for a compact tumor (in the absence of the immune system), low- level CTL recruitment, and low CTL death probability. 3. An Example in Cancer Biology We emphasize again that the same computer program is used to implement these simulations as those considered in the previous figures; varying system parameters is all that is re- quired to consider quite a different experiment when using the in silico modeling technology. In silico models such as that of Mallet and de Pillis can produce an array of different outputs. In this particular work, the authors focused on presenting growth curves and two- dimensional spatial snapshots in time of growing tumors that were compared with experimental results. Figures 3 and 4, for example, show a growth curve and two-dimensional snapshot of a tumor growing in the absence of the immune system. This result was used as a baseline to compare with ex- perimental and previous mathematical results prior to in- vestigating the effects of the immune system with this new model. Note, in Figure 3, the initially exponential growth phase (cycle 0–200), before a phase of linear growth (cycle 200–800). These growth characteristics mimic the growth rates described in the experimental work of Folkman and Hochberg [21] and mathematically by Greenspan [3]. Figure 4 is a snapshot in time (800 cell cycles) of the same simulation where we see a roughly circular tumor with a radius of about 200 cells growing steadily outward toward the sources of the nutrient. Higher tumor cell densities are seen at the periph- ery of the tumor while in the center, a necrotic core is begin- ning to form with some necrotic material already appearing. The example of an in silico model presented in this sec- tion employed a moderately complex, hybrid cellular auto- mata-partial differential equation methodology to describe interactions between the host immune system and a growing tumor. In the absence of a simulated immune system, the model was capable of reproducing both compact-circular and wild papillary tumor morphologies. Morphology change was directly related to the relative rates of consumption of the survival and mitosis nutrients by both tumor and host tissue cells, and the results presented correspond qualitatively with the experimental literature (such as Folkman and Hochberg [21]). When the model allowed for a simulated immune system, with different choices of T-lymphocyte recruitment and/or death parameters, oscillatory growth curves were 7 ISRN Oncology [4] H. P. Greenspan, “On the self inhibited growth of cell cul- tures,” Growth, vol. 38, no. 1, pp. 81–95, 1974. observed for nearly all parameter sets. 4. Conclusions In silico experimental modeling of cancer involves combining findings from biological literature with computer-based models of biological systems in order to conduct investiga- tions of hypotheses entirely in the computer laboratory. In this paper we have presented a discussion of the concept of in silico modeling and how in silico models are constructed in practice. We have presented an example of in silico modeling that is relevant to the study of cancer and discussed its appli- cation and use as a hypothesis-generating tool as a precursor to or in combination with traditional clinical and laboratory research. This type of computational tool, when used in transdisciplinary research teams, has the potential to allow researchers to refine their experimental programs with an aim to reducing costs and increasing research efficiency, and we advocate increased use of such strategies by research groups. [12] M. Utley, C. Paschalides, and T. Treasure, “Informing decisions concerning adjuvant chemotherapy following surgical resec- tion for non-small cell lung cancer: a mathematical modelling study,” Lung Cancer, vol. 53, no. 2, pp. 153–156, 2006. [13] L. G. de Pillis, A. E. Radunskaya, and C. L. Wiseman, “A vali- dated mathematical model of cell-mediated immune response to tumor growth,” Cancer Research, vol. 65, no. 17, pp. 7950– 7958, 2005. [14] A. A. Patel, E. T. Gawlinski, S. K. Lemieux, and R. A. Gatenby, “A cellular automaton model of early tumor growth and in- vasion: the effects of native tissue vascularity and increased anaerobic tumor metabolism,” Journal of Theoretical Biology, vol. 213, no. 3, pp. 315–331, 2001. [15] L. Zhang, J. R. Conejo-Garcia, D. Katsaros et al., “Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer,” The New England Journal of Medicine, vol. 348, no. 3, pp. 203– 213, 2003. [16] J. C. Schmollinger, R. H. Vonderheide, K. M. Hoar et al., “Melanoma inhibitor of apoptosis protein (ML-IAP) is a target for immune-mediated tumor destruction,” Proceedings of the National Academy of Sciences of the United States of America, vol. 100, no. 6, pp. 3398–3403, 2003. Acknowledgment The authors wish to thank Masoum Bagher Oskouei for assistance and discussions with regard to the second section of the paper. [17] R. Soiffer, T. Lynch, M. Mihm et al., “Vaccination with irradi- ated autologous melanoma cells engineered to secrete human granulocyte-macrophage colony-stimulating factor generates potent antitumor immunity in patients with metastatic melanoma,” Proceedings of the National Academy of Sciences of the United States of America, vol. 95, no. 22, pp. 13141–13146, 1998. 3. An Example in Cancer Biology Depending on the strength of the immune system recruitment and death para- meters, the tumor growth either increased without bound or resulted in destruction of the invasive growth. The model was also able to reproduce experimentally observed immune cell infiltration of growing tumors. [5] H. P. Greenspan, “On the growth and stability of cell cultures and solid tumors,” Journal of Theoretical Biology, vol. 56, no. 1, pp. 229–242, 1976. [6] R. P. Araujo and D. L. S. McElwain, “A history of the study of solid tumour growth: the contribution of mathematical modelling,” Bulletin of Mathematical Biology, vol. 66, no. 5, pp. 1039–1091, 2004. g g The different sets of parameter values used in the simula- tion of the Mallet and de Pillis model are the primary meth- od for computationally mimicking different strengths of im- mune systems of, for example, healthy individuals, capable of early tumor detection and destruction, and individuals in poor immune health, for whom tumors grow easily. In sum- mary, even though the update rules proposed in the Mallet and de Pillis model were relatively simple and the number of cell types considered was far from exhaustive, the authors created an in silico model that was able to produce results in qualitative agreement with both the experimental and theoretical literature and which could be improved upon to provide useful preclinical results of relevance for further model development for guiding experimental work related to various treatment and vaccination strategies. [7] T. Alarc´on, H. M. Byrne, and P. K. Maini, “A cellular auto- maton model for tumour growth in inhomogeneous environ- ment,” Journal of Theoretical Biology, vol. 225, no. 2, pp. 257– 274, 2003. [8] D. G. Mallet and L. G. de Pillis, “A cellular automata model of tumor-immune system interactions,” Journal of Theoretical Biology, vol. 239, no. 3, pp. 334–350, 2006. [9] L. G. de Pillis, D. G. Mallet, and A. E. Radunskaya, “Spatial tumor-immune modeling,” Computational and Mathematical Methods in Medicine, vol. 7, no. 2-3, pp. 159–176, 2006. [10] S. C. Ferreira, M. L. Martins, and M. J. Vilela, “Reaction- diffusion model for the growth of avascular tumor,” Physical Review E, vol. 65, no. 2, Article ID 021907, 8 pages, 2002. [11] S. C. Ferreira, M. L. Martins, and M. J. Vilela, “Morphology transitions induced by chemotherapy in carcinomas in situ,” Physical Review E, vol. 67, no. 5, Article ID 051914, 9 pages, 2003. References [1] S. A. Rosenberg, J. C. Yang, P. F. Robbins et al., “Cell transfer therapy for cancer: lessons from sequential treatments of a pa- tient with metastatic melanoma,” Journal of Immunotherapy, vol. 26, no. 5, pp. 385–393, 2003. [18] V. A. Kuznetsov and G. D. Knott, “Modeling tumor regrowth and immunotherapy,” Mathematical and Computer Modelling, vol. 33, no. 12-13, pp. 1275–1287, 2001. [2] A. C. Burton, “Rate of growth of solid tumours as a problem of diffusion,” Growth, Development and Aging, vol. 30, no. 2, pp. 157–176, 1966. [19] L. G. de Pillis and A. Radunskaya, “A mathematical tumor model with immune resistance and drug therapy: an optimal control approach,” Journal of Theoretical Medicine, vol. 3, no. 2, pp. 79–100, 2001. [3] H. P. Greenspan, “Models for the growth of a solid tumor by diffusion,” Studies in Applied Mathematics, vol. 52, pp. 317– 340, 1972. ISRN Oncology ISRN Oncology ISRN Oncology 8 [20] L. G. de Pillis and A. Radunskaya, “The dynamics of an opti- mally controlled tumor model: a case study,” Mathematical and Computer Modelling, vol. 37, no. 11, pp. 1221–1244, 2003. [21] J. H. Folkman and M. Hochberg, “Self regulation of growth in three dimensions,” Journal of Experimental Medicine, vol. 138, no. 4, pp. 745–753, 1973.

https://openalex.org/W4252821615

https://izvestiya.tinro-center.ru/jour/article/download/90/91

Russian

Distribution of organic matter, phytopigments and heterotrophic bacteria along the salinity gradient in the estuaries of small rivers of the Tatar Strait basin

Izvestiâ TINRO/Izvestiâ Tihookeanskogo naučno-issledovatelʹskogo rybohozâjstvennogo centra

cc-by

* Гаретова Людмила Александровна, кандидат биологических наук, старший научный сотрудник, e-mail: [email protected]; Левшина Светлана Ивановна, кандидат географи­ ческих наук, старший научный сотрудник, e-mail: [email protected]; Фишер Наталья Константиновна, кандидат биологических наук, старший научный сотрудник, e-mail: fisher@ ivep.as.khb.ru; Сиротский Сергей Егорович , кандидат биологических наук, заведующий лабо­ раторией; Шестеркин Владимир Павлович, кандидат географических наук, и.о. заведующего лабораторией, e-mail: [email protected]. Garetova Ludmila A., Ph.D., senior scientist, e-mail: [email protected]; Levshina Svetlana I., Ph.D., senior scientist, e-mail: [email protected]; Fisher Natalia K., Ph.D., senior scientist, e-mail: [email protected]; Sirotsky Sergey E., Ph.D., head of laboratory; Shesterkin Vladimir P., Ph.D., acting head of laboratory, e-mail: [email protected]. Известия ТИНРО Известия ТИНРО 2016 Том 184 Том 184 УДК [574.5:579.68](265.546) * Гаретова Людмила Александровна, кандидат биологических наук, старший научный сотрудник, e-mail: [email protected]; Левшина Светлана Ивановна, кандидат географи­ ческих наук, старший научный сотрудник, e-mail: [email protected]; Фишер Наталья Константиновна, кандидат биологических наук, старший научный сотрудник, e-mail: fisher@ ivep.as.khb.ru; Сиротский Сергей Егорович , кандидат биологических наук, заведующий лабо­ раторией; Шестеркин Владимир Павлович, кандидат географических наук, и.о. заведующего лабораторией, e-mail: [email protected]. G L d il A Ph D i i i il i @i khb L hi S l Л.А. Гаретова, С.И. Левшина, Н.К. Фишер, С.Е. Сиротский, В.П. Шестеркин* Институт водных и экологических проблем ДВО РАН, 680000, г. Хабаровск, ул. Дикопольцева, 56 РАСПРЕДЕЛЕНИЕ ОРГАНИЧЕСКОГО ВЕЩЕСТВА, ФИТОПИГМЕНТОВ И ГЕТЕРОТРОФНЫХ БАКТЕРИЙ ВДОЛЬ ГРАДИЕНТА СОЛЕНОСТИ В ЭСТУАРИЯХ МАЛЫХ РЕК БАССЕЙНА ТАТАРСКОГО ПРОЛИВА Л.А. Гаретова, С.И. Левшина, Н.К. Фишер, С.Е. Сиротский, В.П. Шестеркин* Институт водных и экологических проблем ДВО РАН, 680000, г. Хабаровск, ул. Дикопольцева, 56 Garetova Ludmila A., Ph.D., senior scientist, e-mail: [email protected]; Levshina Svetlana I., Ph.D., senior scientist, e-mail: [email protected]; Fisher Natalia K., Ph.D., senior scientist, e-mail: [email protected]; Sirotsky Sergey E., Ph.D., head of laboratory; Shesterkin Vladimir P., Ph.D., acting head of laboratory, e-mail: [email protected]. Введение Эстуарии — переходные зоны между морскими и пресными водами с выражен­ ным градиентом солености и изменчивым гидрологическим режимом. Они являются наиболее продуктивными зонами мирового океана и характеризуются наибольшим биоразнообразием (Telesh et al., 2013). В то же время эстуарии — зона транспорта химических загрязняющих веществ и природного материала из наземных экосистем в открытый океан (McLusky, Eliott, 2004). Одна из особенностей эстуариев — взаимное влияние реки и моря на все происходящие процессы, которое определяет двусторонний встречный характер основных потоков вещества и энергии. Экологическое состояние эстуариев зависит от геоморфологических, физических, химических и биологических процессов, а также антропогенного воздействия. Устойчивость эстуариев к различного рода воздействиям частично обусловлена их размерами. В последнее время экосистемы малых эстуариев все больше привлекают внимание исследователей (Барабанщиков, Колпаков, 2012; Callaway et al., 2014; Jickells et al., 2014; Pye, Blott, 2014). Предпо­ лагается, что даже небольшое вмешательство может оказывать большое влияние на эстуарии, имеющие относительно малые площади. В южную часть Татарского пролива впадает множество малых и средних рек. Современный облик устьевых областей многих из них начал формироваться в 40-е гг. прошлого века в период строительства железной дороги Комсомольск-Совгавань. Же­ лезнодорожная магистраль, опоясывая побережье, затронула устья многих малых рек (Мучке, Токи, Большая и Малая Дюанка, Чистоводный и др.). Строительство мостов через устья рек сопровождалось сужением внешних участков эстуариев, что привело к образованию полузамкнутых лагун или приустьевых озер. Нарушение естественного водообмена приводит к интенсификации процессов осадконакопления и, как следствие, к обмелению эстуариев, изменению динамического и теплового режимов. Накопление органического вещества (ОВ) в донных отложениях способствует эвтрофикации малых эстуариев. В функционировании эстуариев важное место занимают процессы образования и разложения ОВ. Потенциальными продуцентами ОВ в эстуариях являются фитоплан­ ктон и фитобентос, макрофиты (водоросли и травы), а также микроорганизмы, которые одновременно могут быть и продуцентами ОВ, и его деструкторами. Существенное значение в формировании пула ОВ в зонах смешения речных и морских вод имеет тер­ ригенное ОВ речного стока (Пересыпкин и др., 2011). В нерестово-нагульных лососевых озерах дополнительным сезонным источником ОВ становятся сненки производителей лососей (Лепская и др., 2005). Антропогенная составляющая ОВ в эстуариях зачастую представлена нефтепродуктами (Немировская, 2004). р ф р у ( р ) Исследования потоков ОВ и связанных с ними биотических компонентов экоси­ стем проводятся в эстуариях многих рек мира. На территории России наиболее изучены в этом отношении эстуарии рек Северного Ледовитого океана (Широколобова, 2009; Hessen et al., 2010; Немировская, 2011; Столяров, 2011; Holmes et al., 2011), эстуарии Балтийского моря (Алимов, Голубков, 2008; Экосистема ..., 2008). РАСПРЕДЕЛЕНИЕ ОРГАНИЧЕСКОГО ВЕЩЕСТВА, ФИТОПИГМЕНТОВ И ГЕТЕРОТРОФНЫХ БАКТЕРИЙ ВДОЛЬ ГРАДИЕНТА СОЛЕНОСТИ В ЭСТУАРИЯХ МАЛЫХ РЕК БАССЕЙНА ТАТАРСКОГО ПРОЛИВА Приведены данные по пространственному распределению органического вещества, его растворенной и взвешенной форм, углеводородов, пигментов фитопланктона (хлоро­ филлов «а», «b», «с», суммы каротиноидов), численности эколого-трофических групп бактериопланктона в эстуариях двух малых рек в период совпадения речной межени и высокого прилива. Показано влияние морфолого-гидрологических характеристик устьевых областей малых рек на величину градиента солености в переходных зонах река–море. Дана оценка качества водной среды и уровня углеводородного загрязнения эстуариев малых рек Мучке и Токи. у р р у Ключевые слова: эстуарии, растворенное, взвешенное органическое вещество, углеводороды, фитопигменты, хлорофилл «а», гетеротрофные, нефтеокисляющие бактерии. Garetova L.A., Levshina S.I., Fisher N.K., Sirotsky S.E., Shesterkin V.P. Distribution of organic matter, phytopigments and heterotrophic bacteria along the salinity gradient in the estu­ aries of small rivers of the Tatar Strait basin // Izv. TINRO. — 2016. — Vol. 184. — P. 219–235. organic matter, phytopigments and heterotrophic bacteria along the salinity gradient in the estu­ aries of small rivers of the Tatar Strait basin // Izv. TINRO. — 2016. — Vol. 184. — P. 219–235. Distribution of total organic matter and its dissolved and suspended forms was investigated in two small estuaries with different morphometric characteristics, as Muchke and Toki Lakes, in late July 2014 at the low tide phase. The time of survey was distinguished by low river dis­ charge and high tides. Content of all forms of organic matter and hydrocarbons was higher in the estuarine waters, as compared with fresh river water, regardless of salinity. The dissolved form prevailed (79.7–98.6 % of the total organic matter content) with concentrations of 5.2–35.7 mg/l. Chlorophyll a dominated (up to 87 %) among phytopigments in both estuaries that corresponded to active development of microalgae. Toki Lake was distinguished by higher productivity and microplankton abundance. On the contrary, allochthonous organic matter was more abundant in Muchke Lake, where interrelation between destructors (heterotrophic bacteria number) and 219 primary producers (Chl a concentration) was observed. Oil-oxidizing microorganisms were up to 60 % of the heterotrophic bacteria number. The hydrocarbons content exceeded the maximal permissible level: 3.6 and 4.5 MPL in Muchke and Toki Lakes, respectively. Both estuaries were subjected to chronic hydrocarbon pressure. Key words: chlorophyll a, dissolved organic matter, suspended organic matter, estuary, heterotrophic bacterium, oil-oxidizing bacterium, hydrocarbon, phytopigment. Введение На дальневосточ­ ном побережье России большое внимание уделяется эстуариям рек юга Приморского края (Семкин и др., 2012, 2015; Колпаков, 2013; Колпаков, Бегун, 2014; Чернова и др., 2014). Эстуарии малых рек Татарского пролива до настоящего времени остаются малоизученными. у Цель исследования — выяснение основных биогеохимических закономерностей распределения и состава ОВ, пигментов фитопланктона, численности эколого-тро­ 220 фических групп бактериопланктона в эстуариях малых рек Мучке и Токи бассейна Татарского пролива. Материалы и методы Малые реки Токи и Мучке протяженностью до 40 км берут начало на восточном макросклоне Северного Сихотэ-Алиня и впадают в Татарский пролив в районе пос. Ванино (рис. 1). Русла рек порожистые, с быстрым течением. Глубина рек в межень не превышает 0,3 м. Устьевая область р. Токи включает в себя одноименное озеро, сообщающееся с бухтой короткой (до 30 м) протокой шириной 12 м. Оз. Токи имеет площадь водного зеркала 0,23 км2, в отлив зона осушки составляет около 20 %. Оз. Мучке расположено в 1,1 км выше места впадения р. Мучке в одноименную бухту и имеет площадь водного зеркала 0,59 км2. В районе исследования приливы имеют неправильный полусуточный характер, высота 0,3–1,0 м. Во время прилива морские воды проникают в устья рек, во время отлива речные и морские воды срабатываются в бухты. Исследуемые водные объекты испытывают антропогенную нагрузку: русла пересекаются автомобильным и железнодорожным мостами, на водосборе р. Токи на­ ходится полигон хранения твердых отходов пос. Ванино, на южном берегу оз. Мучке находится нефтехранилище, а в бухте Мучке действует морской угольный терминал. Рис. 1. Карта-схема района исследования: 1 — станции отбора проб; 2 — зона осушки; 3 — железная дорога; 4 — автомобильная дорога; 5 — полигон хранения твердых отходов Fig. 1. Scheme of surveyed area: 1 — samplings; 2 — intertidal zone; 3 — railroad; 4 — autoroad; 5 — solid disposal tip Рис. 1. Карта-схема района исследования: 1 — станции отбора проб; 2 — зона осушки; 3 — железная дорога; 4 — автомобильная дорога; 5 — полигон хранения твердых отходов Fig. 1. Scheme of surveyed area: 1 — samplings; 2 — intertidal zone; 3 — railroad; 4 — autoroad; 5 — solid disposal tip Схема расположения гидролого-гидрохимических и микробиологических станций, выполненных в ходе изучения эстуариев, представлена на рис. 1. Съемка выполнялась 23 и 25 июля 2014 г. на 6 станциях каждого из эстуариев. В эстуарии р. Мучке наблю­ дениями были охвачены речная зона (ст. М0) и верхняя часть внутреннего эстуария (ст. М1–М5). В эстуарии р. Токи выполнялись речные станции (ст. Т0, Т1), собственно эстуарные (ст. Т2–Т4) и морская (ст. Т5, внешний эстуарий). Пробы воды отбирались из поверхностного горизонта (15–20 см) в фазу отлива. Непосредственно в момент отбора проб на всех станциях определяли температуру, соленость, рН при помощи кондуктометра WQC-24 (DKK-TOA Corporation, Япония). В полевой лаборатории осуществляли фиксацию проб для химических анализов, фильтрование проб для определения взвешенных веществ и пигментов фитопланктона. 221 Содержание взвешенных веществ (ВВ) в воде определяли гравиметрическим методом*. * РД 52.24.468-2005. Взвешенные вещества и общее содержание примесей в водах. Методика выполнения измерений массовой концентрации гравиметрическим методом. Ростов н/Д: Гидрохим. ин-т, 2007. 11 с. ** РД 52.24.476-2007. Массовая концентрация нефтепродуктов в водах. Методика вы­ полнения измерений ИК-фотометрическим методом. Ростов н/Д: Гидрохим. ин-т, 2007. 13 с. Материалы и методы Определение содержания общего и растворенного органического углерода (соответственно Собщ и Ср) осуществляли методом термокаталитического окисления с ИК-регистрацией (анализатор ТOC-Vе, Shimadzu, Япония). Количество взвешен­ ного органического углерода (Св) определяли по И.В. Тюрину с фотометрическим окончанием по Д.С. Орлову и Н.М. Гриндель (Бельчикова, 1975). Для отделения взве­ шенной фазы пробы воды фильтровали (под вакуумом) с использованием ядерных фильтров с размером пор 0,45 мкм. Содержание углеводородов (УВ) определяли ИК- спектрометрическим методом с использованием концентратомера КН-2 (Сибэкоприбор, Россия) по стандартной методике**. Фотосинтетические пигменты определяли согласно стандарту (ГОСТ 17.1.4.02-90) спектрофотометрическим методом на спектрофотометре Shimadzu модели UV-1650 PC. Микробиологические посевы производили в полевой лаборатории не позднее 1 ч после отбора проб согласно общепринятым в водной микробиологии методам (Куз­ нецов, Дубинина, 1989). Численность эвтрофной группы сапрофитных бактерий (СБ) определяли на стандартном рыбопептонном агаре (РПА). На РПА, разбавленном в 10 раз, определяли численность гетеротрофных бактерий (ГБ), потребляющих умеренные концентрации ОВ. Численность нефтеокисляющих бактерий (НОБ) выявляли на сре­ де Раймонда с нефтью, численность фенолрезистентных бактерий (ФРБ) — на среде РПА:10 с внесением фенола в концентрации 1 г/л. Результаты подсчета выражали в численности колониеобразующих единиц (КОЕ) микроорганизмов в 1 мл воды. В сравнительных целях использовали данные по общей численности бактериопланктона в отливных водах эстуария р. Токи, полученные в летний период 2009–2011 гг. мето­ дом люминесцентной микроскопии с использованием красителя диамина-4,6-фенил- 2-индолдихлоргидрата (DAPI) после фильтрации воды на черных поликарбонатных фильтрах «Nucleopor» с диаметром пор 0,2 мкм (Porter, Feig, 1980). Результаты и их обсуждение Токи была примерно на 1 оС выше в сравнении с морскими водами. Между станциями Т0 и Т3 (расстояние 1,75 км) температура воды увеличивалась более чем на 6 оС и на выходе водных масс из озера составляла 20,2 оС. Вследствие малых глубин оз. Токи (0,3–1,2 м) речные и морские воды хорошо перемешиваются и прогреваются. Значение рН воды р. Мучке было ниже (6,82), чем воды р. Токи (7,27). В зонах смешения речных и морских вод рН закономерно увеличивался по мере увеличения воды в р. Токи была примерно на 1 оС выше в сравнении с морскими водами. Между станциями Т0 и Т3 (расстояние 1,75 км) температура воды увеличивалась более чем на 6 оС и на выходе водных масс из озера составляла 20,2 оС. Вследствие малых глубин оз. Токи (0,3–1,2 м) речные и морские воды хорошо перемешиваются и прогреваются. воды в р. Токи была примерно на 1 оС выше в сравнении с морскими водами. Между станциями Т0 и Т3 (расстояние 1,75 км) температура воды увеличивалась более чем на 6 оС и на выходе водных масс из озера составляла 20,2 оС. Вследствие малых глубин оз. Токи (0,3–1,2 м) речные и морские воды хорошо перемешиваются и прогреваются. Значение рН воды р. Мучке было ниже (6,82), чем воды р. Токи (7,27). В зонах смешения речных и морских вод рН закономерно увеличивался по мере увеличения солености (Лысенко, 2011). При этом в оз. Мучке рН увеличивался незначительно (от 7,27 до 7,58), а в оз. Токи в градиенте солености от 0,28 до 15,34 епс (единиц практической солености) более существенно — от 6,57 до 7,35. Среднее значение рН в оз. Токи (6,9) было ниже, чем в пресной воде реки и морской воде одноименной бухты. Снижение рН обычно характеризует восстановительные условия, которые создаются при доминировании анаэробных процессов трансформации ОВ в донных отложениях. Главным морфологическим отличием, обусловливающим разницу в величине горизонтального градиента солености на обследованных участках эстуариев, является положение внутренних (озерных) частей эстуариев в пределах устьевых областей рек. Определенный вклад в величину градиента солености вносят площадь водного зеркала, глубина, очертания береговой линии. На обследованном участке эстуария р. Мучке (табл. 1) фронт солености в фазу отлива обнаруживался в устье реки (ст. М1) на расстоянии 2,1 км от взморья. По раз­ резу «река–море» между станциями М0 и М5 (расстояние 1,4 км) соленость на период съемки увеличивалась от 0 до 4,55 епс. Такой градиент солености в эстуарии р. Результаты и их обсуждение В соответствии с морфолого-гидрологическими признаками оба исследованных водных объекта относятся к озерно-лагунным эстуариям (Михайлов, Горин, 2012) (табл. 1). В зависимости от водного режима реки и высоты приливов разница в глу­ бинах на одних и тех же станциях может составлять до 1,5 м (Гаретова, Каретникова, 2010). При совпадении речной межени и высокого прилива в последней декаде июля 2014 г. наблюдалась пространственная неоднородность гидролого-гидрохимических показателей по разрезам «от реки к морю» в обоих водных объектах (табл. 1). В эсту­ арии р. Мучке максимальная глубина (3 м) отмечена на середине озера (ст. М3). По станциям внутреннего эстуария р. Токи глубины составляли 0,3–1,2 м при максимуме в устье р. Токи. у р Температура речных вод не превышала 15 оС, что характерно для предгорных рек Сихотэ-Алиня. На пресноводных станциях рек Мучке и Токи (М0 и Т0) выявлены различия по величине минерализации (на порядок). По составу воды гидрокарбонат­ но-кальциевые, маломинерализованные, с повышенной цветностью и содержанием железа, что характерно для рек, дренирующих болотистые водосборы (Форина, Ше­ стеркин, 2010). р ) Распределение температуры в эстуариях Мучке и Токи имеет общие черты, главной из которых является ее существенное увеличение в озерах по сравнению с речными водами. В эстуарии р. Мучке температура воды на отрезке между станциями М0 и М5 (расстояние 1,4 км) увеличивалась от 14,7 до 22,3 оС (на 7,6 оС). Температура 222 Таблица 1 Гидролого-гидрохимические параметры эстуариев рек Мучке и Токи в фазу отлива (июль 2014 г.) Table 1 Hydrological and hydrochemical parameters of the Muchke and Toki estuaries in the low tide phase (J l 2014) ц Гидролого-гидрохимические параметры эстуариев рек Мучке и Токи в фазу отлива (июль 2014 г.) Table 1 H d l i l d h d h i l t f th M hk d T ki t i i th l tid h ogical and hydrochemical parameters of the Muchke and Toki estuaries in the low tide phase (July 2014) Hydrological and hydrochemical parameters of the Muchke and Toki estuaries in the low tide phase (July 2014) Станция Место отбора проб L, км Глубина, м T, oC S, епс pH M, г/л М0 Русло р. Мучке (автомоб. мост) 2,50 0,25 14,7 0,00 6,82 0,029 М1 Устье р. Мучке 2,10 0,40 16,0 0,47 6,62 0,439 М2 Южный берег озера 1,50 0,45 22,6 2,97 7,37 2,770 М3 Середина озера 1,50 3,0 21,4 3,77 7,47 3,460 М4 Северный берег озера 1,50 1,0 22,9 3,91 7,35 3,580 М5 Выход из озера (железнодор. Результаты и их обсуждение мост) 1,10 0,70 22,3 4,55 7,58 4,130 Т0 Русло р. Токи (автомоб. мост) 2,0 0,40 14,5 0,00 7,27 0,280 Т1 Русло р. Токи 0,95 1,0 15,5 0,28 6,67 0,350 Т2 Устье р. Токи 0,45 1,20 17,1 0,47 6,57 0,520 Т3 Середина озера 0,25 0,52 20,8 10,58 7,07 9,010 Т4 Выход из озера (железнодор. мост) 0,03 0,30 20,2 15,34 7,35 12,650 Т5 Бухта Токи – 1,45 13,6 19,74 7,26 15,900 Примечание. L — расстояние от морской границы до станции отбора, S — соленость, М — минерализация. воды в р. Токи была примерно на 1 оС выше в сравнении с морскими водами. Между станциями Т0 и Т3 (расстояние 1,75 км) температура воды увеличивалась более чем на 6 оС и на выходе водных масс из озера составляла 20,2 оС. Вследствие малых глубин оз. Токи (0,3–1,2 м) речные и морские воды хорошо перемешиваются и прогреваются. Значение рН воды р. Мучке было ниже (6,82), чем воды р. Токи (7,27). В зонах смешения речных и морских вод рН закономерно увеличивался по мере увеличения солености (Лысенко, 2011). При этом в оз. Мучке рН увеличивался незначительно (от 7,27 до 7,58), а в оз. Токи в градиенте солености от 0,28 до 15,34 епс (единиц практической солености) более существенно — от 6,57 до 7,35. Среднее значение рН в оз. Токи (6,9) было ниже, чем в пресной воде реки и морской воде одноименной бухты. Снижение рН обычно характеризует восстановительные условия, которые создаются при доминировании анаэробных процессов трансформации ОВ в донных отложениях. воды в р. Токи была примерно на 1 оС выше в сравнении с морскими водами. Между станциями Т0 и Т3 (расстояние 1,75 км) температура воды увеличивалась более чем на 6 оС и на выходе водных масс из озера составляла 20,2 оС. Вследствие малых глубин оз. Токи (0,3–1,2 м) речные и морские воды хорошо перемешиваются и прогреваются. Значение рН воды р. Мучке было ниже (6,82), чем воды р. Токи (7,27). В зонах смешения речных и морских вод рН закономерно увеличивался по мере увеличения солености (Лысенко, 2011). При этом в оз. Мучке рН увеличивался незначительно (от 7,27 до 7,58), а в оз. Токи в градиенте солености от 0,28 до 15,34 епс (единиц практической солености) более существенно — от 6,57 до 7,35. Среднее значение рН в оз. Токи (6,9) было ниже, чем в пресной воде реки и морской воде одноименной бухты. Снижение рН обычно характеризует восстановительные условия, которые создаются при доминировании анаэробных процессов трансформации ОВ в донных отложениях. воды в р. Результаты и их обсуждение Мучке обусловлен слабым водообменном с морем за счет узости и относительно большой длины протоки (1,1 км), соединяющей озеро с морем, а также его вытянутостью по разрезу «река–море». р р у р р Распределение солености в эстуарной области р. Токи имеет иной характер по сравнению с эстуарием р. Мучке. Граница фронта солености в период съемки находи­ лась в речной зоне (ст. Т1) на расстоянии 0,95 км со стороны моря. Горизонтальный градиент солености более выражен по сравнению с эстуарием р. Мучке: изменение солености между станциями Т1 и Т4 (расстояние 0,92 км) составляло более чем 15 епс. Такая разница в величине градиента солености по сравнению с эстуарием р. Мучке обусловлена интенсивным водообменном с морем вследствие близости внутреннего эстуария р. Токи к взморью, малой его глубины и площади водного зеркала, а также за счет сжатости по разрезу «река–море». 223 Согласно теории строения и функционирования «маргинальных фильтров» (Лисицын, 2008), обследованный участок эстуария р. Мучке включает в себя гра­ витационную зону (S = 0,01–0,10 епс), где обычно вследствие замедления течения происходит осаждение песчано-алевритовых фракций, и физико-химическую зону (S = 0,1–5,0 епс), где при флоккуляции и коагуляции растворенные ОВ переходят во взвешенную форму. В эстуарии р. Токи, кроме указанных зон, рассматривается био­ логическая зона (S = 5–17 епс). Основным признаком работы маргинального фильтра является неконсервативное поведение веществ, которое проявляется в отклонении от линейной зависимости в графике «соленость–концентрация вещества» при разбавле­ нии речных вод морскими (Kaul, Froelich, 1984; Звалинский и др., 2005). р р ( р ) Содержание ВВ в эстуариях рек Мучке и Токи (табл. 2) варьирует в диапазоне 9,3–44,3 мг/л, что значительно выше по сравнению с аналогичными водными объектами северо-востока Приморья, где в летний период массовая концентрация ВВ составляет 1,5–10,7 мг/л (Луценко и др., 2014). Количество взвеси зависит от гидрологического режима, типа дренируемых почв и геоморфологических особенностей водосбора. В эстуариях рек Мучке и Токи основным источником взвеси являются речные воды с содержанием ВВ более 30 мг/л. Во внутреннем эстуарии р. Мучке содержание ВВ изменялось от 17,1 до 44,3 мг/л (в среднем 25,5 мг/л) при максимальном содержании на проточном участке (ст. М5) (табл. 2). В оз. Токи содержание взвеси колебалось от 13,3 до 24,3 мг/л (в среднем 17,1 мг/л) при максимальном количестве на границе фронта солености (ст. Т1). В обоих эстуариях отмечено снижение содержания ВВ на слабопроточных станциях (М2, М3 и Т2, Т3) по сравнению с речными водами и увеличение его содержания на мелководных проточных участках (ст. М5 и Т4) за счет взмучивания осадков. Результаты и их обсуждение Таблица 2 Содержание органических веществ в воде эстуариев рек Мучке и Токи, мг/л Table 2 Content of organic substances in water of the Muchke and Toki estuaries, mg/l Станция Дата отбора ВВ Собщ Ср Св УВ М0 23.07.14 36,3 6,0 5,2 0,9 0,16 М1 « 24,9 20,6 20,3 0,3 0,10 М2 « 18,1 22,1 20,5 1,6 0,17 М3 « 17,1 25,6 23,8 1,9 0,18 М4 « 23,3 37,0 35,8 1,3 0,22 М5 « 44,3 11,5 9,4 2,2 0,25 Т0 25.07.14 20,7 13,8 13,1 0,7 – Т1 « 30,6 9,0 7,1 1,8 0,10 Т2 « 13,3 9,2 7,6 1,6 0,34 Т3 « 13,7 13,1 11,0 2,2 0,27 Т4 « 24,3 15,2 12,8 2,4 0,21 Т5 « 9,3 11,3 10,0 1,3 0,22 Таблица 2 Содержание органических веществ в воде эстуариев рек Мучке и Токи, мг/л Table 2 Общий пул ОВ, соотношение в нем растворенного и взвешенного ОВ зависит от гидрологических условий, биологической продуктивности рассматриваемого водного объекта и биогеохимических условий его водосбора. Содержание общего органического углерода в воде р. Мучке (ст. М0) в 2 раза ниже, чем на пресноводном участке р. Токи (ст. Т0) (табл. 2). Во внутренней части эстуария р. Мучке среднее содержание Собщ составляло 21,5 мг/л и было выше, чем в оз. Токи (среднее 12,1 мг/л). Такие величины содержания Собщ значительно выше таковых для лагунных озер северо-восточного побережья Приморья, где в летний период они не превышали 11 мг/л (Луценко и др., 2014). К растворенному органическому веществу (РОВ) обычно относят всё ОВ воды, проходящее через мембраны фильтров с размером пор 0,45 мкм. Эта фракция ОВ включает не только собственно растворенные вещества, но и весь спектр коллоидов. Резервуар РОВ представляет собой смесь био- и геомолекул (геополимеров) с разной К растворенному органическому веществу (РОВ) обычно относят всё ОВ воды, проходящее через мембраны фильтров с размером пор 0,45 мкм. Эта фракция ОВ включает не только собственно растворенные вещества, но и весь спектр коллоидов. Резервуар РОВ представляет собой смесь био- и геомолекул (геополимеров) с разной 224 устойчивостью к биоразложению и биоассимиляции (Романкевич и др., 2008, 2009). В составе РОВ морской воды обнаруживаются органические соединения, входящие в состав бактерий, простейших, фито- и зоопланктона. Среди таких соединений при­ сутствуют белки, углеводы, липиды, пигменты, нуклеиновые кислоты, а также про­ дукты деструкции растений и животных суши (Романкевич и др., 2009), поступающие в море с речным стоком и при диффузии из донных отложений. РОВ обеспечивает энергетические потребности осмотрофов и играет основную роль в функционировании бактериальной петли в водных экосистемах. Результаты и их обсуждение устойчивостью к биоразложению и биоассимиляции (Романкевич и др., 2008, 2009). В составе РОВ морской воды обнаруживаются органические соединения, входящие в состав бактерий, простейших, фито- и зоопланктона. Среди таких соединений при­ сутствуют белки, углеводы, липиды, пигменты, нуклеиновые кислоты, а также про­ дукты деструкции растений и животных суши (Романкевич и др., 2009), поступающие в море с речным стоком и при диффузии из донных отложений. РОВ обеспечивает энергетические потребности осмотрофов и играет основную роль в функционировании бактериальной петли в водных экосистемах. р д Содержание РОВ (Ср) в эстуарии р. Мучке варьировало от 5,2 до 35,8 мг/л при средней величине 20,5 мг/л, что в 2 раза превышает таковое для эстуария р. Токи, где содержание Ср изменялось от 7,1 до 13,1 мг/л (среднее 10,2 мг/л). Следует отметить, что данные показатели в эстуариях рек Мучке и Токи значительно выше, чем в эсту­ ариях рек Уссурийского залива (Семкин и др., 2012) и северо-восточного побережья Японского моря (Луценко и др., 2014), где содержание РОВ в летний период составляло соответственно 4,8–7,7 и 1,9–8,6 мг/л. Смешение речных и морских вод, как правило, сопровождается интенсификацией биохимических процессов продукции и деструкции ОВ. На рис. 2 показано распреде­ ление Ср в градиентах солености исследованных эстуариев. В обоих случаях поведение Ср носило неконсервативный характер, но имело различия, обусловленные морфолого- гидрологическими особенностями обследованных водных объектов. 0 20 40 60 80 100 120 0 5 10 15 20 25 30 35 40 0 0,47 2,77 3,77 3,91 4,55 0 0,28 0,47 10,58 15,34 19,74 Ср/Собщ, % Ср, мг/л Соленость епс а б 1 2 Рис. 2. Изменение содержания растворенного органического вещества в зависимости от солености в эстуариях рек Мучке (а) и Токи (б): 1 — Ср; 2 — доля Ср от Собщ Fig. 2. Change of dissolved organic matter content in dependence on salinity in the Muchkе (a) and Toki (б) estuaries: 1 — DOC, mg/l; 2 — DOC, % of total organic matter 0 20 40 60 80 100 120 0 5 10 15 20 25 30 35 40 0 0,47 2,77 3,77 3,91 4,55 0 0,28 0,47 10,58 15,34 19,74 Ср/Собщ, % Ср, мг/л Соленость, епс а б 1 2 б а Соленость, епс Соленость, епс Рис. 2. Изменение содержания растворенного органического вещества в зависимости от солености в эстуариях рек Мучке (а) и Токи (б): 1 — Ср; 2 — доля Ср от Собщ Fig. 2. Результаты и их обсуждение Change of dissolved organic matter content in dependence on salinity in the Muchkе (a) and Toki (б) estuaries: 1 — DOC, mg/l; 2 — DOC, % of total organic matter В эстуарии р. Мучке (рис. 2, а) отклонение от линейности в сторону увеличения концентрации Ср наблюдалось в области солености от 0 до 3,91 епс. При относи­ тельно низком (5,2 мг/л) содержании Ср в русле р. Мучке такое масштабное (в 5 раз) увеличение содержания Ср в воде оз. Мучке, скорее всего, обеспечивается за счет как поверхностного стока с антропогенно-измененной территории водосбора, так и про­ дуцирования ОВ непосредственно в озере. Одновременно происходило увеличение доли Ср в составе Собщ. Снижение как абсолютного содержания Ср, так и его доли от Собщ на проточном участке эстуария р. Мучке при солености 4,55 епс, вероятнее всего, обусловлено коагуляцией коллоидной формы РОВ, что характерно для физико-хими­ ческой зоны маргинальных фильтров (Лисицын, 2008). 225 Вдоль градиента солености эстуария р. Токи распределение Ср имело свои осо­ бенности (рис. 2, б). В интервале солености от 0 до 0,28 епс содержание Ср и его доля в составе Собщ резко снижались, что характерно для зон смешения, где речные воды являются основными поставщиками РОВ терригенного генезиса. Вероятно, на момент исследования данный диапазон солености в эстуарии р. Токи соответствовал гравитаци­ онной зоне маргинального фильтра, где происходит уменьшение содержания как взве­ шенных, так и растворенных форм ОВ (Лисицын, 2008). При дальнейшем увеличении солености от 0,47 до 15,30 епс содержание Ср и его доля в составе Собщ увеличивались уже за счет РОВ, образующегося непосредственно в оз. Токи. Во внешнем эстуарии (S = 19,74 епс) содержание Ср снижается за счет разбавления морскими водами при сохранении высокой его доли (88,5 %) в составе Собщ. общ Источниками Св в водах эстуариев рек Мучке и Токи являются фитопланктон, бактерии, остатки древесных тканей, травянистых растений и макрофитов, а также такие гидрофобные соединения, как углеводороды. Содержание Св в речных водах существенно не различалось и составляло менее 1 мг/л. Во внутреннем эстуарии р. Мучке содержание Св колебалось от 0,3 до 2,2 мг/л (в среднем 1,4 мг/л), в оз. Токи — от 1,6 до 2,4 мг/л (в среднем 2,0 мг/л). Доля Св в составе ВВ составляла от 1,2 до 16,0 %. На слабопроточных мелководных участках эстуариев (ст. М2, М3 и Т2, Т3) при снижении содержания ВВ доля Св в его составе увеличивалась (табл. 2), что, вероятнее всего, обусловлено интенсификацией продук­ ционных процессов под влиянием солености и температуры. Результаты и их обсуждение В то же время наблюда­ емое на проточных станциях М5 и Т4 одновременное увеличение содержания ВВ и Св обусловлено отливным ресуспендированием, которое характерно для мелководных проточных участков эстуариев (Alvarez, Jones, 2002). р у у р ( , , ) Содержание УВ в воде оз. Мучке составляло 0,10–0,25 мг/л (в среднем 0,18), в оз. Токи — 0,10–0,34 мг/л (в среднем 0,23). В обоих случаях отмечается превышение ПДК соответственно в 3,6 и 4,5 раза. На рис. 3 показаны различия в распределении взвешенных ОВ в обследован­ ных эстуариях. В градиенте солености эстуария р. Мучке (0–4,55 епс) содержание УВ имеет прямую зависимость от концентрации Св, что указывает на их единое происхождение. На рис. 3 (а) видно, что Св и входящие в его состав УВ поступают в эстуарий с водами р. Мучке. При увеличении солености от 0 до 0,47 епс происходит одновременное снижение содержания Св и УВ вследствие их осаждения. Однако при дальнейшем увеличении солености содержание данных компонентов в воде эстуария не только не снижается, но и увеличивается. Во внутренней части эстуария р. Токи вдоль градиента солености 0,28–5,30 епс наблюдалось увеличение содержания Св и УВ (рис. 3, б). Уменьшение их содержания отмечалось только во внешнем эстуарии при солености 19,74 епс за счет разбавления морскими водами. Различия в распределении данных компонентов в эстуариях, вероятно, обуслов­ лены тем, что в оз. Мучке УВ представлены взвешенной, мало трансформированной фракцией, что характерно для свежего загрязнения нефтепродуктами (Немировская, Сивков, 2012). В оз. Токи доминируют растворенные, трансформированные УВ, имею­ щие преимущественно биогенный генезис и диффундирующие из донных отложений за счет малых глубин озера (Гаретова, 2014). Обычно среднее отношение РОВ к взвешенному ОВ как в океане, так и в озерах составляет 10 : 1 (Wetzel, 1984). По станциям эстуария р. Мучке данное соотношение варьирует в значительном диапазоне от 4,3 : 1,0 до 28,4 : 1,0, в эстуарии р. Токи — от 4 : 1 до 17 : 1. Значения отношений СР/СВ для внутренних эстуариев рек Мучке и Токи составляют соответственно 14 : 1 и 5 : 1, что может указывать на более высокую про­ дуктивность Токи. Главным критерием продуктивности водных экосистем является первичная про­ дукция как первоисточник энергии для гетеротрофных организмов. Сходство генези­ са пигментов и ОВ в водоеме и одновременное участие этих веществ в деструкции отражают синхронность гидрологических и биологических процессов, что придает 226 Рис. 3. Изменение содержания взвешенных компонентов в градиентах солености эстуариев рек Мучке (а) и Токи (б): 1 — ВВ; 2 — Св; 3 — УВ Fig. 3. Результаты и их обсуждение Change of suspended components content along salinity gradient in the Muchkе (a) and Toki (б) estuaries: 1 — suspended matter, mg/l; 2 — POC, mg/l; 3 — hydrocarbons, mg/l×10 0 0,5 1 1,5 2 2,5 3 3,5 4 0 5 10 15 20 25 30 35 40 45 50 0 0,47 2,77 3,77 3,91 4,55 0 0,28 0,47 10,58 15,34 19,74 УВ, мг/лх10 ВВ, Св, мг/л Соленость, епс а б 1 2 3 0 0,5 1 1,5 2 2,5 3 3,5 4 0 5 10 15 20 25 30 35 40 45 50 0 0,47 2,77 3,77 3,91 4,55 0 0,28 0,47 10,58 15,34 19,74 УВ, мг/лх10 ВВ, Св, мг/л а б 1 2 3 а Соленость, епс Рис. 3. Изменение содержания взвешенных компонентов в градиентах солености эстуариев рек Мучке (а) и Токи (б): 1 — ВВ; 2 — Св; 3 — УВ Fig. 3. Change of suspended components content along salinity gradient in the Muchkе (a) and Toki (б) estuaries: 1 — suspended matter, mg/l; 2 — POC, mg/l; 3 — hydrocarbons, mg/l×10 Соленость, епс Рис. 3. Изменение содержания взвешенных компонентов в градиентах солености эстуариев рек Мучке (а) и Токи (б): 1 — ВВ; 2 — Св; 3 — УВ Fig. 3. Change of suspended components content along salinity gradient in the Muchkе (a) and Toki (б) estuaries: 1 — suspended matter, mg/l; 2 — POC, mg/l; 3 — hydrocarbons, mg/l×10 Соленость, епс пигментным характеристикам фитопланктона значение интегральных экосистемных показателей (Бульон, 1983). Соотношения между различными пигментами фитоплан­ ктона позволяют судить о преобладании той или иной группы водорослей в водной среде. Так, основную массу морского фитопланктона составляют диатомовые и периди­ ниевые водоросли, которые содержат хлорофиллы «а» и «с». Фитопланктон лагунных озер восточного Сихотэ-Алиня представляет собой в основном комплекс диатомовых водорослей (Medvedeva, 2001). пигментным характеристикам фитопланктона значение интегральных экосистемных показателей (Бульон, 1983). Соотношения между различными пигментами фитоплан­ ктона позволяют судить о преобладании той или иной группы водорослей в водной среде. Так, основную массу морского фитопланктона составляют диатомовые и периди­ ниевые водоросли, которые содержат хлорофиллы «а» и «с». Фитопланктон лагунных озер восточного Сихотэ-Алиня представляет собой в основном комплекс диатомовых водорослей (Medvedeva, 2001). р ( ) Наиболее важным в реакциях фотосинтеза является хлорофилл «а» (хл. «а»). Его концентрация тесно связана с биомассой фитопланктона и зависит от комплекса факторов. В эстуарии р. Мучке концентрация хл. «а» варьировала от 1,21 до 7,18 мг/м3 (в среднем 3,15 мг/м3) (табл. 3). Для вод эстуария р. Результаты и их обсуждение Токи содержание хл. «а» изменялось в гораздо большем диапазоне (1,54–15,23 мг/м3), его среднее значение (7,38 мг/м3) было в 2 раза выше, чем в эстуарии р. Мучке. Распределение хл. «а» в обследованных эстуариях имело общие черты: как мак­ симальные концентрации хл. «а», так и максимальное его процентное содержание от суммы хлорофиллов были приурочены к озерным мелководьям (ст. М2, М5, Т3, Т4) и соответствовали температурным максимумам (рис. 4), что демонстрирует преимуще­ ственное влияние температуры на развитие фитопланктона и согласуется с литератур­ ными данными (Rice, Stewart, 2013). На выходе из оз. Мучке (ст. М5) содержание хл. «а» в прогретых осолоненных водах было в 3 раза выше, чем в пресной воде речного участка (ст. М0) (рис. 4, а). Интенсивное развитие фитопланктона наблюдалось в оз. Токи в интервале солености 0,45–15,34 епс (рис. 4, б). Увеличение концентрации хл. «а» при увеличении солености наблюдалось в эстуариях рек Артемовка и Шкотов­ ка (Японское море) в летний период (Семкин и др., 2012, 2015). Ранее (Almeida et al., 2002) было показано, что биомасса и продукция фитопланктона может увеличиваться вдоль градиента солености в направлении к внутренней части эстуария в 8 раз. В на­ ших исследованиях увеличение содержания фитопланктона (по концентрации хл. «а») во внутреннем эстуарии р. Токи по сравнению с морем составило более 9 раз, а по сравнению с пресными водами около 7 раз. 227 Таблица 3 Содержание фотосинтетических пигментов в фитопланктоне эстуариев рек Мучке и Токи Table 3 Content of photosynthetic pigments in phytoplankton of the Muchke and Toki estuaries Content of photosynthetic pigments in phytoplankton of the Muchke and Toki estuaries Станция Дата отбора S, епс Содержание пигментов, мг/м3 СК/Схл. «а» Хл. «а» Хл. «b» Хл. «с» Кар. М0 23.07.14 0,00 1,36 0,73 0,75 2,1 1,54 М1 « 0,47 3,46 1,16 1,10 3,4 0,98 М2 « 2,97 7,18 1,13 0,81 5,6 0,78 М3 « 3,77 1,21 0,83 0,47 2,5 2,07 М4 « 3,91 1,38 0,50 0,54 2,5 1,81 М5 « 4,55 4,33 1,01 0,88 6,0 1,38 Т0 25.07.14 0,00 2,19 1,08 0,65 3,0 1,37 Т1 « 0,28 2,47 1,17 0,82 2,9 1,17 Т2 « 0,47 2,21 0,82 0,75 2,6 1,18 Т3 « 10,58 15,23 0,09 2,18 15,1 0,99 Т4 « 15,34 14,79 0,33 1,93 14,3 0,96 Т5 « 19,74 1,54 0,32 0,19 1,8 1,16 Примечание. Кар. — каротиноиды, СК/Схл «а» — соотношение концентрации каротиноидов Примечание. Кар. — каротиноиды, СК/Схл. «а» — соотношение концентрации каротиноидов и концентрации хл. «а». Рис. 4. Результаты и их обсуждение Вместе с тем в бухте Токи при солености воды 19,74 епс вновь отмечалось увеличение доли хл. «b» за счет галофильных жгутиковых, т.е. маркируемые составом фитопигментов изменения в составе фитопланктона зависят от градиента солености в данных эстуариях (Muylaert et al., 2009; Khemakhem et al., 2010). Имеются данные о том, что в эстуариях реакции фитопланктона на перемешивание воды происходят быстрее, чем в открытом море, что влечет за собой более быструю смену разных слагаемых фотосинтезирующего аппарата (MacIntry et al., 2000). Рис. 5. Содержание фотосинтетических пигментов в биомассе фитопланктона: а — эстуа- рий р. Мучке; б — эстуарий р. Токи Fig. 5. Content of photosynthetic pigments in phytoplankton biomass: a — Muchke estuary; б — Toki estuary                   ɋɨɥɟɧɨɫɬɶɟɩɫ ɚɛ ɯɥɫ ɯɥE ɯɥɚ                   ɋɨɥɟɧɨɫɬɶ ɟɩɫ ɚɛ ɯɥɫ ɯɥE ɯɥɚ ɛ ɚ Рис. 5. Содержание фотосинтетических пигментов в биомассе фитопланктона: а — эстуа- рий р. Мучке; б — эстуарий р. Токи i C f h h i i i h l k bi hk Рис. 5. Содержание фотосинтетических пигментов в биомассе фитопланктона: а — эстуа- рий р. Мучке; б — эстуарий р. Токи Fig. 5. Content of photosynthetic pigments in phytoplankton biomass: a — Muchke estuary; б — Toki estuary р ф ф у рий р. Мучке; б — эстуарий р. Токи Fig. 5. Content of photosynthetic pigments in phytoplankton biomass: a — Muchke estuary; б — Toki estuary р р у ; у р р Fig. 5. Content of photosynthetic pigments in phytoplankton biomass: a — Muchke estua б — Toki estuary До настоящего времени общепринятых критериев для типологии эстуариев не разра­ ботано, поэтому оценка трофического статуса исследуемых водных объектов проводилась в соответствии с классификацией для пресноводных экосистем (Бульон, 1983; Сиренко, 1988). По содержанию хл. «а» были дифференцированы участки устьевых областей с раз­ личной продуктивностью фитопланктона. При концентрации хл. «а» в воде до 3 мг/м3 водный объект относится к олиготрофному типу с качеством вод «очень чистые». Та­ кими характеристиками обладают воды верхних участков русел рек Мучке и Токи. В оз. Мучке концентрация хл. «а» составляла в среднем около 5 мг/м3, что соответствует мезотрофному уровню водного объекта с качеством вод категории «чистые». Отливные воды из оз. Результаты и их обсуждение Изменение содержания хлорофилла «а» в эстуариях рек Мучке (а) и Токи (б): 1 — хл. «а»; 2 — температура; 3 — хл. «а», доля от суммы пигментов Fig. 4. Change of Chl a concentration in the Muchkе (a) and Toki (б) estuaries: 1 — Chl a, µg/l; 2 — water temperature, °C; 3 — Chl a, % of total phytopigments ɚɛ                   ɋɨɥɟɧɨɫɬɶɟɩɫ ɏɥɚɦɤɝɥɌɨɋ            Ⱦɨɥɹɯɥɚ    ɚɛ                   ɋɨɥɟɧɨɫɬɶ ɟɩɫ ɏɥɚɦɤɝɥɌɨɋ            Ⱦɨɥɹɯɥɚ    ɛ ɚ Рис. 4. Изменение содержания хлорофилла «а» в эстуариях рек Мучке (а) и Токи (б): 1 — хл. «а»; 2 — температура; 3 — хл. «а», доля от суммы пигментов Fig. 4. Change of Chl a concentration in the Muchkе (a) and Toki (б) estuaries: 1 — Chl a, µg/l; 2 — water temperature, °C; 3 — Chl a, % of total phytopigments Величина отношения концентрации каротиноидов к концентрации хл. «а» в эсту­ арии р. Мучке составляла 0,78–2,07, а в эстуарии р. Токи — 0,96–1,37 (табл. 3), что находится в пределах условного (< 2) порога, характеризующего активно функциони­ рующие фитопланктонные сообщества (Бульон, 1983). На это же указывает невысокая, не более 26,4 %, доля хл. «с» от суммы фотосинтетических пигментов в данных водных объектах, т.е. фитопланктон в переменных условиях солености эстуариев рек Мучке и Токи находится в активном состоянии. Значительные флюктуации процентного содержания фотосинтетических пигмен­ тов в эстуариях рек Мучке и Токи (рис. 5) свидетельствуют о качественных функци­ ональных изменениях в фитоценозе, скорее всего обусловленных изменением видо­ вого состава фитопланктеров в условиях переменной солености. Определение даже небольшого количества хл. «b» указывает на развитие мелких жгутиковых (зеленых) и сине-зеленых водорослей (Трофимова, 2007). Так, в эстуарии р. Мучке доля хл. «b» 228 варьировала от 12,4 до 33,0 % от суммы хлорофиллов. В градиенте солености эстуария р. Токи (0–19,7 епс) доля хл. «b» варьировала в гораздо большем диапазоне (от 0,5 до 27,5 % от суммы). Снижение содержания хл. «b» отмечалось в интервале солености от 0 до 10,58 епс и, вероятно, было обусловлено снижением доли пресноводных жгутиковых в сообществе микроводорослей. Результаты и их обсуждение Токи характеризуются как «умеренно загрязненные», а его трофический статус укладывается в диапазон концентраций 9–15 мг/м3, соответствующий слабо эвтрофному уровню. р ф у ур Для органических веществ важнейшие закономерности связаны с деятельностью бактериопланктона, поскольку в водных экосистемах одновременно с биосинтезом происходит микробиологическая трансформация автохтонных и аллохтонных веществ. Регулярные наблюдения за распределением бактериопланктона в эстуарии р. Токи проводились в 2009–2011 гг. при различных вариантах сочетания гидрологического ре­ жима реки и моря (Гаретова, 2013). Общая численность бактериопланктона колебалась в значительных пределах (от 0,40 до 2,16 млн кл/мл). Было установлено, что независимо от водного режима реки и моря они увеличивались во внутреннем эстуарии р. Токи по Для органических веществ важнейшие закономерности связаны с деятельностью бактериопланктона, поскольку в водных экосистемах одновременно с биосинтезом происходит микробиологическая трансформация автохтонных и аллохтонных веществ. Регулярные наблюдения за распределением бактериопланктона в эстуарии р. Токи проводились в 2009–2011 гг. при различных вариантах сочетания гидрологического ре­ жима реки и моря (Гаретова, 2013). Общая численность бактериопланктона колебалась в значительных пределах (от 0,40 до 2,16 млн кл/мл). Было установлено, что независимо от водного режима реки и моря они увеличивались во внутреннем эстуарии р. Токи по 229 сравнению с речным и морским участками. В градиенте солености эстуария р. Токи прослеживалась прямая зависимость между общей численностью бактериопланктона, температурой воды и содержанием РОВ (рис. 6). Рис. 6. Зависимость общей численности бактериопланктона (1) от температуры воды (2) и содержания РОВ (3) в градиенте солености эстуария р. Токи (июль 2010 г.) Fig. 6. Correspondence of the total bacterioplankton abundance change along salinity gradient (1) with changes of water temperature (2) and DOM (3) for the Toki estuary in July 2010 0 2 4 6 8 10 12 14 16 18 20 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0 0,87 2,82 5,56 20,7 Т, оС; РОВ, мгС/л ОЧБ, млн кл/мл Соленость, епс 1 2 3 0 2 4 6 8 10 12 14 16 18 20 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0 0,87 2,82 5,56 20,7 Т, оС; РОВ, мгС/л ОЧБ, млн кл/мл Соленость, епс 1 2 3 Рис. 6. Зависимость общей численности бактериопланктона (1) от температуры воды (2) и содержания РОВ (3) в градиенте солености эстуария р. Токи (июль 2010 г.) Fig. 6. Correspondence of the total bacterioplankton abundance change along salinity gradient Рис. 6. Зависимость общей численности бактериопланктона (1) от температуры воды (2) и содержания РОВ (3) в градиенте солености эстуария р. Токи (июль 2010 г.) Fig. 6. Результаты и их обсуждение Correspondence of the total bacterioplankton abundance change along salinity gradient (1) with changes of water temperature (2) and DOM (3) for the Toki estuary in July 2010 щ р ( ) р ур д ( ) и содержания РОВ (3) в градиенте солености эстуария р. Токи (июль 2010 г.) Fig. 6. Correspondence of the total bacterioplankton abundance change along salinity gradient (1) with changes of water temperature (2) and DOM (3) for the Toki estuary in July 2010 р ( ) р у р р ( ) Fig. 6. Correspondence of the total bacterioplankton abundance change along salinity gradient h changes of water temperature (2) and DOM (3) for the Toki estuary in July 2010 Морфологический состав бактериопланктона в зависимости от солености воды раз­ личается. В воде р. Токи доминируют крупные палочки, в водах с умеренной соленостью (3–6 епс) палочки и кокки содержатся в приблизительно равных количествах, при увели­ чении солености более 15 епс увеличивается морфологическое разнообразие бактерий: появляются мелкие палочки, изогнутые палочки, длинные тонкие палочки, очень мелкие кокки. Известно, что кокковые формы преимущественно используют легкоусвояемое, а палочковидные — трудноусвояемое ОВ, а изменение соотношения структурных групп бактерий является показателем микробной сукцессии (Теплинская, 2006). р р у ( ) Численность и соотношение культивируемых эколого-трофических групп бакте­ рий являются важным показателем состояния бактериопланктонного сообщества, его способности к деструкции различных органических веществ. Группа гетеротрофных бактерий, как правило, составляет основу автохтонного микробного сообщества в водных экосистемах. В летнюю межень 2014 г. числен­ ность культивируемых ГБ в воде рек Мучке и Токи (ст. М0 и Т0) не превышала 10 тыс. КОЕ/мл (табл. 4), что является типичным для холодноводных малых рек Сихотэ-Алиня в летний период (Гаретова, 2009). Средняя численность ГБ в оз. Токи была в 2 раза выше, чем в оз. Мучке, что, вероятнее всего, обусловлено чрезвычайной мелководностью оз. Токи. Здесь сочетаются два основных фактора, регулирующих бактериальную активность, — наличие субстрата для развития бактерий (доступное ОВ) и более высокие температуры воды. Группа гетеротрофных бактерий, как правило, составляет основу автохтонного микробного сообщества в водных экосистемах. В летнюю межень 2014 г. числен­ ность культивируемых ГБ в воде рек Мучке и Токи (ст. М0 и Т0) не превышала 10 тыс. КОЕ/мл (табл. 4), что является типичным для холодноводных малых рек Сихотэ-Алиня в летний период (Гаретова, 2009). Средняя численность ГБ в оз. Токи была в 2 раза выше, чем в оз. Мучке, что, вероятнее всего, обусловлено чрезвычайной мелководностью оз. Токи. Результаты и их обсуждение Здесь сочетаются два основных фактора, регулирующих бактериальную активность, — наличие субстрата для развития бактерий (доступное ОВ) и более высокие температуры воды. В целом распределение отдельных эколого-трофических групп бактериопланкто­ на в обследованных эстуариях имеет тенденцию к увеличению численности в мезога­ линных зонах внутренних эстуариев по сравнению с граничными зонами в 2–3 раза, что согласуется с распределением общей численности бактериопланктона в эстуарии р. Токи. Аналогичные данные были получены при исследовании бактериопланктонных сообществ эстуарных областей Баренцева моря (Широколобова, 2009). На проточных участках эстуария р. Мучке (ст. М0 и М5) низкая величина отноше­ ния численности ГБ/СБ (0,98 и 0,35) показывает (табл. 4), что здесь ОВ находится на начальных этапах деструкции с участием копиотрофов. Во внутренней части эстуария преобладают деструкционные процессы с преимущественным участием ГБ, ассими­ лирующих легкоокисляемое ОВ (ГБ/СБ = 5,3–10,5). 230 Таблица 4 Состав микробных сообществ эстуариев рек Мучке и Токи Table 4 Microbial communities composition in the Muchke and Toki estuaries Станция Дата отбора проб Численность микроорганизмов, КОЕ/мл ГБ/СБ ГБ СБ ФРБ НОБ М0 23.07.14 1,2 ˑ 103 1,1 ˑ 103 2,70 ˑ 102 1,1 ˑ 103 0,98 М1 « 8,5 ˑ 103 1,2 ˑ 103 9,10 ˑ 102 5,7 ˑ 103 6,97 М2 « 9,6 ˑ 103 1,8 ˑ 103 1,30 ˑ 102 8,0 ˑ 103 5,29 М3 « 7,1 ˑ 103 0,7 ˑ 103 0,20 ˑ 102 2,3 ˑ 103 10,50 М4 « 6,3 ˑ 103 3,3 ˑ 103 0,30 ˑ 102 1,5 ˑ 103 1,90 М5 « 1,0 ˑ 103 2,5 ˑ 103 0,20 ˑ 102 0,9 ˑ 103 0,35 Т0 25.07.14 6,1 ˑ 103 0,5 ˑ 103 1,00 ˑ 102 4,0 ˑ 103 13,55 Т1 « 12,0 ˑ 103 1,2 ˑ 103 2,10 ˑ 102 5,5 ˑ 103 10,29 Т2 « 23,9 ˑ 103 1,5 ˑ 103 3,80 ˑ 102 5,2 ˑ 103 16,40 Т3 « 7,7 ˑ 103 1,6 ˑ 103 0,05 ˑ 102 0,9 ˑ 103 4,70 Т4 « 9,2 ˑ 103 1,7 ˑ 103 0,25 ˑ 102 0,6 ˑ 103 5,35 Т5 « 0,6 ˑ 103 1,0 ˑ 103 Не обн. 0,3 ˑ 103 0,56 Microbial communities composition in the Muchke and Toki estuaries В микробном сообществе пресных и слабосоленых вод эстуария р. Токи ведущая роль в деструкции ОВ принадлежит ГБ (ГБ/СБ = 10,3–16,4). С увеличением солености активизируется сапротрофная группа бактериопланктона (ГБ/СБ = 5,35–0,56). В эстуарии р. Мучке в диапазоне солености от 0 до 3,9 епс наблюдалась прямая зависимость между численностью ГБ и содержанием хл. Результаты и их обсуждение «а» (рис. 7, а). Вероятно в составе РОВ эстуария р. Мучке преобладает трудноокисляемое терригенное ОВ, по­ этому ведущую роль источника питания для бактериопланктона здесь приобретает ОВ, синтезируемое фитопланктоном, что согласуется с литературными данными (Becquevort et al., 2002; Revilla et al., 2002). 231 Рис. 7. Распределение компонентов в воде эстуариев рек Мучке (а) и Токи (б): 1 — РОВ; 2 — численность гетеротрофных бактерий; 3 — хлорофилл «а» Fig. 7. Spatial distribution of DOM, mgС/l (1), number of heterotrophic bacteria, CFU/ml (2), and Chl a concentration, μg/l (3) in the Muchke (a) and Toki (б) estuaries 0 5 10 15 20 25 30 0 5 10 15 20 25 30 35 40 0 0,47 2,97 3,77 3,91 4,55 0 0,28 0,47 10,58 15,34 19,74 Хл. "а", мкг/л; ГБ, тыс. КОЕ/мл РОВ, мгС/л Соленость,епс а б 1 3 2 0 5 10 15 20 25 30 0 5 10 15 20 25 30 35 40 0 0,47 2,97 3,77 3,91 4,55 0 0,28 0,47 10,58 15,34 19,74 Хл. "а", мкг/л; ГБ, тыс. КОЕ/мл РОВ, мгС/л Соленость,епс а б 1 3 2 б а Соленость,епс Рис. 7. Распределение компонентов в воде эстуариев рек Мучке (а) и Токи (б): 1 — РОВ; 2 — численность гетеротрофных бактерий; 3 — хлорофилл «а» Fig. 7. Spatial distribution of DOM, mgС/l (1), number of heterotrophic bacteria, CFU/ml (2), and Chl a concentration, μg/l (3) in the Muchke (a) and Toki (б) estuaries 231 В эстуарии р. Токи уменьшение содержания РОВ сопровождается увеличением численности ГБ — главных потребителей РОВ (рис. 7, б). Вероятнее всего, в таком мелководном озере в гетеротрофную пищевую цепь активно включается РОВ, образую­ щееся при разложении остатков макрофитов в донных отложениях и диффундирующее в воду за счет малых глубин. Здесь РОВ, продуцируемое фитопланктоном, приобре­ тает второстепенное значение. На примере малых эстуариев Приморья показано, что только за счет макрофитов (без учета других групп первичных продуцентов) обеспе­ чивается уровень продуцирования ОВ до 902,1 гС/м2 (Колпаков, 2013). Известно, что с увеличением трофии водных объектов величина потоков ОВ из донных отложений увеличивается (Белкина, 2011). Численность фенолрезистентных бактерий в исследованных водных объектах была невелика, не более 1 тыс. КОЕ/мл, что указывает на отсутствие хронического загрязнения эстуариев соединениями фенольного ряда. Иная картина наблюдалась в отношении численности нефтеокисляющих бактерий. Результаты и их обсуждение Их содержание в сообществах ГБ рассматриваемых эстуариев достигало 97,7 %, что значительно превышает установленный для незагрязненных углеводородами вод по­ казатель 10 % (Патин, 2001), это может быть обусловлено регулярным поступлением УВ и их стимулирующим влиянием на развитие нефтеокисляющей способности микро­ организмов. Формирование повышенных концентраций УВ в водной среде эстуариев, вероятнее всего, обеспечивает поверхностный сток с автомобильной трассы, территории нефтехранилища на берегу оз. Мучке, территории полигона хранения твердых отходов, расположенного на правом берегу р. Токи. Дополнительный вклад осуществляется за счет диффузии УВ из донных отложений, что согласуется с данными по содержанию и распределению УВ в воде эстуариев рек Мучке и Токи. Заключение Проведена сравнительная оценка распределения абиотических и биотических компонентов в эстуариях малых рек при равном влиянии гидрологического режима и различном пространственным расположением фронта солености. Различия в вели­ чине градиента солености внутренних эстуариев рек Мучке и Токи главным образом определяются их удаленностью от взморья, влияющей на водообмен с морем, а также величиной площади водного зеркала, глубиной, очертаниями береговой линии. На момент совпадения речной межени и высокого прилива в июле 2014 г. обследованный участок эстуария р. Мучке характеризовался олигогалинной зоной с градиентом со­ лености 0–4,55 епс. Величина градиента солености в эстуарии р. Токи составляла 0–19,47 епс (олиго- и мезогалинная зоны). Распределение форм ОВ (Собщ, Ср, Св, УВ) вдоль градиента солености в обследо­ ванных эстуариях характеризуется «неконсервативным» поведением, выражающимся в увеличении содержания данных компонентов во внутренних частях эстуариев за счет увеличения продуцирования автохтонного ОВ, накопления терригенного ОВ, диффузии различных веществ из донных отложений. Органическое вещество в от­ ливных водах эстуариев рек Мучке и Токи находится в основном в растворенной форме (79,7–98,6 % от Собщ). общ Роль фито- и бактерипланктона как источников ОВ в биологической зоне эстуария р. Токи возрастает за счет благоприятного для их развития сочетания гидрологических условий (глубина, температура, соленость). В целом по показателям обилия микроплан­ ктона (концентрация хл. «а», численность микроорганизмов), в 2 раза превышающим таковые в оз. Мучке, оз. Токи характеризуется большей продуктивностью. Прямая зависимость между численностью деструкторов (ГБ) и активностью первичных про­ дуцентов (хл. «а») проявлялась лишь в водных массах оз. Мучке с высоким фоном аллохтонной органики. По содержанию Собщ в воде (среднее 23,4 мг/л) оз. Мучке соответствовало высокоэвтрофному уровню, а по уровню продуцирования фитопланктона, опре­ деляемому по концентрации хл. «а», оно оценивалось мезотрофным уровнем 232 (средняя концентрация хл. «а» 4 мг/м3). Такие различия в оценке трофического статуса, вероятнее всего, обусловлены доминированием в воде эстуария р. Мучке аллохтонного ОВ. Слабый водообмен с морем приводит к накоплению органики в воде и донных отложениях, увеличению внутренней нагрузки и дальнейшему эвтрофированию оз. Мучке. (средняя концентрация хл. «а» 4 мг/м3). Такие различия в оценке трофического статуса, вероятнее всего, обусловлены доминированием в воде эстуария р. Мучке аллохтонного ОВ. Слабый водообмен с морем приводит к накоплению органики в воде и донных отложениях, увеличению внутренней нагрузки и дальнейшему эвтрофированию оз. Мучке. Оценки экологического состояния оз. Токи по всем показателям совпадают. Воды оз. Токи характеризуются как «умеренно загрязненные» (Собщ = 9,0–15,2 мг/л), а содержание хл. «а» (9–15 мг/м3) соответствует эвтрофному уровню. Интенсивный водообмен с морем лагунной экосистемы эстуария р. Заключение Токи способствует аутвеллин­ гу (выносу) органических веществ и биотических компонентов в море, снижению внутренней биогенной нагрузки и снижению антропогенного загрязнения эстуария. Судя по концентрациям УВ (более 4 ПДК) в водных потоках из внутренних частей эстуариев рек Мучке и Токи, а также по высоким показателям численности нефтеокисляющих бактерий, благодаря работе маргинального фильтра НУ погло­ щаются далеко не полностью и поступают в прибрежную зону Татарского пролива. Список литературы Алимов А.Ф., Голубков С.М. Изменения в экосистемах восточной части Финского залива // Вестн. РАН. — 2008. — Т. 78, № 3. — С. 223–230. Барабанщиков Е.И., Колпаков Н.В. Состав, распределение, динамика и продукция зоо­ планктона эстуариев Приморья // Водные биологические ресурсы северной части Тихого океана: состояние, мониторинг, управление : мат-лы Всерос. науч. конф., посвящ. 80-летнему юбилею ФГУП «КамчатНИРО». — Петропавловск-Камчатский : КамчатНИРО, 2012. — С. 448–460. Белкина Н.А. Роль донных отложений в процессах трансформации ОВ и биогенных элементов в озерных экосистемах // Тр. КНЦ РАН. — 2011. — № 4. — С. 35–45. Бельчикова Н.П. Определение гумуса почвы по методу И.В. Тюрина // Агрохимические оды исследования почв. — М. : Наука, 1975. — С. 56–62. у Бульон В.В. Первичная продукция планктона внутренних водоемов : моногр. — М. : Наука, 1983. — 150 с. Гаретова Л.А. Биогенные углеводороды и их влияние на оценку нефтяного загрязнения (на примере устьевой области реки Токи, Хабаровский кр.) // Чтения памяти В.Я. Леванидова. — Вла­ дивосток : Дальнаука, 2014. — Вып. 6. — С. 152–158. у Гаретова Л.А. Характеристика стока органических веществ и бактериопланктона в устьевой области малой реки Токи (Татарский пролив) // Жизнь пресных вод. — Владивосток : Дальнаука, 2013. — Вып. 1. — С. 174–186. Д у Гаретова Л.А. Эколого-микробиологическая оценка лососевых рек Хабаровского края // Изв. ТИНРО. — 2009. — Т. 157. — С. 175–181. Гаретова Л.А., Каретникова Е.А. Гидрохимические и микробиологические показатели в оценке экологического состояния малых эстуарных систем (на примере оз. Токи, Татарский пролив) // Изв. ТИНРО. — 2010. — Т. 162. — С. 294–305. Звалинский В.И., Недашковский А.П., Сагалаев С.Г. и др. Биогенные элементы и первичная продукция в эстуарии реки Раздольной (Амурский залив Японского моря) // Биол. моря. — 2005. — Т. 31, № 2. — C. 107–116. р Колпаков Н.В. Продукция макрофитов в эстуариях рек Приморья // Изв. ТИНРО. — 2013. — Т. 174. — С. 135–148. Колпаков Н.В., Бегун А.А. Состав и распределение микроводорослей в эстуарии реки Суходол (Уссурийский залив, залив Пера Великого) в осенний период. 1. Фитопланктон // Изв. ТИНРО. — 2014. — Т. 176. — С. 115–126. Кузнецов С.И., Дубинина Г.А. Методы изучения водных микроорганизмов : моногр. — Наука, 1989. — 228 с. Лепская Е.В., Уколова Т.К., Калинина И.Б., Свириденко В.Д. Органическое вещество озер Курильское и Паланское (Камчатка), его связь с элементами биоты // Чтения памяти В.Я. Леванидова. — Владивосток : Дальнаука, 2005. — Вып. 3. — С. 199–213. Лисицын А.П. Список литературы Маргинальные фильтры и биофильтры Мирового океана // Океанология на старте XXI века. — М. : Наука, 2008. — С. 159–224. Луценко Т.Н., Чернова Е.Н., Лысенко Е.В. Органическое вещество малых озер Северо- Востока Приморья // Чтения памяти В.Я. Леванидова. — Владивосток : Дальнаука, 2014. — Вып. 6. — С. 408–414. 233 Лысенко Е.В. Макрокомпонентный состав вод ряда озер восточного Сихотэ-Алиня // 6-я регион. школа-семинар молодых ученых, аспирантов и студентов «Территориальные ис­ следования: цели, результаты, перспективы» : тез. докл. — Биробиджан, 2011. — С. 31–33. р у р р Михайлов В.Н., Горин С.Л. Новые определения, районирование и типизация устьевых областей рек и их частей — эстуариев // Вод. ресурсы. — 2012. — Т. 39, № 3. — С. 243–257. р у р р ур Немировская И.А. Особенности распределения углеводородов в дельте Северной Двины во время половодья // Геохимия. — 2011. — № 8. — С. 863–874. Немировская И.А. Углеводороды в океане (снег–лед–вода–взвесь–донные осадки) : моногр. — М. : Науч. мир, 2004. — 328 с. Немировская И.А., Сивков В.В. Особенности распределения углеводородов в юго-вос­ точной части Балтийского моря // Океанол. — 2012. — Т. 52, № 1. — С. 40–53. Немировская И.А., Сивков В.В. Особенности распределения углеводородов в юго-вос­ точной части Балтийского моря // Океанол. — 2012. — Т. 52, № 1. — С. 40–53. Патин С.А. Нефть и экология континентального шельфа : моногр. — М. : ВНИРО, 2001 247 с Патин С.А. Нефть и экология континентального шельфа : моногр. — М. : ВНИРО, 2001. — 247 с. Пересыпкин В.И., Смуров А.В., Шульга Н.А. и др. Состав органического вещества воды, взвеси и донных осадков залива Нячанг (Вьетнам, Южно-Китайское море) // Океанол. — 2011. — Т. 51, № 6. — С. 1020–1029. Романкевич Е.А., Ветров А.А., Пересыпкин В.И. Органическое вещество Мирового океана // Геология и геофизика. — 2009. — Т. 50, № 4. — С. 401–411. Романкевич Е.А., Ветров А.А., Пересыпкин В.И. Цикл углерода в современном оке­ ане и актуальные проблемы биогеохимии // Океанология на старте 21-го века. — М. : Наука, 2008. — С. 78–107. Семкин П.Ю., Тищенко П.Я., Тищенко П.П. и др. Особенности гидрохимии эстуа­ риев рек Артёмовки и Шкотовки (Уссурийский залив, Японское море) летом 2011 г. // Изв. ТИНРО. — 2012. — Т. 171. — С. 267–284. Семкин П.Ю., Тищенко П.Я., Ходоренко Н.В., Звалинский В.И. Продукционно-де­ струкционные процессы в эстуариях рек Артемовка и Шкотовка (Уссурийский залив) в летний сезон // Вод. ресурсы. — 2015. — Т. 42, № 3. — С. Список литературы Seasonal and annual fluxes of nutrients and organic matter from large rivers to the Arctic Ocean and surrounding seas // Estuaries and Coasts. — 2011. — Vol. 1, № 34. — P. 369–382. 011. — Vol. 1, № 34. — P. 369–382. Jickells T.D., Andrews J.E., Parkes D.J. et al. Nutrien transport through estuaries: the im­ portance of the estuarine geography // Estuar. Coast. Shelf Sci. — 2014. — Vol. 150. — P. 215–229. Jickells T.D., Andrews J.E., Parkes D.J. et al. Nutrien transport through estuaries: the im­ portance of the estuarine geography // Estuar. Coast. Shelf Sci. — 2014. — Vol. 150. — P. 215–229. Kaul L.W., Froelich P.N. Modeling estuarine nutrient geochemistry in a simple system // Jickells T.D., Andrews J.E., Parkes D.J. et al. Nutrien transport through estuaries: the im portance of the estuarine geography // Estuar. Coast. Shelf Sci. — 2014. — Vol. 150. — P. 215–229. Kaul L.W., Froelich P.N. Modeling estuarine nutrient geochemistry in a simple system // Geochim. Cosmogonica Acta. — 1984. — Vol. 48. — P. 1417–1433. Kaul L.W., Froelich P.N. Modeling estuarine nutrient geochemistry in a simple system // Geochim. Cosmogonica Acta. — 1984. — Vol. 48. — P. 1417–1433. Khemakhem H., Elloumi J., Ayadi H. et al. The concept of ecological succession applied to phytoplankton over four consecutive years in five ponds featuring a salinity gradient // Estuar. Coast. Shelf Sci. — 2010. — Vol. 88, № 1. — С. 33–44. MacIntry H.L., Kana T.M., Geider R.J. The effect of water motion on short-term rates of photosynthesis by marine phytoplankton // Trends Plant Sci. — 2000. — Vol. 5, № 1. — P. 12–17. y , , photosynthesis by marine phytoplankton // Trends Plant Sci. — 2000. — Vol. 5, № 1. — P. 12–17. McLusky D.S., Eliott M. The Estuarine ecosystem ecology, treats and management. — 3rd Edition. — N.Y. : Oxford Univ. Press Inc., 2004. — 214 p. McLusky D.S., Eliott M. The Estuarine ecosystem ecology, treats and management. — 3rd Edition. — N.Y. : Oxford Univ. Press Inc., 2004. — 214 p. Medvedeva L.A. Biodiversity of aquatic algal communities in the Sikhote-Alin Biosphere Reserve (Russia) // Cryptogamie, Algol. — 2001. — Vol. 22, № 1. — P. 60–100. Muylaert K., Sabbe K., Vyverman W. Changes in phytoplankton diversity and community composition along the salinity gradient of the Schelde estuary (Belgium / The Netherlands) // Estuar. Coast. Список литературы 311–321. у Сиренко Л.А. Проблемы евтрофирования водоемов // Экологическая химия водной сре­ ды. — М. : Наука, 1988. — С. 125–147. у Столяров А.П. Структурно-функциональная организация эстуарных экосистем Белого ря : автореф. дис. ... д-ра биол. наук. — М. : МГУ, 2011. — 44 с. Теплинская Н.Г. Бактерии пелагиали и бентали // Северо-западная часть Черного моря: биология и экология. — Киев : Наук. думка, 2006. — С. 146–174. у у , Трофимова В.В. Фотосинтетические пигменты фитопланктона эстуарных пелагических экосистем Баренцева моря : автореф. дис. ... канд. биол. наук. — Мурманск, 2007. — 30 с. Форина Ю.А., Шестеркин В.П. Особенности гидрохимического состава речных вод вос­ точного макросклона Северного Сихотэ-Алиня // География и природ. ресурсы. — 2010. — № 3. — С. 81–87. Чернова Е.Н., Шулькин В.М., Лысенко Е.В. и др. Гидрохимические и биогеохимические особенности пресных и солоноватоводных озер Восточного Сихотэ-Алиня // Изв. ТИНРО. — 2014. — Т. 178. — С. 157–172. Широколобова Т.И. Бактериопланктонные сообщества эстуарных зон и прибрежных экосистем Баренцева моря : автореф. дис. ... канд. биол. наук. — Мурманск, 2009. — 27 с. Экосистема эстуария реки Невы: биологическое разнообразие и экологические проблемы : моногр. / под ред. А.Ф. Алимова, С.М. Голубкова. — М. : Тов-во науч. изд. КМК, 2008. — 477 с. Almeida М.А., Cunha М.А., Alcantara F. Seasonal change in the proportion of bacterial and phytoplankton production along a salinity gradient in a shallow estuary // Hydrobiologia. — 2002. — Vol. 475. — P. 251–262.ll Alvarez L.G., Jones S.E. Factors influencing suspended sediment flux in the upper gulf of California // Estuar. Coast. Shelf Sci. — 2002. — Vol. 54, № 4. — P. 747–759. Becquevort S., Bouvier T., Lancelot C. et al. The seasonal modulation of organic matter uti­ lization by bacteria in the Danube-Black sea mixing zone // Estuar. Coast. Shelf Sci. — 2002. — Vol. 54, № 3. — P. 337–354. , Callaway R., Grenfell S., Lonborg Ch. Small estuaries: Ecology, environmental drivers and management challenger // Estuar. Coast. Shelf Sci. — 2014. — Vol. 150. — P. 193–195. Hessen D.O., Carroll J., Kjeldstad B. et al. Input of organic carbon as determinant of nutrient fluxes, light climate and productivity in the Ob and Yenisey estuaries // Estuar. Coast. Shelf Sci. — 2010. — Vol. 88, № 1. — P. 53–62. 234 Holmes R.M., Bulygina E., McClelland J.W. et al. Список литературы Shelf Sci. — 2009. — Vol. 82, № 2. — P. 335–340.l , Porter K.G., Feig Y.S. The use of DAPI for identifying and counting aquatic microflora // Limnol. Oceanogr. — 1980. — Vol. 25, № 5. — P. 943–948. Pye K., Blott S.J. The geomorphology of British estuaries: the effects of geological controls, an­ tectdent conditions and human activities // Estuar. Coast. Shelf Sci. — 2014. — Vol. 150. — P. 196–214. Revilla M., Ansotegui A., Iriarte A. et al. Microplankton metabolism along a trophic gradient in a shallow temperate estuary // Estuaries. — 2002. — Vol. 25, № 1. — P. 6–18. Rice E., Stewart G. Analisis of interdecadal trend in chlorophyll and temperature in the central basin of Long Island sound // Estuar. Coast. Shelf Sci. — 2013. — Vol. 128. — P. 64–75. Telesh I., Schubert H., Skarlato S. Life in the salinity gradient: Discovering mechanisms ind a new biodiversity pattern // Estuar. Coast. Shelf Sci. — 2013. — Vol. 135. — P. 317–327. Wetzel R.G. Detritial dissolved and particulate organic carbon functions in aquatic ecosystem // Bull. Mar. Sci. — 1984. — Vol. 35. — P. 503–509. Поступила в редакцию 16.11.15 г. Поступила в редакцию 16.11.15 г. 235 235

https://openalex.org/W2983288321

https://link.springer.com/content/pdf/10.1007/s10579-019-09477-1.pdf

English

Reproduction, replication, analysis and adaptation of a term alignment approach

Language resources and evaluation

cc-by

Lang Resources & Evaluation (2020) 54:767–800 https://doi.org/10.1007/s10579-019-09477-1 ORIGINAL PAPER - REPLICABILITY & REPRODUCIBILITY 1 Introduction The issue of reproducibility has been on the radar of researchers at least for the past 25 years, particularly in the life science research (e.g. Yentis et al. 1993; Prinz et al. 2011; Camerer et al. 2016). More recently, many other disciplines have started to acknowledge the crisis of reproducibility, among them also human language technology research (Pedersen 2008; Kano et al. 2009; Fokkens et al. 2013; Branco et al. 2017; Wieling et al. 2018). However, the basic terminology has remained confusing with different authors using different terms for the same concepts which is why Cohen et al. (2018) describe the three dimensions of reproducibility in natural language processing (NLP) and provide a set of definitions for the various concepts used when discussing reproducibility in NLP. They first differentiate between the concepts of replicability (or repeatability), which they define as the ability to repeat the experiment described in a study, and reproducibility, which describes the outcome—whether the replicability efforts lead to the same conclusions. Then they further break down reproducibility into reproducibility of a conclusion (defined as an explicit statement in the paper arrived at on the basis of the results of the experiments), reproducibility of a finding (a relationship between the values for some reported figure of merit) and reproducibility of a value (actual measured or calculated numbers). In this paper we extend our reproducibility study (Repar et al. 2018), presented at the Workshop on Research Results Reproducibility and Resources Citation (4REAL Workshop, Branco et al. (2018)) organized within the scope of the 11th Language Resources and Evaluation Conference (LREC 2018). Our original motivation came from our interest and need for a terminology alignment tool, and the paper by Aker et al. (2013) titled ‘‘Extracting Bilingual Terminology from Parallel Corpora’’ seemed a perfect candidate for reproduction with nearly perfect results, coverage of the Slovenian-English pair (which were the languages of our interest) and what seemed like a well described and simple to replicate method. The authors treat aligning terms in two languages as a binary classification problem. They use an SVM binary classifier (Joachims 2002) and training data terms taken from the Eurovoc thesaurus (Steinberger et al. 2002) and construct two types of features: dictionary-based (using word alignment dictionaries created with Giza?? (Och and Ney 2003)) and cognate-based (effectively utilizing the similarity of terms across languages). Reproduction, replication, analysis and adaptation of a term alignment approach Andrazˇ Repar1,2 • Matej Martinc1,2 • Senja Pollak2,3 Published online: 18 November 2019  The Author(s) 2019 Abstract In this paper, we look at the issue of reproducibility and replicability in bilingual terminology alignment (BTA). We propose a set of best practices for reproducibility and replicability of NLP papers and analyze several influential BTA papers from this perspective. Next, we present our attempts at replication and reproduction, where we focus on a bilingual terminology alignment approach described by Aker et al. (Extracting bilingual terminologies from comparable cor- pora. In: Proceedings of the 51st annual meeting of the association for computational linguistics, vol. 1 402–411, 2013) who treat bilingual term alignment as a binary classification problem and train an SVM classifier on various dictionary and cognate-based features. Despite closely following the original paper with only minor deviations—in areas where the original description is not clear enough—we obtained significantly worse results than the authors of the original paper. We then analyze the reasons for the discrepancy and describe our attempts at adaptation of the approach to improve the results. Only after several adaptations, we achieve results which are close to the results published in the original paper. Finally, we perform the experiments to verify the replicability and reproducibility of our own code. We publish our code and datasets online to assure the reproducibility of the results of our experiments and implement the selected BTA models in an online & Andrazˇ Repar [email protected] Matej Martinc [email protected] Senja Pollak [email protected] 1 Jozˇef Stefan Postgraduate School, Ljubljana, Slovenia 2 Jozˇef Stefan Institute, Ljubljana, Slovenia 3 Usher institute, Medical school, University of Edinburgh, Edinburgh, UK 123 123 768 A. Repar et al. platform making them easily reusable even by the technically less-skilled researchers. platform making them easily reusable even by the technically less-skilled researchers. Keywords Bilingual term alignment  Reproducibility  Machine learning  Cognates Keywords Bilingual term alignment  Reproducibility  Machine learning  1 Introduction Given that the results looked very promising—precision on the held-out set was 1 or close to 1 for many language pairs, we thought we could use the approach in our work and we set out to replicate it. We expected a straightforward process, but it turned out to be anything but: the results of our experiments were very vastly different from the original paper. For example, while the original paper Reproduction, replication, analysis and adaptation... 769 reports an extremely high precision (1 or close to 1) for the language pairs we have focused on, our experiments showed a precision below 0.05. Based on the reproducibility dimensions mentioned above, in our original reproducibility experiment from Repar et al. (2018) we were not able to reproduce any of the three dimensions: the values and findings in our experiments were vastly different, and—had we stopped at this point—we would have concluded that the proposed machine learning approach is not suitable for bilingual terminology alignment. Only after a great deal of tweaking and optimization have we managed to get to a respectable precision level (similar to the results in the original paper). In the present paper, we aim to explore the issue of reproducibility and replicability in the field of terminology alignment further. To do so, we extend the work in Repar et al. (2018) with the following: – an overview of bilingual terminology extraction and alignment approaches in terms of replicability and reproducibility. – an overview of bilingual terminology extraction and alignment approaches in terms of replicability and reproducibility. – extending the original reproducibility experiment to two additional languages, resulting in Slovenian, French and Dutch as target languages from three different language families. – extending the original reproducibility experiment to two additional languages, resulting in Slovenian, French and Dutch as target languages from three different language families. – providing very detailed description of feature construction. – additional filtering and refinement of the cognate-based features. – a reproducibility experiment with source code from Repar et al. (2018). – implementation of our code into an online data mining platform ClowdFlows – a discussion on good practices for reproducibility and replicability in NLP. This paper is organized as follows: After the introduction in Sect. 1, we present the related work and the analysis of bilingual terminology alignment papers from the point of view of replicability and reproducibility (Sect. 2). 1 Introduction Section 3 contains the main replicability and reproducibility experiments, and is followed by Sect. 4, which describes our attempts at improving the results of the replicated approach, while Sect. 5 contains the results of manual evaluation. Section 6 describes the reproducibility experiment using our code from Repar et al. (2018) and Sect. 7 the implementation of the system in the ClowdFlows platform, for making it accessible to a wider community. Section 8 contains the conclusions and presents ideas for future work. The code and datasets of our experiments are published online, to enable future reproducibility and replicability.1 1 http://source.ijs.si/mmartinc/4real2018. 2.1 Related work We start by providing a clarification regarding the terminology used in this paper. Following the distinction between two basic approaches made by Foo (2012): – extract-align where we first extract monolingual candidate terms from both sides of the corpus and then align the terms, and – extract-align where we first extract monolingual candidate terms from both sides of the corpus and then align the terms, and – align-extract where we first align single and multi-word units in parallel sentences and then extract the relevant terminology from a list of candidate term pairs. we propose the following two definitions: we propose the following two definitions: – Bilingual terminology extraction is the process which, given the input of related specialized monolingual corpora, results in the output of terms aligned between two languages. The process can either start with extracting monolingual candidate terms and aligning them between two languages (i.e. extract-align) or with aligning phrases and then extracting terms (i.e. align-extract) or any other sequence of actions. – Bilingual terminology alignment is the process of aligning terms between two candidate term lists in two languages. – Bilingual terminology alignment is the process of aligning terms between two candidate term lists in two languages. Bilingual terminology alignment has a narrower focus than bilingual terminology extraction, but the two terms are often used interchangeably in various papers. For example, the title of the paper we were trying to replicate ‘‘Extracting bilingual terminologies from comparable corpora’’ is somewhat misleading in this regard, since the paper primarily deals with bilingual terminology alignment, while they utilize monolingual terminology extraction (specifically the approach by Pinnis et al. (2012) without any modifications) only in the manual evaluation experiments. Bilingual terminology alignment has a narrower focus than bilingual terminology extraction, but the two terms are often used interchangeably in various papers. For example, the title of the paper we were trying to replicate ‘‘Extracting bilingual terminologies from comparable corpora’’ is somewhat misleading in this regard, since the paper primarily deals with bilingual terminology alignment, while they utilize monolingual terminology extraction (specifically the approach by Pinnis et al. (2012) without any modifications) only in the manual evaluation experiments. The primary purpose of bilingual terminology extraction is to build a term bank—i.e. a list of terms in one language along with their equivalents in the other language. 2 Overview of bilingual terminology extraction and alignment approaches In this section we first look at the related work on bilingual terminology extraction and alignment and then analyze several related papers from the viewpoint of replicability and reproducibility. 1 http://source.ijs.si/mmartinc/4real2018. 123 123 A. Repar et al. 770 2 However, some publicly available parallel corpora do exist. A good overview can be found at the OPUS web portal (Tiedemann 2012). 2.1 Related work With regard to the input text, we can distinguish between alignment on the basis of a parallel corpus and alignment on the basis of a comparable corpus. For the translation industry, bilingual terminology extraction from parallel corpora is extremely relevant due to the large amounts of sentence-aligned parallel corpora available in the form of translation memories (in the TMX file format). Consequently, initial attempts at bilingual terminology extraction involved parallel input data (Kupiec 1993; Daille et al. 1994; Gaussier 1998), and the interest of the community continued until today (Ha et al. 2008; Ideue et al. 2011; Macken et al. 2013; Haque et al. 2014; Arcˇan et al. 2014; Baisa et al. 2015). However, most parallel corpora are owned by private companies,2 such as language service providers, who consider them to be their intellectual property and are reluctant to share them publicly. For this reason (and in particular for language pairs not 12 123 Reproduction, replication, analysis and adaptation... 771 involving English) considerable efforts have also been invested into researching bilingual terminology extraction from comparable corpora (Fung and Yee 1998; Rapp 1999; Chiao and Zweigenbaum 2002; Cao and Li 2002; Daille and Morin 2005; Morin et al. 2008; Vintar 2010; Bouamor et al. 2013; Hazem and Morin 2016, 2017). involving English) considerable efforts have also been invested into researching bilingual terminology extraction from comparable corpora (Fung and Yee 1998; Rapp 1999; Chiao and Zweigenbaum 2002; Cao and Li 2002; Daille and Morin 2005; Morin et al. 2008; Vintar 2010; Bouamor et al. 2013; Hazem and Morin 2016, 2017). Despite the problem of bilingual term alignment lending itself well to the binary classification task, there have been relatively few approaches utilizing machine learning. For example, similar to Aker et al. (2013), Baldwin and Tanaka (2004) generate corpus-based, dictionary-based and translation-based features and train an SVM classifier to rank the translation candidates. Note that they only focus on multi-word noun phrases (noun ? noun). A similar approach, again focusing on noun phrases, is also described by Cao and Li (2002). Finally, Nassirudin and Purwarianti (2015) also reimplement Aker et al. (2013) for the Indonesian-Japanese language pair and further expand it with additional statistical features. In the best scenario, their accuracy, precision and recall all exceed 90% but the results are not directly comparable since Nassirudin and Purwarianti (2015) use tenfold cross- validation while Aker et al. (2013) use a held-out test set. 3 The selection process was as follows: the starting point were selected seminal papers on the field, as well as two queries in the ACL Anthology database: ‘‘term alignment’’ and ‘‘bilingual terminology extraction’’. We analyzed the papers found by these two queries as well as additional papers mentioned in the related works sections of these papers and the main criterion for including a paper in our analysis was that it primarily deals with bilingual terminology extraction (and not for example latent semantic analysis, such as Bader and Chew (2008)). However, no strict systematic review with inclusion and exclusion criteria was made, as such a survey would be beyond the needs of this paper. 2.1 Related work In addition, Nassirudin and Purwarianti (2015) have a balanced test set while Aker et al. (2013) use a very unbalanced one (ratio of positive vs. negative examples 1:2000). 2.2 Analysis of past papers on bilingual terminology extraction from the viewpoint of reproducibility and replicability In an ideal reproducibility and replicability scenario, a scientific paper would contain an accurate and clear description of the datasets used and experiments conducted and the authors would provide a single link containing all the datasets (versions, subsets etc.) used for the experiments along with the experiment source code (or alternatively, an online tool to run the experiments). These could then be used to replicate the experiments and reproduce the results using the descriptions provided in the paper. We have analyzed several3 bilingual terminology extraction papers from the past 25 years from the point of view of dataset, code and tool availability. The summary of results is available in Table 1. 2.2.1 Dataset availability In terms of dataset availability, we looked at whether the paper contains some description of how the datasets were constructed and which could (theoretically) be used to reconstruct the datasets. Note that under ‘‘dataset’’, we include corpora, gold 12 3 772 A. Repar et al. Table 1 An analysis of bilingual terminology extraction papers from the point of view of reproducibility and replicability Paper Dataset Code Tool Google Scholar citations as of September 2019 Kupiec (1993) Links No No 333 Daille et al. (1994) No No No 268 Fung and Yee (1998) Description No No 427 Gaussier (1998) No No No 84 Rapp (1999) Description No No 552 Chiao and Zweigenbaum (2002) Description No No 135 Cao and Li (2002) Description No No 141 Morin et al. (2007) No No No 113 Daille and Morin (2005) Obsolete No Obsolete 56 Morin et al. (2008) Links No Obsolete 22 Ha et al. (2008) Description No No 4 Lee et al. (2010) Description No No 22 Vintar (2010) No No Obsolete 53 Ideue et al. (2011) No No Yes a 9 Macken et al. (2013) No No No 48 Bouamor et al. (2013) Description No No 24 Aker et al. (2013) Links No No 36 Arcˇan et al. (2014) Links No No 18 Haque et al. (2014) Links No No 11 Kontonatsios et al. (2014) Description No No 14 Baisa et al. (2015) No No Yes 5 Hazem and Morin (2016) Links No No 12 Hazem and Morin (2017) Links No No 2 aA Perl module (Term Extract) was used, however the link leads to a Japanese website able 1 An analysis of bilingual terminology extraction papers from the point of view of reproducibility d replicability standard termlists, seed dictionaries and all other linguistic resources needed to conduct the experiments in the paper. For example, we consider the following paragraph from Rapp (1999) to be a valid description of a dataset: As the German corpus, we used 135 million words of the newspaper Frankfurter Allgemeine Zeitung (1993 to 1996), and as the English corpus 163 million words of the Guardian (1990 to 1994). On the other hand, this paragraph from Ideue et al. (2011) is not considered a valid description: We extracted bilingual term candidates from a Japanese-English parallel corpus consisting of documents related to apparel products. 2.2.1 Dataset availability In the former example, dataset reconstruction would be difficult but not impossible, while in the latter it is impossible. An even better option is to link to actual datasets or refer to papers where datasets are described and linked, which is why we also looked for dataset links and/or references in the analyzed papers. Note that there are several examples where links are provided only for a selection of the datasets used in the experiments (e.g., Morin et al. (2008)). 123 773 Reproduction, replication, analysis and adaptation... As evident from Table 1, dataset availability is the least problematic aspect of reproducibility and replicability in terminology (extraction and) alignment papers with approximately two thirds of the analyzed papers (15 out of 23) either containing a description of the resources used for the experiments, providing links to them or refering to papers where they are described. We expected the earlier papers to have less information on datasets than latter ones, but this turned out not to be the case. In fact, the earliest paper analyzed— Kupiec (1993)—provides a reference to a publicly available corpus (Canadian Hansards (Gale and Church 1993)). The first paper to have a separate section with data/resource description is Rapp (1999) and from this point on, almost all papers have such a section—usually titled ‘‘Data and Resources’’, ‘‘Resources and Experimental Setup’’, ‘‘Linguistic resources’’ or similar. However, it is rarely documented what version of the dataset was used and whether an entire dataset was used or only a part of it (as in random selection, train- test split, etc.). In most cases, little information is provided on the actual subsets used for the experiments. Another aspect of dataset use is the languages: when one of the languages involved is English, it is much easier to find datasets than for other language combinations. Finally, there is also the issue of keeping the links active. For example, many of the links in Daille and Morin (2005) and Morin et al. (2008) are not active anymore while Bouamor et al. (2013) state that the corpora and terminology gold standard lists created for the paper will be shared publicly, but no links are provided. The most significant problem encountered during our analysis was the fact that terminology alignment is most often not the sole focus of a paper, such as in Haque et al. 5 https://hlt-mt.fbk.eu/technologies/bittercorpus. 4 http://www.lina.univ-nantes.fr/?Reference-Term-Lists-of-TTC.html. 4 http://www.lina.univ-nantes.fr/?Reference-Term-Lists-of-TTC.html. 5 https://hlt-mt.fbk.eu/technologies/bittercorpus. 7 https://www.nlm.nih.gov/research/umls/. 2.2.1 Dataset availability (2014), where the experiments start with monolingual terminology extraction from two languages and the extracted terms are then aligned. As terminology extraction and alignment go hand-in-hand, it may often be impossible to make a clear distinction between the terminology extraction and terminology alignment datasets. This means that the dataset results in Table 1 are not a true apple-to-apple comparison: one paper might link to the parallel corpus used to extract terms from, while another to a gold standard termlist. Our main criterion was whether the dataset description (or link) could be used to replicate the experiments described in the paper. An ideal terminology (extraction and) alignment dataset would therefore consist of a bilingual or multilingual (parallel or comparable) corpus along with reference (gold standard) term lists containing terms that can be found in the corpus. Such corpora are TTC wind energy and TC mobile technology4, which contain data for six languages (English, French, German, Spanish, Russian, Latvian, Chinese), or the Bitter corpus5, which contains data for the EN-IT language pair. The first was used in Hazem and Morin (2016), while the second one by Arcˇan et al. (2014). Since such datasets are scarce, researchers employ various methodologies for constructing their own datasets. One method, used by Aker et al. (2013), is to take one of the available multilingual translation memories containing EU documentation (such as 123 774 A. Repar et al. Europarl (Koehn 2005) or DGT (Steinberger et al. 2013)) as the corpus and a glossary (e.g., IATE (Johnson and Macphail 2000)) or thesaurus (e.g., Eurovoc (Steinberger et al. 2002)) as the terminology gold standard list. Another strategy, used by Hazem and Morin (2017), is to collect a comparable corpus manually (i.e. scientific articles in French and English from the Elsevier6 website) and a domain specific terminological resource (i.e. UMLS7) as a reference termlist. Hazem and Morin (2017) also filter out those terms from the termlist that do not appear often enough in their corpus. In other cases (e.g., Haque et al. (2014)), the datasets are not available because the papers were written as part of industrial projects and the datasets are private. 2.2.2 Code and tool availability We have discovered that no paper has made experiment code available and only a few provide access or links to tools where the experiments were conducted. But even when links to tools are provided, reproducibility and replicability may be hindered: for example, the link provided in Ideue et al. (2011) leads to a Japanese website. Another issue is the long-term availability of resources. For example, Daille and Morin (2005) conducted their experiments in ACABIT, an open source terminology extraction software. However, the link given in the paper does not work anymore. From the analyzed papers, the only example of bilingual term extraction and alignment tool, which is publicly available, is the Sketch Engine term extraction module, described by Baisa et al. (2015). None of the papers analyzed in this section fulfill the ideal scenario described at the start of this section (i.e. a single link with code and all datasets) which severly hinders any replicability attempts as will be evident from our own experiments described in this paper. 3 Replicating a machine learning approach to bilingual term alignment and reproducing its results This section describes our efforts in replicating a machine learning approach to bilingual term alignment described in Aker et al. (2013),by which we extend our initial experiments and analysis (Repar et al. 2018). Section 3.1 describes the original approach and Sect. 3.2 contains an overview of our attempts to replicate it. 6 https://www.elsevier.com/. 3.1 Description of the original approach The original approach designed by Aker et al. (2013) was developed to align terminology from comparable (or parallel) corpora using machine-learning techniques. They use terms from the Eurovoc (Steinberger et al. 2002) thesaurus and train an SVM binary classifier (Joachims 2002) (with a linear kernel and the 123 Reproduction, replication, analysis and adaptation... 775 trade-off between training error and margin parameter c = 10). The task of bilingual alignment is treated as a binary classification—each term from the source language S is paired with each term from the target language T and the classifier then decides whether the aligned pair is correct or incorrect. They then extract features (dictionary and cognate-based) to be used by the classifier. They run their experiments on the 21 official EU languages covered by Eurovoc with English always being the source language (20 language pairs altogether). They evaluate the performance on a held-out term pair list from Eurovoc using recall, precision and F- measure for all 20 languages. Next, they propose an experimental setting for a simulation of a real-world scenario where they collect English-German comparable corpora of two domains (IT, automotive) from Wikipedia, perform monolingual term extraction using the system by Pinnis et al. (2012) followed by the bilingual alignment procedure described above and manually evaluate the results (using two evaluators). They report excellent performance on the held-out term list with many language pairs reaching 100% precision and the lowest recall being 65%. For Slovenian, which is of our main interest, as well as for the additional target languages that we selected, namely French and Dutch, the reported results were excellent with perfect or nearly perfect precision and good recall for all three language pairs. The reported results of the manual evaluation phase were also good, with two evaluators agreeing that at least 81% of the extracted term pairs in the IT domain and at least 60% of the extracted term pairs in the automotive domain can be considered exact translations. 3.1.1 Features Aker et al. (2013) use two types of features that express correspondences between the words (composing a term) in the target and source language (for a detailed description see Table 2: – 7 dictionary-based (using Giza??) features which take advantage of dictionar- ies created from large parallel corpora of which 6 are direction-dependent (source-to-target or target-to-source) and 1 direction-independent—resulting in altogether 13 features, and – 5 cognate-based (on the basis of Gaizauskas et al. (2012)) which utilize string- based word similarity between languages. – 5 cognate-based (on the basis of Gaizauskas et al. (2012)) which utilize string- based word similarity between languages. To match words with morphological differences, they do not perform direct string matching but utilize Levenshtein Distance. Two words were considered equal if the Levenshtein Distance (Levenshtein 1966) was equal or higher than 0.95. For closed- compounding languages, they check whether the compound source term has an initial prefix that matches the translation of the first target word, provided that translation is at least 5 characters long. 123 776 A. Repar et al. Table 2 Features used in the experiments Feature Cat Description isFirstWordTranslated Dict Checks whether the first word of the source term is a translation of the first word in the target term (based on the Giza?? dictionary) isLastWordTranslated Dict Checks whether the last word of the source term is a translation of the last word in the target term percentageOfTranslatedWords Dict Ratio of source words that have a translation in the target term percentageOfNotTranslatedWords Dict Ratio of source words that do not have a translation in the target term longestTranslatedUnitInPercentage Dict Ratio of the longest contiguous sequence of source words which has a translation in the target term (compared to the source term length) longestNotTranslatedUnitInPercentage Dict Ratio of the longest contiguous sequence of source words which do not have a translation in the target term (compared to the source term length) Longest Common Subsequence Ratio (LCSSR) Cogn Measures the longest common non-consecutive sequence of characters between two strings (divided by the length of the longest string) Longest Common Substring Ratio (LCSTR) Cogn Measures the longest common consecutive string (LCST) of characters that two strings have in common (divided by the length of the longest string) Dice similarity Cogn 2*LCST / (len(source) ? 8 For combined features, a word is considered as covered if it can be found in the corresponding set of Giza?? translations or if one of the cognate-based measures (Longest Common Subsequence, Longest Common Substring, Levenshtein Distance, Needleman-Wunsch Distance, Dice) is 0.70 or higher (set experimentally by Aker et al. (2013)). Additional features are also constructed by: – Using language pair specific transliteration rules to create additional cognate- based features. The purpose of this task was to try to match the cognate terms while taking into account the differences in writing systems between two languages: e.g. Greek and English. Transliteration rules were created for both directions (source-to-target and target-to-source) separately and cognate-based features were constructed for both directions - resulting in additional 10 cognate-based features with transliteration rules. – Combining the dictionary and cognate-based features in a set of combined features where the term pair alignment is correct if either the dictionary or the cognate-based method returns a positive result. This process resulted in additional 10 combined features.8 – Combining the dictionary and cognate-based features in a set of combined features where the term pair alignment is correct if either the dictionary or the cognate-based method returns a positive result. This process resulted in additional 10 combined features.8 At the end of the feature construction phase, there were 38 features: 13 dictionary- based, 5 cognate-based, 10 cognate-based features with transliteration rules and 10 combined features. 3.1.1 Features len(target)) Needlemann-Wunsch distance Cogn LCST / min(len(source), len(target)) Normalized Levenshtein distance (nLD) Cogn 1 - LD / max(len(source), len(target)) isFirstWordCovered Comb A binary feature indicating whether the first word in the source term has a translation or transliteration in the target term isLastWordCovered Comb A binary feature indicating whether the last word in the source term has a translation or transliteration in the target term percentageOfCoverage Comb Returns the percentage of source term words which have a translation or transliteration in the target term percentageOfNonCoverage Comb Returns the percentage of source term words which have neither a translation nor transliteration in the target term difBetweenCoverageAndNonCoverage Comb Returns the difference between the last two features Note that some features are used more than once because they are direction-dependent 1 3 Reproduction, replication, analysis and adaptation... 777 Additional features are also constructed by: 3.1.2 Data source and experiments Using Giza??, Aker et al. (2013) create source-to-target and target-to-source word alignment dictionaries based on the DGT translation memory (Steinberger et al. 2013). The resulting dictionary entries consist of the source word s, its translation t and the number indicating the probability that t is an actual translation of s. To improve the performance of the dictionary-based features, the following entries were removed from the dictionaries: – entries where probability is lower then 0.05. – entries where probability is lower then 0.05. – entries where the source word was less than 4 characters and the target word more than 5 characters long and vice versa in order to avoid translations of stop word to content words.) – entries where the source word was less than 4 characters and the target word more than 5 characters long and vice versa in order to avoid translations of stop word to content words.) The next step is the creation of term pairs from the Eurovoc (Steinberger et al. 2002) thesaurus, which at the time consisted of 6797 terms. Each non-English language was paired with English. The test set consisted of 600 positive (correct) term pairs—taken randomly out of the total 6797 Eurovoc term pairs—and around 1.3 million negative pairs which were created by pairing each source term with 200 distinct incorrect random target terms. Aker et al. (2013) argue that this was done to simulate real-world conditions where the classifier would be faced with a larger number of negative pairs and a comparably small number of positive ones. The 600 positive term pairs were further divided into 200 pairs where both (i.e. source and target) terms were single words, 200 pairs with a single word only on one side and 12 3 778 A. Repar et al. 200 pairs with multiple-word terms on both sides. The remaining positive term pairs (approximately 6200) were used as training data along with additional 6200 negative pairs. These were constructed by taking the source side terms and pairing each source term with one target term (other than the correct one). Using this approach, Aker et al. (2013) achieve excellent results with 100% precision and 66% recall for Slovenian and French and 98% precision and 82% recall for Dutch. 9 Note that our original replication paper Repar et al. (2018) wrongly states that we did not utilize the compounding solution implemented by Aker et al. (2013) for addressing compouding issues in languages such as German. In fact, we did implement it and used it in all experiments. 10 http://source.ijs.si/mmartinc/4real2018/blob/master/term_list_sl.csv. 3.2 Replication of the approach The first step in our approach was to replicate the algorithm described by Aker et al. (2013). The initial premise is the same: given two lists of terms from the same domain in two different languages, we would like to align the terms in the two lists to get one bilingual glossary to be used in a variety of settings (computer-assisted translation, machine translation, ontology creation etc.). We followed the approach described above faithfully except in the following aspects9: – Instead of the entire set of Eurovoc languages, we have initially focused only on the English-Slovenian language pair (Repar et al. 2018). In the current paper, we add two additional language pairs (English-French, English-Dutch) to see whether our findings can be generalised across different languages. We selected languages from different language families, as the importance of cognates is dependent on the similarity between languages (for example, Dutch and English (being both Germanic languages) presumably have a higher number of cognates). – We use newer datasets. The Eurovoc thesaurus version that we used contained 7,083 terms for Slovenian10 and 7,181 terms for French11 and Dutch.12 Similarly, the DGT translation memory contains additional content not yet present in 2013.13 For English-Slovenian, we at first used the entire DGT corpus up to and including the DGT-TM-release 2017 for deriving GIZA alignments. Later we also experimented with precomputed dictionaries by Aker et al – We use newer datasets. The Eurovoc thesaurus version that we used contained 7,083 terms for Slovenian10 and 7,181 terms for French11 and Dutch.12 Similarly, the DGT translation memory contains additional content not yet present in 2013.13 For English-Slovenian, we at first used the entire DGT corpus up to and including the DGT-TM-release 2017 for deriving GIZA alignments. Later we also experimented with precomputed dictionaries by Aker et al. (2014). When performing the experiments on the other languages pairs, we did not create our own GIZA alignment, but only used the precomputed ones by Aker et al. (2014). Later we also experimented with precomputed dictionaries by Aker et al. (2014). When performing the experiments on the other languages pairs, we did not create our own GIZA alignment, but only used the precomputed ones by Aker et al. (2014). – Since no particular cleaning of training data (e.g., manual removal of specific entries) is described in the paper for the languages of our interest, we do not perform any. 13 The versions of the resources used in Aker et al. (2013) were not documented or made available. 12 http://source.ijs.si/mmartinc/4real2018/blob/master/term_list_nl.csv. 11 http://source.ijs.si/mmartinc/4real2018/blob/master/term_list_fr.csv. 3.2.1 Problems with replicating the approach While the general approach is clearly laid out in the article, there are several spots where further clarification would be welcome: – There is no sufficient information about the Giza?? settings or whether the input corpora have been lemmatized. In order to improve term matching, we experimented with and without lemmatization of the Giza?? input corpora. – There is no sufficient information about the Giza?? settings or whether the input corpora have been lemmatized. In order to improve term matching, we experimented with and without lemmatization of the Giza?? input corpora. – There is no information about the specific character mappings rules other than a general principle of one character in the source being mapped to one or more character in the target. Since the authors cover 20 languages, it is understandable that they cannot include the actual mapping rules in the article. Therefore, we have created our own mapping rules for English-Slovenian and English-French according to the instructions in the original paper: – Mapping the English term to the Slovenian writing system (the character before the colon is replaced by the sequence of characters after the colon): x:ks, y:j, w:v, q:k. y j q – Mapping the Slovenian term to the English writing system: cˇ:ch, š:sh, zˇ:zh – Mapping the French term to the English writing system: we deleted all accents e.g., e´:e, eˆ:e. – Mapping the Dutch term to the English writing system: we deleted all accents and replace the digraph ij with two separate letters ij. – Instead of the unclear Needleman–Wunsch distance formula from Aker et al. (2013) LCST min½lenðsourceÞþlenðtargetÞ (which implies that we should take the minimum value of the sum of the length of the target and source term) we opted for LCST min½lenðsourceÞ;lenðtargetÞ as in Nassirudin and Purwarianti (2015). ½ ð Þ; ð g Þ – We were not completely certain how to treat examples such as ‘‘passport—potni list’’, where a single-word source term is translated by a multi-word target term and both combinations (passport—potni and passport—list) can be found in the Giza?? dictionary. In this case, our implementation returns values of 1 for both isFirstWordTranslated and isLastWordTranslated features despite the fact that the source term only has one word. – There was a slight ambiguity on how to calculate cognate-based features: on the level of words or on the level of entire terms. 3.2 Replication of the approach – Since no particular cleaning of training data (e.g., manual removal of specific entries) is described in the paper for the languages of our interest, we do not perform any. We think that regardless of these differences, the experiments should yield similar results. We think that regardless of these differences, the experiments should yield similar results. 123 Reproduction, replication, analysis and adaptation... 779 3.2.2 Results The evaluation on the test set created as described in the original paper by Aker et al. (2013) shows that compared to the results reported by the authors (see line 1 in Tables 3, 4 and 5), our results are significantly worse. Despite all our efforts to follow the original approach, we were unable to match the results achieved in the original paper when running the algorithm without any changes to the original approach. When trying to follow the original paper’s methodology, precision is only 3.59% and recall is 88% for the English-Slovenian language pair. The results for the other two language pairs are comparable (see line 2 in Tables 3, 4 Table 5 for details). ) In Sect. 4, we provide the results of detailed analysis and additional experiments that we performed in order to reach results comparable to the original approach. 3.2.1 Problems with replicating the approach We opted for the second, since the names of the cognate-based features did not imply that cognates are calculated on the word level (as was the case with the dictionary-based features) and since there was no mention in the original paper on how to combine cognate-based scores for specific word pairs in the multi-word term pairs in order to get a final cognate score for the whole term pair. – In the original article, the isFirstWordCovered feature is described as ‘‘a binary feature indicating whether the first word in the source term has a translation (i.e. has a translation entry in the dictionary regardless of the score) or transliteration (i.e. if one of the cognate metric scores is above 0.7) in the target term.’’ While 12 123 23 780 A. Repar et al. the dictionary-based part is clear, for calculating the cognate-based feature values (e.g., of the first word in the source term), the values of the cognate metric scores concern the entire target term. As we did not find this fully intuitive, and we believe other interpretations are possible, we experimented with these settings in the adaptation of the approach (see Sect. 4.8). To avoid ambiguities, we provide a separate document with examples of constructed features, together with the code (http://source.ijs.si/mmartinc/ 4real2018/blob/master/feature_examples.docx). 3.2.3 Attempts at establishing contact with the authors When replicating an existing paper, especially when the code is not made available, contacting the authors for clarification (or for providing/running the code) is the most obvious step when encountering the problems or ambiguities. However, due to busy schedules of researchers, change of professional paths or other similar reasons, getting detailed help might be impossible. This is true for our case as well. Initially, we were hopeful of getting useful feedback, as the authors already provided the software to other researchers in the past (see Arcˇan et al. (2014)). However, despite a friendly response, we have been able to get only a limited number of answers and many questions remained unanswered, and the auhors have not been able to share their code. We have first contacted the original authors of the paper when we were running the experiments reported in Repar et al. (2018) and did receive some answers confirming our assumptions (e.g. regarding mapping terms to the different writing systems and that the test set data was selected individually for each language pair), but several other issues remained unaddressed (in particular, what was the exact train and test data selection strategy for the EN-SL language pair). Further inquiries proved unsuccessful due to time constraints on the part of the original authors. As we expanded the paper with additional languages and experiments, we again contacted the main author, provided him the code and the paper and asked for help in 123 Reproduction, replication, analysis and adaptation... 781 Table 3 Results on the English–Slovenian term pair No. Config EN-SL Training set size Pos/neg ratio Precision Recall F-score 1 Reported by Aker et al. (2013) 12,400 1:1 1 0.6600 0.7900 2 Replicated approach 12,966 1:1 0.0359 0.8800 0.0689 3 Giza?? terms only 8306 1:1 0.0645 0.9150 0.1205 4 Giza?? cleaning 12,966 1:1 0.0384 0.7789 0.0731 4a Lemmatization 12,966 1:1 0.0373 0.8150 0.0713 5 Training set 1:200 1,303,083 1:200 0.4299 0.7617 0.5496 6 Training set filtering 1 6426 1:1 0.5969 0.64167 0.6185 7 Training set filtering 2 35,343 1:10 0.9042 0.5350 0.6723 8 Training set filtering 3 645,813 1:200 0.9342 0.4966 0.6485 9 Term length filtering 6426 1:1 0.8144 0.4900 0.6119 10 Cognates approach 672,345 1:200 0.8732 0.5167 0.6492 No. 1 presents the results reported by the authors, No. 2 our replication of the approach and No. 3.2.3 Attempts at establishing contact with the authors 3–10 our modifications of the first replicated approach with the aim of improving the results Table 3 Results on the English–Slovenian term pair Table 4 Results on the English–French language pair No. Config EN-FR Training set size Pos/neg ratio Precision Recall F-score 1 Reported by Aker et al. (2013) 12,400 1:1 1 0.6600 0.7900 2 Replicated approach 13,160 1:1 0.0323 0.8483 0.0622 3 Giza?? terms only 8892 1:1 0.0437 0.8433 0.0830 4 Giza?? cleaning 13,160 1:1 0.0317 0.7917 0.0610 5 Training set 1:200 1,322,580 1:200 0.5273 0.6767 0.5927 6 Training set filtering 1 2650 1:1 0.4623 0.5517 0.5030 7 Training set filtering 2 14,575 1:10 0.9422 0.3533 0.5139 8 Training set filtering 3 266,325 1:200 0.9791 0.3117 0.4728 9 Term length filtering 2650 1:1 0.6791 0.3950 0.4995 10 Cognates approach 311,952 1:200 0.8603 0.3900 0.5367 No. 1 presents the results reported by the authors, No. 2 our replication of the approach and No. 3–10 our modifications of the first replicated approach with the aim of improving the results Table 4 Results on the English–French language pair o. 1 presents the results reported by the authors, No. 2 our replication of the approach and No. 3–10 our odifications of the first replicated approach with the aim of improving the results identification of any possible mistakes leading to the results, however, we were ultimately not able to get any information which would explain the differences. We think the original paper is generally well-written and that the main reason for occasional lack of clarity is its scope: as the authors deal with more than 20 language pairs, it would be impossible to provide specific information regarding all of them. Providing more examples would be useful, but still the code and the exact dataset are in our opinion the only way to be able to fully replicate the experiments. 12 3 A. Repar et al. 782 Table 5 Results on the English–Dutch language pair No. Config EN-NL Training set size Pos/neg ratio Precision Recall F-score 1 Reported by Aker et al. (2013) 12,400 1:1 0.9800 0.8200 0.8000 2 Replicated approach 13,160 1:1 0.0227 0.8850 0.0442 3 Giza?? terms only 7310 1:1 0.0636 0.9317 0.1191 4 Giza?? 3.2.3 Attempts at establishing contact with the authors cleaning 13,160 1:1 0.0340 0.8500 0.0654 5 Training set 1:200 1,322,580 1:200 0.5053 0.6300 0.5608 6 Training set filtering 1 4250 1:1 0.5122 0.4917 0.5017 7 Training set filtering 2 23,375 1:10 0.6842 0.4333 0.5306 8 Training set filtering 3 427,125 1:200 0.9356 0.3633 0.5234 9 Term length filtering 4250 1:1 0.7621 0.3683 0.4966 10 Cognates approach 468,933 1:200 0.9101 0.5233 0.6646 No. 1 presents the results reported by the authors, No. 2 our replication of the approach and No. 3–10 our modifications of the first replicated approach with the aim of improving the results Table 5 Results on the English–Dutch language pair No. 1 presents the results reported by the authors, No. 2 our replication of the approach and No. 3–10 our modifications of the first replicated approach with the aim of improving the results 123 4.2 Giza11 cleaning The output of the Giza?? tool contained a lot of noise and we thought it could perhaps have a detrimental effect on the results. There is no mention of any sophisticated Giza?? dictionary cleaning in the original paper beyond removing all entries where probability is lower then 0.05 and entries where the source word is less than 4 characters and the target word more than 5 characters in length and vice versa (introduced to avoid stopword-content word pairs). For clean Giza?? dictionaries, we used the resources described in Aker et al. (2014), available via the META-SHARE repository14 (Piperidis et al. 2014), specifically, the transliteration- based approach which yielded the best results according to the cited paper. For Slovenian and Dutch, precision and F-score improved marginally at a cost of a lower recall, while for French, precision, recall and F-score all decreased. For details, see line 4 in Tables 3, 4 and 5. 4 Analysis and adaptation: experiments for improving the replicated approach The results in our replicated experiments differ dramatically from the results obtained by Aker et al. (2013). Their approach yields excellent results with perfect or almost perfect precision and respectable recall for all three languages under our consideration. For the EN-SL language pair, the reported results have the precision of 100% and the recall of 66%, meaning that with 600 positive term pairs in the test set, their classifier returns only around 400 positive term pairs. In contrast, in our replication attempts the classifier returned a lot of falsely classified positive term pairs. In addition to 526 true positive examples (out of a total of 600), the classifier also returns 14,194 misclassified examples—incorrect term pairs wrongly classified as correct. Similar statistics can be observed for the other two language pairs. These results are clearly not useful for our goals which is to use the methods to continuously populate a termbase with as little manual intervention as possible. In this section we present the analysis of ambiguities in the description of the approach and the issues spotted when inspecting the results of the replicated approach, and propose several methods aiming at improving the results. To do so, we have performed experiments with regard to the following aspects: – Giza?? terms only: using only those terms that can be found in the Giza?? training corpora (i.e. DGT). Gi ?? l i – Giza?? terms only: using only those terms that can be found in the Giza?? training corpora (i.e. DGT). – Giza?? cleaning. – Lemmatization. – Changing the ratio of positive/negative examples in the training set. – Training set filtering. – Training set filtering. 123 783 Reproduction, replication, analysis and adaptation... The experiments have been initially presented for Slovenian in our short paper in the 4REAL workshop (Repar et al. 2018). Here, we provide additional analysis and extend the experiments to the other two languages under consideration. The results are reported in Sect. 4.1 to 4.5. In the 4REAL paper, precision was already relatively high (see for example line 8 in Table 3), which is why our additional experiments focused on improving recall. We implemented several additional approaches as reported in Sect. 4.6 to 4.8: – Removing the Needleman–Wunsch Distance feature. – Term length filtering. – Adding new cognate-based features. 4.1 Giza11 terms only We thought that one of the reasons for low results can be that not all EUROVOC terms actually appear in the Giza?? training data (i.e. DGT translation memory). The terms that do not appear in the Giza?? training data could have dictionary- based features similar to the generated negative examples, which could affect the precision of a classifier that was trained on those terms. We found that only 4,153 out of 7,083 Slovenian terms of the entire EUROVOC thesaurus do in fact appear in a DGT translation memory. Using only these terms in the classifier training set did provide modest improvements of precision, recall and F-score across all three languages. For details, see line 3 in Tables 3, 4 and 5. 14 http://metashare.tilde.com, last accessed: February 14, 2019. 4.4 Changing the ratio of positive/negative examples in the training set In the original paper, the training set is balanced (i.e. the ratio of positive vs. negative examples is 1) but the test set is not (the ratio is around 1:2000). Since our classifier had low precision and relatively high recall, we figured that an unbalanced training set with much more negative than positive examples could improve the former. To test this, we experimented with training the classifier on unbalanced train sets with different ratios between positive and negative examples. The general tendency we noticed during experimentation is that a very unbalanced train set (ratio of 1:200 between positive and negative examples15) greatly improves the precision of the classifier at a cost of somewhat lower recall, when compared to balanced train set or less unbalanced train set (e.g., ratio of 1:10 between positive and negative examples). For details, see line 5 in Tables 3, 4 and 5. 4.3 Lemmatization The original paper does not mention lemmatization which is why we assumed that all input data (Giza?? dictionaries, EUROVOC thesaurus) is not lemmatized. They 12 123 784 A. Repar et al. state that to capture words with morphological differences, they don’t perform direct string matching but utilize Levenshtein Distance and two words are considered equal if the Levenshtein Distance (Levenshtein 1966) is equal or higher than 0.95. This led us to believe that no lemmatization was used. Nevertheless, we thought lemmatizing the input data could potentially improve the results which is why we adapted the algorithm to perform lemmatization (using Lemmagen (Jursˇicˇ et al. 2010)) of the Giza?? input data and the EUROVOC terms. We have also removed the Levenshtein distance string matching and replaced it with direct string matching (i.e. word A is equal to word B, if word A is exactly the same as B), which drastically improved the execution time of the software. We considered lemmatization as a factor that could explain the difference in results obtained by us and Aker et al. (2013), but our experiments on lemmatized and unlemmatized clean Giza?? dictionaries show that lemmatization does not have a significant impact on the results. Compared to the configuration with unlemmatized clean Giza?? dictionaries, in the configuration with lemmatized Giza?? dictionaries precision was slightly lower (by 0.1%), recall was a bit higher (by around 4%) and F-score was lower by 0.2%. For details, see Table 3, line 4a. As lemmatization significantly slows down the experimentation, we tested the results first on Slovenian, where the influence of the lemmatization should be the largest as it is a morphologically-rich language. As lemmatization did not improve the results, we did not repeat the experiments for French and Dutch. 15 1:200 imbalance ratio was the largest imbalance we tried, since the testing results indicated that no further gains could be achieved by further increasing the imbalance. 4.5 Training set filtering The original paper mentions that their classifier initially achieved low precision on Lithuanian language training set, which they were able to improve by manually removing 467 positive term pairs that had the same characteristics as negative examples from the training set. No manual removal is mentioned for Slovenian, French and Dutch. 123 785 Reproduction, replication, analysis and adaptation... We have performed an error analysis and found that many incorrectly classified term pairs are cases of partial translation where one unit in a multi-word term has a correct Giza?? dictionary translation in the corresponding term in the other language. Some EN-SL examples can be seen in Table 6, and similar errors were observed for for the other two language pairs. Based on this problem of partial translations, leading to false positive examples, we focused on the features that would eliminate these partial translations from the training set. After a systematic experimentation, we noticed that we can drastically improve precision if we only keep positive term pairs with the following feature values in the training set: – isfirstwordTranslated = True. – islasttwordTranslated = True. – percentageOfCoverage [ 0:66. – isfirstwordTranslated-reversed = True. – islasttwordTranslated-reversed = True. – percentageOfCoverage-reversed [ 0:66. Using this approach, we managed to greatly increase precision at a cost of significant drop in recall values for all three languages. For details see line 6 (Training set filtering 1) in Tables 3, 4 and 5. When combining this approach with an unbalanced dataset described in the previous section, we managed to improve precision even further, but again at a cost of lower recall. For details, see lines 7 and 8 (Training set filtering 2 and 3) in Tables 3, 4 and 5. 4.6 Cognate feature analysis and removing the Needleman–Wunsch Distance feature Note that due to character mapping rules (see Section 3.2.1.), the word ‘‘predstavnisˇka’’ was transformed into ‘‘predstavnishka’’ Table 7 Cognate-based features values (showing issues with NWD) The first two term pairs are actual cognates with all five cognate-based features having high values. The last two pairs are not cognates and show the issues with the Needleman-Wunsch Distance (NWD), which is the only measure that keeps a high value. Note that due to character mapping rules (see Section 3.2.1.), the word ‘‘predstavnisˇka’’ was transformed into ‘‘predstavnishka’’ For this reason, we ran our experiments without the NWD feature, but the results did not improve since the SVM classifier is known to be capable of handling noisy features. 4.6 Cognate feature analysis and removing the Needleman–Wunsch Distance feature We performed an analysis of the results on the English–Slovenian language pair achieved with the best configuration for precision (line 8—Training set filtering 3 in Table 3) in our experiments (Repar et al. 2018) and discovered that cognate term pairs were not being considered by the classifier. In a way, this was expected since in the previous step we have filtered the training set based on mostly dictionary- based features. When analyzing the performance of the cognate-based features, we found that four (Longest Common Subsequence Ratio (LCSSR) Longest Common Substring Ratio (LCSTR), Dice Similarity (Dice), Normalized Levenshtein Distance (nLD)) out of five perform as expected with cognate term pairs having high values, but Needleman-Wunsch Distance (NWD) did not. As already mentioned in the beginning, the formula provided by the authors for computing NWD feature possibly contained an error, therefore we opted for the implementation as mentioned in Nassirudin and Purwarianti (2015). Table 7 shows the behaviour of the five cognate-based features. When we are dealing with actual cognates, all five features have high values, but when the two terms in questions are not cognates, only NWD stays high. 12 123 A. Repar et al. 786 Table 6 Examples of negative term pairs misclassified as positive EN SL Giza?? Agrarian reform Kmetijski odpadki Agrarian, kmetijske, 0.29737 Brussels region Obmocˇje proste trgovine Region, obmocˇje, 0.0970153 Energy transport Nacionalni prevoz Transport, prevoz, 0.442456 Fishery product Tekstilni izdelek Product, izdelek, 0.306948 Column 1 contains the English term, column 2 contains the Slovenian term and column 3 contains the Giza?? dictionary entry (from the non-clean version, see Sect. 4.2) responsible for positive dictionary- based features Table 6 Examples of negative term pairs misclassified as positive Table 7 Cognate-based features values (showing issues with NWD) EN SL LCSSR LCSTR Dice nLD NWD hospitalisation hospitalizacija 0.73 0.60 0.60 0.73 0.6 monopsony monopson 0.89 0.89 0.94 0.89 1.00 fish predstavnisˇka demokracija 0.12 0.12 0.20 0.12 0.75 Yemen osna obremenitev 0.25 0.25 0.38 0.25 0.80 The first two term pairs are actual cognates with all five cognate-based features having high values. The last two pairs are not cognates and show the issues with the Needleman-Wunsch Distance (NWD), which is the only measure that keeps a high value. 4.7 Term length filtering Based on error analysis, one of the major issues confusing the classifier were training examples with differing word lengths. E.g., the source term in the example would have one word, but the target term would have two. An analysis of the terms in Eurovoc for the three language pairs in question showed that 26% of the EN-SL term pairs, 34% of the EN-FR term pairs and 48% of the EN-NL term pairs have different word lengths of the source and target terms (the reason for the high ratio in EN-NL is the use of compounds in Dutch). This turned out to be one of the characteristics leading to low classification performance: for Slovenian with the replicated configuration (line 2 in Table 3) the classifier returned a total of 14,721 positively classified examples. 14,193 out of these were false positives— incorrectly aligned term pairs. A further 13376 out of these had different lengths of the source and target terms. A visual inspection of feature values indicated that there is often no clear difference between positive and negative term pairs (see Table 8). 123 123 787 Reproduction, replication, analysis and adaptation... Since this was an issue, we experimented with additional term length filtering. We took the positively classified examples from the training set filtering 1 approach as described in Sect. 4.5 (see line 6 in the tables) and added an additional filter: if the two terms do not have the same number of words, we change the prediction from positive to negative. Using this additional filter, we achieved good precision for Slovenian (81%), and respectable for French (68%) and Dutch (76%). On the other hand, recall values were badly affected, since one third of positive term pairs in the constructed test set are terms of different word length (meaning that highest possible theoretical recall with this approach is 66%). Recall was again best for Slovenian with a value close to 50% and a bit worse for French and Dutch with a value at around 40% and 37% respectively. Consequently, F-scores were the highest for Slovenian and lower for Dutch and French. For details, see line 9 in Tables 3, 4 and 5. From the original paper it is clear, that authors were aware of the possible complexity of terms of unequal length, as they consider terms of different lengths in the test set construction. 4.7 Term length filtering So, we exclude the possibility that authors did not have such examples in the test set. 4.8 Cognate-based feature approach The analysis showed that all Training set filtering approaches tend to overestimate the importance of Giza?? features and underestimate cognate-based features. This results in a low recall for correct cognate term pairs, which are rarely classified as positive, if their Giza?? based feature values do not show similarity with Giza?? based feature values for non-cognate correct term pairs. For example, Giza?? dictionary does not contain a Slovenian translation pacifizem for the English term pacifism, which means that the values of features isFirstWordTranslated, isLastWordTranslated, isFirstWordTranslated-reversed and isLastWordTrans- lated-reversed are False and the values for features percentageOfCoverage and percentageOfCoverage-reversed are zero, therefore the classifier would have a strong inclination to classify this correct term pair as incorrect, even though cognate based feature values clearly indicate that these two terms are cognates. In order to improve the detection of cognate terms, we first propose two new cognate based features: – isFirstWordCognate: a binary feature which returns True if the longest common consecutive string (LCST) of the first words in the source and target terms divided by the length of the longest of the two words is greater than or equal to a threshold value of 0.7 and both words are longer than 3 characters. For example, the value of the feature for the English-Slovenian term pair Klaipeda county - Klaipedsko okrozˇje would be True because the LCST for the first words in both terms is Klaiped, which has a length of 7. The length of the longest of the two first words in the terms (Klaipedsko) is 10 and 7 divided by 10 is 0.7, which is equal to the threshold value. – isLastWordCognate: a binary feature which returns True if the longest common consecutive string (LCST) of the last words in the source and target terms – isLastWordCognate: a binary feature which returns True if the longest common consecutive string (LCST) of the last words in the source and target terms 12 123 123 788 A. Repar et al. 4.8 Cognate-based feature approach Table 8 A comparison of dictionary feature values Source term raw material provision additional resources provision Target term surovine racˇunovodska rezervacija surovine urbanisticˇni predpisi Correctly aligned True True False False isFirstWordTranslated 1 0 0 0 isLastWordTranslated 1 1 1 1 pctOfTransWords 0.5 1 0.5 1 pctOfNotTransWords 0.5 0 0.5 0 longestTransUnitInPct 0.5 1 0.5 1 longestNotTransUnitInPct 0.5 0 0.5 0 isFirstWordTranslated_R 0 0 0 0 isLastWordTranslated_R 1 1 1 1 pctOfTransWords_R 1 0.5 1 0.5 pctOfNotTransWords_R 0 0.5 0 0.5 longestTransUnitInPct_R 1 0.5 1 0.5 longestNotTransUnitInPct_R 0 0.5 0 0.5 The first two term pairs are correctly alligned term pairs from the training set (line 2 in Table 3), the second two are not correctly alligned term pairs. We can observe that the dictionary feature values are very similar—compare for example raw material/surovine and additional resources/surovine Table 8 A comparison of dictionary feature values 12 3 Reproduction, replication, analysis and adaptation... 789 divided by the length of longest of the two words is greater than or equal to a threshold value of 0.7 and both words are longer than 3 characters. For example, the value of the feature for the English-Slovenian term pair Latin America - Latinska Amerika would be True because the LCST for the last words in both terms is Ameri, which has a length of 5. The length of the longest of the two last words in the terms is 7 and 7 divided by 5 is 0.714, which is greater than the threshold value. As having the same number of words in the source and target term could play a role in classification, we also add three new features responsible for encoding term length information: – sourceTargetLengthMatch: a binary feature that returns True if the number of words in source and target terms match. – sourceTargetLengthMatch: a binary feature that returns True if the number of words in source and target terms match. – sourceTermLength: returns the number of words in the source term. – targetTermLength: returns the number of words in the target term. Analysis of the filtered training set showed that it contained a small number of positive cognate based term pair examples, therefore the first step was to include more of them into the dataset. We build three separate datasets, each of them filtered according to the following feature values: – isFirstWordCognate = True and isLastWordCognate = True. 4.8 Cognate-based feature approach – isFirstWordTranslated = True and isLastWordCognate = True. – isFirstWordCognate = True and isLastWordTranslated = True. The terms from these three datasets are added to the original filtered train set (we make sure that each positive term pair is represented in the new dataset only once by removing all the duplicates). The new dataset contains two distinct groups of terms, one with favorable Giza?? based features (and unfavorable cognate based features) and one with favorable cognate based features (and in some cases unfavorable Giza?? based features). Since this new dataset structure represents a classic ‘‘exclusive or’’ (XOR) problem which a linear classifier is unable to solve, we also replace the linear kernel of our SVM classifier with the Gaussian one. Using this approach, precision was close to 90% (Slovenian, French) or just over 90% (Dutch), recall was just over 50% for Slovenian, around 52% for Dutch and close to 40% for French. For details, see line 10 in Tables 3, 4 and 5. 5.1 Replicating the manual evaluation experiments from the original paper Similar to the original paper, we also performed manual evaluation. We selected a random subset of term pairs classified as positive by the classifier (using the Training set filtering 3 configuration (line 8 in Table 3) that yielded the best precision). While the authors of the original approach extract monolingual terms using the term extraction and tagging tool TWSC (Pinnis et al. 2012), we use a workflow for monolingual term extraction by Pollak et al. (2012). Both use a similar approach - terms are first extracted using morphosyntactic patterns and then filtered using statistical measures: TWSC uses pointwise mutual information and TF*IDF, while Pollak et al. (2012) is based on an approach by Vintar (2010) and compares the relative frequencies of words composing a term in the domain-specific (i.e. the one we are extracting terminology from) corpus and a general language corpus. g gy p g g g p In contrast to the original paper where they extracted terms from domain-specific Wikipedia articles (for the English-German language pair), we are using two translation memories—one containing finance-related content, the other containing IT content. Another difference is that extraction in the original paper was done on comparable corpora, but we extracted terms from parallel corpora - which is why we expected our results to be better. Each source term is paired with each target term (just as in the original paper - if both term lists contained 100 terms, we would have 10,000 term pairs) and extract the features for each term pair. The term pairs were then presented to the classifier that labeled them as correct or incorrect term translations. Afterwards, we took a random subset of 200 term pairs classified as correct and showed them to an experienced translator16 fluent in both languages who evaluated them according to the criteria set out in the original paper: – 1—Equivalence: The terms are exact translations/transliterations of each other (e.g., type—tip). – 1—Equivalence: The terms are exact translations/transliterations of each other (e.g., type—tip). – 2—Inclusion: Not an exact translation/transliteration, but an exact transla- tion/transliteration of one term is entirely contained within the term in the other language (e.g., end date—datum). – 2—Inclusion: Not an exact translation/transliteration, but an exact transla- tion/transliteration of one term is entirely contained within the term in the other language (e.g., end date—datum). 4.9 Best results Overall, the setting with the best precision is Train set filtering 3. Compared to the replicated approach (line 2 in Tables 3, 4 and 5), it has an unbalanced dataset of 1:200 (see Section 4.4) and employs the term filtering strategy described in Sect. 4.5. However, for a small gain in recall at the price of a slight decrease in precision, a good alternative is the Cognates approach (line 10 in Tables 3, 4 and 5), which is 12 123 790 A. Repar et al. based on the Train set filtering 3 approach and additionally includes the cognate detection strategies described in Sect. 4.8. based on the Train set filtering 3 approach and additionally includes the cognate detection strategies described in Sect. 4.8. 16 The original paper used two annotators, hence two lines for each domain in Table 4. 5 Manual evaluation The first part of this section contains the manual evaluation replicated from Aker et al. (2013), already reported in Repar et al. (2018), while the second part is novel and contains an evaluation using a new dataset and has a specific focus on cognate term pairs. – 1—Equivalence: The terms are exact translations/transliterations of each othe (e.g., type—tip). 18 http://source.ijs.si/mmartinc/4real2018/tree/master/datasets/karst_corpus. 5.1 Replicating the manual evaluation experiments from the original paper – 3—Overlap: Not category 1 or 2, but the terms share at least one translated/transliterated word (e.g., user id—uporabniško ime). 123 Reproduction, replication, analysis and adaptation... 791 – 4—Unrelated: No word in either term is a translation/transliteration of a word in the other (e.g., level—uporabnik17). – 4—Unrelated: No word in either term is a translation/transliteration of a word in the other (e.g., level—uporabnik17). The results of the manual evaluation can be found in Table 9. Manual evaluation showed that 72% of positive term pairs in the Finance domain, and 79% of positive term pairs in the IT domain were correctly classified by the classifier. The differences between the Finance and IT datasets can be partially explained by the Finance dataset containing more MWE terms than the IT dataset (84 vs. 51 for SL and 78 vs. 49 for EN). On the one hand, this means that the chances of aligning a single word term in one language with a multi word term in another language is greater, hence the greater number of partial translations in Finance (category 2 - Inclusion), while on the other, single word terms means less characters for the algorithm to work with, hence the greater number of outright mistakes in IT (category 4 - Unrelated). Compared to the original paper, we believe these results are comparable when taking into account the different monolingual extraction procedures , the different language pairs and the human factor related to different annotators. 17 ‘‘uporabnik’’ means ‘‘user’’. 5.2 Evaluation on a Karst terminology gold-standard As mentioned in Sect. 4, the best configuration in terms of precision used in Repar et al. (2018) (line 8 in Tables 3, 4 and 5) overestimates dictionary-based and underestimates cognate-based features. To alleviate this, we added additional features and filtering strategies to our approach to try to improve cognate term pair alignment (see lines 9 and 10 in the results tables). However, evaluating its performance on EUROVOC is difficult as many terms have favorable dictionary- based features due to the fact that both the Giza?? dictionary and EUROVOC are made from the same content (i.e. EU documentation). For the evaluation in this section, we therefore selected a domain, with a content type which is unlikely to be found in DGT (Steinberger et al. 2013), i.e. karstology, which is the science in the field of geomorphology, specializing in the study of karst formations. To evaluate our bilingual term alignment approach, we used a gold standard of EN-SL aligned karst terminology,18 which was manually created by the authors of the karstology corpus (Vintar and Grcˇic´-Simeunovic´ 2016). The gold standard consists of 52 English-Slovenian term pairs. For the evaluation experiment, we aligned all Slovenian term with all English terms, resulting in a dataset of 52 positive examples and 2652 negative examples. With the best configuration for precision (line 8 in Table 3), selected also as the best configuration in Repar et al. (2018), precision was 100%, but recall was only 40.4%. Many term pairs containing cognates such as ‘‘eogenetic cave—eogenetska jama’’, ‘‘epigenic aquifer—epigeni vodonosnik’’ or ‘‘karst polje—krasˇko polje’’, were not aligned. With the final cognate approach (line 10 in Table 3), we managed to retain 100% precision and raise the recall to 50% by finding 7 additional cognate term pairs (aggressive 12 123 792 A. Repar et al. water—agresivna voda, eogenetic cave—eogenetska jama, precipitation—precip- itacija, ponor cave—ponorna jama, epigenic aquifier—epigeni vodonosnik, karst polje—kraško polje, linear stream cave—linearna epifreaticˇna jama). However one half of correct term pairs remain undiscovered. We believe this is due to 1) domain- specific words which are not cognates and are missing from the Giza?? dictionary (e.g., porous aquifer—medzrnski vodonosnik and denuded cave—brezstropa jama), and 2) valid cognate words which do not meet the threshold described in Sect. 4.8 (e.g. oxidization—oksidacija, percolation—perkolacija and liquefication— likvifakcija).19 19 It might also make sense to include morphological information as a feature of the machine learning algorithm, since all these word have endings typical of cognates in their respective languages. 20 2 Master students (one in Economy and one in Computer science) and 1 first year PhD student in ICT. 6 Replicability and reproducibility of our own terminology alignment results As mentioned before, availability of the source code can drastically improve the reproducibility of experiments, since very detailed descriptions of procedures used in the experiments are beyond the scope of most papers because of length limitations and negative effects on the readability of the paper. Since we wanted to ensure the full reproducibility of our approach, we decided to publish the source code for all the conducted experiments and results that are published in the paper. As we were aware that just the presence of source code itself does not guarantee complete reproducibility, we decided that the published code should comply to the following three criteria: – Instructions on how to use the code should be as unambiguous, simple and clear as possible. – Instructions on how to use the code should be as unambiguous, simple and clear as possible. p – Code should be bug free and running it according to the instructions should yiel the exact same results as published in the paper. – Running the code should require as little time and technical skills as possible In order to validate that the published code complies to these criteria, we asked three students20 to try to reproduce the results published in the paper (Repar et al. 2018) and after that answer the following questions related to the proposed criteria: – Did you manage to reproduce the results? – If not, what do you think was the main problem? – If yes, how much time did you need for replicating the experiment? – Were the instructions clear? – Did you run into any specific problems during any part of the replicability attempt? If yes, please describe it. – Do you have any suggestions on how to further improve the reproducibility of the results? 123 Reproduction, replication, analysis and adaptation... 793 Table 9 Manual evaluation results Ann. stands for ‘‘Annotator’’ since the original paper uses two annotators Domain 1 2 3 4 Reported in Aker et al. (2013) IT, Ann. 1 0.81 0.06 0.06 0.07 IT, Ann. 2 0.83 0.07 0.07 0.03 Auto, Ann. 1 0.66 0.12 0.16 0.06 Auto, Ann. 2 0.60 0.15 0.16 0.09 Replication Finance 0.72 0.09 0.12 0.07 IT 0.79 0.01 0.09 0.12 We also imposed a time limit of 8 hours (one working day) for the entire replicability attempt. If that limit was reached, the replicability attempt would count as unsuccessful. 21 Note that the execution time of term alignment increases rapidly with the increase in number of terms, e.g., alignment of hundred terms takes around five minutes, while it takes about one hour for alignment of thousand terms. – Instructions should clearly specify on which operating system and in which programming environment the reported results were produced. We have updated the usage instructions for our source code to comply with these criteria. 6 Replicability and reproducibility of our own terminology alignment results The feedback we got was interesting and made us reconsider the initial source code criteria. Two out of three students managed to reproduce all the published results in less than an hour without any major problems. They did however point out some mistakes and ambiguities in the instructions on how to run the code. These were mostly connected with the programming environment used by the students, one of them using PyPI Python package manager for acquiring dependencies while the other one used the Conda environment, for which the usage instructions were not published. The third student managed to reproduce the results in about two hours and reported some major problems with dependencies installation. He was the only person trying to reproduce the experiments in the Windows environment while the other two students used a Linux operating system, and he reported problems with the Python implementation of the Lemmagen lemmatizer (Jursˇicˇ et al. 2010), which he was unable to install properly on the Windows platform. He managed to overcome the problem by manually removing the dependency from the code, by which he limited the flexibility of the published source code (he could only use it for the classification on the pre-generated train and test sets) but did not make the reproduction impossible. While he was successful at reproducing the results for eight out of nine experiments published in the paper, he also reported a slight deviation (by less than 0.05 percentage point) from the reported recall and F-score in one of the experiments. Although we are not sure what is the exact reason for this deviation, we suspect it could be connected to the difference in operating systems. These experiments show that programming environment and the choice of the operating system can have an unexpected negative impact on the reproducibility. While attaching code usage instructions for every possible programming environ- ment and operating system is practically impossible, we do believe that the results of this experiment show that a published source code should comply to one additional criteria: 123 794 A. Repar et al. – Instructions should clearly specify on which operating system and in which programming environment the reported results were produced. – Instructions should clearly specify on which operating system and in which programming environment the reported results were produced. 7 Reusability of our code in the ClowdFlows online platform Because we want to make sure that our terminology alignment system is also available to a wider audience of users with lower level of technical skills (e.g., translators or linguists) and because we want to encourage a very simple reusability of our system, we have implemented the system into a cloud-based visual programming platform ClowdFlows (Kranjc et al. 2012). The ClowdFlows platform employs a visual programming paradigm in order to simplify the representation of complex data mining procedures into visual arrangements of their building blocks. Its graphical user interface is designed to enable the users to connect processing components (i.e. widgets) into executable pipelines (i.e. workflows) on a design canvas by a drag and drop technique, reducing the complexity of composition and execution of these workflows. The platform also enables online sharing of the composed workflows. We took pretrained models of our terminology alignment system for English- Slovenian, English-French and English-Dutch alignment and packed them in a widget Terminology alignment, so it can be used out-of-the-box. The widget takes two columns of the Pandas dataframe (McKinney 2011) containing the source and target terms as inputs and returns a dataframe containing aligned term pairs. The user needs to define the names of the columns in the dataframe containing source and target language termlists, and the language of alignment as parameters. The user can also switch between configurations Training set filtering 3 with the best precision and Cognates approach with the on average best F-score for all three languages while still having good precision by either enabling or disabling the Maximize recall widget parameter. Such an end to end system for bilingual terminology alignment in ClowdFlows is implemented at: http://clowdflows.org/ workflow/13789/.21 Another widget called Terminology alignment evaluation is used for determining the performance of the system (if we have a gold standard available), taking as input the dataframe produced by the Terminology alignment widget and a dataframe containing true alignments, and outputting the performance score in terms of precision, recall and F-score. Workflow in Fig. 1 (available at http://clowdflows.org/workflow/13753/) is a ClowdFlows implementation for terminology alignment and evaluation. The source and target terminologies are both loaded from a CSV file with the help of the Load Corpus From CSV widget and fed as input to the Terminology alignment widget, 123 Reproduction, replication, analysis and adaptation... 795 Fig. 1 ClowdFlows implementation of the system for terminology alignment and evaluation available at http://clowdflows.org/workflow/13753/ Fig. 7 Reusability of our code in the ClowdFlows online platform 1 ClowdFlows implementation of the system for terminology alignment and evaluation available at http://clowdflows.org/workflow/13753/ which returns a dataframe with alignments. These are written to a CSV file with the Corpus to CSV widget and also fed to the Terminology alignment evaluation widget together with the dataframe containing true alignments (which was also loaded from a CSV file with the Load Corpus From CSV widget) in order to estimate the performance of the system. In addition, term alignment widget can also be incor- porated into a bilingual terminology extraction workflow (Pollak et al. 2012). The workflow with the newly added term alignment widget, is available at http:// clowdflows.org/workflow/13723/), where a user can now input text from a specific domain in Slovenian and English and get aligned terminology as output. 8 Conclusions and future work Based on our research and attempts at replicating a bilingual terminology alignment paper reproducing its results, we propose a set of best practices any bilingual terminology extraction paper (and more generally every NLP paper) should fulfill to facilitate reproducibility and replicability of the experiments: – Dataset availability. Availability of datasets (i.e. gold standard term lists, corpora) is an essential prerequisite for successful replication. – Dataset availability. Availability of datasets (i.e. gold standard term lists, corpora) is an essential prerequisite for successful replication. – Experiment code availability. The main task of reproducibility and replicability experiments is often to reconstruct the experiments in computer code. It is a cumbersome process which inevitably requires that the reproducer/replicator makes educated guesses at some point since a detailed description of the code is beyond the scope of most papers. Having the original code available greatly increases the ease of reproducibility and replicability experiments. – Tool availability. Availability of a tool or application (online or offline) where experiments can be conducted eases reproducibility and replicability, but also enables the reusability of results by a larger community. – Tool availability. Availability of a tool or application (online or offline) where experiments can be conducted eases reproducibility and replicability, but also enables the reusability of results by a larger community. – Finally, releasing intermediate results, configuration settings and the actual outcomes of individual experiments, while not essential, would provide future researchers with an even greater possibility of successful reproduction of the paper’s results. 123 123 796 A. Repar et al. A prerequisite for successful reproduction and replication is a clearly written research paper. However as is evident from our example, it is often difficult to include all necessary implementation notes given the length restrictions of the paper. For this reason, another best practice would be to provide relevant implementation examples alongside the code (which is what we did for feature construction.22) Finally, as the experiment in Sect. 6 showed, even code itself is sometimes not enough without additional implementation notes and information on the operating systems and software used. In addition, testing the code by non- authors is strongly recommended. Our attempts focused on the approach to bilingual term alignment using machine learning by Aker et al. (2013). They approach term alignment as a bilingual classification task—for each term pair, they create various features based on word dictionaries (i.e. created with Giza?? 22 http://source.ijs.si/mmartinc/4real2018/blob/master/feature_examples.docx. 8 Conclusions and future work from the DGT translation memory) and word similarities across languages. They evaluated their classifier on a held-out set of term pairs and additionally by manual evaluation. Their results on the held-out set were excellent, with 100% precision and 66% recall for the English-Slovenian and English-French language pair and 98% precision and 82% recall for English-Dutch. Our reproduction attempt focused on three language pairs: English-Slovenian, English-Dutch and English-French (in contrast with the original article where they had altogether 20 language pairs) and we were unable to reproduce the results following the procedures described in the paper. In fact, our results have been dramatically different from the original paper with precision being less than 4% and recall close to 90% for all three language pairs under consideration. We then tested several different strategies for improving the results ranging from Giza?? dictionary cleaning, lemmatization, different ratios of positive and negative examples in the training and test sets, training set filtering based on feature values and term length, and adding new cognate-based features. The most effective strategies employed unbalanced training set and training set filtering based on certain feature values which resulted in precision exceeding 90% for all three language combinations (Training set filtering 3 configuration, line 8 in Tables 3, 4 and 5). It is possible that in the original experiments authors performed a similar training set filtering strategy, because the original paper mentions that their classifier initially achieved low precision on Lithuanian language training set, which they were able to improve by manually removing positive term pairs that had the same characteristics as negative examples from the training set. However, no manual removal is mentioned for Slovenian, Dutch or French. Further attempts were directed at boosting recall and the performance of cognate-based features. By adding additional cognate-based features, we were able to improve recall by around 16% for Dutch, 8% for French and by around 2% for Slovenian (over the Training set filtering 3 configuration) at a cost of a moderate drop in precision. For evaluation we focused only on Slovenian, which is our native language and of primarily interest for our applied tasks. We performed manual evaluation similar to the original paper and reached roughly the same results with our adapted approach. In addition, because we discovered that Eurovoc data is of limited use for 123 Reproduction, replication, analysis and adaptation... 8 Conclusions and future work 797 evaluating the performance of cognate-based features, we ran experiments on an English-Slovenian karstology gold standard term list. With the Cognates approach configuration (line 10 in Tables 3, 4 and 5), we improved recall by 11% (compared to the Training set filtering 3 configuration) and a qualititive analysis of the results showed that the new strategies for boosting the performance of cognate-based features do indeed result in more cognate term pairs being properly aligned. This paper demonstrates some of the obstacles for research reproducibility and replicability, with the prime one being code unavailability. Had we had access to the code of the original experiments, it is highly likely that replicating the original paper would be a trivial matter. Also in this particular case, the discrepancy in the results could be attributed to the scope of the original paper - with more than 20 languages—which is also a demonstration of very impressive approach—it would be impossible to describe procedures for all of them. We weren’t able to reproduce the results of the original paper, but after developing the optimization approaches described above over the course of several months, we were able to reach a useful outcome at the end. We believe that providing supplementary material online, i.e. the code and datasets, is the only way of assuring complete reproducibility of results. For this reason, in order to help with any future reproducibility/replicability attempts of our paper, we are publishing the code at: http://source.ijs.si/mmartinc/ 4real2018. In terms of future work, we plan to expand the feature set by introducing the features derived from the distributions in parallel corpora (e.g. co-frequency, logDice and other measures, see Baisa et al. (2015)), as well as investigate novel methods using cross-lingual embeddings. In terms of reproducibility, we plan to extend the study to a systematic comparison of different term alignment and term extraction methods. Acknowledgements This paper is supported by European Union’s Horizon 2020 research and innovation programme under Grant Agreement No. 825153, project EMBEDDIA (Cross-Lingual Embeddings for Less-Represented Languages in European News Media). The authors acknowledge also the financial support from the Slovenian Research Agency for research core funding for the programme Knowledge Technologies (No. P2-0103). The authors also acknowledge the project TermFrame—Terminology and Knowledge Frames across Languages (No. J6-9372), which was financially supported by the Slovenian Research Agency. 8 Conclusions and future work We would also like to thank the company Iolar, for allowing us to use the data from the translation memories in one of the experiments. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, dis- tribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Aker, A., Paramita, M., & Gaizauskas, R. (2013). Extracting bilingual terminologies from comparable corpora. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics: Vol. 1. Long Papers (pp 402–411). References Aker, A., Paramita, M., & Gaizauskas, R. (2013). Extracting bilingual terminologies from comparable corpora. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics: Vol. 1. Long Papers (pp 402–411). 123 798 A. Repar et al. Aker, A., Paramita, M. L., Pinnis, M., & Gaizauskas, R. (2014). Bilingual dictionaries for all EU languages. In Proceedings of 9th International Conference on Language Resources and Evaluation. (pp 2839–2845). Arcˇan, M., Turchi, M., Tonelli, S., & Buitelaar, P. (2014). Enhancing statistical machine translation with bilingual terminology in a CAT environment. https://doi.org/10.13140/2.1.1019.8404. Bader, B. W., & Chew. P. A. (2008). Enhancing multilingual latent semantic analysis with term alignment information. In Proceedings of the 22nd International Conference on Computational Linguistics: Vol. 1. Association for Computational Linguistics (pp 49–56). g f p g pp Baisa, V., Ulipova´, B., & Cukr, M. (2015). Bilingual terminology extraction in Sketch Engine. In 9th Workshop on Recent Advances in Slavonic Natural Language Processing. (pp 61–67). Baldwin, T., & Tanaka, T. (2004). Translation by machine of complex nominals: Getting it right. In Proceedings of the Workshop on Multiword Expressions: Integrating Processing. (pp 24–31). Bouamor, D., Semmar, N., Zweigenbaum, P. (2013). Context vector disambiguation for bilingual lexicon extraction from comparable corpora. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics: Vol. 2: Short Papers. (pp 759–764). Branco, A., Calzolari, N., & Choukri, K. (eds) (2018). 4REAL 2018—Workshop on Replicability and Reproducibility of Research Results in Science and Technology of Language, ELRA. Branco, A., Cohen, K. B., Vossen, P., Ide, N., & Calzolari, N. (2017). Replicability and reproducibility of research results for human language technology: introducing an LRE special section. Camerer, C. F., Dreber, A., Forsell, E., Ho, T. H., Huber, J., Johannesson, M., et al. (2016). Evaluatin replicability of laboratory experiments in economics. Science, 351(6280), 1433–1436. Cao, Y,, & Li, H. (2002). Base noun phrase translation using web data and the EM algorithm. Proceedings of the 19th International Conference on Computational Linguistics: Vol. 1. (pp 1–7 Cao, Y,, & Li, H. (2002). Base noun phrase translation using web data and the EM algorithm. In Proceedings of the 19th International Conference on Computational Linguistics: Vol. 1. (pp 1–7). Chiao, Y. C., & Zweigenbaum, P. (2002). Looking for candidate translational equivalents in specialized, comparable corpora. In Proceedings of the 19th International Conference on Computational Linguistics: Vol. 2. (pp 1–5). References g f f p g (pp ) Chiao, Y. C., & Zweigenbaum, P. (2002). Looking for candidate translational equivalents in specialized, comparable corpora. In Proceedings of the 19th International Conference on Computational Linguistics: Vol. 2. (pp 1–5). Cohen, K. B., Xia, J., Zweigenbaum, P., Callahan, T., Goss, F., Ide, N., Ne´ve´ol, A., Grouin, C., & Hunter, L. E. (2018). Three dimensions of reproducibility in natural language processing. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018). (pp 156– 165). Daille, B., Gaussier, E., & Lange´, J. M. (1994). Towards automatic extraction of monolingual and bilingual terminology. In Proceedings of the 15th Conference on Computational Linguistics: Vol. 1. (pp 515–521). Daille, B., & Morin, E. (2005). French-English terminology extraction from comparable corpora. Natur Language Processing - IJCNLP, 2005, 707–718. Fokkens, A., Van Erp, M., Postma, M., Pedersen, T., Vossen, P., & Freire, N. (2013). Offspring from reproduction problems: What replication failure teaches us. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics: Vol. 1: Long Papers. (pp 1691–1701). Foo, J. (2012). Computational terminology: Exploring bilingual and monolingual term extraction. PhD thesis, Linko¨ping University Electronic Press. Fung, P., & Yee, L. Y. (1998). An IR approach for translating new words from nonparallel, comparable texts. In Proceedings of the 17th International Conference on Computational Linguistics: Vol. 1. (pp 414–420). Gaizauskas, R., Aker, A., & Yang Feng, R. (2012). Automatic bilingual phrase extraction from comparable corpora. In 24th International Conference on Computational Linguistics. (pp 23–32). Gale, W., & Church, K. (1993). A program for aligning sentences in bilingual corpora. Computational Linguistics, 19(1), 75–102. Gaussier, E. (1998). Flow network models for word alignment and terminology extraction from bilingual corpora.InProceedingsof the 17thInternational Conference on Computational Linguistics:Vol.1.(pp 444–450). Ha, L. A., Fern, G., Mitkov, R., Corpas, G. (2008). Mutual bilingual terminology extraction. In Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC’08). (pp 1818–1824). Haque, R., Penkale, S., & Way, A. (2014). Bilingual termbank creation via log-likelihood comparison and phrase-based statistical machine translation. In Proceedings of the 4th International Workshop on Computational Terminology (Computerm). (pp 42–51). 123 Reproduction, replication, analysis and adaptation... 799 Hazem, A., & Morin, E. (2016). Efficient data selection for bilingual terminology extraction from comparable corpora. In: Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers. (pp 3401–3411). Hazem, A., & Morin, E. (2017). References Bilingual word embeddings for bilingual terminology extraction from specialized comparable corpora. In Proceedings of the 8th International Joint Conference on Natural Language Processing: Vol. 1: Long Papers. (pp 685–693). Ideue, M., Yamamoto, K., Utiyama, M., & Sumita., E. (2011). A comparison of unsupervised bilingual term extraction methods using phrase tables. In Proceedings of the 13th Machine Translation Summit. (pp 346–351). pp Joachims, T. (2002). Learning to classify text using support vector machines: Methods, theory and algorithms. Alphen aan den Rijn: Kluwer Academic Publishers. Johnson, I., & Macphail, A. (2000). IATE–Inter-Agency Terminology Exchange: Development of a single central terminology database for the institutions and agencies of the European Union. In Proceedings of the Workshop on Terminology resources and computation, LREC 2000 Conference. Jursˇicˇ, M., Mozeticˇ, I., Erjavec, T., & Lavracˇ, N. (2010). Lemmagen: Multilingual lemmatisation wi induced ripple-down rules. Journal of Universal Computer Science, 16(9), 1190–1214. Kano, Y., Baumgartner, W. A, Jr., McCrohon, L., Ananiadou, S., Cohen, K. B., Hunter, L., et al. (2009). U-compare: Share and compare text mining tools with uima. Bioinformatics, 25(15), 1997–1998. Koehn, P. (2005). Europarl: A parallel corpus for statistical machine translation. In Proceedings of the 10th Machine Translation Summit: Vol. 5. (pp 79–86). Kontonatsios, G., Korkontzelos, I., Tsujii, J., & Ananiadou, S. (2014). Combining string and context similarity for bilingual term alignment from comparable corpora. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). (pp 1701–1712). Kranjc, J., Podpecˇan, V., & Lavracˇ, N. (2012). Clowdflows: A cloud based scientific workflow platform. In Proceedings of Machine Learning and Knowledge Discovery in Databases, ECML/PKDD (2). Springer. (pp 816–819). Kupiec, J. (1993). An algorithm for finding noun phrase correspondences in bilingual corpora. Proceedings of the 31st Annual Meeting on Association for Computational Linguistics. (pp 17–22 Lee, L., Aw, A., Zhang, M., & Li, H. (2010). Em-based hybrid model for bilingual terminology extraction from comparable corpora. In Proceedings of the 23rd International Conference on Computational Linguistics: Posters, Association for Computational Linguistics. (pp 639–646). Levenshtein, V. I. (1966). Binary codes capable of correcting deletions, insertions and reversals. Soviet Physics Doklady, 10, 707. Macken, L., Lefever, E., & Hoste, V. (2013). Texsis: Bilingual terminology extraction from parallel corpora using chunk-based alignment. Terminology International Journal of Theoretical and Applied Issues in Specialized Communication, 19(1), 1–30. pp p McKinney, W. (2011). Pandas: A foundational Python library for data analysis and statistics. Python for High Performance and Scientific Computing. References (pp 1–9). Morin, E., Daille, B., Takeuchi, K., Kageura, K. (2007). Bilingual terminology mining-using brain, not brawn comparable corpora. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics. (pp 664–671). Morin, E., Daille, B., Takeuchi, K., & Kageura, K. (2008). Brains, not brawn: The use of smart comparable corpora in bilingual terminology mining. ACM Transactions on Speech and Language Processing, 7(1), 1. Nassirudin, M., & Purwarianti, A. (2015). Indonesian-Japanese term extraction from bilingual corpora using machine learning. In International Conference on Advanced Computer Science and Information Systems (ICACSIS), 2015. (pp 111–116). Och, F. J., & Ney, H. (2003). A systematic comparison of various statistical alignment model Computational Linguistics, 29(1), 19–51. Pedersen, T. (2008). Empiricism is not a matter of faith. Computational Linguistics, 34(3), 465–470. ( ) p p g ( ) Pinnis, M., Ljubesˇic´, N., Stefanescu, D., Skadina, I., Tadic´, M., & Gornostaya, T. (2012). Term extraction, tagging, and mapping tools for under-resourced languages. In Proceedings of the 10th Conference on Terminology and Knowledge Engineering (TKE 2012). (pp 20–21). Piperidis, S., Papageorgiou, H., Spurk, C., Rehm, G., Choukri, K., Hamon, O., Calzolari, N., Del Gratta, R., Magnini, B., & Girardi, C. (2014). Meta-share: One year after. In Proceedings of the 9th International Conference on Language Resources and Evaluation. (pp 1532–1538). 12 123 800 A. Repar et al. Pollak, S., Vavpeticˇ, A., Kranjc, J., Lavracˇ, N., & Vintar V,. (2012). NLP workflow for on-line definition extraction from English and Slovene text corpora. 11th Conference on Natural Language Processing, KONVENS 2012 - Empirical Methods in Natural Language Processing (pp. 53–60). Vienna: Austria. Prinz, F., Schlange, T., & Asadullah, K. (2011). Believe it or not: How much can we rely on published data on potential drug targets? Nature Reviews Drug Discovery, 10(9), 712. Rapp, R. (1999). Automatic identification of word translations from unrelated english and german corpora. In Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics on Computational Linguistics. (pp 519–526). Repar, A., Martinc, M., & Pollak, S. (2018). Machine learning approach to bilingual terminology alignment: Reimplementation and adaptation. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018). f g g ( ) Steinberger, R., Eisele, A., Klocek, S., Pilos, S., Schlu¨ter, P. (2013). DGT-TM: A freely available translation memory in 22 languages. In Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC’2012). Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Steinberger, R., Pouliquen, B., & Hagman, J. (2002). Cross-lingual document similarity calculation using the multilingual thesaurus eurovoc. Computational Linguistics and Intelligent Text Processing. (pp 101–121). Tiedemann, J. (2012). Parallel data, tools and interfaces in opus. In: Chair) NCC, Choukri K, Declerck T, Dogan MU, Maegaard B, Mariani J, Odijk J, Piperidis S (eds) Proceedings of the Eight International Conference on Language Resources and Evaluation (LREC’12), European Language Resources Association (ELRA), Istanbul, Turkey. Vintar, Sˇ., & Grcˇic´-Simeunovic´ (2016). Definition frames as language-dependent models of knowledge transfer. Fachsprache : internationale Zeitschrift fu¨r Fachsprachenforschung, - didaktik und Terminologie. (pp 43–58). ˇ Vintar, Sˇ. (2010). Bilingual term recognition revisited: The bag-of-equivalents term alignment approach and its evaluation. Terminology International Journal of Theoretical and Applied Issues in Specialized Communication, 16(2), 141–158. Wieling, M., Rawee, J., & van Noord, G. (2018). Reproducibility in computational linguistics: Are w willing to share? Computational Linguistics, 44(4), 641–649. Yentis, S., Campbell, F., & Lerman, J. (1993). Publication of abstracts presented at anaesthesia meetings. Canadian Journal of Anaesthesia, 40(7), 632–634. 123 123

https://openalex.org/W2938317040

https://corpus.ulaval.ca/bitstreams/3b599fbe-0a38-48fd-8065-b6d44b58dc4f/download

English

A Conceptual Model for Forest Naturalness Assessment and Application in Quebec’s Boreal Forest

Forests

cc-by

Received: 14 March 2019; Accepted: 7 April 2019; Published: 11 April 2019 Abstract: Research Highlights: To inform eco-designers in green building conception, we propose a conceptual model for the assessment of the impact of using wood on the quality of ecosystems. Background and Objectives: The proposed model allows the assessment of the quality of ecosystems at the landscape level based on the condition of the forest and the proportion of different practices to characterize precisely the forest management strategy. The evaluation provides a numerical index, which corresponds to a suitable format to inform decision-making support tools, such as life cycle analysis. Materials and Methods: Based on the concept of naturalness, the methodology considers five naturalness characteristics (landscape context, forest composition, structure, dead wood, and regeneration process) and relies on forest inventory maps and data. An area within the boreal black spruce-feathermoss ecological domain of Quebec (Canada) was used as a case study for the development of the methodology, designed to be easily exportable. Results: In 2012, the test area had a near-natural class (naturalness index NI = 0.717). Simulation of different management strategies over 70 years shows that, considering 17.9% of strict protected areas, the naturalness index would have lost one to two classes of naturalness (out of five classes), depending on the strategy applied for the regeneration (0.206 ≤∆NI ≤0.413). Without the preservation of the protected areas, the management strategies would have further reduced the naturalness (0.274 ≤∆NI ≤0.492). Apart from exotic species plantation, the most sensitive variables are the percentage of area in irregular, old, and closed forests at time zero and the percentage of area in closed forests, late successional species groups, and modified wetlands after 70 years. Conclusions: Despite the necessity of further model and parameter validation, the use of the index makes it possible to combine the effects of different forestry management strategies and practices into one alteration gradient. Keywords: naturalness; forest management intensity; land use intensity; quality of ecosystems; boreal forest Forests 2019, 10, 325; doi:10.3390/f10040325 Manuele Margni 2 1 Department of Wood and Forestry Sciences, Université Laval, Quebec City, QC G1V 0A6, Canada; [email protected] (L.B.); [email protected] (R.B.); Evelyne.Thiff[email protected] (É.T.) y ( ) 2 CIRAIG, Polytechnique Montréal, Department of Mathematical and Industrial Engineering, Montreal, QC H3C 3A7, Canada; [email protected] * Correspondence: [email protected]; Tel.: +1-418-424-0422 y ( ) 2 CIRAIG, Polytechnique Montréal, Department of Mathematical and Industrial Engineering, Montreal, QC H3C 3A7, Canada; [email protected] * Correspondence: [email protected]; Tel.: +1-418-424-0422 * Correspondence: [email protected]; Tel.: +1-418-424-0422 Article Sylvie Côté 1,*, Louis Bélanger 1 , Robert Beauregard 1, Évelyne Thiffault 1 and Manuele Margni 2 1. Introduction Quantitative tools to discriminate between different wood supplies depending on forest management and wood procurement practices are needed to inform architects and designers planning the eco-design of buildings. Using the science of applied ecology, such tools should make it possible to evaluate and compare the impact of different forestry strategies and the combination of practices on the quality of forest ecosystems. This study aims to develop a methodology to characterize the potential impacts on ecosystem quality of different forestry management practices, in the perspective of describing the intensity of Forests 2019, 10, 325; doi:10.3390/f10040325 www.mdpi.com/journal/forests www.mdpi.com/journal/forests 2 of 29 Forests 2019, 10, 325 land use as driven by forestry. The methodology is based on the naturalness concept and relies on forest inventory maps and data. Our methodology allows the evaluation of combinations of practices and provides one numerical index, a suitable format for further use in decision-making support tools for eco-design and green building conception, such as life cycle analysis (LCA) [1]. An area within the boreal black spruce ecological domain of Quebec (Canada) was used as a case study for the development of the methodology. The specific objectives of the study are to: (1) Develop a naturalness evaluation model using the example of the boreal black spruce-feathermoss ecological bioclimatic domain of Quebec (Canada). (1) Develop a naturalness evaluation model using the example of the boreal black spruce-feathermoss ecological bioclimatic domain of Quebec (Canada). (1) Develop a naturalness evaluation model using the example of the boreal black spruce-feathermoss ecological bioclimatic domain of Quebec (Canada). (2) Apply the model over time on three forest management units (3 FMU) to analyze the variability of the naturalness evaluation associated with changes in forest management strategies and practices. (2) Apply the model over time on three forest management units (3 FMU) to analyze the variability of the naturalness evaluation associated with changes in forest management strategies and practices. (3) Perform a sensitivity analysis of the model to (hypothetical) high pressure levels and to identify the naturalness evaluation associated with changes in forest management strategies and practices. (3) Perform a sensitivity analysis of the model to (hypothetical) high pressure levels and to identify the most sensitive variables. (3) Perform a sensitivity analysis of the model to (hypothetical) high pressure levels and to identify the most sensitive variables. The need for evaluating the quality of ecosystems in relation with their anthropic uses presents many challenges. 1. Introduction As land use, and particularly land use change, is one of the main drivers of biodiversity loss [2], there is a desire to express its impact on the quality of ecosystems in terms of biodiversity damage in LCA [3,4]. The latest proposed LCA approach uses potential species loss from land use as an indicator; for forestry, it considers two land use intensities (intensive and extensive) [3]. This proposal raises two issues: Biodiversity data and indicators’ availability, and land use intensity evaluation. Concerning biodiversity, potential species loss is still proposed as the biodiversity indicator even if it does not reflect the multidimensional character of biodiversity and might lead to inappropriate conclusions [5]. As stated by Souza et al. [6], the biodiversity models proposed up to now do not grasp the full reach of the phenomena involved, such as functional effects and impacts on populations. Furthermore, there are data gaps in biodiversity: Biodiversity data are often fragmentary (they do not include all taxa) and of varying quality (all biomes are not evenly studied, especially the boreal biome for which data are particularly scarce). For instance, boreal forests are underrepresented in global biodiversity databases (see GLOBIO [7]; PREDICTS [8]). Concerning the intensity, forest management strategies generally include a mix of practices that have different impacts on the ecosystem, and the intensity is related to the recurrence of treatments over the same area planned in the silvicultural scenario. Because of these issues, we propose an alternative approach to evaluate ecosystem quality related to forest management, one that focuses on habitat characteristics and the concept of naturalness. g p Many authors have proposed to use the concepts of naturalness and hemeroby in impact evaluation of land use (such as forestry) on the quality of ecosystems in LCA [9–13]. Naturalness is defined as “the similarity of a current ecosystem state to its natural state” [14], whereas hemeroby expresses “distance to nature” in landscape ecology [11]. The use of these concepts can provide a management guide that overcomes the challenge of data gaps in biodiversity. Even if the concepts of naturalness and hemeroby are closely related, one is not the exact inverse of the other. There is also divergence concerning the highest degree of alteration that should be included [14]. 1. Introduction To clarify, we associate the naturalness concept with forest ecosystems, as shown in Figure 1; its lower class, i.e., the most altered state, corresponds to artificial forests [15,16] created by humans and showing deep modifications to the ecosystem and its species composition [15]. On the other hand, in the hemeroby scale, the alteration gradient is further developed and extended to sealed soils, and constructed, degraded, or devastated areas [11], with some authors even distinguishing dumpsites and partially built areas from sealed soils [9]. As stated by Winter [14], “greater naturalness is characterized by a large number of adapted, specialized and often endangered plant and animal species”. Thus, in order to prevent or limit forest biodiversity loss due to forestry, the emphasis should be put on maintaining or restoring a high degree of naturalness. The concept of naturalness is well adapted to evaluate forestry management practices, but its application to evaluate the full alteration range of different land uses beyond forestry will require further work for proper insertion in the hemeroby concept that addresses a larger alteration gradient. Since the scope of 3 of 29 Forests 2019, 10, 325 Forests 2019 this paper focuses on the impacts of forest management practices, the evaluation is restricted to the naturalness part of the alteration gradient. a larger alteration gradient. Since the scope of this paper focuses on the impacts of forest management practices, the evaluation is restricted to the naturalness part of the alteration gradient. Fi 1 N t l d h b l th lt ti di t ( d t d f Wi t t l [17]) Figure 1. Naturalness and hemeroby along the alteration gradient (adapted from Winter et al. [17]). Figure 1. Naturalness and hemeroby along the alteration gradient (adapted from Winter et al. [17]). Figure 1. Naturalness and hemeroby along the alteration gradient (adapted from Winter et al. [17]). The use of an index to evaluate the departure from the natural state along the alteration gradient avoids the problem of multiple classification of naturalness encountered in the literature [18] and allows inclusion of realistic forest management strategies, which involve a mix of different practices. Naturalness evaluation based on habitat characteristics is possible with actual data, and the future evolution of characteristics can be predicted. 1. Introduction However, impacts on biodiversity are more challenging to assess considering the need of biodiversity indicators that encompass multidimensional characteristics of the biodiversity concept [5] and the uncertainty related to the timelag existing between habitat destruction and species extinction or extirpation [19]. Forest naturalness can be defined and evaluated using three interdependent approaches based The use of an index to evaluate the departure from the natural state along the alteration gradient avoids the problem of multiple classification of naturalness encountered in the literature [18] and allows inclusion of realistic forest management strategies, which involve a mix of different practices. Naturalness evaluation based on habitat characteristics is possible with actual data, and the future evolution of characteristics can be predicted. However, impacts on biodiversity are more challenging to assess considering the need of biodiversity indicators that encompass multidimensional characteristics of the biodiversity concept [5] and the uncertainty related to the timelag existing between habitat destruction and species extinction or extirpation [19]. Forest naturalness can be defined and evaluated using three interdependent approaches based on structure (i.e., spatial arrangement of the various components of the ecosystem [20]), composition, and processes [21]. Generally, naturalness assessment results from the comparison between the actual condition and a reference state [22] using either historical inventory data, prior to commercial forest exploitation, or modelling studies of forest dynamics evaluating the range of natural variability [23]. Where historic data are not available, the reference state, which corresponds to the most natural state, can be associated with a position along the alteration gradient [17]. Many methods have been proposed to assess naturalness [14], which is coherent with the fact that the choice of variables for naturalness studies must be adapted to regional conditions and knowledge [17]. The method developed here uses condition and pressure indicators. Indicators correspond to specific “elements of the forest system (e.g., species, processes and habitats) that correlate with many Forest naturalness can be defined and evaluated using three interdependent approaches based on structure (i.e., spatial arrangement of the various components of the ecosystem [20]), composition, and processes [21]. Generally, naturalness assessment results from the comparison between the actual condition and a reference state [22] using either historical inventory data, prior to commercial forest exploitation, or modelling studies of forest dynamics evaluating the range of natural variability [23]. 1. Introduction We also propose an original method for handling condition and pressure indicators in the index calculation. ecological condition to another [25]. We also propose an original method for handling condition and pressure indicators in the index calculation. 1. Introduction Where historic data are not available, the reference state, which corresponds to the most natural state, can be associated with a position along the alteration gradient [17]. Many methods have been proposed to assess naturalness [14], which is coherent with the fact that the choice of variables for naturalness studies must be adapted to regional conditions and knowledge [17]. other unmeasured elements of the system” [24]. Condition or state indicators describe the current status or condition of a characteristic; pressure indicators represent the level of a pressure that affects the condition of a characteristic (i.e., an action that is causing the condition to degrade or improve) [24]. Thus, condition indicators are related to the concept of naturalness (i.e., the similarity of a current ecosystem state to its natural state), whereas pressure indicators are rather related to the hemeroby concept (i.e., distance to nature). However, pressure indicators can still be used to evaluate naturalness considering their effects on the condition of a characteristic. We developed our conceptual model for naturalness assessment at the landscape level in a way that it could be easily adapted to other contexts and available data. Our method explores the The method developed here uses condition and pressure indicators. Indicators correspond to specific “elements of the forest system (e.g., species, processes and habitats) that correlate with many other unmeasured elements of the system” [24]. Condition or state indicators describe the current status or condition of a characteristic; pressure indicators represent the level of a pressure that affects the condition of a characteristic (i.e., an action that is causing the condition to degrade or improve) [24]. Thus, condition indicators are related to the concept of naturalness (i.e., the similarity of a current ecosystem state to its natural state), whereas pressure indicators are rather related to the hemeroby concept (i.e., distance to nature). However, pressure indicators can still be used to evaluate naturalness considering their effects on the condition of a characteristic. We developed our conceptual model for naturalness assessment at the landscape level in a way that it could be easily adapted to other contexts and available data. Our method explores the application of non-linear relationships to integrate the notion of ecological thresholds in the naturalness assessment; habitat thresholds correspond to points or zones at which relatively rapid changes occur from one 4 of 29 Forests 2019, 10, 325 ecological condition to another [25]. 2.1. Conceptual Model Designing a model to feed decision support systems relying on science-based evidence requires condensing and summarizing original information from studies and reviews in a form accessible to decision-makers [26]. This challenging exercise involves a choice of critical criteria relevant to the decision; in this case, assessing the impact of forest management practices on ecosystem quality. The model we propose determines an aggregated naturalness index (NI) based on five forest naturalness characteristics: (1) Landscape context, (2) composition, (3) structure, (4) dead wood (DW), and (5) regeneration process (RP). The landscape context characteristic refers to forest habitat at the landscape level; composition corresponds to tree species composition; structure considers age structure as well as physical vertical and horizontal structure; dead wood focuses on coarse woody debris; finally, the regeneration process characteristic refers to the forest renewal mode (see Appendix A for more details about indicators and measures for each characteristic). The conceptual model developed for naturalness assessment in the black spruce and feathermoss domain of Quebec’s boreal forest is presented in Figure 2. Figure 2. A naturalness evaluation conceptual model for the black spruce and feathermoss boreal forest. PNI: Partial naturalness index for naturalness characteristic; pni: partial naturalness index for condition indicator; NDP: Naturalness degradation potential; FO: Function of. Figure 2. A naturalness evaluation conceptual model for the black spruce and feathermoss boreal forest. PNI: Partial naturalness index for naturalness characteristic; pni: partial naturalness index for condition indicator; NDP: Naturalness degradation potential; FO: Function of. For each naturalness characteristic, the selection of condition indicators was based on ecological issues relevant to this region [27]. Corresponding measures for each indicator (Table 1) were identified from available ecoforest maps and from relevant modelling studies. 5 of 29 Forests 2019, 10, 325 Table 1. Naturalness characteristics, indicators and measures. 2.1. Conceptual Model 6 of 29 6 of 29 Forests 2019, 10, 325 Forests 2019, 10, x FOR Figure 3. Generic curves used for modelling. (a) Sigmoid curve for the partial naturalness index of condition indicators. (b) Linear curve for naturalness degradation potential evaluation for pressure indicators proportional to the area. (c) Logarithmic curve for naturalness degradation potential l ti f hi h t ti l i t i di t Figure 3. Generic curves used for modelling. (a) Sigmoid curve for the partial naturalness index of condition indicators. (b) Linear curve for naturalness degradation potential evaluation for pressure indicators proportional to the area. (c) Logarithmic curve for naturalness degradation potential evaluation for high potential impact pressure indicators. Figure 3. Generic curves used for modelling. (a) Sigmoid curve for the partial naturalness index of condition indicators. (b) Linear curve for naturalness degradation potential evaluation for pressure ndicators proportional to the area (c) Logarithmic curve for naturalness degradation potential Figure 3. Generic curves used for modelling. (a) Sigmoid curve for the partial naturalness index of condition indicators. (b) Linear curve for naturalness degradation potential evaluation for pressure indicators proportional to the area. (c) Logarithmic curve for naturalness degradation potential evaluation for high potential impact pressure indicators. high potential impact pressure indicators. Table 2. Characteristic_PNI equations for each naturalness characteristic. Table 2. Characteristic_PNI equations for each naturalness characteristic. aturalness Characteristic Characteristic_PNI 1 Equation Landscape context Context_PNI = CF_pni × (1 − (ant_NDP + Wm_NDP + W_CC_NDP)) Forest Composition Compo_PNI = ((CT_pni+LS_pni)/2) × (1 − (exo_NDP + CS_NDP)) Structure Struc_PNI = ((OF_pni + IR_pni)/2) × (1 − HS_NDP) Dead wood DW_PNI = 1 − DW_NDP _ q Naturalness Characteristic Characteristic_PNI 1 Equation Landscape context Context_PNI = CF_pni × (1 −(ant_NDP + Wm_NDP + W_CC_NDP)) Forest Composition Compo_PNI = ((CT_pni+LS_pni)/2) × (1 −(exo_NDP + CS_NDP)) Structure Struc_PNI = ((OF_pni + IR_pni)/2) × (1 −HS_NDP) Dead wood DW_PNI = 1 −DW_NDP Regeneration process RP_PNI = 1 −RP_NDP 1 PNI/pni: partial naturalness index; NDP: naturalness degradation potential; see Table 1 for variables definitions. u _ I ((O _p i I _p i) ) ( _ ) Dead wood DW PNI = 1 − DW NDP 1 PNI/pni: partial naturalness index; NDP: naturalness degradation potential; see Table 1 for variables definitions. Regeneration process RP_PNI 1 RP_NDP 1 PNI/pni: partial naturalness index; NDP: naturalness degradation potential; see Table 1 for variables definitions. 2.1. Conceptual Model Naturalness Characteristic Condition Indicator(s) Condition Measure(s) Pressure Indicator(s) Pressure Measure(s) Landscape context (Context_PNI 1) Forest habitat (closed forest) (CF_pni) % of terrestrial area of forest >40 years old (CF) Land use (ant_NDP) % of terrestrial area with anthropization (ant) Wetlands (Wm_NDP) % of modified wetlands (Wm) (W_CC_NDP) % of humid area in clear cut (W_CC) Forest composition (Compo_PNI) Cover type (CT_pni) % of forest area with coniferous cover type (CT) Companion species (CS_NDP) Recognized companion species diminution (CS) Late successional species (LS_pni) % of forest area in late successional species groups (LS) Exotic species (exo_NDP) % of forest area of exotic species stands (exo) Structure (Struc_PNI) Age structure (OF_pni) % of forest area of old forests (>100 years old) (OF) Horizontal structure (HS_NDP) HS NDP_factor by silvicultural treatment weighed by % of forest area Vertical structure (IR_pni) % of forest area of irregular forests (IR) Dead wood (DW_PNI) Coarse woody debris (DW_NDP) DW NDP_factor by silvicultural treatment weighed by % of forest area Regeneration process (RP_PNI) Regeneration process (RP_NDP) RP NDP_factor by silvicultural treatment weighed by % of forest area 1 PNI: Partial naturalness index for naturalness characteristic; pni: Partial naturalness index for condition indicator; NDP: Naturalness degradation potential. Table 1. Naturalness characteristics, indicators and measures. 1 PNI: Partial naturalness index for naturalness characteristic; pni: Partial naturalness index for condition indicator; NDP: Naturalness degradation potential. First, the measures of condition are used to evaluate partial naturalness indexes (PNI/pni) using a sigmoidal curve (Figure 3a). Measures of pressure are then used to evaluate naturalness degradation potentials (NDP), using either linear or logarithmic curves (Figure 3b,c) or territory specific NDP factors related to practices weighed by the percentage of area as described in the section test area. Then, for each naturalness characteristic, i, the partial naturalness index (Characteristic_PNIi) is calculated as follows (see Table 2 for the calculation details of each characteristic): Characteristic_PNIi = ( 1 n X n j=1Condition_pnij) × (1 − X m k=1NDPk) (1) (1) where PNI/pni = partial naturalness index; NDP = naturalness degradation potential; n = number of condition indicators, j, for each characteristic, i (up to two); m: number of NDP, k, for each characteristic, i. where PNI/pni = partial naturalness index; NDP = naturalness degradation potential; n = number of condition indicators, j, for each characteristic, i (up to two); m: number of NDP, k, for each characteristic, i. 2.1. Conceptual Model Therefore, we can consider that the probability of persistence of a species is generally high in the “near-natural” class; it then declines in the “semi-natural” class as some very specialized species might be affected. Sensitive species will be lost in the “altered” class and many species will be lost in the “very altered” class (Figure 3a). PNIs and exo_NDP was evaluated by linear interpolation between curve points (see Supplementary Materials Table S1: PNIs and exo_NDP determination). Measures of pressures are used to evaluate a naturalness degradation potential (NDP) [9]. In the model, the total naturalness degradation (ΣNDP) for each naturalness characteristic was applied as a relative reduction (expressed in the percent of naturalness degradation) of the corresponding mean of condition_pni (Formula 1). Higher values of NDP represent a higher potential of naturalness degradation, corresponding to the red class. There are four possible approaches to determine an NDP. The first one considers a unique degradation factor for the whole area. The second considers that NDP is proportional to the area under pressure using a linear relationship (Figure 3b). The third approach relates to potentially high impact interventions, and considers that NDP is evaluated using a logarithmic curve (Figure 3c). A practice is considered to have a high potential impact when a small proportion of impacted area can have detrimental effects over a wider area. For example, if the proportion of exotic species stands reaches 60%, this corresponds to a very high potential of naturalness degradation resulting from the modification of the forest matrix. The fourth approach for NDP evaluation is used for variables that cannot be measured or derived from a forest cartography or inventory (i.e., horizontal structure (HS), dead wood (DW), and regeneration process (RP)). For each of these variables, the pressure level associated with silvicultural treatments was rated to reflect the effect of the disturbance intensity on the variable considered, using degradation factors based either on data (for dead wood) or on expert opinion. Expert opinion is often used in decision support systems applied to environmental management either to compensate for the lack of data or to interpret scientific results in order to provide guidelines based on science [26]. The evaluation of NDP factors related to the fourth evaluation approach (see examples in Table 3, Table 4, and Table 5) could be further developed using participatory methods, such as the Delphi survey [32], involving a team of professionals. 2.1. Conceptual Model The naturalness index and the partial naturalness indexes of both levels (characteristic and The naturalness index and the partial naturalness indexes of both levels (characteristic and condition) range from 1 (natural) to 0 (very altered). To ease the interpretation and the discussion, we divided this range in five equal classes with an associated colour code: Natural (dark green): 1–0.8; Forests 2019, 10, 325 7 of 29 near natural (light green): 0.799–0.6; semi-natural (yellow): 0.599–0.4; altered (orange): 0.399–0.2; very altered (red): 0.199–0 (see Figure 3a). The evaluation of a condition indicator is based on a comparison with historical values. For each condition indicator, a partial naturalness index (condition_pni) is evaluated using a sigmoid curve relating the measure, corresponding to the actual proportion of the historical area percentage (or at a given time in the future, for forecasting scenarios) to the pni (Figure 3a). The sigmoid curve is considered a good representation of the type of relationship existing between the amount of habitat and species’ response [28] and reflects the presence of thresholds. In our approach, habitat thresholds were used to determine changes between naturalness classes in order to put in relation the degree of ecosystem alteration and its potential effect on biodiversity. Many authors reported an important reduction in biodiversity when the amount of habitat is below 30% of the historical level [25,29,30]; hence, the upper limit of the “very altered” class was set at 30% of the historical level. As naturalness classes correspond to equal divisions of the alteration gradient, the sigmoid curve was centered at 50% of the historical level. Therefore, on the opposite side of the gradient, the lower limit of the “natural” class was set at 70% of the historical level for each condition indicator; the lower and upper limits for the “semi-natural” class were therefore set at 43.5% and 56.5% of the historical level. This range corresponds roughly to the range of the mean thresholds (absence, colonization, extinction, and persistence) for the amount of habitat of 44% to 61% of historical levels observed for breeding birds by Zuckerberg and Porter [28]. The upper limit of the “altered” class is slightly above the threshold of 40% of the historical level values observed for the persistence of some specialized species [31]. 2.1. Conceptual Model For dead wood and regeneration process, to overcome the absence of condition measures, the NDP is applied on the value corresponding to the natural state (condition_pni = 1); therefore, in these cases: Characteristic_PNI = 1 −NDP. 8 of 29 Forests 2019, 10, 325 Table 3. Naturalness degradation potentials (NDPs) for long lived companion species. Long Lived Companion Species Status NDP_Factors Recognized species in diminution 0.2 Recognized extirpated species (theoretical) 0.6 bl 4 N t l d d ti t ti l (NDP) f h i t l t t (HS) b il i lt l Table 3. Naturalness degradation potentials (NDPs) for long lived companion species. Table 4. Naturalness degradation potential (NDP) for horizontal structure (HS) by silvicultural treatments in Quebec’s boreal forest. Table 4. Naturalness degradation potential (NDP) for horizontal structure (HS) by silvicultural treatments in Quebec’s boreal forest. Practice NDP_Factors % Forest_Area NDPx Plantation—thinning 1 0.47% 0.0047 Plantation 0.9 4.62% 0.0416 Thinning (natural), strip cutting 0.8 0.46% 0.0037 Precommercial thinning (natural), release 0.75 1.68% 0.0126 Salvage logging 0.6 0.07% 0.0004 Careful logging (CL) and clear cut 0.35 13.93% 0.0488 CLASS, variable retention cut (2% vol) 0.3 0.00% 0.0000 Partial cutting 0.2 0.12% 0.0002 Undisturbed or natural disturbances 0 78.65% 0.0000 Actual HS_NDP 0.1120 Note: NDP_factors: naturalness degradation potential factors related to practices; % for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice; CLASS: careful logging around small merchantable stems. Note: NDP_factors: naturalness degradation potential factors related to practices; % for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice; CLASS: careful logging around small merchantable stems. Note: NDP_factors: naturalness degradation potential factors related to practices; % for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice; CLASS: careful logging around small merchantable stems. Table 5. Naturalness degradation potential (NDP) for dead wood (DW) by silvicultural treatments in Quebec’s boreal forest. Table 5. Naturalness degradation potential (NDP) for dead wood (DW) by silvicultural treatments in Quebec’s boreal forest. Table 5. Naturalness degradation potential (NDP) for dead wood (DW) by silvicultural treatments in Quebec’s boreal forest. 2.1. Conceptual Model For each condition indicator, find a reference value using either historical studies or old forest inventories and maps (for the 3 FMU: SIFORT1 maps (forest information system by tessellation) and Bouchard et al. 2015 [33] for OF). 5. For each condition indicator, find a reference value using either historical studies or old forest inventories and maps (for the 3 FMU: SIFORT1 maps (forest information system by tessellation) and Bouchard et al. 2015 [33] for OF). 6. For each condition indicator, evaluate actual measures using the latest forest inventory map (for the 3 FMU: SIFORT4 maps). 6. For each condition indicator, evaluate actual measures using the latest forest inventory map (for the 3 FMU: SIFORT4 maps). 7. For each condition indicator, set pni curves for the studied area (by changing the reference values for each condition indicator in Table S1) and enter the actual measure to calculate the corresponding pni (by changing the measured values for each condition indicator in Table S1). 7. For each condition indicator, set pni curves for the studied area (by changing the reference values for each condition indicator in Table S1) and enter the actual measure to calculate the corresponding pni (by changing the measured values for each condition indicator in Table S1). 8. For pressure measures, identify the appropriate approach for NDP evaluation related to each naturalness characteristic. Identify curves, set factors based on studies or expert opinion, and get the measures of the area by practice from forest inventory maps (for the 3 FMU: Ecoforest 4 maps, CS_NDP in Table 3, exo_NDP curve in Table S1, and NDP tables (factors and area) for HS, DW, and RP in Table S1). 9. Calculate the PNI for each naturalness characteristic using Equation (1) (for the 3 FMU: Formu by characteristic are detailed in Table 2). 9. Calculate the PNI for each naturalness characteristic using Equation (1) (for the 3 FMU: Formulas by characteristic are detailed in Table 2). 10. Calculate the NI, which corresponds to the arithmetic mean of the five characteristic_PNI. by characteristic are detailed in Table 2). 10. Calculate the NI, which corresponds to the arithmetic mean of the five characteristic_PNI. 10. Calculate the NI, which corresponds to the arithmetic mean of the five characteristic_PN 2.1. Conceptual Model Practice NDP_Factors % Forest_Area NDPx Biomass harvesting 1 0.00% 0.0000 Thinnings (in natural or plantation) 0.95 2.15% 0.0205 Plantation—no thinnings 0.85 4.62% 0.0392 Partial cut 0.75 0.58% 0.0044 Salvage logging 0.7 0.07% 0.0005 Careful logging (CL) 0.65 13.93% 0.0906 Variable retention cut (2% vol) 0.6 0.00% 0.0000 Undisturbed or natural disturbances 0 78.65% 0.0000 DW_NDP 0.1551 Actual DW_PNI 0.8449 Note: NDP_factors: naturalness degradation potential factors related to practices; % for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice. Note: NDP_factors: naturalness degradation potential factors related to practices; % for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice. The naturalness index (NI) calculation then results from the arithmetic mean of the five PNI by characteristic. For the assessment of a given forest management strategy, the calculation should cover a complete harvest cycle (i.e., forest rotation), simultaneously considering the effects over time of the harvest on condition indicators and of silvicultural treatments on pressure measures. The generic procedure for naturalness assessment (and the corresponding files used for the 3 FMU) is as follows: 1. Define the territory for which the analysis will be performed and anticipate aggregation of results if the studied area covers multiple data sources. Identify ecological issues for the studied area based on literature and/or stakeholder consultations 2. Identify ecological issues for the studied area based on literature and/or stakeholder consultations. 3. Pinpoint potential measures available for reference and actual data of the condition and pressure based on literature, forest inventories, and maps. 3. Pinpoint potential measures available for reference and actual data of the condition and pressure based on literature, forest inventories, and maps. 4. For each naturalness characteristic, identify condition indicators and corresponding measures which can to be used to assess ecological issues. 9 of 29 Forests 2019, 10, 325 5. For each condition indicator, find a reference value using either historical studies or old forest inventories and maps (for the 3 FMU: SIFORT1 maps (forest information system by tessellation) and Bouchard et al. 2015 [33] for OF). 5. For each condition indicator, find a reference value using either historical studies or old forest inventories and maps (for the 3 FMU: SIFORT1 maps (forest information system by tessellation) and Bouchard et al. 2015 [33] for OF). 5. 2.2. Test Area The proposed approach for naturalness assessment was applied to a public forest territory formed by three forest management units (FMU) located in the western black spruce feathermoss bioclimatic sub-domain, near the locality of Chibougamau in Northern Quebec region (Figure 4). These 3 FMU (no 2663, 2665, and 2666) cover a total area of 1,305,200 ha, which is larger than the home range of the boreal caribou, an umbrella species for the boreal forest [34]. Historical data were taken from the first Quebec forest inventory, corresponding to the 1965 to 1974 period, using Quebec’s SIFORT system (tessellation of provincial forest inventory maps), from which the 6% of harvested areas and other anthropic disturbances were removed. Current data, corresponding to the 2011 to 2013 period, were taken from the fourth inventory program. The territory used for the analysis covers the whole area included in the perimeter of the FMU (without cutting tessell in SIFORT maps), including the surrounding strict protected areas (IUCN categories I to III) associated with these units. The percentage of forested area over the territory of analysis was calculated for measures of forest condition (CT, LS, OF, and IR) and the percentage of terrestrial area over the territory of analysis for context measures (CF, W_CC, Wm, ANT) was obtained from SIFORT maps (SIFORT1 for “reference” measures, except OF, and SIFORT4 for “actual” measures). Percentages of forested area by origin considering silvicultural treatments in the portion admissible for wood production necessary for weighing NDP_factors were measured with the ecoforest map, which provides polygonal data that are more precise. Each measure was then used to evaluate corresponding pnis or NDPs using curves and tables set for the territory. 10 of 29 10 of 29 Forests 2019, 10, 325 Forests 2019, 10, x FO Figure 4. Test area localization. Figure 4. Test area localization. Figure 4 Test area localization Figure 4. Test area localization. The curves elaborated for pni and NDP evaluation specifically for the 3 FMU are presented in Figure 5 and the NDP factors used in Tables 3,4,5,6. Model adaptation to other territories will require calibration of the curves (Figure 5) using appropriate historical/reference values. In our study, all historical values were based on forest inventory data, except for the proportion of forest land covered by old forests (>100 years old), which was based on the modelling study of Bouchard et al. [33]. 2.2. Test Area For landscape context evaluation, the NDP related to clearcut on wetlands was arbitrarily set at 50% of the percentage of wetland area affected by clearcut, as this disturbance was considered less damaging than the drainage of wetlands (NDP_W_CC = 50% × % of wetland area with clearcut). To allow for proper evaluation of extreme scenarios, a two-sided curve was developed for cover type and late successional species groups in order to consider the loss of dominant characteristics on one side, and loss of secondary characteristics on the other. As the reduction of long-lived companion species cannot be measured precisely using inventory data, a reduction factor of 0.2, corresponding to a decrease of one naturalness class, was applied as NDP when diminution was recognized by forest managers; a factor of 0.6 was also tested to evaluate the effect of an hypothetical species extirpation. Improvement of that measure might be possible in the future with more detailed forest composition characterisation performed in more recent forest inventories in Quebec. As dead wood data are not currently available from Quebec’s forest inventory, the evaluation was derived from pressure measures resulting from silvicultural treatments, by applying NDP factors weighed by the proportion of forest area by treatments. These factors were estimated based on dead wood data for coarse woody debris measured after a range of silvicultural treatments (careful clearcut logging, The curves elaborated for pni and NDP evaluation specifically for the 3 FMU are presented in Figure 5 and the NDP factors used in Tables 3–6. Model adaptation to other territories will require calibration of the curves (Figure 5) using appropriate historical/reference values. In our study, all historical values were based on forest inventory data, except for the proportion of forest land covered by old forests (>100 years old), which was based on the modelling study of Bouchard et al. [33]. For landscape context evaluation, the NDP related to clearcut on wetlands was arbitrarily set at 50% of the percentage of wetland area affected by clearcut, as this disturbance was considered less damaging than the drainage of wetlands (NDP_W_CC = 50% × % of wetland area with clearcut). To allow for proper evaluation of extreme scenarios, a two-sided curve was developed for cover type and late successional species groups in order to consider the loss of dominant characteristics on one side, and loss of secondary characteristics on the other. 2.2. Test Area As the reduction of long-lived companion species cannot be measured precisely using inventory data, a reduction factor of 0.2, corresponding to a decrease of one naturalness class, was applied as NDP when diminution was recognized by forest managers; a factor of 0.6 was also tested to evaluate the effect of an hypothetical species extirpation. Improvement of that measure might be possible in the future with more detailed forest composition characterisation performed in more recent forest inventories in Quebec. As dead wood data are not currently available from Quebec’s forest inventory, the evaluation was derived from pressure measures resulting from silvicultural treatments, by applying NDP factors weighed by the proportion of forest area by treatments. These factors were estimated based on dead wood data for coarse woody debris measured after a range of silvicultural treatments (careful clearcut logging, plantation, precommercial thinning, biomass harvesting) compared with naturally disturbed forests at the Montmorency Research Forest [35,36]. Forests 2019, 10, 325 plantation, preco 11 of 29 orests 11 of 29 forests (a) (b) (c) (d) (e) Figure 5. Pni determination curves used for condition indicators’ evaluation for three forest management units (3FMU) in the boreal black spruce-feathermoss bioclimatic domain: (a) Coniferous cover type; (b) late successional species groups; (c) old forests; (d) irregular stands; (e) closed forests. Figure 5. Pni determination curves used for condition indicators’ evaluation for three forest management units (3FMU) in the boreal black spruce-feathermoss bioclimatic domain: (a) Coniferous cover type; (b) late successional species groups; (c) old forests; (d) irregular stands; (e) closed forests. (a) (b) (d) (b) (c) (d) (d) (c) (e) ure 5. Pni determination curves used for condition indicators’ evaluation for three forest nagement units (3FMU) in the boreal black spruce-feathermoss bioclimatic domain: (a) Coniferous re 5. Pni determination curves used for condition indicators’ evaluation for three forest management s (3FMU) in the boreal black spruce-feathermoss bioclimatic domain: (a) Coniferous cover type; (c) (c) (e) (e) (e) Figure 5. Pni determination curves used for condition indicators’ evaluation for three forest management units (3FMU) in the boreal black spruce-feathermoss bioclimatic domain: (a) Coniferous cover type; (b) late successional species groups; (c) old forests; (d) irregular stands; (e) closed forests. Figure 5. 2.3. Description of Tests Data from 2012 were used to assess the actual naturalness of the 3 FMU. Two series of tests were then performed: (1) Scenario tests on the 3 FMU over time, using three base management scenarios to analyse the sensitivity of the evaluation system to changes in forest management strategies and practices; and (2) hypothetical tests to verify the sensitivity of the model to high pressure levels. Base management scenarios were: (1) Regeneration through careful clearcut logging on 100% of the harvested area (CL); (2) regeneration through careful clearcut logging on 50% of the harvested area, combined with plantation with thinning on the remaining 50% (CL-PL); and (3) regeneration through plantation on 100% of the harvested area (PL). Careful clearcut logging corresponds to the cut with regeneration and soil protection (CPRS) required by law for clearcut operations in Quebec [37]. For each scenario, the possibility of biomass harvest (bh) over the whole harvested area was also considered [35]. For the plantation, indigenous (PL) or exotic species (PLexo), with a rotation of 70 years for the entire planted portion, were tested. Within the 3 FMU, 17.9% of the forested area have a “strict protected area” status. In order to evaluate its effect on naturalness, the same set of scenarios was applied on the 3 FMU hypothesizing the absence of protected areas. In that case, 95.7% of the forest area would be available for harvest compared to a proportion of 77.8% for the scenario with protected areas. The hypothesis used for the evaluation of naturalness over time for the 3MU are listed in Appendix B. The spreadsheets used for simulation of the 3 FMU through time are provided as Supplementary Materials (Table S2: Area by age class evolution by a 10 year period and Table S3: Composition and irregular evaluation over time) and the procedure for evaluation over time is detailed in Appendix C. A sensitivity analysis was performed on the 3 FMU CL-PL scenarios (including protected areas), in order to identify the most sensitive variables of the model at time 0, 30, and 70 (corresponding to actual, mid-rotation, and end-rotation): A variation of ±5% was tested independently for each input variable (percentage of area or NPD factor for CS). Scenario results used for this test have been adjusted for exotics, anthropized, and modified wetlands by setting these reference values at 5% to test the influence of a ±5% variation. 2.2. Test Area Pni determination curves used for condition indicators’ evaluation for three forest management units (3FMU) in the boreal black spruce-feathermoss bioclimatic domain: (a) Coniferous cover type; (b) late successional species groups; (c) old forests; (d) irregular stands; (e) closed forests. 12 of 29 Forests 2019, 10, 325 Table 6. Naturalness degradation potential (NDP) for regeneration process (RP) by silvicultural treatments in Quebec’s boreal forest. Practice NDP_Factors % Forest_Area NDPx Exotic plantations, afforestation 1 0.00% 0.0000 Plantation 0.9 5.08% 0.0458 Seeding 0.7 0.25% 0.0017 In-fill planting 0.6 0.50% 0.0030 Salvage logging 0.55 0.07% 0.0004 Clearcut and final cut 0.5 9.86% 0.0493 Commercial thinning (natural) 0.45 0.02% 0.0001 Careful logging (CL) 0.4 5.10% 0.0204 Partial cut 0.3 0.56% 0.0017 Undisturbed or natural disturbances 0 78.55% 0.0000 RP_NDP 0.1224 Actual RP_PNI 0.8776 Note: NDP_factors: naturalness degradation potential factors related to practices; %for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice. Note: NDP_factors: naturalness degradation potential factors related to practices; %for_area: percentage of forested area (in 2012); NDPx: Portion of the naturalness degradation potential for the xth practice. 2.3. Description of Tests Other analyses were carried out to verify the impact of specific assumptions. The effect of the hypothesis used for forest composition after careful logging or plantation was estimated by replacing the COMPO_PNI value at T70 by the actual value (COMPO_PNI at T0). A fire cycle of 245 years was used for natural disturbance inclusion in the aging simulation, based on a study located East of the study area [38]. To verify the effect of that factor on the proportions of closed and old forests, Forests 2019, 10, 325 13 of 29 the naturalness at T70 was evaluated by setting the fire cycle at 150 years, i.e., the average value for the Western Black spruce-feathermoss domain [39]. An exploratory analysis was also performed to check the model’s behaviour after the first rotation. The model was applied for the PL, CL-PL, and CL scenarios with and without strict protected areas beyond the first rotation, up to T150, keeping the same hypothesis for composition after CL and PL (as the composition after the second cutting cycle is still not known for these forests). For the hypothetical extreme tests, eight scenarios were considered (Table 7) with an increasing percentage of exotic species from 0% to 100% of the forest area (0%, 7%, 15%, 30%, 50%, 80%, and 100%). For example, scenario 1 considers 80% of plantation (PL) and 20% of careful logging (CL) along with the use of herbicide in plantations, leading to a coniferous cover of 85% and a proportion of late successional species groups equal to 85% (minus the percentage of exotic species), with a maximal LS value set to 30%. Scenario 2 is similar to scenario 1, but without drainage of wetlands. The hypotheses used for extremes scenarios are listed in Appendix D. Table 7. Extreme scenarios’ descriptions. Table 7. Extreme scenarios’ descriptions. Practice or Variable 1 Sce1 Sce2 Sce3 Sce4 Sce5 Sce6 Sce7 Sce8 Scenario 80PL 80PL-0DR 80PL-0DR- noBH 10PC 10PC-0DR- noBH 100Herb. 0Herb. 100PL %PL 80 80 80 80 80 80 80 100 %CL 20 20 20 10 10 20 20 0 %PC 0 0 0 10 10 0 0 0 Herb. 2.3. Description of Tests In PL In PL In PL In PL In PL All No All %DR 50 0 0 50 0 50 50 50 %CT 85 85 85 85 85 100 60 100 %LS 85-PLexo; ≤30 id id id id 85-PLexo; ≤50 15 100-PLexo; ≤30 CS Rarefied(R) R R R R R R Disappeared %OF 0 0 0 15 15 0 0 0 %IR 0 0 0 10 10 0 0 0 BH Y Y N Y N Y Y Y 1 Sce1: Scenario #1 and so on; PL: plantation; CL: careful logging; PC: partial cutting; DR: drained; CT coniferous cover type; LS: late successional species groups; CS: long lived companion species; OF: old forests; IR: irregular stands; BH: biomass harvest; PLexo: plantation of exotic species; Herb.: use of herbicides; id: idem as preceding. 1 Sce1: Scenario #1 and so on; PL: plantation; CL: careful logging; PC: partial cutting; DR: drained; CT coniferous cover type; LS: late successional species groups; CS: long lived companion species; OF: old forests; IR: irregular stands; BH: biomass harvest; PLexo: plantation of exotic species; Herb.: use of herbicides; id: idem as preceding. 1 Sce1: Scenario #1 and so on; PL: plantation; CL: careful logging; PC: partial cutting; DR: drained; CT coniferous cover type; LS: late successional species groups; CS: long lived companion species; OF: old forests; IR: irregular stands; BH: biomass harvest; PLexo: plantation of exotic species; Herb.: use of herbicides; id: idem as preceding. 3. Results Model results for the 3 FMU give a naturalness index (NI) of 0.717 for the year 2012, which corresponds to the near-natural class (Table 8). This naturalness level is explained by the logging of 22.3% of the area, and the rarefaction of some long-lived companion species. The main alteration is related to structure, resulting from the reduction of irregular stands (from 31.5% in the 1970s to less than 7% in 2012) and the loss of old forests (from 49.3% to 21.5% during the same period) relative to the historical state. Detailed results are provided as Supplementary Materials: Table S3 for 3FMU scenarios with protected areas; Table S4 for 3FMU scenarios without protected areas; Table S5 for extremes scenarios. Table 8. Actual (2012) results for the three forest management units. Characteristic PNI Naturalness Class Landscape context 0.870 Natural Composition 0.761 Near-natural Structure 0.232 Altered Dead wood 0.845 Natural Regeneration process 0.877 Natural Naturalness index (NI) 0.717 Near-natural Table 8. Actual (2012) results for the three forest management units. 14 of 29 Forests 2019, 10, 325 3.1. Naturalness of Forest Management Scenarios over Time 3.1. Naturalness of Forest Management Scenarios over Time 3.1. Naturalness of Forest Management Scenarios over Time 3.1.1. Naturalness Evolution of the Test Area Results of the naturalness assessment of the 3 FMU over time are given in Table 9 and illustrated in Figure 6. Results of the naturalness assessment of the 3 FMU over time are given in Table 9 and illustrated in Figure 6. Table 9. Results of the 3 FMU scenarios over time. Scenarios 1 with 17.9% protected areas Time since 2012 % logged pro CL pro CL-PL pro PL pro CL_bh pro CL-PL_bh pro PL_bh pro CL-PLexo_bh pro PLexo_bh pro 0 22.31 0.717 0.717 0.717 0.717 0.717 0.717 0.717 0.717 10 30.09 0.692 0.685 0.679 0.678 0.682 0.687 0.651 0.626 20 37.86 0.666 0.653 0.639 0.638 0.647 0.655 0.598 0.562 30 45.64 0.634 0.616 0.596 0.593 0.606 0.618 0.545 0.501 40 53.42 0.596 0.573 0.546 0.543 0.560 0.574 0.489 0.439 50 61.11 0.560 0.533 0.502 0.498 0.517 0.533 0.441 0.386 60 68.97 0.535 0.503 0.467 0.462 0.485 0.503 0.404 0.344 70 76.75 0.511 0.474 0.431 0.426 0.452 0.473 0.369 0.304 Scenarios without protected areas Time since 2012 % logged nopro CL nopro CL-PL nopro PL nopro CL_bh nopro CL-PL_bh nopro PL_bh nopro CL-PLexo_bh nopro PLexo_bh nopro 0 22.31 0.717 0.717 0.717 0.717 0.717 0.717 0.717 0.717 10 31.88 0.680 0.672 0.664 0.663 0.668 0.674 0.632 0.604 20 41.44 0.643 0.625 0.609 0.607 0.617 0.629 0.562 0.524 30 51.01 0.593 0.571 0.548 0.545 0.560 0.573 0.491 0.444 40 60.57 0.540 0.514 0.483 0.479 0.498 0.514 0.422 0.368 50 70.14 0.505 0.473 0.437 0.432 0.454 0.471 0.373 0.313 60 79.71 0.475 0.438 0.395 0.389 0.415 0.435 0.332 0.267 70 89.27 0.443 0.402 0.351 0.344 0.375 0.397 0.290 0.225 1 % logged: cumulative % of forest area regenerated through cutting; CL: 100% Careful logging; CL-PL: 50% careful logging with 50% plantation; PL: 100% plantation; PLexo: plantation of exotic species; bh: biomass harvest; pro: with 17.9% in protected areas; nopro: without protected areas. Table 9. Results of the 3 FMU scenarios over time. Scenarios 1 with 17.9% protected areas 1 % logged: cumulative % of forest area regenerated through cutting; CL: 100% Careful logging; CL-PL: 50% careful logging with 50% plantation; PL: 100% plantation; PLexo: plantation of exotic species; bh: biomass harvest; pro: with 17.9% in protected areas; nopro: without protected areas. Figure 6. Cont. Figure 6. Cont. 15 of 29 Forests 2019, 10, 325 Figure 6. Results for the 3 FMU over time: (a) With protected areas; (b) without protected areas. 3.1.1. Naturalness Evolution of the Test Area Scenario description: CL: 100% careful logging, CL-PL: 50% careful logging and 50% plantation of indigenous species, CL-PLexo: 50% careful logging and 50% plantation of exotic species, PL: 100% plantation of indigenous species, PLexo: 100% plantation of exotic species, bh: with biomass harvest, pro: with strict protected areas, nopro: without strict protected areas. Figure 6. Results for the 3 FMU over time: (a) With protected areas; (b) without protected areas. Scenario description: CL: 100% careful logging, CL-PL: 50% careful logging and 50% plantation of indigenous species, CL-PLexo: 50% careful logging and 50% plantation of exotic species, PL: 100% plantation of indigenous species, PLexo: 100% plantation of exotic species, bh: with biomass harvest, pro: with strict protected areas, nopro: without strict protected areas. Practicing careful logging only (CL) for the next 70 years on the 3 FMU, taking into account 17.9% of protected areas, would lead to a loss of one naturalness class relative to the current state. With 50% of careful logging and 50% of plantation of indigenous species (CL-PL), which corresponds roughly to the scenario currently applied, the study area would become semi-natural around 2045 and remain in this class for the rest of the period, while it would be from 2040 with 100% plantation of indigenous species (PL) (Figure 6a). In general, for a given ratio of protected areas, the naturalness declines with the proportion of plantation, and more sharply if exotic species are used. After 70 years (which roughly approaches the time required to complete the first cutting cycle), all scenarios that include protected areas would lead the studied 3 FMU to be classified as semi-natural except for those using exotic species, which would lead to an altered state. However, without protected areas (Figure 6b), only two scenarios would lead to a semi-natural class: CL or CL-PL, all the others would be in the altered class (0.4 > NI > 0.2), and the scenario considering 100% of plantation in exotic species over 70 years would be close to the very altered class. After 70 years, the scenario without protected areas corresponds to a rejuvenation of almost 90% of the territory. After the first cutting cycle, the age structure of the forest would be closer to normalization in the harvested area (i.e., each age class would be more evenly represented among forest stands), and the pressures would cover the whole production area, so the naturalness would stop its decline. 3.1.1. Naturalness Evolution of the Test Area Therefore, if ratios per practice and pressures are maintained, and assuming that stand composition following the harvest of the secondary forest would remain unchanged, values for naturalness would tend to stabilize (NI150, given on an indicative basis in Tables S3 and S4: with protection: PL = 0.381, CL-PL = 0.444, CL = 0.495; without protection: PL = 0.282, CL-PL = 0.351, CL = 0.406). With the hypothesis used, forest rejuvenation through 100% careful logging (CL) produces a reduction of the naturalness index over time that is less important (∆NI70: pro = −0.206; nopro = −0.274) than regeneration through plantation with indigenous species only (PL) (∆NI70: pro = −0.286; nopro = −0.366). Use of exotic species combined with biomass harvest (PLexo_bh) (∆NI70: pro = −0.413; nopro = −0.492) would produce around twice as much alteration as natural regeneration through logging itself. Forests 2019, 10, 325 16 of 29 Among the tests performed, it is the application of a forest management regime over the first cutting cycle that has the most important effect on the naturalness index (mean ∆NI70: pro = −0.287; nopro = −0.364). Compared with natural regeneration through CL, the regeneration mode has an important effect when exotic species are used (∆NI70: pro = −0.207; nopro = −0.218), but a lesser impact when indigenous species are planted over the whole area (∆NI70: PLpro = −0.08; PLnopro = −0.092; CL-PLpro = −0.037; CL-PLnopro = −0.042). Protection of 17.9% of the forest area has a noticeable effect in limiting the loss of naturalness in the 3FMU (mean ∆NI70 = 0.077). g Biomass harvesting would cause a reduction of the naturalness index of 0.038 and 0.047 after 70 years (Table 9) for the 100% careful logging scenarios with and without protected areas, respectively. This practice has a lesser effect in scenarios with higher levels of plantation: The NI reduction would be 0.005 and 0.007 for the 100% plantation scenario (Table 9) with and without protected areas, respectively, since it is assumed that site preparation prior to plantation would impact the dead wood. With the hypothesis used, the biomass harvest over the entire area would have as much impact on the naturalness as planting indigenous species over 50% of the harvested area. A better evaluation of the effect of the biomass harvest should include small woody debris for DW_NDP factors’ evaluation. 3.1.1. Naturalness Evolution of the Test Area y When comparing scenarios with and without 17.9% of strict protected areas (Table 9), after 70 years from 2012, the model determines that the CL-PL scenario with protected areas leads to a higher level of naturalness than the scenario with 100% CL without protected areas (NI70: 0.474 vs 0.443). The same observation is applied for 100% plantation (PL) with protected areas, which performs better than the CL-PL scenario without protected areas (NI70: 0.431 vs 0.402). CL only without protection performs slightly better than PL only with protected areas (NI70: 0.443 vs 0.431). The CL-PLexo (plantation of exotic species over 50%) with biomass harvest and protected areas has a higher naturalness than PL of indigenous species over 100% without biomass harvest, but no protected areas (NI70: 0.369 vs 0.351). It is important to underline the fact that with protected areas, after 70 years, 77% of the forest area will have been rejuvenated after harvesting, as opposed to 89% for the scenario without protected areas. With protected areas, after 70 years of regeneration through CL, a reduction corresponding to one naturalness class is observed (NI0 = 0.717; NI70 CL = 0.511 for a NI loss of 0.206). With plantation of exotic species and biomass harvest, the difference represents more than two classes (NI0 = 0.717; NI70 PLexo_bh = 0.304 for an NI loss of 0.413 over 70 years). Without strict protected areas, losses are more important as a larger area is available for cutting. 3.1.2. Sensitive Variables and Exploratory Analysis The results of the sensitivity analysis performed on the 3FMU for the scenario, CL-PL, with strict protected areas for current, 30 year, and 70 year periods are provided in Figure 7. Due to the use of non-linear models, a uniform variation of input parameters (5%) can have a non-linear effect on the results, depending on the curve slope around the parameter value. Therefore, the sensitivity of results can also vary over time (Figure 7). Results proved to be most sensitive to the proportion of forest area covered by exotic species. However, exotic species have never been used in the area under study. Beside the exotics, the most sensitive variables are the percentage of area in irregular (IR), old (OF), and closed forests (CF) at T0; percentage of area in closed (CF), old (OF), and irregular (IR) forests at T30; and percentage of area of closed forests (CF), late successional species groups (LS), and modified wetlands (Wm) at T70. As LS is among the most sensitive variables after 70 years, the hypothesis applied for composition after CL or PL might have an important impact on the resulting naturalness index. Assuming no effect on composition (by replacing the COMPO_PNI estimated with the actual value), the NI70 would be higher (∆NI70: PLpro: 0.037; PLnopro: 0.046; CL-PLpro: 0.040 CL-PLnopro: 0.051; CLpro: 0.05; CLnopro: 0.064), in the same range as planting indigenous species over 50% of the harvested area. Therefore, a better assessment of forest composition through time as influenced by silvicultural treatments would be important to improve the reliability of results. 17 of 29 Forests 2019, 10, 325 2019, 10, x FOR PEER REVIEW 1 (a) (b) (c) igure 7. Results of the sensitivity analysis testing a variation of 5% of the adjusted parameter valu a) Adjusted values at T0; (b) adjusted values at T30; (c) adjusted values at T70; HS: Horizontal structur T: coniferous cover type, LS: late successional species groups, W_CC: clear cut on wetland, R egeneration process, DW: dead wood, ant: anthropized land; WM: modified wetland, CS: companio pecies, CF: closed forest, OF: old forest, IR: irregular stands, exo: exotic species. 3.1.2. Sensitive Variables and Exploratory Analysis -6 -4 -2 0 2 4 6 8 exo OF CS ant RP LS HS ± 5% parameter variation on T0 adjusted values %-5 %+5 -6 -4 -2 0 2 4 6 8 exo CF OF IR LS Wm ant CS DW RP W_CC CT HS ± 5% parameter variation on T30 years adjusted values %-5 %+5 -6 -4 -2 0 2 4 6 8 exo CF LS Wm ant CS DW OF IR RP W_CC CT HS ± 5% parameter variation on T70 years adjusted values %-5 %+5 igure 7. Results of the sensitivity analysis testing a variation of 5% of the adjusted parameter valu a) Adjusted values at T0; (b) adjusted values at T30; (c) adjusted values at T70; HS: Horizontal structur T: coniferous cover type, LS: late successional species groups, W_CC: clear cut on wetland, R egeneration process, DW: dead wood, ant: anthropized land; WM: modified wetland, CS: companio pecies, CF: closed forest, OF: old forest, IR: irregular stands, exo: exotic species. (a) -6 -4 -2 0 2 4 6 8 exo OF CS ant RP LS HS ± 5% parameter variation on T0 adjusted values %-5 %+5 (b) -6 -4 -2 0 2 4 6 8 exo CF OF IR LS Wm ant CS DW RP W_CC CT HS ± 5% parameter variation on T30 years adjusted values %-5 %+5 (b) (c) -6 -4 -2 0 2 4 6 8 exo CF LS Wm ant CS DW OF IR RP W_CC CT HS ± 5% parameter variation on T70 years adjusted values %-5 %+5 ± 5% parameter variation on T70 years adjusted values (c) Figure 7. Results of the sensitivity analysis testing a variation of 5% of the adjusted parameter value: (a) Adjusted values at T0; (b) adjusted values at T30; (c) adjusted values at T70; HS: Horizontal structure, CT: coniferous cover type, LS: late successional species groups, W_CC: clear cut on wetland, RP: regeneration process, DW: dead wood, ant: anthropized land; WM: modified wetland, CS: companion species, CF: closed forest, OF: old forest, IR: irregular stands, exo: exotic species. Figure 7. 3.1.2. Sensitive Variables and Exploratory Analysis Results of the sensitivity analysis testing a variation of 5% of the adjusted parameter value: (a) Adjusted values at T0; (b) adjusted values at T30; (c) adjusted values at T70; HS: Horizontal structure, CT: coniferous cover type, LS: late successional species groups, W_CC: clear cut on wetland, RP: regeneration process, DW: dead wood, ant: anthropized land; WM: modified wetland, CS: companion species, CF: closed forest, OF: old forest, IR: irregular stands, exo: exotic species. Forests 2019, 10, 325 18 of 29 Considering that a fire cycle of 150 instead of 245 years induces a reduction of the NI70 of 0.026 (±0.001) in the scenarios with protected areas and a reduction of 0.030 in the scenarios without protected areas, the model therefore seems relatively robust to the hypothesis used for the fire cycle. Considering that a fire cycle of 150 instead of 245 years induces a reduction of the NI70 of 0.026 (±0.001) in the scenarios with protected areas and a reduction of 0.030 in the scenarios without protected areas, the model therefore seems relatively robust to the hypothesis used for the fire cycle. The weight of 0.5 given to estimate the effect on the naturalness of clearcuts on wetlands has a very limited effect considering that clearcuts only affect 7.8% of the wetlands. Hypothesizing 100% instead of 50% would have reduced the resulting NI70 by 0.0059, which is marginal. The results suggest that the model structure, which applies pressure as a reduction of the condition, produces a degradation of the naturalness over the first cutting cycle. The exploratory analysis reveals that beyond that point, results tend to stabilize (see results at T150 in Table 10): The strategy is applied over the whole managed territory and therefore the composition after regeneration treatment related to age (here younger and older than 20 years old) becomes constant. The age structure is gradually normalized over the managed area, but continues to evolve with stand aging in the excluded (protected) areas. Applying our hypothesis, we observe a slight reduction of the naturalness between T70 and T150 in all scenarios. Table 10. Results of the 3 FMU scenarios at T150. 3.1.2. Sensitive Variables and Exploratory Analysis Protection CL CL-PL PL CL_bh CL-PL_bh PL_bh CL-PLexo_bh PLexo_bh Pro 0.495 0.444 0.381 0.441 0.412 0.374 0.328 0.260 Nopro 0.406 0.351 0.282 0.339 0.313 0.272 0.231 0.164 CL: 100% Careful logging; CL-PL: 50% careful logging with 50% plantation; PL: 100% plantation; PLexo: plantation of exotic species; bh: biomass harvest; pro: with 17.9% in protected areas; nopro: without protected areas. Table 10. Results of the 3 FMU scenarios at T150. 3.2. Naturalness of High Pressure Management Scenarios 3.2. Naturalness of High Pressure Management Scenarios Extreme scenarios lead to a naturalness index corresponding to altered and very altered classes (Table 11, Figure 8). Higher levels of alteration are associated with an important use of exotic species combined with the loss of companion species. Absence of drainage and, to a lesser extent, application of measures leading to the presence of old forests and irregular stands make it possible to sustain a higher level of naturalness. Scenarios including a homogenous coniferous cover type (scenarios 6 and 8) and a scenario corresponding to a degraded composition with low coniferous cover and late successional representation (scenario 7) lead to a very altered class even without using exotic species. For less extreme combinations, the naturalness class is altered if no exotic species are used, and very altered if a small proportion of exotics is present. Table 11. Extreme scenarios’ results. See also Table 7 for a detailed scenario description. Variable 1 Sce1 Sce2 Sce3 Sce4 Sce5 Sce6 Sce7 Sce8 %exotics 80PL 80PL-0DR 80PL-0DR- noBH 10PC 10PC-0DR- noBH 100Herb. 0Herb. 100PL 0 0.215 0.247 0.269 0,231 0.292 0.167 0.177 0.085 7 0.174 0.206 0.228 0,191 0.251 0.141 0.148 0.065 15 0.149 0.181 0.203 0.165 0.226 0.125 0.130 0.053 30 0.120 0.152 0.173 0.136 0.197 0.106 0.109 0.050 50 0.090 0.122 0.144 0.106 0.167 0.085 0.087 0.046 80 0.074 0.106 0.128 0.090 0.151 0.074 0.074 0.040 100 0.036 1 SceX: Scenario number, PL: plantation; PC: partial cutting; DR: drained; BH: biomass harvest; Herb.: use of herbicides; Sce1: 80%PL 20%CL Herb. In PL 50%DR CT = 85% LS = (85% −PLexo) ≤30 CS = rarefied OF = 0% IR = 0% BH; Sce2: Sce1 without DR; Sce3: Sce1 without DR nor BH; Sce4: Sce1, but 10CL and 10PC so OF = 15% and IR = 10%; Sce5: Sce4 without DR nor BH; Sce6: Sce1, but 100%Herb so CT = 100% and LS = (85% −PLexo) ≤50; Sce7: Sce1 without Herb in PL so CT = 60% and LS = 15%; Sce7: 100%PL 100%Herb. 50%DR CT = 100% LS = (100% − PLexo) ≤30 CS = disappeared OF = 0% IR = 0% BH. Table 11. Extreme scenarios’ results. See also Table 7 for a detailed scenario description. 19 of 29 Forests 2019, 10, 325 Figure 8. Results for the eight extreme scenarios. See also Table 7 for detailed scenario descriptions. Sce1: 80%PL 20%CL Herb. 4. Discussion Decision support systems depend upon summaries and systematic reviews available at the time of their conception and are therefore subject to improvements as new information becomes accessible [26]. Nevertheless, the conceptual model developed in this study provides a basic frame for naturalness assessment, although indicators, measures, and curves might be revised when better information and data become available. 3.2. Naturalness of High Pressure Management Scenarios In PL 50%DR CT = 85% LS = (85% −PLexo) ≤30 CS = rarefied OF = 0% IR = 0% BH; Sce2: Sce1 without DR; Sce3: Sce1 without DR nor BH; Sce4: Sce1, but 10CL and 10PC so OF = 15% and IR = 10%; Sce5: Sce4 without DR nor BH; Sce6: Sce1, but 100%Herb so CT = 100% and LS = (85% −PLexo) ≤50; Sce7: Sce1 without Herb in PL so CT = 60% and LS = 15%; Sce7: 100%PL 100%Herb. 50%DR CT = 100% LS = (100% −PLexo) ≤30 CS = disappeared OF = 0% IR = 0% BH. Di i Figure 8. Results for the eight extreme scenarios. See also Table 7 for detailed scenario descriptions. Sce1: 80%PL 20%CL Herb. In PL 50%DR CT = 85% LS = (85% −PLexo) ≤30 CS = rarefied OF = 0% IR = 0% BH; Sce2: Sce1 without DR; Sce3: Sce1 without DR nor BH; Sce4: Sce1, but 10CL and 10PC so OF = 15% and IR = 10%; Sce5: Sce4 without DR nor BH; Sce6: Sce1, but 100%Herb so CT = 100% and LS = (85% −PLexo) ≤50; Sce7: Sce1 without Herb in PL so CT = 60% and LS = 15%; Sce7: 100%PL 100%Herb. 50%DR CT = 100% LS = (100% −PLexo) ≤30 CS = disappeared OF = 0% IR = 0% BH. 4.1. Conceptual Model Evaluating the intensity of silvicultural management makes it possible to quantify land-use intensity in forests. Gossner et al. [40] showed that biodiversity can be related with land use measures, such as naturalness based on trees species composition, dead wood, and other structural characteristics, or stand management intensity based on tree species, stand age, and aboveground living and dead woody biomass. Therefore, a methodology combining the condition evaluation of composition and structure, and pressure measures resulting from silvicultural practices represents a good proxy for the evaluation of the effects of forestry on biodiversity resulting from a combination of various practices. Further research is needed to verify to which extent the utilization of condition measures can detect improvements of naturalness resulting from restoration or enhanced ecological management strategies. The inclusion of pressure measures can adequately reflect the effects of mitigation measures, although the model should be further validated. Our results suggest that the presence of strict protected areas in a forest landscape compensates, to some extent, for the impacts of intensive management [41]. The model developed in this paper was shown to be sensitive enough to characterize the naturalness of different forestry management systems and therefore discriminate between different wood supplies from a variety of forest management practices. The results demonstrate that the proposed naturalness assessment model can be useful to evaluate the land use intensity of forestry practices at a finer level than existing approaches to inform decision-making in life cycle assessment. For instance, global guidance for life cycle impact assessments [3] currently considers only two levels Forests 2019, 10, 325 20 of 29 of intensity for forestry: Intensive and extensive forestry. For their part, Chaudhary and Brooks [42] proposed to divide secondary vegetation in four classes: Plantations, clearcut, selective logging, and reduced impact logging. In contrast, our approach makes it possible to take into account the condition of the forest as well as the proportion of different practices to more precisely characterize the forest management strategy and its impacts on ecosystem quality. The application of pressure as a reduction of the condition contributes to an important reduction of the naturalness resulting from the progression of the first cutting cycle over the area. Such an important reduction of the quality observed as a result of the initial land use transformation process is coherent with the conceptual model proposed by LCA developers for ecosystem quality evolution related to land use [43]. 4.1. Conceptual Model However, tests over time highlight some specificities of forestry land use reflected by the model. Contrary to most land uses, the first cutting cycle of a forest land corresponds to a progressive transformation from the natural state to a naturalness level related to the forest management strategy applied. In Quebec’s boreal forests, this initial transformation may take up to 100 years. During the subsequent rotations, the naturalness index tends to stabilize as a result of the normalization of the forest, supposing that sustainable management is used. However, our model indicates a trend toward a slow erosion of the ecosystem quality over time during the sustainable production phase. If the land use ever stops and constraints are relaxed (although future land use changes are more likely to progress toward land uses of higher hemeroby), the naturalness should progressively improve as condition indicators will gradually recover with the aging of the forest. However, we do not know if condition indicators will ever come back to the natural state after relaxation. Nevertheless, some pressures will remain (ex: Drained wetlands or other anthropic features, like permanent roads, energy transportation lines, etc.), so theoretically the ecosystem should never recover completely. Given the model’s sensitivity to age related variables, a better integration of plantations would be necessary to improve the results. It would be interesting to explore the application of the naturalness assessment model using data from sustainable harvest calculation systems to reflect the effect of forest management strategy implementation considering simultaneously shorter rotation for plantations, application of the modelled composition and growth, simulation of harvest spatially applied to the admissible area, and a different handling method of natural disturbances [44]. The natural assessment model developed in our study was designed to be easily adapted to other regions using the conceptual model. All five naturalness characteristics should be considered, indicators should be reviewed to include all regionally important ecological issues, and measures should be identified among available data. Curves for condition indicators would then have to be calibrated using the specific historical values of the studied region and NDP factors should be adjusted to reflect regional practices’ effects. Our model could also be integrated in LCA models and used to inform building and construction eco-designers beyond the outcomes of this specific case study. To do so, further work still needs to be done to generate regionalized results across Canadian FMUs and ecosystems. 4.1. Conceptual Model Depending on the availability of historical data, naturalness assessment could be performed for ecological domains or sub-domains (ex. Western black spruce feathermoss sub-domain); aggregated results could be calculated by region or country to allow for the assessment of harvested wood products in a broader context, where the exact provenance of the wood is not known. 4.2. Naturalness Assessment Application to the 3 FMU The results of the case study inform us that wood coming from the 3 FMU has less negative impacts on the quality of ecosystems if the management strategy relies on natural regeneration through careful logging instead of plantation, especially if exotics are involved. It is still possible to limit the potential loss of specialized species resulting from sustainable forest management provided that the proportion of strict protected areas is sufficient to mitigate the degradation of the condition indicators. The following observations raise questions that should be further addressed in LCA. The actual naturalness of the 3 FMU is near-natural as a result of a rejuvenation through the harvest of 22.3% of 21 of 29 21 of 29 Forests 2019, 10, 325 the forest area, including plantation of indigenous species over 5% of the forest area. The difference between management scenarios results from the cumulative effects of practices over time, mainly those inducing rejuvenation, the model being sensitive to age-related variables (OF and CF). Given that the naturalness index tends to stabilize after the completion of the first cutting cycle, the naturalness assessment of a forest management strategy requires an evaluation over the whole cutting cycle. This corresponds also to the potential impact of the forest regime. However, actual naturalness could be used if the objective is to characterize the naturalness of the forest from which the wood is currently procured. The actual naturalness is the result of the practices applied up to now. It does not necessarily correspond to the level resulting from sustainable management, which is more consistent with the evaluation over a whole cutting cycle. In a territory including forests that have never been harvested, such as the 3 FMU, actual naturalness gives an optimistic portrait and does not correspond to the potential impact of the present activities. Some important limitations of the use of the model in Quebec’s boreal forest need to be stressed. The model could not be used to evaluate the naturalness index in the test area beyond the first cutting cycle, as no data was available to describe the future evolution of the composition of secondary forests in that area. Pni and NDP evaluation curves and factors should be validated according to expert opinion. The uncertainty of the results increases over time; the evolution of forest composition in older secondary forests (>40 years) in this ecological domain has yet to be verified. 4.2. Naturalness Assessment Application to the 3 FMU The future evolution of the natural disturbance regime under a changing climate is also unknown. Therefore, its effects on age structure are unknown as well as the proportion of future regeneration failures, which affects the closed forest coverage. Nevertheless, the importance of the proportion of old forests and irregular stands, and eventually closed forests, is coherent with present concerns related to age structure when attempting to apply ecosystem-based management [27]. 5. Conclusions Despite the necessity of further model and parameter validation, the model developed in this paper makes it possible to assess along a single alteration gradient the impact on ecosystem quality of different forestry management systems, simultaneously considering the condition of the forest and the mix of forestry practices involved. Therefore, the model is sensitive enough to differentiate between forest management strategies. The capacity of the model to reach a very altered class was tested with hypothetical high pressure levels associated with the use of exotic species. Tests over time showed that the results are coherent with the conceptual model proposed by LCA developers for ecosystem quality evolution related to land use and highlight some specificities of the forest land use related to forestry. For instance, the initial land use transformation caused by forestry is gradual and the resulting level of naturalness depends upon the management strategy. The results of this research work set the basis to inform building and construction designers on the potential impact on ecosystem alteration associated to harvested wood products at the landscape level as a function of forest management strategy, considering the condition of the forest and the nature of the adopted forestry practices. The naturalness index will have to be assessed at a regional level and scaled for the wood productivity and eventually aggregated at an upper geographical level in order to be used in life cycle impact assessment methodologies. Whether the naturalness index could be used as a mid-point indicator or is related to the damage category is still an open question and depends on biodiversity data that are available to generate the correlation. Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4907/10/4/325/s1, Table S1: Pni and exo_NDP determination, Table S2: Area by age class evolution by 10 years period, Table S3: Composition and irregular evaluation over time, Table S4: Detailed results over time for the 3FMU with protected areas, Table S5: Detailed results over time for the 3FMU without protected areas, Table S6: Detailed results for the hypothetical extreme scenarios, Data files: Maps: SIFORT1_3MU, SIFORT4_3MU and Ecoforest4_3MU. 22 of 29 Forests 2019, 10, 325 Author Contributions: Conceptualization, S.C. and L.B.; methodology, S.C. and L.B.; validation, S.C.; formal analysis, S.C.; investigation, S.C.; resources, R.B.; data curation, S.C. and É.T. for dead wood.; writing—original draft preparation, S.C.; writing—review and editing, S.C., L.B., M.M., R.B. and É.T.; visualization, S.C.; supervision, R.B., L.B. Appendix A Naturalness Characteristics, Indicators and Measures The five naturalness characteristics selected for the naturalness assessment corresponds to (1) landscape context, (2) forest composition, (3) structure, (4) dead wood and (5) regeneration process. Grouping indicator evaluation under naturalness characteristics limit the number of indicators and can facilitate the adaptation of the method to other regions according to available measures and regional ecological issues. Tree species and forest structure are among the most studied traits of naturalness [14]. Compared with others naturalness assessments [14], the naturalness characteristic “Landscape context” has been added to take into account the proportion of forest habitat at the landscape level. Dead wood was distinguished from the structure because forestry practices can have distinct or even opposite effects (ex: biomass harvest impact dead wood directly; partial cut can promote irregular structure but can reduce dead wood amount if harvest reduces mortality). Similarly, old forests and irregular stands are evaluated distinctly because stands over 90 years old are not necessarily irregular. Very old forests generally exhibit irregular vertical structure and important amount of dead wood, but this is not the case for 100 years old stands. The use of three different indicators for old forests (>100 years old), irregular stands and dead wood allows considering the effects of mitigation measures. g g g This section provides information about indicators and measures used for the naturalness assessment, and the calculation for each characteristic. 5. Conclusions and M.M.; project administration, R.B. and M.M.; funding acquisition, M.M. Funding: This research was funded by the Natural Sciences and Engineering Research Council of Canada through a grant to Manuele Margni, grant number CRD-462197-13, with the collaboration of Cecobois, Canadian Wood Council, Desjardins, GIGA, Hydro-Québec and Pomerleau. Acknowledgments: The authors would like to thank M. Denis Chiasson from Barrette Chapais Ltée for providing information related to the studied area, M. David Baril from BFEC for providing links to required information about sustainable yield calculation, Martin Barrette from Forest Research Direction of the Québec’s Ministère des Forêts de la Faune et des Parcs for his valuable contribution the concept of Figure 1, and M. Stefano Biondo and his team form Laval University GeoStat Center for their help with maps data handling. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. Appendix A.2 Tree Species Composition Tree species composition considers two condition indicators: the cover type (CT) (coniferous, deciduous or mixed) which characterizes the forest matrix, and the late successional species groups (LS). Both indicators are measured using percentage of forested area (Table 1). The distribution of forest cover type represents a basic forest characteristic used for forest ecosystem management [49]; it represents an indicator of ecosystem diversity that is essential for biodiversity according to Quebec’s forest sustainable management criteria and indicators [50]. The cover type distribution depends on successional status related to the major natural disturbance regime, i.e., fire in Quebec’s boreal forest [51]. The percentage of forested area of the dominant forest cover type, namely coniferous for Northeastern America [39], is used as the indicator. This measure allows the detection of a shift toward mixed or deciduous cover types related to forest exploitation, a phenomenon observed in Quebec’s boreal forests [52]. In other parts of the world, as in southern Finland, forest management practices such as the use of herbicides reduce the non-dominant cover types (i.e., hardwood and mixed), diminishing tree species diversity [53]. The alteration of natural tree species composition of forest stands results primarily from forest management and past land use, which increase the abundance of stands in their early successional stages [54]. In Quebec’s boreal forests, early successional stages include intolerant hardwood species such as trembling aspen (Populus tremuloides) and paper birch (Betula papyrifera). In the northernmost areas of the boreal forest, the late successional species groups correspond to pure black spruce or old black spruce/balsam fir stands [55]. The increase in extent and frequency of disturbances can lead to a reduction of the abundance of late successional species groups in the landscape. Elsewhere, as in eastern Finland for example, forest management activities can reduce the occurrence of stands in their earlier stages of succession [56]. The pressure measure on forest composition considers exotic species stands (exo), and the decrease of the abundance of long-lived companion species (CS). Even if exotic species are absent from the test area, this indicator has been included to allow model adaptation to other regions and perform the test to high levels of pressure. The NDP from exotic stands is evaluated using the logarithmic curve (Figure 3c). Appendix A.2 Tree Species Composition The use of exotic species is considered to have a high potential impact because of the associated risks of genetic pollution or hybridization with indigenous species [57], species naturalization or even invasion [58], and alteration of natural disturbance regime resulting from interaction with disturbance agents such as insects [59]. Higher risks associated with this factor justifies the use of the logarithmic curve. The rarefaction of late successional companion species resulting from forest management activities represents another issue related to forest composition. In Quebec’s boreal forests it is the case for eastern white cedar (Thuya occidentalis) and white spruce (Picea glauca) [27]. Theoretically, high pressure values for both measures applied on the mean condition indicator could yield negative values; to avoid inconsistencies, the minimal value for compo_PNI is set to 0 when the calculation leads to negative values. Appendix A.1 Landscape Context Landscape context refers to the amount of forest habitat at the landscape level. This characteristic considers closed forest (CF) habitat (density ≥25%) over 40 years old as a condition indicator (Table 1), measured using the percentage of terrestrial area of forest stands [38]. This measure evaluates the extent of conditions suitable for forest dependent species, such as certain birds or litter invertebrates [19], or for species needing large undisturbed areas such as the woodland caribou [45]. The measure is sensitive to overabundance of young stands resulting from disturbances and/or regeneration failure following disturbance [46]. Habitat loss has a greater effect on biodiversity than habitat fragmentation [47] to some extent; the latter becomes more important when the amount of habitat represents a low proportion of the landscape, below a critical habitat threshold of 30% [29], at which point many species are lost. Therefore, if the proportion of forested area is equal or below 30%, the inclusion of fragmentation in landscape context evaluation should be considered. However, the model should not be applied if the proportion of anthropized land reaches 70%: such a situation would extend outside of the proposed naturalness gradient. The pressure on landscape context considers both land use change, identified as one of the main drivers of biodiversity loss by the Millennium Ecosystem Assessment [2], and alteration of wetlands, recognized as an important habitat to protect in the latest Quebec’s regulation [48]. The pressure measures are evaluated as the percentage of terrestrial area in anthropic land use, the percentage 23 of 29 Forests 2019, 10, 325 of wetlands with an anthropically modified condition (ex: drainage or transmission line) and the percentage of wetlands with clearcut. The resulting context_PNI (Table 2) could have a negative value when pressures are severe enough, reflecting an extension of the naturalness gradient into the hemeroby gradient. However, in our model, negative values were set to zero. Forests 2019, 10, 325 Forests 2019, 10, 325 24 of 29 Forest age structure corresponds to the relative abundance of stands belonging to different development stages or age classes [33]. In natural forests, the age structure is the result of natural disturbances regimes, mainly driven by fire in Quebec’s boreal forest [49]. Disturbance regime modelling of Quebec’s boreal forest show that the median proportion of old forests (>100 years old) varies from 49 to 77% [39]. The proportion of forests older than 100 years has never been under 30% in the natural boreal forest [62]. Old forests represent a critical habitat for maintaining biodiversity due to the presence of bigger older trees, internal structural complexity and high abundance of dead wood including large pieces, which are important characteristics for many specialized species [63]. In the crownland managed forest of Quebec, almost 1% of the productive forests is harvested annually [44], reducing the proportion of the old forests [64]. Therefore, the measure used for the condition indicator relates to old forests (OF) and corresponds to the percentage of forest area comprising stands over 100 years old. Vertical structure is measured with the percentage of forest area covered by irregular stands (IR). With age, old forests progressively develop an irregular vertical structure (associated with cohort replacement and gap dynamics due to the death of old trees) along with the presence of old high trees and an abundance of dead wood [55,65]. This structural complexity produces habitat diversification and favors biodiversity, especially among vascular plants, terrestrial mosses, liverworts and lichens [62]. We used two different condition indicators for structure: old forests and irregular stands. 100 years old forests do not necessarily have an irregular structure and the proportion of multi-storied and irregular stands was more important before the onset of the commercial exploitation of the forest. Cohort based management had been proposed to answer this issue and the inclusion of the irregular stands as an indicator allows to consider stands from second and third cohorts [64]. Horizontal structure refers to spatial distribution of stems. Plantation and thinning homogenize the horizontal structure by regularising spatial stem distribution and density [15]. This results in a lower variation within the stand, and therefore a lower variety of microhabitats, suitable for a narrower array of species. Horizontal structure is evaluated through pressure measures resulting from silvicultural treatments using percentage of forest area by treatments multiplied by corresponding NDP factors. Forests 2019, 10, 325 The resulting NDP is then applied as a reduction percentage on the mean of the age and vertical structure pni’s (Table 2). Appendix A.4 Dead Wood About one fifth of all forest species are dependent on decaying wood [56]. The relationship between dead wood (DW) and species diversity is higher in boreal forests than in temperate forests [66]. Decaying wood plays a substantial role in many ecological processes [67]: it affects carbon storage, energy flow and nutrient cycles, contributes to the water-holding capacity of the soil, sustains ectomycorrhizal formation and activity and offers a substrate for seedlings establishment. Decaying wood hosts a large number of epixylic bryophytes and lichens, polypores and other decomposer fungi and invertebrates. Coarse woody debris are particularly impacted by wood harvesting [67]. As no dead wood data is available from forest inventory, the PNI evaluation uses pressure related to human intervention associated with specific silvicultural treatments (PNI = 1 −NDP) (Table 5), considering their respective effects on coarse woody debris, weighed by the percentage of forest area affected by each treatment. Appendix A.3 Structure In general, natural stands tend to be structurally heterogeneous, both vertically and horizontally. Structural complexity may determine habitat availability and may thus influence diversity of plant, animal and microbial communities [60]. Relevant measures of forest structure include canopy cover, vertical structure and size or age distribution of trees [61]. The following condition indicators were retained for structure: age structure and vertical structure (Table 1). Horizontal structure was evaluated using a pressure indicator resulting from silvicultural treatments. Appendix A.5 Regeneration Process Regeneration process (RP) refers to a chain of events necessary to ensure the renewal of the forest and focusses on the method of arrival or persistence of a species on a site during or after disturbance [68]. Regeneration process considers the regeneration mode in fire dependent ecosystems as a benchmark for natural state (RP_PNI = 1) and, as for dead wood, applies NDP factors associated with silvicultural treatments, weighed by the proportion of forest area subject to each treatment. The determination of NDP factors (based on expert opinion) compares silvicultural treatments with fire effects considering: seedling provenance (natural or artificial) and their genetic variability; adaptation of regeneration to 25 of 29 25 of 29 Forests 2019, 10, 325 fire; regeneration density; protection/destruction of natural advance regeneration; seed tree abundance; effects on forest floor that can be either positive (when necessary to control paludification on sensitive sites) or negative. Appendix B Hypothesis Used for the 3MU Naturalness over Time Evaluation • Logging of 1% of the forest area per year corresponding to the forest area included in the area subject to sustainable allowable cut estimation. The harvest was applied on the oldest age class available. Scenarios were applied on 10-years time-steps for a total length of 70 years, corresponding roughly to the first cutting cycle, as 22.5% of the forest area has already been harvested once in 2012; • A rate of 0.408% per year [38] of rejuvenation from natural disturbances was applied by distributing the rejuvenation proportionally to the area of age classes ≥30 years old; • A rate of 0.408% per year [38] of rejuvenation from natural disturbances was applied by distributing the rejuvenation proportionally to the area of age classes ≥30 years old; • For aging, 2 matrix of area by age classes has been used: rejuvenation from natural disturbances only, and natural disturbances plus harvest (over 1% of the admissible area applied on the oldest age class available at each period). Then the resulting matrix has been weighed according to the percentage of excluded area from the production forest depending on the scenario: 22.2% for the scenarios with protected areas, and 4.3% without protected areas; • Anthropic land use, drainage modification and unproductive area were held constant • No modification of forest composition resulting from natural disturbance was included. Composition after logging was based on composition observed on the forest map, distinguishing forest under and over 20 years old: forest under 20 years old: CL: CT = 6.5% coniferous, %LS = 1.1%; PL: CT = 42.8% coniferous, %LS = 0.4%; Forest over 20 years old: CL: CT = 48.1% coniferous, %LS = 13.1%; PL: CT = 86.6% coniferous, %LS = 9.7%. References 1. Jolliet, O.; Saadé, M.; Crettaz, P.; Shaked, S. Analyse du Cycle de vie. Comprendre et Réaliser un Écobilan, 2nd ed.; Presses Polytechniques et Universitaires Romandes: Lausanne, Switzerland, 2010. 2. MEA. Ecosystems and Human Well-being: Biodiversity Synthesis; World Resources Institute: Washington, DC, USA, 2005; p. 86. 3. Life Cycle Initiative. Global Guidance for Life Cycle Impact Assessment Indicators. Volume 1; UNEP DTIE: Paris, France, 2016; p. 159. 4. Curran, M.; De Souza, D.M.; Anton, A.; Teixeira, R.F.M.; Michelsen, O.; Vidal-Legaz, B.; Sala, S.; Mil, I.; Canals, L. How Well Does LCA Model Land Use Impacts on Biodiversity?—A Comparison with Approaches from Ecology and Conservation. Environ. Sci. Technol. 2016, 50, 2782–2795. [CrossRef] [PubMed] 5. Gaudreault, C.; Wigley, T.B.; Margni, M.; Verschuyl, J.; Vice, K.; Titus, B. Addressing Biodiversity Impacts of Land Use in Life Cycle Assessment of Forest Biomass Harvesting. Wiley Interdiscip. Rev. Energy Environ. 2016. [CrossRef] 6. Souza, D.M.; Teixeira, R.F.; Ostermann, O.P. Assessing biodiversity loss due to land use with Life Cycle Assessment: Are we there yet? Glob. Chang. Biol. 2015, 21, 32–47. [CrossRef] 7. Globio. Available online: https://www.globio.info/what-is-globio/reference-database/per-pressure (accessed on 6 February 2019). 8. PREDICTS. Available online: https://www.predicts.org.uk/pages/outputs.html (accessed on 6 February 2019). 8. PREDICTS. Available online: https://www.predicts.org.uk/pages/outputs.html (accessed on 6 February 2019). 9. Brentrup, F.; Küsters, J.; Lammel, J.; Kuhlmann, H. Life Cycle Impact Assessment of Land Use Based on the H b C t I t J Lif C l A 2002 7 339 348 [C R f] 9. Brentrup, F.; Küsters, J.; Lammel, J.; Kuhlmann, H. Life Cycle Impact Assessment of Land Use Based on the Hemeroby Concept. Int. J. Life Cycle Assess. 2002, 7, 339–348. [CrossRef] 0. Michelsen, O. Assessment of land use impact on biodiversity. Int. J. Life Cycle Assess. 2008, 13, 22 [CrossRef] 1. Fehrenbach, H.; Grahl, B.; Giegrich, J.; Busch, M. Hemeroby as an impact category indicator for the integra of land use into life cycle (impact) assessment. Int. J. Life Cycle Assess. 2015, 20, 1511–1527. [CrossRef] 12. Rossi, V.; Lehesvirta, T.; Schenker, U.; Lundquist, L.; Koski, O.; Taylor, R.; Humbert, S. Capturing the potential biodiversity effects of forestry practices in life cycle assessment. Int. J. Life Cycle Assess. 2017. [CrossRef] 13. Farmery, A.K.; Jennings, S.; Gardner, C.; Watson, R.A.; Green, B.S. Naturalness as a basis for incorporating marine biodiversity into life cycle assessment of seafood. Int. J. Life Cycle Assess. 2017, 22, 1571–1587. [CrossRef] 14. Winter, S. Appendix C Naturalness Assessment over Time Procedure 1. Starting from actual area distribution by age classes, simulation by time step of aging, natural disturbances effects and rejuvenation from cutting (For the 3FMU: Table S2), to calculate OF and CF for each period; 2. Find hypothesis for composition indicators after silvicultural treatments included in the forest management strategy (For the 3FMU: CT and LS after CL and PL, under and over 20 years old, were compiled from ecoforest maps), calculate CT, LS and IR resulting from practices application for each period (For the 3FMU: Table S3); 3. For each condition indicator, transform the percentage of area computed by period in pni (For the 3FMU: Using Table S1); 4. Considering the proportion gradually rejuvenated as a result of practices application, evaluation of NDP (For the 3FMU: Using NDP appropriate grid or curve for the scenario from Table S1), to calculate exo_NDP, HS_NDP, DW_NDP and RP_NDP; 5. Complete NDP evaluation using hypothesis about their evolution for other pressure measures (For the 3FMU: CS, Wm, W_CC and ANT assumed to be constant); 6. Calculate NI by time step for each scenario (For the 3FMU: Tables S4 and S5). 6. Calculate NI by time step for each scenario (For the 3FMU: Tables S4 and S5). pp yp • These scenarios use a rotation period of 65 years; • The proportions of unproductive and anthropic lands correspond to those from the 3 FMU; • No old forests nor irregular stands are considered except when partial cuts are used; • Recognized reduction of companion species is included. • Drainage of wetlands is set to 50% when applied. • No excluded area is considered. pp yp • These scenarios use a rotation period of 65 years; • The proportions of unproductive and anthropic lands correspond to those from the 3 FMU; • No old forests nor irregular stands are considered except when partial cuts are used; • Recognized reduction of companion species is included. • Drainage of wetlands is set to 50% when applied. • No excluded area is considered. The proportions of unproductive and anthropic lands correspond to those from the 3 FMU; No old forests nor irregular stands are considered except when partial cuts are used; 26 of 29 26 of 29 Forests 2019, 10, 325 23. Landres, P.B.; Morgan, P.; Swanson, F.J. Overview of the use of natural variability concepts in managing ecological systems. Ecol. Appl. 1999, 9, 1179–1188. [CrossRef] References 28. Zuckerberg, B.; Porter, W.F. Thresholds in the long-term responses of breeding birds to forest cover and fragmentation. Biol. Conserv. 2010, 143, 952–962. [CrossRef] 29. Rompré, G.; Boucher, Y.; Bélanger, L.; Côté, S.; Robinson, W.D. Conservation de la biodiversité dans les paysages forestiers aménagés: Utilisation des seuils critiques d’habitat. For. Chron. 2010, 86, 572–579. [CrossRef] 30. Estavillo, C.; Pardini, R.; da Rocha, P.L.B. Forest Loss and the Biodiversity Threshold: An Evaluation Considering Species Habitat Requirements and the Use of Matrix Habitats. PLoS ONE 2013, 8, e82369. [CrossRef] 31. Rompré, G.; Robinson, W.D.; Desrochers, A.; Angehr, G. Predicting declines in avian species richness under nonrandom patterns of habitat loss in a Neotropical landscape. Ecol. Appl. 2009, 19, 1614–1627. [CrossRef] 31. Rompré, G.; Robinson, W.D.; Desrochers, A.; Angehr, G. Predicting declines in avian species richness under nonrandom patterns of habitat loss in a Neotropical landscape. Ecol. Appl. 2009, 19, 1614–1627. [CrossRef] 32. Eycott, A.E.; Marzano, M.; Watts, K. Filling evidence gaps with expert opinion: The use of Delphi analysis in least-cost modelling of functional connectivity. Landsc. Urban Plan. 2011, 103, 400–409. [CrossRef] p p p pp 32. Eycott, A.E.; Marzano, M.; Watts, K. Filling evidence gaps with expert opinion: The use of Delphi analysis in least-cost modelling of functional connectivity. Landsc. Urban Plan. 2011, 103, 400–409. [CrossRef] 33. Bouchard, M.; Boucher, Y.; Belleau, A.; Boulanger, Y. Modélisation de la Variabilité Naturelle de la Structure d’âge des Forêts du Québec. Gouvernement du Québec, Ministère des Forêts, de la Faune et des Parcs, Direction de la Recherche Forestière: Québec, QC, Canada, 2015; p. 32. 34. Bichet, O.; Dupuch, A.; Hébert, C.; Le Borgne, H.; Fortin, D. Maintaining animal assemblages through single-species management: The case of threatened caribou in boreal forest. Ecol. Appl. 2016, 26, 612–623. [CrossRef] 35. Thiffault, E.; Béchard, A.; Paré, D.; Allen, D. Recovery rate of harvest residues for bioenergy in boreal and temperate forests: A review. Wiley Interdiscip. Rev. Energy Environ. 2014, 4, 429–451. [CrossRef] 36. Senez-Gagnon, F.; Thiffault, E.; Paré, D.; Achim, A.; Bergeron, Y. Dynamics of detrital carbon pools following harvesting of a humid eastern Canadian balsam fir boreal forest. For. Ecol. Manag. 2018, 430, 33–42. [CrossRef] 37. Harvey, B.; Brais, S. Effects of mechanized careful logging on natural regeneration and vegetation competition in the southeastern Canadian boreal forest. Can. J. For. Res. 2002, 32, 653–666. [CrossRef] 38. Boucher, Y.; Perrault-Hébert, M.; Fournier, R.; Drapeau, P.; Auger, I. References Forest naturalness assessment as a component of biodiversity monitoring and conservation management. Forestry 2012, 85, 293–304. [CrossRef] 15. Barrette, M.; Leblanc, M.; Thiffault, N.; Paquette, A.; Lavoie, L.; Bélanger, L.; Bujold, F.; Côté, L.; Lamoureux, J.; Schneider, R.; et al. Issues and solutions for intensive plantation silviculture in a context of ecosystem management. For. Chron. 2014, 90, 748–762. [CrossRef] 16. Angermeier, P.L. The Natural Imperative for Biological Conservation. Conserv. Biol. 2000, 14, 373–381. [CrossRef] 17. Winter, S.; Fischer, H.S.; Fischer, A. Relative Quantitative Reference Approach for Naturalness Assessments of forests. For. Ecol. Manag. 2010, 259, 1624–1632. [CrossRef] 18. Machado, A. An index of naturalness. J. Nat. Conserv. 2004, 12, 95–110. [CrossRef] 19. Angelstam, P.; Boutin, S.; Schmiegelow, F.; Villard, M.-A.; Drapeau, P.; Host, G.; Innes, J.; Isachenko, G.; Kuuluvainen, T.; Mönkkönen, M.; et al. Targets for Boreal Forest Biodiversity Conservation: A Rationale for Macroecological Research and Adaptive Management. Ecol. Bull. 2004, 487–509. 20. Angelstam, P.; Roberge, J.-M.; Axelsson, R.; Elbakidze, M.; Bergman, K.-O.; Dahlberg, A.; Degerman, E.; Eggers, S.; Esseen, P.-A.; Hjältén, J.; et al. Evidence-Based Knowledge Versus Negotiated Indicators for Assessment of Ecological Sustainability: The Swedish Forest Stewardship Council Standard as a Case Study; The Royal Swedish Academy of Sciences: Stockholm, Sweden, 2013; Volume 42, pp. 229–240. Academy of Sciences: Stockholm, Sweden, 2013; Volume 42, pp. 229–240. 1. Brümelis, G.; Jonsson, B.G.; Kouki, J.; Kuuluvainen, T.; Shorohova, E. Forest Naturalness in Northern Eur Perspectives on Processes, Structures and Species Diversity. Silva Fennica 2011, 45, 807–821. [CrossRef] 2. McRoberts, R.E.; Winter, S.; Chirici, G.; LaPoint, E. Assessing Forest Naturalness. For. Sci. 2012, 58, 294– [CrossRef] 23. Landres, P.B.; Morgan, P.; Swanson, F.J. Overview of the use of natural variability concepts in managing ecological systems. Ecol. Appl. 1999, 9, 1179–1188. [CrossRef] 27 of 29 Forests 2019, 10, 325 24. Hagan, J.M.; Whitman, A.A. Biodiversity Indicators for Sustainable Forestry: Simplifying Complexity. J. For. 2006, 104, 203–210. 25. Huggett, A.J. The concept and utility of ecological thresholds in biodiversity conservation. Biol. Conserv. 2005, 124, 301–310. [CrossRef] 26. Dicks, L.V.; Walsh, J.C.; Sutherland, W.J. Organising evidence for environmental management decisions: A ‘4S’ hierarchy. Trends Ecol. Evol. 2014, 29, 607–613. [CrossRef] [PubMed] 27. Jetté, J.-P.; Leblanc, M.; Bouchard, M.; Villeneuve, N. Intégration des Enjeux Écologiques dans les Plans D’aménagement Forestier Intégré, Partie I—Analyse des Enjeux; Gouvernement du Québec, Ministère des Ressources Naturelles, Direction de L’aménagement de de L’environnement Forestiers: Québec, QC, Canada, 2013; p. 150. References Critères et INDIcateurs D’aménagement Durable des Forêts. Available online: http://mffp.gouv.qc.ca/ publications/enligne/forets/criteres-indicateurs/1/e11.asp (accessed on 17 July 2018). 51. Bergeron, Y.; Gauthier, S.; Kafka, V.; Lefort, P.; Lesieur, D. Natural fire frequency for the eastern Canadian boreal forest: Consequences for sustainable forestry. Can. J. For. Res. 2001, 31, 384–391. [CrossRef] 52. Laquerre, S.; Leduc, A.; Harvey, B.D. Augmentation du couvert en peuplier faux-tremble dans les pessières noires du nord-ouest du Québec après coupe totale. Écoscience 2009, 16, 483–491. [CrossRef] 53. Vanha-Majamaa, I.; Lilja, S.; Ryömä, R.; Kotiaho, J.S.; Laaka-Lindberg, S.; Lindberg, H.; Puttonen, P.; Tamminen, P.; Toivanen, T.; Kuuluvainen, T. Rehabilitating boreal forest structure and species composition in Finland through logging, dead wood creation and fire: The EVO experiment. For. Ecol. Manag. 2007, 250, 77–88. [CrossRef] 54. Bonˇcina, A.; Klopˇciˇc, M.; Simonˇciˇc, T.; Dakskobler, I.; Ficko, A.; Rozman, A. A general framework to describe the alteration of natural tree species composition as an indicator of forest naturalness. Ecol. Indic. 2017, 77, 194–204. [CrossRef] 55. Rheault, H.; Bélanger, L.; Grondin, P.; Ouimet, R.; Hébert, C.; Dussault, C. Stand composition and structure as indicators of epixylic diversity in old-growth boreal forests. Écoscience 2009, 16, 183–196. [CrossRef] 56. Uotila, A.; Kouki, J.; Kontkanen, H.; Pulkkinen, P. Assessing the naturalness of boreal forests in eastern Fennoscandia. For. Ecol. Manag. 2002, 161, 257–277. [CrossRef] 57. Meirmans, P.G.; Lamothe, M.; Gros-Louis, M.-C.; Khasa, D.; Périnet, P.; Bousquet, J.; Isabel, N. Complex patterns of hybridization between exotic and native North American poplar species. Am. J. Bot. 2010, 97, 1688–1697. [CrossRef] 58. Carnus, J.-M.; Parrotta, J.; Brockerhoff, E.; Arbez, M.; Jactel, H.; Kremer, A.; Lamb, D.; O’Hara, K.; Walters, B. Planted Forests and Biodiversity. J. For. 2006, 104, 65–77. [CrossRef] 59. Branco, M.; Brockerhoff, E.G.; Castagneyrol, B.; Orazio, C.; Jactel, H. Host range expansion of native insects to exotic trees increases with area of introduction and the presence of congeneric native trees. J. Appl. Ecol. 2015, 52, 69–77. [CrossRef] 0. Ferris, R.; Humphrey, J.W. A review of potential biodiversity indicators for application in British for Forestry 1999, 72, 313–328. [CrossRef] 1. Newton, A.C.; Kapos, V. Biodiversity indicators in national forest inventories. Unasylva 2002, 53, 56–64 62. Bergeron, Y.; Fenton, N.J. Boreal forests of eastern Canada revisited: Old growth, nonfire disturbances, forest succession, and biodiversity. Botany 2012, 90, 509–523. [CrossRef] 63. Franklin, J.F.; Cromack, J.K.; Denison, W.; McKee, A.; Maser, C.; Sedell, J.; Swanson, F.; Juday, G. References Cumulative patterns of logging and fire (1940–2009): Consequences on the structure of the eastern Canadian boreal forest. Landsc. Ecol. 2017, 32, 361–375. [CrossRef] 39. Boucher, Y.; Bouchard, M.; Grondin, P.; Tardif, P. Le Registre des États de Référence: Intégration des Connaissances sur la Structure, la Composition et la Dynamique des Paysages Forestiers Naturels du Québec Méridional. Gouvernement du Québec, Ministère des Ressources Naturelles et de la Faune, Direction de la Recherche Forestière: Québec, QC, Canada, 2011; p. 21. 40. Gossner, M.M.; Schall, P.; Ammer, C.; Ammer, U.; Engel, K.; Schubert, H.; Simon, U.; Utschick, H.; Weisser, W.W. Forest management intensity measures as alternative to stand properties for quantifying effects on biodiversity. Ecosphere 2014, 5, 1–111. [CrossRef] 41. Messier, C.; Tittler, R.; Kneeshaw, D.D.; Gelinas, N.; Paquette, A.; Berninger, K.; Rheault, H.; Meek, P.; Beaulieu, N. TRIAD zoning in Quebec: Experiences and results after 5 years. For. Chron. 2009, 85, 885–896. [CrossRef] 42. Chaudhary, A.; Brooks, T.M. Land Use Intensity-specific Global Characterization Factors to Assess Product Biodiversity Footprints. Environ. Sci. Technol. 2018. [CrossRef] 43. Milà i Canals, L.; Bauer, C.; Depestele, J.; Dubreuil, A.; Freiermuth Knuchel, R.; Gaillard, G.; Michelsen, O.; Müller-Wenk, R.; Rydgren, B. Key Elements in a Framework for Land Use Impact Assessment Within LCA (11 pp). Int. J. Life Cycle Assess. 2007, 12, 5–15. [CrossRef] 28 of 29 Forests 2019, 10, 325 44. BFEC. Manuel de Détermination des Possibilités Forestières 2013–2018. Structure D’âge; Gouvernement du Québec.: Québec, QC, Canada, 2013; pp. 115–122. 45. Environnement Canada. Évaluation Scientifique aux Fins de la Désignation de L’habitat Essentiel de la Population Boréale du Caribou des Bois (Rangifer Tarandus Caribou) au Canada; Environnement Canada: Gatineau, Québec, QC, Canada, 2011; p. 116. 46. Payette, S.; Bhiry, N.; Delwaide, A.; Simard, M. Origin of the lichen woodland at its southern range limit in eastern Canada: The catastrophic impact of insect defoliators and fire on the spruce-moss forest. Can. J. For. Res. 2000, 30, 288–305. [CrossRef] 47. Fahrig, L. Relative effects of habitat loss and fragmentation on population extinction. J. Wildlife Manag. 1997, 61, 603–610. [CrossRef] 48. Loi Concernant la Conservation des Milieux Humides et Hydriques; LRQ No132; Éditeur officiel du Québec: Québec, QC, Canada, 2017; p. 38. 49. Gauthier, S.; Vaillancourt, M.-A.; Leduc, A.; De Grandpré, L.; Kneeshaw, D.; Morin, H.; Drapeau, P.; Bergeron, Y. Aménagement Écosystémique en Forêt Boréale; Presses de l’Université du Québec: Sainte-Foy, QC, Canada, 2008; 600p. 50. MFFP. References Ecological Characteristics of Old-Growth Douglas-Fir Forests; PNW-118; USDA Forest Service: Portland, OR, USA, 1981; p. 48. 64. Bergeron, Y.; Leduc, A.; Harvey, B.D.; Gauthier, S. Natural fire regime: A guide for sustainable management of the Canadian boreal forest. Silva Fennica 2002, 36, 81–95. [CrossRef] 65. Harper, K.A.; Bergeron, Y.; Drapeau, P.; Gauthier, S.; De Grandpré, L. Structural development following fire in black spruce boreal forest. For. Ecol. Manag. 2005, 206, 293–306. [CrossRef] 29 of 29 Forests 2019, 10, 325 29 of 29 66. Lassauce, A.; Paillet, Y.; Jactel, H.; Bouget, C. Deadwood as a surrogate for forest biodiversity: Meta-analysis of correlations between deadwood volume and species richness of saproxylic organisms. Ecol. Indic. 2011, 11, 1027–1039. [CrossRef] 67. Siitonen, J.; Martikainen, P.; Punttila, P.; Rauh, J. Coarse woody debris and stand characteristics in mature managed and old-growth boreal mesic forests in southern Finland. For. Ecol. Manag. 2000, 128, 211–225. [CrossRef] 8. Hicks, R.R. Ecology and Management of Central Hardwood Forests; Wiley & Sons: New York, NY, USA, 199 © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

https://openalex.org/W4327728381

https://jis.tu.edu.iq/index.php/jis/article/download/105/95

Arabic

The DISAGREEMENT of IMAM ABU HANIFAH and His COMPANIONS to The COMPANIONS of IBN MAS'UD, The BOOK of FASTING is a MODEL - COMPRATIVE JURISPRUDENCE

Islamic sciences journal/Mağallaẗ al-ʿulūm al-islāmiyyaẗ

cc-by

The DISAGREEMENT of IMAM ABU HANIFAH and His COMPANIONS to The COMPANIONS of IBN MAS'UD, The BOOK of FASTING is a MODEL - COMPRATIVE JURISPRUDENCE Ammar Ali Kareem * 1 Dr.Yassin Hassan Hamad ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 https://doi.org/10.25130/jis.19.10.4.13 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 https://doi.org/10.25130/jis.19.10.4.13 ISJ ABSTRACT college of Islamic Sciences Tikri University. Praise be to Allah, Lord of the Worlds, and prayers and peace be upon our master Muhammad and his family and companions: After: ISLAMIC SCIENCES JOURNAL (ISJ ISLAMIC SCIENCES JOURNAL (ISJ) The research in Islamic jurisprudence is renewed by the renewal of time and the renewal of facts and events to solve what the society is facing problems that occur over time and the jurisprudence of its flexibility and renewal by dissolving that contract by the diligence of the jurisprudents of the nation. y g j p The importance of this study lies in the statement that the disagreement between Abu Haneefah (may Allaah have mercy on him) and Ibn Mas'ud ( may Allah Pleased with him) was not in the matters of faith, but rather in the matters of diligence. It is consistent with the circle of thought . It also explains the clear face of the Islamic jurisprudence, which is characterized by a variety of opinions and divergent views and a statement that intolerance to doctrinal views is not the approach of our dear imams. The research plan was formed from the introduction: the importance of the topic and the choice of it : Ruling on fasting on the day of doubt, and the second requirement: Ruling on the day of doubt if the sight of the crescent before or after sunset is established. SCIENCES JOURNAL (ISJ ISLAMIC KEY WORDS: Disagreement , Fasting , Imam Abu Hanifah , Ibn Mas'ud , Juristic study. ARTICLE HISTORY: Received: 2/06/2019 Accepted: 25/06/2019 Available online: 0/0/2019 * Corresponding author: E-mail: Kareem6677 @gmail.com المقجمة الحطااااااد ذَ الَااااااسَز مااااااا فساااااا ل و اااااادر فياااااادل, الااااااسز أ ساااااان كاااااا اااااادء مجااااااو ثاااااان ىاااااادل, الااااااسز مااااااا ا نساااااث عمطاااااو البيااااااث, الااااسز عظاااااة ال جااااا ه لاااااو و اااااب مااااان افعااااا ل, صاااااا ئ الطاااان والعلااااااء, الكتيااااا بالمياااااا ة لطاااااان ا جااااااى و ااااااك , والصااااااالة والسااااااالم عمااااااى ساااااايد الثجمااااااين وامااااااام الطعجااااااين الطبعاااااا ث ر طااااااة لمعااااااالطين ساااااايدنا و بيبظااااااا دمحم الظبااااااد ا مااااااد صااااااا ئ الاااااا د ا ليااااااد الطكاااااا م, الطظاااااامه عاااااان ال ظااااااة وا فعاااااا اء, الصااااااا ال عااااااد ا مااااااين, صااااااا ئ الطجااااااام وصاااااااب الشااااااتاعة وصااااااا ئ الحاااااا , ماااااان ا بعااااااو اىعاااااادل وماااااان عصاااااااه لاااااان ظاااااا الطظااااى, وعمااااى آلااااو وصاااابو الليبااااين اللاااااى ين الهاااا َ الطيااااامين وماااان اىعاااادل حيااااد ين وا عتااااى اثاااا ىن وعااااى عمااااى .سظظين بالظ اجس إلى م الد ن وماااان ىااااسا الطظلمااااا الج آنااااد والظباااا ز جااااا ء ت ىااااسه الدراسااااة م التااااات ا مااااام احااااد ظيتااااة وصااااا بيو لمصااااحاحد ال مي احن مسع ًكعاب الص م إنط ذجا- راسة فجيية مجارنة). وماااان ىااااسا الطظلمااااا الج آنااااد والظباااا ز جااااا ء ت ىااااسه الدراسااااة م التااااات ا مااااام احااااد ظيتااااة وصااااا بيو لمصااااحاحد ال مي احن مسع ًكعاب الص م إنط ذجا- راسة فجيية مجارنة). والاااااسز لمااااا عماااااى العظااااا اث ظجااااادح فاااااد ذىظاااااو اث ىظاااااا م التاااااات ل ماااااام احاااااد ظيتاااااة ر طاااااو لمصاااااحاحد ال ميااااا عباااااد احااااان مساااااع , الط التاااااة ىظاااااا اااااأ د بطعظاااااى ا جعياااااا , والط التاااااات العاااااد كاناااااة ىاااااد فاااااد الطسااااا العااااد وجااااد فييااااا ا لااااة اىااااا ذىظااااو ا اااا ل, فااااد الطسااااا ا جعيا ااااة و ااااأث ا مااااا م احاااا د ظيتااااة ر طااااو انااااو م عيااااد والط عيااااد جمااااد وياااا ه فكاااااث عيااااد ويااااأ د فااااد بعااااى إجعيا ا ااااو م التااااات لشااااي و طااااا حاااان ساااااميطاث ر طاااااو , وكااااااث ااااا زَ باااااو اث عياااااد فاااااد بعاااااى الطساااااا العاااااد نااااا فيياااااا العاااااد اااااال احااااان مسع . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 https://doi.org/10.25130/jis.19.10.4.13 مخالفات االمام ابي حنيفة وصاحبيو لمرحابي الجميل ابن مدعهد كتاب الرهم أنمهذجا دراسة فقهية مقارنة عمار عمي كريم و أ.د. ياسين حدن حمج جامعة تكريت كلية العلوم االسالمية فاااااااث البحااااااإل فااااااد التجااااااو ا سااااااالمد مع ااااااد حع ااااااد المماااااااث و ااااااد ال ااااااا وا ااااااداث لحاااااا مااااااا ععاااااا الط عطاااااا ماااااان مشاك حدث بط ور الممن ويأ د التجو لط ونعو و د ه بح مك العجد با جيا من ب فجياء ا مة . ا فاااااااث البحااااااإل فااااااد التجااااااو ا سااااااالمد مع ااااااد حع ااااااد المماااااااث و ااااااد ال ااااااا وا ااااااداث لحاااااا مااااااا ععاااااا الط عطاااااا ماااااان مشاك حدث بط ور الممن ويأ د التجو لط ونعو و د ه بح مك العجد با جيا من ب فجياء ا مة . كطاااان اىطيااااة ىااااسه الدراسااااة فااااد حياااااث اث ااااالس احااااد ظيتااااة ر طااااو حاااان مسااااع لاااان كاااان فااااد مسااااا العجياااادة , وانطااااا ىاااااد فاااااد ا ماااا ر ا جعيا اااااة ف نطاااااا ىااااا عطشااااى اااااطن ا ااااا ة العتكاااا والعااااادح , كطاااااا أنياااااا اااا ال جاااااو الظاصااااا لمتجاااااو ا ساااااالمد الااااااسز عطياااااام حعظاااااا ا ا اااااا اا وا ااااااعالس اصراء واياااااااث أث الععصااااائ لاااااارراء التجييااااااة لااااااي ماااااان مااااااظي أ طعظااااااا الكاااااا ام , ك نااااة لااااة البحااااإل ماااان الطجدمااااة : وفييااااا اىطيااااة الط اااا ا وساااائ ا عياااااره , و العطييااااد : وذكاااا ت ؼيااااو ع يااااف التجااااو, ع يااااف العبااااااا ات, الط التاااااااات :, وملمبااااااااث الطلمااااااائ : ا وا كااااااان صااااااا م ااااااا م الشااااااا ك , والطلمااااااائ : الثااااااااند كااااااان ااااااا م الشاااااااك اذا ثبعة رؤية اليالا ب المواا او بعده نيارا. الكممات:المفتاحية المخالفات ,الص م ,ا مام أح ظيتة , احن مسع , راسة فجيية . الكممات:المفتاحية المخالفات ,الص م ,ا مام أح ظيتة , احن مسع , راسة فجيية . 582 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 1 ) - س رة الظساء : اص ة88 . 5 ) - س رة ص : اص ة52 . المقجمة كطاااان اىطيااااة ىااااسه الدراسااااة فااااد حياااااث اث ااااالس احااااد ظيتااااة ر طااااو حاااان مسااااع لاااان كاااان فااااد مسااااا : العجيدة , وانطا ىد فد ا م ر ا جعيا ة ف نطا ى عطشى طن ا اااااا ة ال ااااااا م والعتكاااااا والعاااااادح فااااااد آ ااااااات و لااااااة ذلااااااك لااااااو عااااااالى : ﭽ ﮊ ﮋ ﮌ ﮍ ﮎ ﮏ ﮐ ﮑ ﮒ ﮓ ﮔ ﮕ ﮖ ﮗ ﮘ ﮙ ﮚ ﮛ ﮜ ﮝ ﮞ ﮟ ﮠ ﮡ ﮢ ﮣ ﮤ ﮥ ﮦ ﮧ ﮨ ﮩ ﮪ ﭼ 1 ) , : و لو عالى ﭽ ﭲ ﭳ ﭴ ﭵ ﭶ ﭷ ﭸ ﭹ ﭺ ﭻ ﭼ( 7 ). وإ ااااافة إلااااى ذلاااااك فاااااث البحااااإل فاااااد التجااااو ا سااااالمد مع اااااد حع ااااد المماااااث و اااااد ال ااااا وا اااااداث لحاااا مااااا ععاااا الط عطاااا ماااان مشاااااك حاااادث بطاااا ور الااااممن ويااااأ د التجااااو لط ونعااااو و ااااد ه بحاااا مااااك العجااااد . با جيا من ب فجياء ا مة فقيطااااة التجااااو ا سااااالمد ااااسوب و ظيااااد و عجااااا م حعجااااا م المماااااث ف نيااااا ا اااا ة فااااد اذىاااااث التجياااااء . وطالب العمن 5 ) - س رة ص : اص ة52 . 782 1 ) - : ظ :م آة ا ص ا44 / 1 ،) 5 ) – : ظ : الع ي لطعن العظقي11 / 1 .) . 8 ) - : س رة ى اص ة21 . 4 ) - : س رة الظساء اص ة88 . 4 ) - : س رة الظساء اص ة88 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 وكاااااسلك اااااي ال جاااااو الظاصااااا لمتجاااااو ا ساااااالمد حعظااااا ا ا ااااا اا وا اااااعالس اصراء وايااااااث أث الععصااااائ لرراء التجيية . لي من مظي أ طعظا الك ام واياااااااااث اث م افجااااااااة الطااااااااسىئ الحظتااااااااد لطع اااااااان آراء احاااااااان مسااااااااع عظااااااااد العجميااااااااد حاااااااا وافجاااااااا ه اجعيا ا حدلي انين الت ه أ ضا فد مسا كطا مبين فد ىس ا البحإل . : خطة البحث وكاااااسلك اااااي ال جاااااو الظاصااااا لمتجاااااو ا ساااااالمد حعظااااا ا ا ااااا اا وا اااااعالس اصراء وايااااااث أث الععصااااائ لرراء التجيية . لي من مظي أ طعظا الك ام واياااااااااث اث م افجااااااااة الطااااااااسىئ الحظتااااااااد لطع اااااااان آراء احاااااااان مسااااااااع عظااااااااد العجميااااااااد حاااااااا وافجاااااااا ه اجعيا ا حدلي انين الت ه أ ضا فد مسا كطا مبين فد ىس ا البحإل . : خطة البحث ا عضة طبيعة البحإل اث أ سطو إلى مجدمة و طييد وملمبين , فكانة عمى:الظح ا د الطجدمة : وفييا اىطية الط ا. وسئ ا عياره العطييد ع يف التجو, ع يف العبا ات, الط التات. ع يف التجو, ع يف العبا ات, الط التات. الطلمئ : ا وا كن ص م م الشك الطلمئ : الثاند كن م الشك اذا ثبعة رؤية اليالا ب المواا او بعده نيارا تم هيج باااا الاااااد ا فاااااد فصااااا ا ىااااسه ال ساااااالة الطع ا اااااعة ااااا ز حاااااد اث ا ظاااااوا فاااااد ىاااااسا العطيياااااد ع ياااااف بعى ا لتاظ ال ار ة فد عظ اث ال سالة . باااا الاااااد ا فاااااد فصااااا ا ىااااسه ال ساااااالة الطع ا اااااعة ااااا ز حاااااد اث ا ظاااااوا فاااااد ىاااااسا العطيياااااد ع ياااااف بعى ا لتاظ ال ار ة فد عظ اث ال سالة . باااا الاااااد ا فاااااد فصااااا ا ىااااسه ال ساااااالة الطع ا اااااعة ااااا ز حاااااد اث ا ظاااااوا فاااااد ىاااااسا العطيياااااد ع ياااااف بعى ا لتاظ ال ار ة فد عظ اث ال سالة . - : تعريف الفقو في المغة واالصطالح التجاااااو فاااااد المهاااااة : التيااااان 1 ) َ, جااااااا: ف ا جااااا و تجَااااا و َكع َمااااان عمَااااا ن ، أز فيااااان ملمجااااااً، سااااا اء أكااااااث التيااااان قيجااااااً أم ساااااالحياً، ويجاااااااا: فجااااااو تجااااااو مثاااااا كاااااا م كاااااا م، أز صااااااار التجااااااو لااااااو ساااااا ية. ويجاااااااا: تجااااااو ال جاااااا تجياااااااً: أز عاطى التجو 5 ). ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 : تعريف الفقو في المغة واالصطالح التجاااااو فاااااد المهاااااة : التيااااان 1 ) َ, جااااااا: ف ا جااااا و تجَااااا و َكع َمااااان عمَااااا ن ، أز فيااااان ملمجااااااً، سااااا اء أكااااااث التيااااان قيجااااااً أم ساااااالحياً، ويجاااااااا: فجااااااو تجااااااو مثاااااا كاااااا م كاااااا م، أز صااااااار التجااااااو لااااااو ساااااا ية. ويجاااااااا: تجااااااو ال جاااااا تجياااااااً: أز عاطى التجو 5 ). اا التجاااااو فاااااد المهاااااة : التيااااان 1 ) َ, جااااااا: ف ا جااااا و تجَااااا و َكع َمااااان عمَااااا ن ، أز فيااااان ملمجااااااً، سااااا اء أكااااااث التيااااان قيجااااااً أم ساااااالحياً، ويجاااااااا: فجااااااو تجااااااو مثاااااا كاااااا م كاااااا م، أز صااااااار التجااااااو لااااااو ساااااا ية. ويجاااااااا: تجااااااو ال جاااااا تجياااااااً: أز عاطى التجو 5 ). و د جاء لتظ التجو بطعظى التين فد الج آ :ث الك ين والسظة الظب ية فطاااااااااان الجاااااااااا آ:ث الكاااااااااا ين لااااااااااو عااااااااااالى ﭽ ﭴ ﭵ ﭶ ﭷ ﭸ ﭹ ﭺ ﭻ ﭼ ﭿﭽ ﭾ ﮀ ﮁ ﮂ ﮃ ﮄ ﮅ ﮆ ﮇ ﮈ ﭼ 8 ). ﭼ :و لاااااااااااااااااااااااااااااو عاااااااااااااااااااااااااااااالى ﭽ ﯜ ﯝ ﯞ ﯟ ﯠ ﯡ ﯢ ﯣ ﯤ ﯥ ﯦ ﯧ ﯨ ﯩ ﯪ ﯫ ﯬ ﯭ ﯮ ﯯ ﯰ ﯱ ﯲ ﯳ ﯴ ﯵ ﯶ ﯷ ﯸ ﯹ ﯺ ﯻ ﯼ ﯽ ﯾ ﯿ ﰀ ﰁ ﰂ ﰃ ﰄ ﭼ 4 ) . 4 ) - : س رة الظساء اص ة88 . 788 : و لاااااااااااااااااااااااااااااو عاااااااااااااااااااااااااااااالى ﯦﭽ ﯧ ﯨ ﯩ ﯪ ﯫ ﯬ ﯭ ﯮ ﯯ ﯰ ﯱ ﯲ ﯳ ﯴ ﯵ ﯶ ﯷ ﯸ ﯹ ﯺ ﯻ ﯼ ﯽ ﯾﭼ 1 ). أ ب- الجااااا ا الثااااااند : جااااا از صَ ااااا و م َااااا و مث الشت اااااك ث حثظثيتاااااةث العتلَااااا ُّاث فاااااد التااااا والظتااااا ، وااااااو ااااااا عَلااااااء وَعطااااا حااااان عباااااد الوعَمثيااااام ، واحااااا ظيتاااااة ، ودمحم حااااان الحسااااان الشااااايباند 2 ) , والمتيواااااإل وَا وَووزَاعثاااااد 1 ) ، وَإثسو اااااحَا ر طيااااان 5 ) . َوَى اااا َ أَنتااااو حوعَطَاااا أَثو َك اااا ثَ ىَااااسَا الويَاااا و م مثاااانو رَمَضَ اااااثَ وَي حوعَطَاااا أَثو َك اااا ثَ مثاااانو َ ااااعوبَاثَ ، فَمَاااا و صَ ااااام َلَاااادَار الصت ااااا و م حَااااايونَ أَثو َك ااااا ثَ مثااااانو رَمَضَ ااااااثَ ، وَاَااااايونَ أَثو َ َك ااااا ثَ مثااااانو َ اااااعوبَاثَ ، وَلَااااا و أَفولَااااا َ لَااااادَارَ الوتثلوااااا حَااااايونَ أَثو َك ااااا ث َفثد رَمَضَ اثَ وَاَيونَ أَثو َك ثَ فثد َ عوبَاث 8 ). 1 - ا ااااااا اا التجيااااااااء : ا تاااااااا التجيااااااااء ر طيااااااان عماااااااى جااااااا از الظتااااااا بااااااا رمضااااااااث لطااااااان اععاااااااا صااااااايام ااااااا م الثالثين من عباث ل عا اذا ا ع ث بالظية ولن شيد ح ؤيعو وعدم اععباره من رمضاث 4 ) . : ا انين ا عمت ا فد ص م م الشك عمى طسة ا اا 1 - ا ااااااا اا التجيااااااااء : ا تاااااااا التجيااااااااء ر طيااااااان عماااااااى جااااااا از الظتااااااا بااااااا رمضااااااااث لطااااااان اععاااااااا صااااااايام ااااااا م الثالثين من عباث ل عا اذا ا ع ث بالظية ولن شيد ح ؤيعو وعدم اععباره من رمضاث 4 ) . : ا انين ا عمت ا فد ص م م الشك عمى طسة ا اا أ- الج ا ا وا : مة صيام الويَ و م التسثز شَ كُّ ؼثيوث مثنو رَمَضَ اثَ واو اا احن مسع 2 ). َم َ افج ْت ال ك َ و أََ سَوتَ ةَ َ ْت و َو َت ا ْت و َو أ- الج ا ا وا : مة صيام الويَ و م التسثز شَ كُّ ؼثيوث مثنو رَمَضَ اثَ واو اا احن مسع 2 ). ووافجاااااو ع طَااااا َ ، وَعَمثاااااد ، وَعَبوااااادث ْتث حوااااانث ََبتااااااس ، وَعَبوااااادث ْتث حوااااانث ع طَااااا َ ، وَ سَ وتَاااااةَ، وَأَنَااااا ث حوااااانث مَالثاااااك رَ ث ااااادَ ْت عَظوي نو 2 ) , واح س 8 ) , وا مام مالك ر طو 8 ). 1 ) - س رة الع اة : اص ة155 . : 5 )- عطدة الجارز ح صحي الب ارز11 / 582 .) 8 )- حدا الصظا فد يئ الش ا ط الكعاب الع اد5 / 88 . ) 4 )- م اىئ ال مي فد ح م عص مي5 / 828 ) , حدا الصظا فد يئ الش ا5 / 88 , ) الحاوز الكبي8 / 412 ) , :مسا ا مام أ طد روا ة احظو عبد ص188 . ) 2 )- السظن الصه ل لمبييجد5 / 82 . ) 2 )- السظن الصه ل لمبييجد نس ة ا ع طد8 / 521 . ) 8 ) - حدا الصظا فد يئ الش ا5 / 88 .) 8 ) - فجو العبا ات عمى الطسىئ الطالكد1 / 812 . ) 2 )- حدا الصظا فد يئ الش ا5 / 88 ) : و لاااااااااااااااااااااااااااااو عاااااااااااااااااااااااااااااالى ﯦﭽ ﯧ ﯨ ﯩ ﯪ ﯫ ﯬ ﯭ ﯮ ﯯ ﯰ ﯱ ﯲ ﯳ ﯴ ﯵ ﯶ ﯷ ﯸ ﯹ ﯺ ﯻ ﯼ ﯽ ﯾﭼ 1 ). أ : و لاااااااااااااااااااااااااااااو عاااااااااااااااااااااااااااااالى ﯦﭽ ﯧ ﯨ ﯩ ﯪ ﯫ ﯬ ﯭ ﯮ ﯯ ﯰ ﯱ ﯲ ﯳ ﯴ ﯵ ﯶ ﯷ ﯸ ﯹ ﯺ ﯻ ﯼ ﯽ ﯾﭼ 1 ). ﯾﭼ 1 ). ا لمطمب األول حكم صهم يهم الذك ﯾﭼ 1 ). ا لمطمب األول حكم صهم يهم الذك : ااا م الشااااك ىااا الويَاااا و م التااااسثز َعَحَااادتث الظتاااااس ؼثيااااوث حث ؤويَاااةث الويثااااالَ اث وَلَااانو َثوب ااااةو ر ؤويَع ااااو أَو َ ااايثدَ وَا ثااااد فَاااا تتو َ يَا َ و أَوو َ اىثدَاثث فَاسث جَاثث فَ تتو َ يَا َ ي طَا 5 ) . : ااا م الشااااك ىااا الويَاااا و م التااااسثز َعَحَااادتث الظتاااااس ؼثيااااوث حث ؤويَاااةث الويثااااالَ اث وَلَااانو َثوب ااااةو ر ؤويَع ااااو أَو َ ااايثدَ وَا ثااااد فَاااا تتو َ يَا َ و أَوو َ اىثدَاثث فَاسث جَاثث فَ تتو َ يَا َ ي طَا 5 ) . َوَى اااا َ أَنتااااو حوعَطَاااا أَثو َك اااا ثَ ىَااااسَا الويَاااا و م مثاااانو رَمَضَ اااااثَ وَي حوعَطَاااا أَثو َك اااا ثَ مثاااانو َ ااااعوبَاثَ ، فَمَاااا و صَ ااااام َلَاااادَار الصت ااااا و م حَااااايونَ أَثو َك ااااا ثَ مثااااانو رَمَضَ ااااااثَ ، وَاَااااايونَ أَثو َ َك ااااا ثَ مثااااانو َ اااااعوبَاثَ ، وَلَااااا و أَفولَااااا َ لَااااادَارَ الوتثلوااااا حَااااايونَ أَثو َك ااااا ث َفثد رَمَضَ اثَ وَاَيونَ أَثو َك ثَ فثد َ عوبَاث 8 ). : و تصي الطسألة ى ا د 1 - ا ااااااا اا التجيااااااااء : ا تاااااااا التجيااااااااء ر طيااااااان عماااااااى جااااااا از الظتااااااا بااااااا رمضااااااااث لطااااااان اععاااااااا صااااااايام ااااااا م الثالثين من عباث ل عا اذا ا ع ث بالظية ولن شيد ح ؤيعو وعدم اععباره من رمضاث 4 ) . : ا انين ا عمت ا فد ص م م الشك عمى طسة ا اا أ- الج ا ا وا : مة صيام الويَ و م التسثز شَ كُّ ؼثيوث مثنو رَمَضَ اثَ واو اا احن مسع 2 ). ووافجاااااو ع طَااااا َ ، وَعَمثاااااد ، وَعَبوااااادث ْتث حوااااانث ََبتااااااس ، وَعَبوااااادث ْتث حوااااانث ع طَااااا َ ، وَ سَ وتَاااااةَ، وَأَنَااااا ث حوااااانث مَالثاااااك رَ ث ااااادَ ْت عَظوي نو 2 ) , واح س 8 ) , وا مام مالك ر طو 8 ). : و لاااااااااااااااااااااااااااااو عاااااااااااااااااااااااااااااالى ﯦﭽ ﯧ ﯨ ﯩ ﯪ ﯫ ﯬ ﯭ ﯮ ﯯ ﯰ ﯱ ﯲ ﯳ ﯴ ﯵ ﯶ ﯷ ﯸ ﯹ ﯺ ﯻ ﯼ ﯽ ﯾﭼ 1 ). أ ب- الجااااا ا الثااااااند : جااااا از صَ ااااا و م َااااا و مث الشت اااااك ث حثظثيتاااااةث العتلَااااا ُّاث فاااااد التااااا والظتااااا ، وااااااو ااااااا عَلااااااء وَعطااااا حااااان عباااااد الوعَمثيااااام ، واحااااا ظيتاااااة ، ودمحم حااااان الحسااااان الشااااايباند 2 ) , والمتيواااااإل وَا وَووزَاعثاااااد 1 ) ، وَإثسو اااااحَا ر طيااااان 5 ) . 782 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ج-الجاااا ا الثالااااإل : َ ثاااائ َعويثااااين الظ ثيتااااةث فثااااد ك اااا ث صَ اااا و م وَاجثاااائ ، وَى اااا َ أَثو َعوعَجثاااادَ أَنتااااو َص اااا ، َ م وَاااادًا مثاااانو رَمَضَ اااااث أَوو مثنو َضَ ا ثوث، أَوو مثنو كَتتارَ ثوث، أَوو نَسورثهث ، واو اا ا مام ا طد ر طو 8 ). ولو ثالث روا ات : ا 1 - َفَ ثذَا لَنو َح و وثَ مَظو َ ثهث سَ حَاب أَوو َع4 لن م الص م 2 ). 5 - إثذَا َااَ وثَ مَظو َ ثهث سَ حَاب أَوو َعَ صَ ام 2 ). 8 - َ َ ثئ صَ و م و ، وَ َ ومث و عَنو رَمَضَ اثَ إثو صَ امَو 8 ). - الجاااا ا ال اباااا : اااا ز أَث صااااام آ اااا َاااا و م ماااان ااااعوبَاث َلَ عااااا إ ت أَث َافااااا صاااا ما كَاااااثَ َص اااا مو وىاااا ا الشافعد ر طو 8 ). 1 - َفَ ثذَا لَنو َح و وثَ مَظو َ ثهث سَ حَاب أوو َع4 لن م الص م 2 ). 5 - إثذَا َااَ وثَ مَظو َ ثهث سَ حَاب أَوو َعَ صَ ام 2 ). 8 - َ َ ثئ صَ و م و ، وَ َ ومث و عَنو رَمَضَ اثَ إثو صَ امَو 8 ). - الجاااا ا ال اباااا : اااا ز أَث صااااام آ اااا َاااا و م ماااان ااااعوبَاث َلَ عااااا إ ت أَث َافااااا صاااا ما كَاااااثَ َص اااا مو وىاااا ا الشافعد ر طو 8 ). ََ اَ َ ا َإثو أَفولَ َ أَفولَ َ و َ اا و ََ َ ا يو إثو ف : أث الظ ا الج ا ال ا د و- َالجااااا ا ال اااااام : أث الظااااااس فاااااد صااااا مو َبَااااا ثثمَاااااامثيثنو إثثو صَ اااااامَ صَ اااااام ه ، وَإثثو أَفولَااااا َ أَفولَااااا وه وَاثاااااوث َااااااا الوحَسَ ن وَاحون سث ي ثينَ وروا ة عن ا طد ر طو 2 ). ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 او- ا لة اصحاب الج ا ا وا : اسعدا اصحاب ىسا الج ا بالسظة وافعاا الصحابة : اسعد لين بالسظة عَنو عطار حن اس ة - " أَنتو َااَ: "مَنو صَ امَ َ و مَ الشت ك ث فَجَدو عَصَ ى أَبَا الوجَاسث نث 11 ) ) 1 : وجو الد لة عَنو عطار حن اس ة - " أَنتو َااَ: "مَنو صَ امَ َ و مَ الشت ك ث فَجَدو عَصَ ى أَبَا الوجَاسث نث 11 ) ) 1 : وجو الد لة كااااا ه صااااا م ااااا م الشاااااك لعًاااااا أث كااااا ث مااااان رمضااااااث أو عماااااى وجاااااو الط اعااااااة فاااااا أث كااااا ث مااااان رمضااااااث، فيمحاااااا باااااالت ماااااا لاااااي مااااان جظساااااو، فأماااااا إذا أ مااااا الظياااااة لمعلااااا ا، فمااااان حصااااا ؼياااااو معظاااااى الشك، ف نطا نيعو أنو من عباث، في كطا ص مو عن نسر أو ضاء رمضاث 11 ). : ر عميين بي زو/ ) 2 )- الططع فد ح الطجظ ت احن هيش ط8 5 / 8 . ) 11 )– صحي الب ارز ، كعاب الص م ، بَاب َ و اث الظتبثد ث : "إثذَا رَأَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ر ن8 / 58 . 11 )- ح صحي الب ارز حن بلاا4 / 85 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 فَاااااااالوط َا مثظواااااااو صَ ااااااا و م َااااااا و مث الشت اااااااك ث عَااااااانو رَمَضَ ااااااااثَ ثَثت الوطَااااااا و وثزت أَثت الظتبثااااااادت-  َوَسَ ااااااامتن- "نَيَاااااااى عَااااااانو " صَ ااااا و مث َااااا و مث الشت اااااك ث عَااااانو رَمَضَ ااااااثَ وَ َاااااااَ: مَااااانو صَ اااااامَ َااااا و مَ الشت اااااك ث فَجَااااادو عَصَ اااااى أَبَاااااا الوجَاسث ااااانث 1 ) أَزو : صَ اااااامَ عَااااانو َرَمَضَ اث 5 ) َمعظاااااى الحَااااادث إل " َ َسو اااااعَجوبثم ا رَمَضَ ااااااثَ بثصَ ااااا و مث َااااا و م مثااااانو َ اااااعوبَاث " 8 ), ،عماااااى نثيتاااااة ا ث واااااعثالَ ط ل مضااااااث َ حااااس ا مثطتااااا صااااظعة الظتصَ ااااارَل فثااااد الم ثيَااااا َة عمااااى مَااااا افوعاااا عَمَاااايويثن حاااا أ ين الوتَاسث ااااد، فَكَاااااث َااااأوم بط التااااة أىااااااا الوكعااااااااب وَكَااااااااثَ أَو حااااااائ م َافجَاااااااة أىااااااا الوكعااااااااب ؼثيطَاااااااا لااااااان ااااااا م ؼثياااااااوث بثشَ ااااااادو ء، ثااااااانت أَمااااااا بعاااااااد ذَلثاااااااك بط التعين 4 ). َمعظاااااى الحَااااادث إل " َ َسو اااااعَجوبثم ا رَمَضَ ااااااثَ بثصَ ااااا و مث َااااا و م مثااااانو َ اااااعوبَاث " 8 ), ،عماااااى نثيتاااااة ا ث واااااعثالَ ط ل مضااااااث َ حااااس ا مثطتااااا صااااظعة الظتصَ ااااارَل فثااااد الم ثيَااااا َة عمااااى مَااااا افوعاااا عَمَاااايويثن حاااا أ ين الوتَاسث ااااد، فَكَاااااث َااااأوم بط التااااة أىااااااا الوكعااااااااب وَكَااااااااثَ أَو حااااااائ م َافجَاااااااة أىااااااا الوكعااااااااب ؼثيطَاااااااا لااااااان ااااااا م ؼثياااااااوث بثشَ ااااااادو ء، ثااااااانت أَمااااااا بعاااااااد ذَلثاااااااك بط التعين 4 ). فَ ثث ي : ىَسَا الظتيود لمعتحو ثين أَو لمعظميو؟ ر عميين : لمكَ َاىَة 2 ). ا لة اصحاب الج ا الثاند : اسعدل ا بالسظة و: أفعاا الصحابة اا الحااااااد إل ا وا- عَاااااان أحااااااد ى َيواااااا َة َااااااااَ: َااااااااَ رَس اااااا ا : " َ جاااااادم ا الشت اااااايو حثيَاااااا و م وَ َ حياااااا مين إ ت " أَث َافا ذَلثك ص ما وَكَاثَ َص مو أ دك ن، ص م ا ل ؤويَعو وأفل وا ل ؤويَعو 2 ). وجاااااو: الد لاااااة لكاااااد َ واااااعَمْ صَ ااااا و م الوتَااااا و بثصَ ااااا و م نتااااا بماااااو وَ َ بعاااااده، حاااااس ا مطتاااااا صاااااظعة الظتصَ اااااارَل فثد الم يَا َة عمى مَا افوع عَمَيوين ح أ ين الوتَاسث د 8 ) . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 اا الحااااااد إل ا وا- عَاااااان أحااااااد ى َيواااااا َة َااااااااَ: َااااااااَ رَس اااااا ا : " َ جاااااادم ا الشت اااااايو حثيَاااااا و م وَ َ حياااااا مين إ ت " أَث َافا ذَلثك ص ما وَكَاثَ َص مو أ دك ن، ص م ا ل ؤويَعو وأفل وا ل ؤويَعو 2 ). وجاااااو: الد لاااااة لكاااااد َ واااااعَمْ صَ ااااا و م الوتَااااا و بثصَ ااااا و م نتااااا بماااااو وَ َ بعاااااده، حاااااس ا مطتاااااا صاااااظعة الظتصَ اااااارَل فد الم يَا َة عمى مَا افوع عَمَيوين ح أ ين الوتَاسد 8 ) . ث ثن ن ى ثد مثي الحااااااد إل الثاااااااند- مااااااا رَ وَزو عاااااان الظبااااااد أَنتااااااو اااااااا " َ صَ ااااااام الويَاااااا و م الااااااسز شَ ااااااكُّ ؼيااااااو ماااااان رَمَضَ اااااااثَ إ ت " َلَ ُّعًااااا 8 ) ،اسو ااااعَثوظَى العتلَاااا ُّاَ وَالوط سو ااااعَثوظَى َااااالث كوط ااااو كواااانَ الوط سو ااااعَثوظَى مظااااو وَأَمتااااا الوحَاااادث إل فَااااالوط َا مظااااو صَ اااا و م َاااا و مث الشت ااااك ث عاااان رَمَضَ اااااثَ ثَثت َالوطَاااا و وثزت أَثت الظبااااد نيااااى عاااان صَ اااا و مث َاااا و مث الشت ااااك ث عاااان رَمَضَ اااااثَ و اااااا" ماااان صَ ااااام م الشت ك ث فَجَدو عَصَ ى أَبَا الوجَاسث نث " 2 ). 1 )- صحي الب ارز ، كعاب الص م ، بَاب َ و اث الظتبثد ث َ"إثذَا ر : أَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ر ن8 / 58 . 5 )– ظ حدا الصظا فد يئ الش ا ط الكعاب الع اد5 / 88 . 8 )– مسظد أحد او الليالسد4 / 822) ، َدث إل َسَ ن صَ حثي . وَرَ وَاه احون مَيوطَةَ، وَاحون باث فد "صح يحيطا"، وَرَ وَاه أَح َاو اللتيَالثسث دُّ فثد " ، وَ َ َسو عَجوبثم ا رَمَضَ اثَ بثصَ ومث َ و م مثنو َ عوبَاثَ ". َااَ: وَاثالو طومَةث فَيَسَا الوحَدث َ إل نَ فثد الوطَسو أَلَةث، وَى َ صَ حثي كَطَا َاا الع ث ومثسثزُّ ، وَسث طَا ، وَثتجَو أَح َا ثن ، وَاحون مَعثين ظ نصئ ال ا ة5 / 488 . ) 1 )- صحي الب ارز ، كعاب الص م ، بَاب َ و اث الظتبثد ث َ"إثذَا ر : أَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ر ن8 / 58 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 وجو الد لة : وَمَعوظَى الوحَدث إلث َ َص م ا رَمَضَ اثَ فثد وَيو ث أوانو 1 ). الحااااد إل الثالااااإل- مااااا رَ وَزو عاااان الظبااااد ، َااااااَ ل جاااا : " ىاااا صااااطة ماااان ساااا ر 5 ), ااااعباث َااااااَ: فَاااا ثذا " فَ ثذا أفل ت فَصن َ و مَيونث 8 ). و ال لة وجو الد لة : وَمَعوظَى الوحَدث إلث َ َص م ا رَمَضَ اثَ فثد وَيو ث أوانو 1 ). الحااااد إل الثالااااإل- مااااا رَ وَزو عاااان الظبااااد ، َااااااَ ل جاااا : " ىاااا صااااطة ماااان ساااا ر 5 ), ااااعباث َااااااَ: فَاااا ثذا " فَ ثذا أفل ت فَصن َ و مَيونث 8 ). :وجو الد لة مَااااا صااااطة ساااا ر ىَااااسَا الشت اااايو التااااسثز ى اااا َ ااااعوبَاث , إنطااااا ىاااا اااادء كاااااث ال جاااا ااااد أوجبااااو عمااااى نتسااااو حظااااسره فااااأم ه بال فاااااء باااااو أو كاااااث ذلااااك عاااااا ة ااااد اععا ىااااا فاااااد صاااايام أوا اااا الشاااااي ر فع كااااو سااااعقباا الشاااااي فاسااااعحئ َ أث جضاااايو , وَي حوطَاااا ىَااااسَا عَمَااااى أَثت ال تج اااا َ كَانَااااةو لَااااو عَااااا َة بثصث اااايَامث سَ اااا رث الشت اااايو ث أَوو َاااادو نَااااسَرَه , فأ ه الظبد عمى ذلك 4 ) . الحااااد إل ال اباااا- مااااا روز عاااان أم سَ ااااامطَة رَ ث ااااد َعَااااالَى عَظويَااااا أَنااااو لااااان كاااان َص اااا م ماااان الس ااااظة اااااي ا كَامثال إ ت عوبَاث صمو ح مضاث 2 ). إنااااو كاااااث صااااا م ااااعباث كمااااو لكااااان ااااد جاااااء مااااان ااااد إل عا شااااة أ ضااااااً مااااا اااادا عماااااى الفااااو، فماااااسلك طاااا عمااااى أنااااو كاااااث صاااا م والبااااو فكأنااااو صاااا م كمااااو وإنااااو صاااامو ح مضاااااث , و ااااد إل العجاااادم حياااا م أو اااا مين لطن ح عاط حمعطو ل مضاث ملما العل ا، و أعمن 2 ). ثانيا- : اسعد لين بأفعاا الصحابة ثانيا- : اسعد لين بأفعاا الصحابة ماااااااا روز عااااااان عَا ثشَ اااااااة وَأَسو اااااااطَاء احوظَعَاااااااا أَحثاااااااد بَكوااااااا ر اااااااد عظيطاااااااا َص ااااااا مَاثث َااااااا و مَ الشت اااااااك ث ، وَكَانَاااااااةو َعَا ثشَ ة ، َج ا : ثَثو أَص مَ َ و مًا مثنو َ عوبَاثَ أَ َئُّ إثلَدت مَنو أَثو أ فولث َ َ و مًا مثنو رَمَضَ اث 8 ). : وجو الد لة ث أ ع اااااا فااااااد صاااااا م رمضاااااااث حياااااا م أ اااااائ إلااااااى ماااااان أث أ ااااااأ ، نااااااد إذا ع مااااااة لاااااان تعظااااااد وإذا " أ ت فا ظد 8 ). ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 5 )– ظ حدا الصظا فد يئ الش ا ط الكعاب الع اد5 / 88 . 8 )– مسظد أحد او الليالسد4 / 822) ، َدث إل َسَ ن صَ حثي . وَرَ وَاه احون مَيوطَةَ، وَاحون باث فد "صح يحيطا"، وَرَ وَاه أَح َاو اللتيَالثسث دُّ فثد " ، وَ َ َسو عَجوبثم ا رَمَضَ اثَ بثصَ ومث َ و م مثنو َ عوبَاثَ ". َااَ: وَاثالو طومَةث فَيَسَا الوحَدث َ إل نَ فثد الوطَسو أَلَةث، وَى َ صَ حثي كَطَا َاا الع ث ومثسثزُّ ، وَسث طَا ، وَثتجَو أَح َا ثن، وَاحون مَعثين ظ نصئ ال ا ة5 / 488 . ) ثُ ث ز ن) 4 )- عطدة الجارز ح صحي الب ارز11 / 588 .) 2 )- ظ : الطصدر نتسو. 2 )- صحي الب ارز1 / 228 ) ، كعاب الص م ، باب: "إثذَا رَأَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ال ن: 1212 . 8 ) - ظ : عطدة الجارز ح صحي الب ارز11 / 585 , 588 . ) 8 )– سظن الدار لظد8 / 111 :), كعاب الصيام, ال ن5121 , إسو ظَا ه صَ حثي م عتصث , ظ :نصئ ال ا ة5 / 442 ) 2 )- صحي الب ارز ، كعاب الص م ، بَاب َ و اث الظتبثد ث : "إثذَا رَأَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ر ن8 / 58 . 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 1 )- نصئ ال ا ة5 / 441 . 5 )- وس ر الشت يو آ ث ه، سطد حسلك سععار الوجَطَ ؼثيوث , ظ : عطدة الجارز ح صحي الب ارز11 / 588 . ) 8 )– صحي مسمن5 / 851 , ) : كعاب الصيام , باب صيام س ر عباث , ال ن511 . 4 )- ظ : عطدة الجارز ح صحي الب ارز11 / 115 ) , ظ : ني ا وطار4 / 818 ) . 2 )– سظن أحد او5 / 811 َ) , كعاب الص م , باب ؼثيطَنو َصث َ عوبَاثَ حث َمَضَ اث ,ال ن : 5882 , رواه الظسا د فد الص م من من الكب ل كطا فد حتة ا اس18 / 81 ) ورواه أ طد2 / 854 ) واحن ميطة5128 ) واحن باث241 م ار ) عن احن ميطة، والحاكن1 / 482 ) وعظو البييجد4 / 818 ) وصححو ووافجو السىبد. ظ :ا كام ال سلى5 / 541 . ) 2 ) - : ظل م عاة الطتا ي ح مشكاة الطصاحي2 / 448 ) البدر العطام ح حم غ الط ام2 / 155 .) .و د إل العجدم حي م أو مين لطن حعاط حمعطو ل مضاث ملما العل ا، و أعمن 8 )– مسظد أ طد ط ال سالة41 / 451 ) , م سو ظَد الص ث د ث جَةث عَا ثشَ ةَ حثظوةث الص ث د ث اث رَ ث دَ عَظويَا , ؼيوث انوجثلَاا ، ظ : العم ي الحبي5 / 415 . ) 5 )- وس ر الشت يو آ ث ه، سطد حسلك سععار الوجَطَ ؼثيوث , ظ : عطدة الجارز ح صحي الب ارز11 / 588 ) 8 )– صحي مسمن5 / 851 , ) : كعاب الصيام , باب صيام س ر عباث , ال ن511 . 4 )- ظ : عطدة الجارز ح صحي الب ارز11 / 115 ) , ظ : ني ا وطار4 / 818 ) . 2 )– سظن أحد او5 / 811 َ) , كعاب الص م , باب ؼثيطَنو َصث َ عوبَاثَ حث َمَضَ اث ,ال ن : 5882 , رواه الظسا د فد الص م من من الكب ل كطا فد حتة ا اس18 / 81 ) ورواه أ طد2 / 854 ) واحن ميطة5128 ) واحن باث241 م ار ) عن احن ميطة، والحاكن1 / 482 ) وعظو البييجد4 / 818 ) وصححو ووافجو السىبد. ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ظ :ا كام ال سلى5 / 541 . ) .و د إل العجدم حي م أو مين لطن حعاط حمعطو ل مضاث ملما العل ا، و أعمن 8 )– مسظد أ طد ط ال سالة41 / 451 ) , م سو ظَد الص ث د ث جَةث عَا ثشَ ةَ حثظوةث الص ث د ث اث رَ ث دَ عَظويَا , ؼيوث انوجثلَاا ، ظ : العم ي الحبي5 / 415 . ) 8 ) - : ظ البدر العطام ح حم غ الط ام2 / 11 . ) 727 1 )– صحي مسمن5 / 821 ) , كعاب الصيام , باب صحي مسمن5 / 822 : ), و ج بث صَ ومث رَمَضَ اثَ لث ؤويَةث الويثالَ اث ، ال ن 1181 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ج- ا لة اصحاب الج ا الثالإل : اسعدل ا بالسظة واأفعاا الصحابة او- : اسعد لين بالسظة َماااااا رواه احوااااانث ع طَااااا َ َاااااااَ: َااااااا "إثنتطَاااااا الشت ااااايو ثسو ااااا وَعثشو ااااا وثَ فَاااااالَ َص ااااا م ا َعتاااااى َااااا َ ووه وَ َ تولث ااااا وا َعتى َ َ ووه فَ ثثو و نت عَمَيوك نو فَا ود ر وا لَو 1 ). ج- ا لة اصحاب الج ا الثالإل : اسعدل ا بالسظة واأفعاا الصحابة او- : اسعد لين بالسظة و ن َماااااا رواه احوااااانث ع طَااااا َ َاااااااَ: َااااااا "إثنتطَاااااا الشت ااااايو ثسو ااااا وَعثشو ااااا وثَ فَاااااالَ َص ااااا م ا َعتاااااى َااااا َ ووه وَ َ تولث ااااا وا َعتى َ َ ووه فَ ثو و نت عَمَيوك نو فَا ود ر وا لَو 1 ). ثَث الصت اااا و م ى اااا َ العتجَاااا ُّب إثلَااااى ، وَالظ ثيتااااة َ اااا ط فثااااد و عااااو اَااااة، وَإثنتطَاااا ا لاااان ااااسك جَاااا َاب ماااان، اكوعثتَاااااء حثسك ثهث فثد الحَدث إل والج ا التص بوعَث ثَ عَمَى نثيتا ثيثنو 5 ) . ثا ا ل ا أ ا ة ف اا ال 1 - َ َما رواه ا ث م ، أَثت ع ط َااَ: مَنو أَكَ َ فَمويَجوىث َ و مًا مَكَانَو 8 ) . 5 - عَااااانو فَاطث طَاااااةَ امو َأَ ثاااااوث، عَااااانو أَسو اااااطَاءَ ر اااااد عظياااااا الَاااااة : "أَفولَ ونَاااااا عَمَاااااى عَيوااااادث رَس ااااا اث ْتث-  - " فثد َ و م غَيون ، ث نت طَمَعَةو الشت طو " . ثي َ لثيثشَ ام : أ مث وا بثالوجَضَ اءث؟ َااَ:" َ ح دت مثنو َضَ اء 4 ). 8 - مااااااا رواه ،َ َاحواااااانث ع طَاااااا َ : إثثو َااااااااَ وثَ مَظو َاااااا ثهث سَ ااااااحَاب أَوو َعَاااااا أَصو اااااابَ َ صَ ااااااا ثطًا، فَجَااااااااَ: لَااااااو رَج اااااا أَصو ااااااب َعوظثد: احونَ ع طَ َ صَ ا ثطًا َظوعَ ث ا وَ وبَارَ ، َااَ: َ َعوظثد أَثت احونَ ع طَ َ كَاثَ عثنُّ صث يَامَو 2 ). : وجو الد لة لطا سبا ج از الصيام فد الي م السز َااَ وثَ مَظو َ ثهث سَ حَاب أَوو َعَ أَصو بَ َ صا طا 2 ). ج از الصيام فد الي م السز َااَ وثَ مَظو َ ثهث سَ حَاب أَوو َعَ أَصو بَ َ صا طا 2 ). ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ا انين ا عمت ا ؼيطا إذا ثبعة رؤية اليالا ب المواا او بعده : و تصي الطسألة كطا أ د 1 - :أقهال الفقهاء ا عم التجياء ر طين , إذا ثبعة رؤية اليالا ب المواا او بعده عمى لين ىطا: : وجو الد لة : وجو الد لة كااااا ه صااااا م ااااا م الشاااااك لعًاااااا أث كااااا ث مااااان رمضااااااث أو عماااااى وجاااااو الط اعااااااة فاااااا أث كااااا ث مااااان رمضااااااث، فيمحاااااا باااااالت ماااااا لاااااي مااااان جظساااااو، فأماااااا إذا أ مااااا الظياااااة لمعلااااا ا، فمااااان حصااااا ؼياااااو معظاااااى الشك، ف نطا نيعو أنو من عباث، في كطا ص مو ع ن نسر أو ضاء رمضاث 1 ). َالحاااااااد إل الثااااااااند: عااااااان احثاااااااد ى َيوااااااا َة أَثت رَس ااااااا اَ ْتث َ َ َاااااااااَ: َ َجَااااااادتم ا الشت ااااااايو َ بصااااااا م َااااااا و م و َااااا و مَيونث إث ت أَثو َافثااااااَ صَ ااااا و مًا كَااااااثَ َص ااااا م و أَ َاااااد ك نو، َ َص ااااا م ا َعتاااااى َااااا َ و ا الويثاااااالَ اَ، وَ ً تولث ااااا و ، ا َعتاااااى َااااا َ ووه َفَ ثثو و نت عَمَيوك نو فَأَكوطثم ا ثَالَ ثثين 5 ). كااااا ه صااااا م ااااا م الشاااااك لعًاااااا أث كااااا ث مااااان رمضااااااث أو عماااااى وجاااااو الط اعااااااة فاااااا أث كااااا ث مااااان رمضااااااث، فيمحاااااا باااااالت ماااااا لاااااي مااااان جظساااااو، فأماااااا إذا أ مااااا الظياااااة لمعلااااا ا، فمااااان حصااااا ؼياااااو معظاااااى الشك، ف نطا نيعو أنو من عباث، في كطا ص مو ع ن نسر أو ضاء رمضاث 1 ). َالحاااااااد إل الثااااااااند: عااااااان احثاااااااد ى َيوااااااا َة أَثت رَس ااااااا اَ ْتث َ َ َاااااااااَ: َ َجَااااااادتم ا الشت ااااااايو َ بصااااااا م َااااااا و م و َااااا و مَيونث إث ت أَثو َافثااااااَ صَ ااااا و مًا كَااااااثَ َص ااااا م و أَ َاااااد ك نو، َ َص ااااا م ا َعتاااااى َااااا َ و ا الويثاااااالَ اَ، وَ ً تولث ااااا و ، ا َعتاااااى َااااا َ ووه َفَ ثثو و نت عَمَيوك نو فَأَكوطثم ا ثَالَ ثثين 5 ). و- . أ لة اصحاب الج ا ال ام : اسعدا اصحاب ىسا الج ا بالطعج ا أث الظاااااس فااااد صاااا مو َبَاااا ثثمَااااامثيثنو إثثو صَ ااااامَ صَ ااااام ه ، وَإثثو أَفولَاااا َ أَفولَاااا وه وي اااائ صاااا مو فااااد الػااااين وي م الصح ، و اا م إث الظاس ب لإلمام إث صام صام ا وإث أفل أفل وا 4 ). 1 )- ح صحي الب ارز حن بلاا4 / 85 .) 5 ) - صحي الب ارز ، كعاب الص م ، بَاب َ واث الظتبثد ث "إثذَا رَأَ وع ن ا ليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ر ن8 / 58 .) 8 ) - ظ : عطدة الجارز ح صحي الب ارز11 / 588 . ) 4 )– ظ ني ا وطار4 / 585 )، عطدة الجارز ح صحي الب ارز1 / 588 ) ، الطهظد حن دامة8 / 181 .) 2 ) - حدا الصظا فد يئ الش ا5 / 85 , ) حدا ة الط عيد ونيا ة الطجعصد5 / 48 , ) الحاوز الكبي8 / 411 , ) الطهظد حن دامة8 / 188 . ) ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 - : ا لة اصحاب الج ا ال اب : اسعدل ا بالسظة والطعج ا الحااااااد إل ا وا :عَاااااانو عطااااااار حاااااان اساااااا - أَنتااااااو َااااااااَ: "مَاااااانو صَ ااااااامَ َاااااا و مَ الشت ااااااك ث فَجَاااااادو عَصَ ااااااى أَبَااااااا الوجَاسث نث" 8 ). الحااااااد إل ا وا :عَاااااانو عطااااااار حاااااان اساااااا - أَنتااااااو َااااااااَ: "مَاااااانو صَ ااااااامَ َاااااا و مَ الشت ااااااك ث فَجَاااااادو عَصَ ااااااى أَبَااااااا الوجَاسث نث" 8 ). 1 )– صحي مسمن5 / 821 ) , كعاب الصيام , باب صحي مسمن5 / 822 : ), و ج بث صَ ومث رَمَضَ اثَ لث ؤويَةث الويثالَ اث ، ال ن 1181 . 5 ) - ظ : عطدة الجارز ح صحي الب ارز11 / 584 ) . 8 )– صحي الب ارز8 / 88 ) , كعاب الص م , باب اذا افل فد رمضاث ثن طعة الشط ، ر ن1222 . 4 )- صحي الب ارز8 / 88 ) , كعاب الص م , باب اذا افل فد رمضاث ثن طعة الشط ، ر ن1222 . 2 )- ظ : مسا ا مام أ طد روا ة أحد او الس سعاند1 / 158 .) د د 2 ) - ظ : البدر العطام ح حم غ الط ام2 / 11 . ) 8 )– صحي الب ارز ،ال ام الطسظد الصحي الط عص ،باب ا الظبد : اذا, ر ن8 / 58 ) 1 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 : وجو الد لة كااااا ه صااااا م ااااا م الشاااااك لعًاااااا أث كااااا ث مااااان رمضااااااث أو عماااااى وجاااااو الط اعااااااة فاااااا أث كااااا ث مااااان رمضااااااث، فيمحاااااا باااااالت ماااااا لاااااي مااااان جظساااااو، فأماااااا إذا أ مااااا الظياااااة لمعلااااا ا، فمااااان حصااااا ؼياااااو معظاااااى الشك، ف نطا نيعو أنو من عباث، في كطا ص مو ع ن نسر أو ضاء رمضاث 1 ). َالحاااااااد إل الثااااااااند: عااااااان احثاااااااد ى َيوااااااا َة أَثت رَس ااااااا اَ ْتث َ َ َاااااااااَ: َ َجَااااااادتم ا الشت ااااااايو َ بصااااااا م َااااااا و م و َااااا و مَيونث إث ت أَثو َافثااااااَ صَ ااااا و مًا كَااااااثَ َص ااااا م و أَ َاااااد ك نو، َ َص ااااا م ا َعتاااااى َااااا َ و ا الويثاااااالَ اَ، وَ ً تولث ااااا و ، ا َعتاااااى َااااا َ ووه َفَ ثثو و نت عَمَيوك نو فَأَكوطثم ا ثَالَ ثثين 5 ). وجو الد: لة َص صَ و مو أصال وَلَ وَافا عَا َة لَو 8 ) . و- . أ لة اصحاب الج ا ال ام : اسعدا اصحاب ىسا الج ا بالطعج ا - : اسعد لين بالطعج ا أث الظاااااس فااااد صاااا مو َبَاااا ثثمَااااامثيثنو إثثو صَ ااااامَ صَ ااااام ه ، وَإثثو أَفولَاااا َ أَفولَاااا وه وي اااائ صاااا مو فااااد الػااااين وي م الصح ، و اا م إث الظاس ب لإلمام إث صام صام ا وإث أفل أفل وا 4 ). 8 - : الع جي بعااااد عاااا ا لااااة ومظا شااااعيا بااااين لاااادل البا ااااإل اث الجاااا ا الثاااااند وىاااا مااااا اااااا بااااو احاااا ظيتااااة ، ودمحم حاااان الحسااااان الشااااايباند ر طيطاااااا وويااااا ىن الماااااساث التاااااا ؼياااااو ااااا ا احااااان مساااااع ر اااااد عظاااااو ىااااا الااااا اج و اعمن. المطمب الثاني حكم يهم الذك اذا ثبتت رؤية الهالل قبل الزوال او بعجه نهارا ا تا التجياء ر طين عالى عمى صيام رمضاث عظد رؤية اليالا 2 ). 2 -ا: الدلة ومناقذتها وَسَ اااابَئ ا وااااعثالَ فثيثنو : َاااا و اعوعثبَااااارث العت و ثاَااااةث ؼثيطَااااا سَ اااابثيم و العت و ثاَااااة وَال ُّج اااا ا إثلَااااى ا وَ وبَااااارث فثااااد َ ذَلثااااكَ، وَلَاااايو فثاااااد ذَلثاااااكَ أَثَااااا عَااااانث الظتبثاااااد ث َ وجَااااا إثلَيواااااوث، لَكثااااانو ر وثزَ عَااااانو ع طَااااا ، أَثَااااا َاثث : أَ َاااااد ى طَا عَاااااام ، وَاصو َااااا م تَس ث ااااا فَسَىَئَ َ و م إثلَى الوعَام وَذَىَئَ َ و م إثلَى الوط تَس ث 8 ). وَسَ اااابَئ ا وااااعثالَ فثيثنو : َاااا و اعوعثبَااااارث العت و ثاَااااةث ؼثيطَااااا سَ اااابثيم و العت و ثاَااااة وَال ُّج اااا ا إثلَااااى ا وَ وبَااااارث فثااااد َ ذَلثااااكَ، وَلَاااايو فثاااااد ذَلثاااااكَ أَثَااااا عَااااانث الظتبثاااااد ث َ وجَااااا إثلَيواااااوث، لَكثااااانو ر وثزَ عَااااانو ع طَااااا ، أَثَااااا َاثث : أَ َاااااد ى طَا عَاااااام ، وَاصو َااااا م تَس ث ااااا فَسَىَئَ َ و م إثلَى الوعَام ث وَذَىَئَ َ و م إثلَى الوط تَس ث ث 8 ). َ فَأَمتاااااا الوعَاااااامُّ: فَطَاااااا رَ وَاه ا وَعوطَاااااش عَااااانو أَحثاااااد وَا ثااااا َ اااااجثياث حوااااانث سَ ااااامَطَةَ َاااااااَ: أَ َانَاااااا كثعَااااااب ع طَااااا وَنَحوااااان ثب َاااانثجثينَ أَثت ا وَىثمث اااا ةَ بَعوضَ ااايَا أَكوبَاااا مثااانو بَعوااااى ، فَااا ثذَا رَأَ وااااع ن الويثاااالَ اَ نَيَااااارًا فَاااالَ تولث اااا وا َعتاااى َشو اااايَدَ رَج اااالَ ثث أَنتي طَاااا ا رَأَ َاه بثا وَمو ث 8 ). فَسَىَئَ َ و م إثلَى الوعَام ث وَذَىَئَ َ و م إثلَى الوط تَس ث ث 8 ). َ فَأَمتاااااا الوعَاااااامُّ: فَطَاااااا رَ وَاه ا وَعوطَاااااش عَااااانو أَحثاااااد وَا ثااااا َ اااااجثياث حوااااانث سَ ااااامَطَةَ َاااااااَ: أَ َانَاااااا كثعَااااااب ع طَااااا وَنَحوااااان ثب َاااانثجثينَ أَثت ا وَىثمث اااا ةَ بَعوضَ ااايَا أَكوبَاااا مثااانو بَعوااااى ، فَااا ثذَا رَأَ وااااع ن الويثاااالَ اَ نَيَااااارًا فَاااالَ تولث اااا وا َعتاااى َشو اااايَدَ رَج اااالَ ثث أَنتي طَاااا ا رَأَ َاه بثا وَمو ث 8 ). َ فَأَمتاااااا الوعَاااااامُّ: فَطَاااااا رَ وَاه ا وَعوطَاااااش عَااااانو أحثاااااد وَا ثااااا َ اااااجثياث حوااااانث سَ ااااامَطَةَ َاااااااَ: أ َانَاااااا كثعَااااااب ع طَااااا وَنَحوااااان ثب َاااانثجثينَ أَثت ا وَىثمث اااا ةَ بَعوضَ ااايَا أَكوبَاااا مثااانو بَعوااااى ، فَااا ثذَا رَأَ وااااع ن الويثاااالَ اَ نَيَااااارًا فَاااالَ تولث اااا وا َعتاااى َشو اااايَدَ رَج اااالَ ثث أَنتي طَاااا ا رَأَ َاه بثا وَمو ث 8 ). ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 أ- َ َالجاااا ا ا وا : إث ر ثاااادَ َاااا و مَ الشت ااااك ث الويثااااالَ اَ بَعواااادَ الاااامتوَااث أَوو َبومَااااو فَي اااا َ لثمتيومَااااةث الوط سو ااااعَجوبَمَةث و َك اااا ث ذَلثااااكَ الويَاااا و م مثنو رَمَضَ اثَ ، وى ا احن مسع 1 ) . َوَىَااااااسَا َاااااا و ا ع طَاااااا َ ، وَاحواااااانث ع طَاااااا َ ، وعثطاااااااث ,وعمااااااد , وَأ نَاااااا ر ااااااد عااااااظين ، وَا وَووزَاعثااااااد ث 5 ) ، وَأَحثاااااا َ َظثيتَة ومحن حن الحسن , ومالك ، وَالشت افثعثد ث ، وا طد ر طين 8 ) . ب- : الجاااااا ا الثاااااااند اث ر ثاااااادَ َاااااا و مَ الشت ااااااك ث الويثااااااالَ اَ إثو َكَاااااااثَ َبواااااا َ الاااااامتوَااث فَي اااااا َ لثمتيومَااااااةث الوطَا ث اااااايَةث وَإثثو كَاااااااثَ بَعوااااااد الااااامتوَااث فَي ااااا َ لثمتيومَاااااةث الوط سو اااااعَجوبَمَةث , روز ذلاااااك عااااان عطااااا , وااااااو ااااااا َ أَح ااااا س ااااا ر طاااااو 4 ) , والثااااا رز 2 ) , وا طد فد روا ة 2 ) . ر طين ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 8 - : الع جي بعااااد عاااا ا لااااة ومظا شااااعيا بااااين لاااادل البا ااااإل اث الجاااا ا الثاااااند وىاااا مااااا اااااا بااااو احاااا ظيتااااة ، ودمحم حاااان الحسااااان الشااااايباند ر طيطاااااا وويااااا ىن الماااااساث التاااااا ؼياااااو ااااا ا احااااان مساااااع ر اااااد عظاااااو ىااااا الااااا اج و اعمن. حكم يهم الذك اذا ثبتت رؤية الهالل قبل الزوال او بعجه نهارا ا تا التجياء ر طين عالى عمى صيام رمضاث عظد رؤية اليالا 2 ). ا انين ا عمت ا ؼيطا إذا ثبعة رؤية اليالا ب المواا او بعده ا عم التجياء ر طين , إذا ثبعة رؤية اليالا ب المواا او بعده عمى لين ىطا: 1 )- ح صحي الب ارز حن بلاا4 / 85 .) 5 ) - صحي الب ارز ، كعاب الص م ، بَاب َ واث الظتبثد ث "إثذَا رَأَ وع ن ا ليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " ر ن8 / 58 .) 8 ) - ظ : عطدة الجارز ح صحي الب ارز11 / 588 . ) 4 )– ظ ني ا وطار4 / 585 )، عطدة الجارز ح صحي الب ارز1 / 588 ) ، الطهظد حن دامة8 / 181 .) 2 ) - حدا الصظا فد يئ الش ا5 / 85 , ) حدا ة الط عيد ونيا ة الطجعصد5 / 48 , ) الحاوز الكبي8 / 411 , ) الطهظد حن دامة8 / 188 . ) 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 1 ) - ظ م اة الطتا ي مشكاة الطصاحي4 / 1881 .) 1 ) - ظ م اة الطتا ي مشكاة الطصاحي4 / 1881 .) 5 ) - . ظ الطصدر الساحا 8 ) - حدا ة الط عيد ونيا ة الطجعصد5 / 48 , ) حدا الصظا فد يئ الش ا5 / 85 ) , الحاوز الكبي8 / 411 ,) الطهظد حن دامة8 / 188 . ) 4 ) - حدا الصظا فد يئ الش ا5 / 85 ) . بيين الحجا ا ح كظم الد ا ا و ا ية الشمبد1 / 851 ). حدا ة ال ط عيد ونيا ة الطجعصد5 / 48 . ) 2 ) – ظ : م عاة الطتا ي ح مشكاة الطصاحي2 / 454 ). العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 .) العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 . ) 2 ) - ظ :الطهظد حن دامة8 / 128 . ) 8 ) - : ظ. الطصدر الساحا 8 ) - . ظ : الطصدر الساحا 2 ) - ظ حدا ة الط عيد ونيا ة الطجعصد5 / 48 . ) ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 أ- : ا لة أصحاب الج ا ا وا لَيو َ فثد ذَلثكَ أَثَ عَنث الظتبثد ث-  - وجَ إثلَيووث1. لكن اسعدا اصحاب ىسا الج ا. بالسظة والطعج ا أو-ا: سعد لين بالسظة عاااااااان الظبااااااااد " أنااااااااو اااااااااا : " ص اااااااا م ا لث ؤويَعثااااااااوث وَأَفولث اااااااا وا لث ؤويَعثااااااااوث 1 ) َأَمَاااااااا َ بثالصت اااااااا و مث، وَالوتثلواااااااا ث بَعوااااااااد ال ُّؤويَةث 5 ). : وجو الد لة نياااااااى أَث َص ااااااا م بااااااا رَمَضَ ااااااااث حيَااااااا و م َعتاااااااى ااااااا ل الوياااااااالَ ا، أَو نتاااااااد الوعااااااادة، ثااااااانت َ تلااااااا وث َعتاااااااى وه 8 ). أ- : ا لة أصحاب الج ا ا وا لَيو َ فثد ذَلثكَ أَثَ عَنث الظتبثد ث-  - وجَ إثلَيووث1. لكن اسعدا اصحاب ىسا الج ا. بالسظة والطعج ا أو-ا: سعد لين بالسظة ال ُّؤويَةث 5 ). : وجو الد لة نياااااااى أَث َص ااااااا م بااااااا رَمَضَ ااااااااث حيَااااااا و م َعتاااااااى ااااااا ل الوياااااااالَ ا، أَو نتاااااااد الوعااااااادة، ثااااااانت َ تلااااااا وث َعتاااااااى وه 8 ). 1 ) - صحي الب ارز8 / 48 ), كعاب الص م , بَاب َ واث الظتبثد ث صَ متى عَمَيووث وَسَ متنَ , , صحي الب ارز8 / 58 ) "إثذَا رَأَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " . ر ن1212 . . 5 ) - - ظ حدا ة الط عيد ونيا ة الطجعصد 5 / 48 . ) 8 ) - عطدة الجارز ح صحي الب ارز11 / 585 . ) 4 ) - ح معاند اصثار8 / 155 :), كعاب اصذاث والظسور, بَاب : ال تج َحومث أَثو َ كَم ثنَ رَج الً َ يو ًا ,ال ن4884 : , ظ يئ ا مالد ال طيسية لمش ز1 / 884 .) 2 ) - ظ حدا الصظا فد يئ الش ا5 / 85 ) - 88 ), ظ : يئ ا مالد ال طيسية لمش ز1 / 884 .) 2 -ا: الدلة ومناقذتها ث وَأَمتااااا الو َاااااصُّ : فَطَااااا رَ وَل الثتاااا ورثزُّ عَظوااااو أَنتااااو حَمَاااا َ ع طَاااا َ حواااا نَ الو َلتااااابث أَثت َ و مًااااا رَأَو ا الويثااااالَ اَ بَعواااادَ الاااامتوَااث َفَاااااأَفولَ وا، فَكَعَااااائَ إثلَااااايويثنو َم ااااا م ي نو وَ َاااااااَ: إثذَا رَأَ واااااع ن الويثاااااالَ اَ نَيَاااااارًا َبوااااا َ الااااامتوَااث فَاااااأَفولث وا، و إثذَا رَأَ وع ط ااااا ه بَعوااااادَ الااااامتوَااث فَالَ تولث وا 2 ) . 1 ) - ظ م اة الطتا ي مشكاة الطصاحي4 / 1881 .) 5 ) - . ظ الطصدر الساحا 8 ) - حدا ة الط عيد ونيا ة الطجعصد5 / 48 , ) حدا الصظا فد يئ الش ا5 / 85 ) , الحاوز الكبي8 / 411 ,) الطهظد حن دامة8 / 188 . ) 4 ) - حدا الصظا فد يئ الش ا5 / 85 ) . بيين الحجا ا ح كظم الد ا ا و ا ية الشمبد1 / 851 ). حدا ة ال ط عيد ونيا ة الطجعصد5 / 48 . ) 2 ) – ظ : م عاة الطتا ي ح مشكاة الطصاحي2 / 454 ). العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 .) العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 . ) 2 ) - ظ :الطهظد حن دامة8 / 128 . ) 8 ) - : ظ. الطصدر الساحا 8 ) - . ظ : الطصدر الساحا 2 ) - ظ حدا ة الط عيد ونيا ة الطجعصد5 / 48 . ) 1 ) - ظ م اة الطتا ي مشكاة الطصاحي4 / 1881 .) 5 ) - . ظ الطصدر الساحا 4 ) - حدا الصظا فد يئ الش ا5 / 85 ) . بيين الحجا ا ح كظم الد ا ا و ا ية الشمبد1 / 851 ). حدا ة ال ط عيد ونيا ة الطجعصد5 / 48 . ) 2 ) – ظ : م عاة الطتا ي ح مشكاة الطصاحي2 / 454 ). العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 .) العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 . ) 2 ) - ظ :الطهظد حن دامة8 / 128 . ) 8 ) - : ظ. الطصدر الساحا 8 ) - . ظ : الطصدر الساحا 2 ) - ظ حدا ة الط عيد ونيا ة الطجعصد5 / 48 . ) 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 1 َ) فَثَبَااااةَ أَثت الشت اااايو َ َاااادو َك اااا ث ثَال ثثااااينَ وَ َاااادو َك اااا ث ثسو ااااعَةً وَعثشو اااا ثينَ وَ َاااادو ر وثزَ عَاااانو أَنَاااا-  - :َأَنتااااو َاااااا "ص ااااطوظَا عَمَااااى عَيواااادث رَس اااا اث ْتث-  - " ثسو ااااعَةً وَعثشو اااا ثينَ َ و مًااااا أَكوثَاااا َ مثطتااااا ص ااااطوظَا ثَالَ ثثااااينَ َ و مًااااا 4 ) َوَلَاااا و صَ ااااام أَىواااا حَمَاااااد ثَالَ ثثاااااينَ َ و مًااااا وَصَ اااااامَ أَىوااااا حَمَااااد آ َااااا َ ثسو اااااعَةً وَعثشو اااا ثينَ َ و مًاااااا فَااااا ثثو كَاااااثَ صَ ااااا و م أَىوااااا ث ذَلثااااكَ ا لوبَمَااااادث حث ؤويَاااااةث الويثااااالَ اث وَثَبَااااةَ ذَلثااااكَ عثظواااادَ َا ث اااايينو، أَوو عَاااادُّوا َ ااااعوبَاثَ ثَالَ ثثااااينَ َ و مًااااا ث اااانت صَ ااااام ا رَمَضَ اااااثَ فَعَمَااااى أَىواااا ث الوبَمَاااادث اصو َاااا ث ، َضَ ااااااء َااااا و م ثَنتي ااااانو أَفولَااااا وا َ و مًاااااا مثااااانو رَمَضَ ااااااثَ لثث ب ااااا تث ال تمَضَ اااااانثيتةث حث ؤويَاااااةث أَىوااااا ث ذَلثاااااكَ الوبَمَااااادث وَعَااااادَم ر ؤويَاااااةث أَىوااااا ث َالوبَمَااااادث َ َجوااااادَح فثاااااد ر ؤويَاااااةث أ ولَ ثاااااكَ، إذو الوعَاااااد م َ عَاااااارث الو ج ااااا َ، وَإثثو كَااااااثَ صَ ااااا و م أَىوااااا ث ذَلثاااااكَ الوبَمَااااادث بثهَيوااااا ث ر ؤويَاااااةث ىثاااااالَ اث رَمَضَ ااااااثَ أَوو لَااااانو َثوب اااااةو ال ُّؤويَاااااة عثظوااااادَ َا ث ااااايينو وَ َ عَااااادُّوا َ اااااعوبَاثَ ثَالَ ثثاااااينَ َ و مًاااااا فَجَااااادو أَسَ ااااااء وا َيواااااإل َجَااااادتم ا َ رَمَضَ اثَ بثصَ و مث و م 2 ). 5 )ا: سعد لين الطعج ا ااااا لين لمطسااااااعجبمية , ولَااااايو َ كَ و ن ااااااو لثموط سو اااااعَجوبَمَةث ثَاحثعًااااااا حث ؤويَعثاااااوث نَيَااااااارًا ثَنتاااااو َ عثبواااااا َةَ حث ؤويَعثاااااوث نَيَااااااارًا وَإثنتطَااااااا ثَبَاااةَ بث ثكوطَاااااث الوعثااادتةث ثَثت الو ثاااالَ سَ عَمَاااى مَاااا صَ ااا تحَ بثاااوث َ فثاااد الوبَااادَا ث ث وَالوتَاااعو ث إنتطَاااا ى ااا َ فثاااد ر ؤويَعثاااوث َااا و مَ الشت اااك ث وَى ااا َاااا و م الثتالَ ثثااااينَ مثاااانو َ ااااعوبَاثَ أَوو مثاااانو رَمَضَ اااااثَ . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 1 - عااااان الثتااااا ورثز ث عااااان احااااا اهين َاااااااَ : كَعَااااائَ ع طَااااا إثلَاااااى ع عوبَاااااةَ حوااااانث فَ و َاااااد إثذَا رَأَ واااااع ن الويثاااااالَ اَ نَيَاااااارًا َبوااااا َ أَثو َااااام واَ الشت اااااطو لثعَطَاااااامث ثَالَ ثثاااااينَ فَاااااأَفولث وا وَإثذَا رَأَ وع ط ااااا ه بَعوااااادَ مَاااااا َااااام و ا الشت اااااطو فَاااااالَ تولث ااااا وا َعتاااااى طوس ااااا ا ,وروز َعن إثسو حَا َ عَنث الوحَ وثث عَنو عَمثد مثثو َ ذَلثك 5 ) . 5 - مَااااااا رَ وَل الثتاااااا ورزُّ عَاااااانو أَحااااااد وَا ثاااااا أَنتااااااو حَمَاااااا َ ع طَاااااا َ حواااااانَ الو َلتااااااابث أَثت َ و مًااااااا رَأَو ا الويااااااالَ اَ بَعواااااادَ الاااااامتوَااث 5 - مَااااااا رَ وَل الثتاااااا ورثزُّ عَاااااانو أَحثااااااد وَا ثاااااا أَنتااااااو حَمَاااااا َ ع طَاااااا َ حواااااانَ الو َلتااااااابث أَثت َ و مًااااااا رَأَو ا الويثااااااالَ اَ بَعواااااادَ الاااااامتوَااث َفَاااااأَفولَ وا، فَكَعَااااائَ إثلَااااايويثنو َم ااااا م ي نو وَ َاااااااَ: إثذَا رَأَ واااااع ن الويثاااااالَ اَ نَيَاااااارًا َبوااااا َ الااااامتوَااث فَاااااأَفولث وا، و إثذَا رَأَ وع ط ااااا ه بَعوااااادَ الااااامتوَااث فَالَ تولث وا 8 ) . ا ثانيااااا- ا سااااعد لين َبااااالطعج ا : فَعثظواااادَ أَحثااااد س اااا َ ى اااا َ لثمتيومَااااةث الوطَا ث اااايَةث بثطَعوظَااااى أَنتااااو عوعَبَاااا أَثت الويثااااالَ اَ َاااادو و جثااااد فثاااااد ا و ف ااااااث لَيومَاااااةَ الو ط عَاااااةث فَهَاااااابَ ، ث ااااانت يَيَااااا َ نَيَاااااارًا فَ ي ااااا ر ه فثاااااد الظتيَاااااارث فثاااااد كوااااانث ي ي ااااا رثهث فثاااااد لَيومَاااااة ثَانث يَاااااة مثااااانو َ احوعثدَاءث الشت يو ث ثَنتو لَ و لَنو َك نو َبو َ لَيومَة لَنو طوكثنو ر ؤويَع و نَيَارًا ثَنتو َ َل َبو َ المتوَااث إ ت أَثو َك ث 4 ). ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 فَاااا ثذَا كَاااااثَ َاااا و م الو ط عَااااةث الوطَااااسوك رث َاااا و مَ الثتالَ ثثااااينَ مثاااا َنو الشت اااايو ث وَر ثااااد َؼثياااااوث الويثاااااالَ ا نَيَاااااارًا ف َعثظوااااادَ أَحثاااااد س ااااا َ ذَلثاااااكَ الويَااااا و م أَوتا الشت ااااايو ث َ عثبوااااا َةَ لثيَاااااسثهث ال ُّؤويَاااااةث وَيَك ااااا ث أَوتا الشت ااااايو ث َااااا و م 1 ) - صحي الب ارز8 / 48 ), كعاب الص م , بَاب َ واث الظتبثد ث صَ متى عَمَيووث وَسَ متنَ , , صحي الب ارز8 / 58 ) "إثذَا رَأَ وع ن اليثالَاَ فَص م ا، وَإثذَا رَأَ وع ط ه فَأَفولث وا " . ر ن1212 . . ي ب رز رز ح / ) 4 ) - ح معاند اصثار8 / 155 :), كعاب اصذاث والظسور, بَاب : ال تج َحومث أَثو َ كَم ثنَ رَج الً َ يو ًا ,ال ن4884 : , ظ يئ ا مالد ال طيسية لمش ز1 / 884 .) 2 ) - ظ حدا الصظا فد يئ الش ا5 / 85 ) - 88 ), ظ : يئ ا مالد ال طيسية لمش ز1 / 884 .) ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 الست اااابوةث سَ اااا َاء و جثاااادَتو ىَااااسثهث ثَثت ر ؤويَعَااااو َاااا و مَ العتاسث اااا ث وَالوعثشو اااا ثينَ لَاااانو َج اااا و أَ َااااد فثييَااااا أَنتااااو لثمو طَا ث اااا يَةث لثاااا َالت َمواااامَمَ أَثو َك ثَ الشت يو ثَطَانثيَةً وَعثشو ثينَ كَطَا نَ ت عَمَيووث بَعوى الوط حَج ثجثينَ ، وَ َ طث َ َ و ل ي نو َ عثبو َةَ حث ؤو يَعثوث نَيَارًا 1 ). 5 - .ا لة الج ا الثاند : اسعدا اصحاب ىسا الج ا بطا رول عن الصحابة , واالطعج ا أو-ا:سعد لين بطا رول عن الصحابة الست اااابةث سَ اااا َاء و جثاااادت ىااااسثهث ث ث ر ؤيعااااو اااا و مَ العاسث اااا ث وَالعثشاااا ثين لاااانو جاااا أ َااااد فثييَااااا أنااااو لثم طَا ث اااا يةث لثاااا ال مااااممَ أث َك ثَ الشت يو ثَطَانثيَةً وَعثشو ثينَ كَطَا نَ ت عَمَيووث بَعوى الوط حَج ثجثينَ ، وَ َ طث َ َ و ل ي نو َ عثبو َةَ حث ؤو يَعثوث نَيَارًا 1 ). 5 - .ا لة الج ا الثاند : اسعدا اصحاب ىسا الج ا بطا رول عن الصحابة , واالطعج ا أو- ا: سعد لين بطا رول عن الصحابة 1 - عااااان الثتااااا ورثز ث عااااان احااااا اهين َاااااااَ : كَعَااااائَ ع طَااااا إثلَاااااى ع عوبَاااااةَ حوااااانث فَ و َاااااد إثذَا رَأَ واااااع ن الويثاااااالَ اَ نَيَاااااارًا َبوااااا َ أَثو َااااام واَ الشت اااااطو لثعَطَاااااامث ثَالَ ثثاااااينَ فَاااااأَفولث وا وَإثذَا رَأَ وع ط ااااا ه بَعوااااادَ مَاااااا َااااام و ا الشت اااااطو فَاااااالَ تولث ااااا وا َعتاااااى طوس ااااا ا ,وروز َعن إثسو حَا َ عَنث الوحَ وثث عَنو عَمثد مثثو َ ذَلثك 5 ) . 5 - مَااااااا رَ وَل الثتاااااا ورثزُّ عَاااااانو أَحثااااااد وَا ثاااااا أَنتااااااو حَمَاااااا َ ع طَاااااا َ حواااااانَ الو َلتااااااابث أَثت َ و مًااااااا رَأَو ا الويثااااااالَ اَ بَعواااااادَ الاااااامتوَااث َفَاااااأَفولَ وا، فَكَعَااااائَ إثلَااااايويثنو َم ااااا م ي نو وَ َاااااااَ: إثذَا رَأَ واااااع ن الويثاااااالَ اَ نَيَاااااارًا َبوااااا َ الااااامتوَااث فَاااااأَفولث وا، و إثذَا رَأَ وع ط ااااا ه بَعوااااادَ الااااامتوَااث فَالَ تولث وا 8 ) . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 وَالوحَاصث اااااا : إذَا ر ثاااااادَ الويثااااااالَ ا َاااااا و مَ الو ط عَااااااةث مَااااااثَالً َبواااااا َ الاااااامتوَااث فَعثظواااااادَ أَحثااااااد س اااااا َ ى اااااا َ لثمتيومَااااااةث الوطَا ث اااااايَةث َبثطَعوظَاااى أَنتاااو عوعَبَااا أَثت الويثاااالَ اَ َااادو و جثااادَ فثاااد ا و ف ااااث لَيومَاااةَ الو ط عَاااةث فَهَاااابَ ، ث ااانت يَيَااا َ نَيَاااارًا ف ي ااا ر ه فثاااد الظتيَاااارث فثاااد كواااانث ي ي اااا رثهث فثااااد لَيومَااااة ثَانثيَااااة مثاااانو احو َ عثاااادَاءث الشت اااايو ث ثَنتااااو لَاااا و لَاااانو َك اااانو َبواااا َ لَيومَااااة لَاااانو طوكثاااانو ر ؤويَع ااااو نَيَااااارًا ثَنتااااو ااااا َل َبوااااا َ الااااامتوَااث إ ت أَثو َك ااااا ثَ لثمَيومَعَااااايونث ، فَاااااالَ م ظَافَااااااةَ حَااااايونَ كَ و نثاااااوث لثمتيومَاااااةث الوطَا ث ااااايَةث وَكَ و نثااااا َوث لثمَيومَعَااااايونث ث َثت الظتيَاااااار َصَ اااااارَ بثطَظومثلَاااااةث لَيومَاااااة ثَانثيَاااااة وَإثذَا كَااااااثَ لثمتيومَاااااةث الوطَا ث ااااايَةث َك ااااا ث َااااا و م الو ط عَاااااةث الوطَاااااسوك ر أَوتا الشت ااااايو ث ، ؼَيَ ثااااائ صَ ااااا و م و ً إثو كَاثَ رَمَضَ اثَ ، وَيَ ثئ فثلو ه إثو كَاثَ َ تا 2 ). ااااااا البااااااج د: ااااااا أحااااا اك حااااان ال يااااان ىاااااسا ثباااااة عااااان عطااااا رواه ااااابا وىااااا م يااااا ا ااااااا وىاااااسا ال الس إنطا ى إذا رؤز فد م ثالثين و ص أث ص أث ك ث ب ذلك 2 ) . 1 ) - : ظ نَيَارًا )الدر الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 . ) 5 ) - : ظ العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 . ) 1 ) - : ظ نَيَارًا )الدر الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 . ) 5 ) - : ظ العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 . ) 8 ) - ظ : حدا ة الط عيد ونيا ة الطجعصد5 / 48 . ) 4 ) - مسظد التارو حن كثي1 / 581 :), كعاب الصيام,ال ن1 محى الص اب فد فضا أمي الط مظين عط حن ال لاب 5 / 812 ) 1 ) - : ظ نَيَارًا )الدر الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 . ) 5 ) - : ظ العطييد لطا فد الط طأ من الطعاند وا سانيد5 / 44 . ) 8 ) - ظ : حدا ة الط عيد ونيا ة الطجعصد5 / 48 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ) 4 ) - مسظد التارو حن كثي1 / 581 :), كعاب الصيام,ال ن1 محى الص اب فد فضا أمي الط مظين عط حن ال لاب 5 / 812 ) :البييجد: السظن4 / 518 ، والطظا ئ ص511، والطعجد اليظدز: ، وى عيف، نجلاعو حين الظ عد وع , ط : ظ الدر )الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 . ) 2 ) - : ظ )الدر الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 .) 2 ) ا شكا ال ا ال تا2 / 454 ) د د 8 ) - ظ : حدا ة الط عيد ونيا ة الطجعصد5 / 48 . ) و/ ) 4 ) - مسظد التارو حن كثي1 / 581 :), كعاب الصيام,ال ن1 محى الص اب فد فضا أمي الط مظين عط حن ال لاب 5 / 812 ) :البييجد: السظن4 / 518 ، والطظا ئ ص511، والطعجد اليظدز: ، وى عيف، نجلاعو حين الظ عد وع , ط : ظ الدر )الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 . ) 2 ) - : ظ )الدر الط عار و ا ية احن عاحد ن ر الطحعار5 / 825 .) 2 ) - م عاة الطتا ي ح مشكاة الطصاحي2 / 454 .) 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 بعاااااد عااااا ا لاااااة ومظا شاااااعيا باااااين لمبا اااااإل اث الجااااا ا ا وا وىااااا ااااا ا احااااان مساااااع وماااااا ذىااااائ اليو ال طي ر ى ال اج السز ال ؼيو اح س ر طو احن مسع . و اعمن الخاتمة بعاااااد عااااا ا لاااااة ومظا شاااااعيا باااااين لمبا اااااإل اث الجااااا ا ا وا وىااااا ااااا ا احااااان مساااااع وماااااا ذىااااائ اليو ال طي ر ى ال اج السز ال ؼيو اح س ر طو احن مسع . و اعمن الخاتمة ا لحطااااد هلل فاااااد ا ولاااااى اص اااا ة ولاااااو الحكااااان وإلياااااو جاااا ا مااااا كماااااو, والصااااالة والساااااالم عماااااى سااااايدنا دمحم وعمااااى آلااااو وأصااااحابو أجطعااااين مااااا أ صاااامة عااااين حظ اااا او سااااطعة اذث ب باااا , اااان بحطااااد بعااااد ال ياااااد الكبياااااا الااااااسز حسلعااااااو معطععااااااا باااااا راء الصااااااحاحد احاااااان مسااااااع ر واحااااااد ظيتااااااة وصااااااا بو, يااااااإل از ت اااااا فا وعمطا و ا عا ول ك . ا عالم : أىن الظعا العد صمة الييا فد ال سالة ظجماااااة فاااااد اسااااال ال ساااااالة حاااااين الطساااااا العاااااد اااااال حياااااا ا ماااااام احاااااد ظيتاااااة وصاااااا بيو لمصاااااحاحد ال ميااااا احااااان مساااااع ر , اععبااااا ت الطساااااالة العاااااد اااااال حياااااا ا اااااد ا طاااااة حااااان مساااااع ر ط فاااااا وجم اااااا من ال الس فد : الطسالة فكاث ال الس مبظيا عمى ا لة العد دميا ك مظين, وىد كال د - مة صيام الويَ و م التسثز شَ كُّ ؼثيوث مثنو رَمَضَ اثَ واو اا احن مسع ر د عظو . : فكاااااث ال ااااالس جاااا از صَ اااا و م َاااا و مث الشت ااااك ث حثظثيتااااةث العتلَاااا ُّاث فااااد التاااا والظتاااا ، واااااو اااااا عَلاااااء وَعطاااا حااااان عبد الوعَمثيم ، واح ظيتة . ، ودمحم حن الحسن الشيباند ر طين ث دحمد - مااااة الصاااايام إث ر ثاااادَ َااااا و مَ الشت ااااك ث الويثااااالَ اَ بَعواااادَ الااااامتوَااث أَوو َبومَااااو فَي اااا َ لثمت يومَااااةث الوط سو اااااعَجوبَمَةث وَ َ َك اااا ث ذَلثااااكَ الويَااااا و م مثنو رَمَضَ اثَ ، وى ا احن مسع . -فكااااااث ال اااااالس جااااا از الصااااا م اث ر ثااااادَ َااااا و مَ الشت اااااك ث الويثاااااالَ اَ إثو كَااااااثَ َبوااااا َ الااااامتوَااث فَي ااااا َ لثمتيومَاااااةث الوطَا ث ااااايَةث وَإثثو كَاثَ بَعودَ المتوَااث فَي َ لثمت يومَةث الوط سو عَجوبَمَةث . وى ا احد س ر طو 728 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 المرادر المرادر  . الج آث الك ين 1 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 ا كااااااام ال ساااااالى: عبااااااد الحااااااا حاااااان عبااااااد الاااااا طن حاااااان عبااااااد حاااااان الحسااااااين حاااااان سااااااعيد إحااااااا اهين ا ز ز، ا ندلساااااااااد ا ااااااااابيمد، الطعااااااااا وس بااااااااااحن ال ااااااااا اط , :الطعااااااااا فى281 ىاااااااااا, حجياااااااااا: طااااااااادز السمتد، صبحد السام ا د , السظة 1412 ىا- 1222 م - الططمكة الع اية السع ة. 5 . العطيياااااد لطاااااا فاااااد الط طاااااأ مااااان الطعااااااند وا ساااااانيد: أحااااا عطااااا سااااا حااااان عباااااد حااااان دمحم حااااان عباااااد :الب حن عاصن الظط ز الج طبد الطع فى428 .ىا, حجيا: مصلتى حن أ طد العم ز 8 . الحاااااااااوز ال كبياااااااا : أحاااااااا الحساااااااان عمااااااااد حاااااااان دمحم حاااااااان دمحم حاااااااان بياااااااائ البصاااااااا ز البهاااااااادا ز، الشاااااااايي بالطااااااااور ز, حجياااااااا: الشاااااااي عماااااااد دمحم معااااااا- :الشاااااااي عاااااااا ا أ طاااااااد عباااااااد الط جااااااا , الطعااااااا فى 421ىا, ط1 , السظة1412 ىا- 1222 م– .لبظاث 4 . الاااااادر الط عااااااار و ا ااااااية احاااااان عاحااااااد ن ر الطحعااااااار): حاااااان عاحااااااد ن، دمحم أمااااااين حاااااان عطاااااا حاااااان عبااااااد :العميم عاحد ن الدمشجد الحظتد, الطع فى1525 ,ىا ط5 , 1415 ىا- 1225 .م, حي وت 2 . السااااااظن الكبااااااا ل: أ طاااااااد حاااااان الحساااااااين حااااااان عمااااااد حااااااان م ساااااااى ال سو اااااا َ ووجث ز ال اسااااااااند، أحااااااا بكااااااا :البييجاااااااد, الطعااااااا فى428ىاااااااا, حجياااااااا: دمحم عباااااااد الجاااااااا ر علاااااااا, ,ط8 , الساااااااظة1454 ىاااااااا- 5118 2 . التجو ا سالمد وأ لعو لمم يمد : أ. . وَهوبَة حن مصلتى المُّ َيومثد ، ط4, مشا ,ال.جاى ة 8 . الطهظاااااااد حااااااان دامااااااا ة : أحااااااا دمحم م فاااااااا الاااااااد ن عباااااااد حااااااان أ طاااااااد حااااااان دمحم حااااااان داماااااااة ال طااااااااعيمد :الطجدسااااد ثاااان الدمشااااجد الحظبمااااد، الشاااايي باااااحن دامااااة الطجدسااااد, الطعاااا فى251 ىااااا, السااااظة1888 ىااااا - 1228م , الجاى ة . 8 . حدا اااااة الط عياااااد ونيا اااااة الطجعصاااااد: أحااااا ال لياااااد دمحم حااااان أ طاااااد حااااان دمحم حااااان أ طاااااد حااااان ر اااااد الج طباااااد :الشيي باحن ر د الحتيد, الطع فى222ىا. 2 . حاااادا الصااااظا فااااد ياااائ الشاااا ا : عااااالء الااااد ن، أحاااا بكاااا حاااان مسااااع حاااان أ طااااد الكاساااااند الحظتااااد :, الطع فى288ىا, ط1, ا لثانية، السظة1412ىا. :, الطع فى288ىا, ط1, ا لثانية، السظة1412ىا. 11 . ساااااااظن أحاااااااد او : أحااااااا او ساااااااميطاث حااااااان ا اااااااعإل حااااااان إساااااااحا , حجياااااااا: دمحم محياااااااد الاااااااد ن عباااااااد :الحطيد, الطع فى582 ىا– .حي وت 11 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 سااااظن الاااادار لظد : أحاااا الحساااان عمااااد حاااان عطاااا حااااان أ طااااد حاااان مياااادز حاااان مسااااع حاااان الظعطاااااث حااااان : ظاااااااار البهااااااادا ز الااااااادار لظد الطعااااااا فى882ىاااااااا, : حجياااااااا اااااااعيئ ا رنااااااا وط، سااااااان عباااااااد الطاااااااظعن ، مبد ط1, السظة 1454 ىا- 5114 م- لبظاث. 15 . , اااااا ح صااااااحي الب ااااااارز حاااااان بلاااااااا: احاااااان بلاااااااا أحاااااا الحساااااان عمااااااد حاااااان ماااااا حاااااان عبااااااد الطمااااااك :الطعاااااا فى442 ", , ىااااااا , حجيااااااا: أحاااااا طااااااين اساااااا حاااااان إحاااااا اهين5 , السااااااظة1458 ىااااااا- 5118 م- . ال يا 11 . ساااااااظن أحاااااااد او : أحااااااا او ساااااااميطاث حااااااان ا اااااااعإل حااااااان إساااااااحا , حجياااااااا: دمحم محياااااااد الاااااااد ن عباااااااد :الحطيد, الطع فى582 ىا– .حي وت 11 . سااااظن الاااادار لظد : أحاااا الحساااان عمااااد حاااان عطاااا حااااان أ طااااد حاااان مياااادز حاااان مسااااع حاااان الظعطاااااث حااااان : ظاااااااار البهااااااادا ز الااااااادار لظد الطعااااااا فى882ىاااااااا, : حجياااااااا اااااااعيئ ا رنااااااا وط، سااااااان عباااااااد الطاااااااظعن 11 . ساااااااظن أحاااااااد او : أحااااااا او ساااااااميطاث حااااااان ا اااااااعإل حااااااان إساااااااحا , حجياااااااا: دمحم محياااااااد الاااااااد ن عباااااااد :الحطيد, الطع فى582 ىا– .حي وت 11 . سااااظن الاااادار لظد : أحاااا الحساااان عمااااد حاااان عطاااا حااااان أ طااااد حاااان مياااادز حاااان مسااااع حاااان الظعطاااااث حااااان : ظاااااااار البهااااااادا ز الااااااادار لظد الطعااااااا فى882ىاااااااا, : حجياااااااا اااااااعيئ ا رنااااااا وط، سااااااان عباااااااد الطاااااااظعن ، مبد ط1, السظة 1454 ىا- 5114 م- لبظاث. 15 . , اااااا ح صااااااحي الب ااااااارز حاااااان بلاااااااا: احاااااان بلاااااااا أحاااااا الحساااااان عمااااااد حاااااان ماااااا حاااااان عبااااااد الطمااااااك :الطعاااااا فى442 ", , ىااااااا , حجيااااااا: أحاااااا طااااااين اساااااا حاااااان إحاااااا اهين5 , السااااااظة1458 ىااااااا- 5118 م- . ال يا 722 722 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 18 . اااااا ح معاااااااند اصثااااااار :أحاااااا ج عتاااااا أ طااااااد حاااااان دمحم حاااااان سااااااالمة حاااااان عبااااااد الطمااااااك حاااااان ساااااامطة ا ز ز الح ااااا ز الطصااااا ز الطعااااا وس باللحااااااوز, :الطعااااا فى851ىاااااا, ججاااااو و ااااادم لاااااو: دمحم زىااااا ز الظ اااااار- )دمحم سيد جا الحا , ط1, السظة - 1414 ،ىا1224 م- من عمطاء ا زى الش يف. 14 . صااااحي مساااامن: مساااامن حاااان الح اااااج أحاااا الحساااان الجشااااي ز ,الظيساااااح رز, حجيااااا: دمحم فاااا ا عبااااد البااااا د :الطع فى521 ىا- . حي وت احمد الح ااااا ز الطصااااا ز الطعااااا وس باللحااااااوز, :الطعااااا فى851ىاااااا, ججاااااو و ااااادم لاااااو: دمحم زىااااا ز الظ اااااار- )دمحم سيد جا الحا , ط1, السظة - 1414 ،ىا1224 م- من عمطاء ا زى الش يف. 14 . صااااحي مساااامن: مساااامن حاااان الح اااااج أحاااا الحساااان الجشااااي ز ,الظيساااااح رز, حجيااااا: دمحم فاااا ا عبااااد البااااا د :الطع فى521 ىا- . حي وت 12 . عطاااادة الجااااارز اااا ح صااااحي الب ااااارز: أحاااا دمحم محطاااا حاااان أ طااااد حاااان م سااااى حاااان أ طااااد حاااان سااااين الهيعاحى الحظتى حدر الد ن العيظى , :الطع فى822ىا , - حي وت. 12 . م: عااااة الطتاااا ي ااا ح مشاااكاة الطصااااحي أحااا الحسااان عبياااد حااان دمحم عباااد الساااالم حااان ااااث دمحم :حاااااان أمااااااااث حاااااان ساااااااام الاااااااد ن ال طاااااااند الطبااااااااركت رز, الطعاااااا فى1414ىاااااااا, ط8 , الساااااااظة1414 ،ىا1284 م. 18 . مساااا ا مااااام أ طااااد روا ااااة احظااااو أحااااد التضاااا صاااال : أحاااا عبااااد أ طااااد حاااان دمحم حاااان ظباااا حاااان :ىالا حن أسد الشيباند, الطع فى541ىا- .اليظد 18 . م اىاااائ ال مياااا فااااد اااا ح م عصاااا مياااا : ااااط الااااد ن أحاااا عبااااد دمحم حاااان دمحم حاااان عبااااد الاااا طن :الل احمساااااااااد الطه ااااااااااد، الطعااااااااا وس بالحلااااااااااب ال ُّعيظاااااااااد الطاااااااااالكد, الطعااااااااا فى224ىاااااااااا, ط8 , الساااااااااظة 1415 ىا- 1225 . م 12 . عطاااادة الجااااارز اااا ح صااااحي الب ااااارز: أحاااا دمحم محطاااا حاااان أ طااااد حاااان م سااااى حاااان أ طااااد حاااان سااااين الهيعاحى الحظتى حدر الد ن العيظى , :الطع فى822ىا , - حي وت. 12 . م: عااااة الطتاااا ي ااا ح مشاااكاة الطصااااحي أحااا الحسااان عبياااد حااان دمحم عباااد الساااالم حااان ااااث دمحم :حاااااان أمااااااااث حاااااان ساااااااام الاااااااد ن ال طاااااااند الطبااااااااركت رز, الطعاااااا فى1414ىاااااااا, ط8 , الساااااااظة1414 ،ىا1284 م. 18 . ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 مساااا ا مااااام أ طااااد روا ااااة احظااااو أحااااد التضاااا صاااال : أحاااا عبااااد أ طااااد حاااان دمحم حاااان ظباااا حاااان :ىالا حن أسد الشيباند, الطع فى541ىا- .اليظد 18 . م اىاااائ ال مياااا فااااد اااا ح م عصاااا مياااا : ااااط الااااد ن أحاااا عبااااد دمحم حاااان دمحم حاااان عبااااد الاااا طن :الل احمساااااااااد الطه ااااااااااد، الطعااااااااا وس بالحلااااااااااب ال ُّعيظاااااااااد الطاااااااااالكد, الطعااااااااا فى224ىاااااااااا, ط8 , الساااااااااظة 18 . مساااا ا مااااام أ طااااد روا ااااة احظااااو أحااااد التضاااا صاااال : أحاااا عبااااد أ طااااد حاااان دمحم حاااان ظباااا حاااان :ىالا حن أسد الشيباند, الطع فى541ىا- .اليظد : ى ا يب د, ا ىالا حن أ ى .ا ي 18 . م اىاااائ ال مياااا فااااد اااا ح م عصاااا مياااا : ااااط الااااد ن أحاااا عبااااد دمحم حاااان دمحم حاااان عبااااد الاااا طن :الل احمساااااااااد الطه ااااااااااد، الطعااااااااا وس بالحلااااااااااب ال ُّعيظاااااااااد الطاااااااااالكد, الطعااااااااا فى224ىاااااااااا, ط8 , الساااااااااظة 1415 ىا- 1225 . م 12 . نصاااااائ ال ا ااااااة: جطاااااااا الااااااد ن أحاااااا دمحم عبااااااد حاااااان ساااااا حاااااان دمحم الميمعااااااد, حجيااااااا: دمحم,ع امااااااة :الطع فى825ىا, ط1 , السظة1418 /ىا1228 . م, السع ة 12 . نصاااااائ ال ا ااااااة: جطاااااااا الااااااد ن أحاااااا دمحم عبااااااد حاااااان ساااااا حاااااان دمحم الميمعااااااد, حجيااااااا: دمحم ,ع امااااااة :الطع فى825ىا, ط1 , السظة1418 /ىا1228 . م, السع ة 51 . الباااادر العطااااام اااا ح حماااا غ الطاااا ام : الحسااااين حاااان دمحم حاااان سااااعيد الطه اااااد ت1112 )ىااااا , حجيااااا: عبااااد ال طن سن محط– مص. 51 . صاااااحي الب اااااارز: دمحم حااااان إساااااطاعي أحااااا عباااااد الب اااااارز ال عتاااااد, حجياااااا: دمحم زىيااااا حااااان ناصااااا الظ اص , ط1 , السظة1455 .ىا, , مشا /م 51 . الباااادر العطااااام اااا ح حماااا غ الطاااا ام : الحسااااين حاااان دمحم حاااان سااااعيد الطه اااااد ت1112 )ىااااا , حجيااااا: عبااااد ال طن سن محط– مص. 51 . صاااااحي الب اااااارز: دمحم حااااان إساااااطاعي أحااااا عباااااد الب اااااارز ال عتاااااد, حجياااااا: دمحم زىيااااا حااااان ناصااااا الظ اص , ط1 , السظة1455 .ىا, , مشا ال طن سن محط– مص. 51 . صاااااحي الب اااااارز: دمحم حااااان إساااااطاعي أحااااا عباااااد الب اااااارز ال عتاااااد, حجياااااا: دمحم زىيااااا حااااان ناصااااا الظ اص , ط1 , السظة1455 .ىا, , مشا 58 . References  Great Quran.  Great Quran. 1. The Middle Judgments: Abdul Haq bin Abdul Rahman bin Abdullah bin Al Hussein bin Saeed Ibrahim Al-Azdi, Andalusian Ashbili, known as Ibn Al-Kharrat,( 581 A.H), investigation: Hamdi Salafi, Subhi al- Samarrai, 1416 A.H – 1995 A.D - Saudi Arabia . 2. The preface for what in Moatta from meanings and Asanids: Abu Omar Ibn Yusuf bin Abdullah bin Mohammed bin Abdul Bar bin Asim al- Namri al-Qurtubi (463 A.H), investigation: Mustafa bin Ahmed Al- Alawi. 3. The large container: Abu Hassan Ali bin Mohammed bin Mohammed bin Habib al-Basri al-Baghdadi, famous by Al-Mawardi, investigation: Sheikh Ali Mohammed Moawad - Sheikh Adel Ahmed Abdul Muqod, (450 A.H), E1, 1419A.H – 1999 A.D - Lebanon. 3. The large container: Abu Hassan Ali bin Mohammed bin Mohammed bin Habib al-Basri al-Baghdadi, famous by Al-Mawardi, investigation: Sheikh Ali Mohammed Moawad - Sheikh Adel Ahmed Abdul Muqod, (450 A.H), E1, 1419A.H – 1999 A.D - Lebanon. 4. Al-Durr al-Mukhtar and footnote of Ibn Abidin (Confused Replay): Ibn Abdeen, Muhammad Amin bin Omar bin Abdul-Aziz, Abdin al-Damshqi al-Hanafi, I 2,(1252 A.H), , 1412 A.H, 1992A.D, Beirut. 4. Al-Durr al-Mukhtar and footnote of Ibn Abidin (Confused Replay): Ibn Abdeen, Muhammad Amin bin Omar bin Abdul-Aziz, Abdin al-Damshqi al-Hanafi, I 2,(1252 A.H), , 1412 A.H, 1992A.D, Beirut. 5. 5. The Great Sunan: Ahmad ibn al-Husayn ibn Ali ibn Musa al- Khasrawarjidi al-Kharasani, Abu Bakr al-Bayhaqi,(458 A.H), investigation: Mohamed Abdelkader Atta, E3, 1424 A.H-2003 A.D. 5. 5. The Great Sunan: Ahmad ibn al-Husayn ibn Ali ibn Musa al- Khasrawarjidi al-Kharasani, Abu Bakr al-Bayhaqi,(458 A.H), investigation: Mohamed Abdelkader Atta, E3, 1424 A.H-2003 A.D. 6. Islamic Jurisprudence and its evidence for al-Zuhaili: Dr.Wahba Ben Mustapha Al-Zahili, E4, Damascus, Cairo. 6. Islamic Jurisprudence and its evidence for al-Zuhaili: Dr.Wahba Ben Mustapha Al-Zahili, E4, Damascus, Cairo. 7. Al-Mugni of Ibn Qudaamah: Abu Muhammad Muwaffaq al-Din Abdullah ibn Ahmad Ibn Muhammad Ibn Qudaamah al-jumaeeli al- Maqdisi and then al-Damaschi al-Hanbali, the famous Ibn Qudaamah al- Maqdisi,( 620 A.H), 1388 A.H - 1968 A.D, Cairo. 7. Al-Mugni of Ibn Qudaamah: Abu Muhammad Muwaffaq al-Din Abdullah ibn Ahmad Ibn Muhammad Ibn Qudaamah al-jumaeeli al- Maqdisi and then al-Damaschi al-Hanbali, the famous Ibn Qudaamah al- Maqdisi,( 620 A.H), 1388 A.H - 1968 A.D, Cairo. 8. ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 52 . نيااااااا ا وطاااااااار: دمحم حااااااان عماااااااد حااااااان دمحم حااااااان عباااااااد الشااااااا كاند اليطظاااااااد, حجياااااااا: عصاااااااام الاااااااد ن :الصبابلد, الطع فى1521ىا, ط1 , السظة1418 ىا- 1228 م– . مص 52 . نيااااااا ا وطاااااااار: دمحم حااااااان عماااااااد حااااااان دمحم حااااااان عباااااااد الشااااااا كاند اليطظاااااااد, ح :الصبابلد, الطع فى1521ىا, ط1 , السظة1418 ىا- 1228 م– . مص 52 . نيااااااا ا وطاااااااار: دمحم حااااااان عماااااااد حااااااان دمحم حااااااان عباااااااد الشااااااا كاند اليطظاااااااد, حجياااااااا: عصاااااااام الاااااااد ن :الصبابلد, الطع فى1521ىا, ط1 , السظة1418 ىا- 1228 م– . مص ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 مساااااااااظد أحاااااااااد او الليالساااااااااد : أحااااااااا او ساااااااااميطاث حااااااااان او حااااااااان ال اااااااااارو الليالساااااااااد البصااااااااا ل , :الطعاااااا فى514ىاااااااا, حج يااااااا ا: الاااااادكع ر دمحم حااااااان عباااااااد الطحساااااان الع كاااااااد , ط1 , الساااااااظة ، 1412 ىاااااااا- 1222 م– مص. 58 . مساااااااااظد أحاااااااااد او الليالساااااااااد : أحااااااااا او ساااااااااميطاث حااااااااان او حااااااااان ال اااااااااارو الليالساااااااااد البصااااااااا ل , :الطعاااااا فى514ىاااااااا, حج يااااااا ا: الاااااادكع ر دمحم حااااااان عباااااااد الطحساااااان الع كاااااااد , ط1 , الساااااااظة ، 1412 ىاااااااا- 1222 م– مص. 54 . مسااااااااظد أ طااااااااد ط ال سااااااااالة : أ طااااااااد حاااااااان ظباااااااا, حج ياااااااا ا : ااااااااعيئ ا رناااااااا وط وآ اااااااا وث ط5, السااااااااظة 1451 ، ىا1222م. 54 . مسااااااااظد أ طااااااااد ط ال سااااااااالة : أ طااااااااد حاااااااان ظباااااااا, حج ياااااااا ا : ااااااااعيئ ا رناااااااا وط وآ اااااااا وث ط5, السااااااااظة 1451 ، ىا1222م. 52 . : مسااااظد التااااارو أحاااا التااااداء إسااااطاعي حاااان عطاااا حاااان كثياااا الج ااااد البصاااا ز ثاااان الدمشااااجد , :الطعاااا فى 884ىا, ي حج ا: عبد الطعلد مع د , ط1, السظة 1411 ىا- 1221م – الطظص رة . 52 . : مسااااظد التااااارو أحاااا التااااداء إسااااطاعي حاااان عطاااا حاااان كثياااا الج ااااد البصاااا ز ثاااان الدمشااااجد , :الطعاااا فى 884ىا, ي حج ا: عبد الطعلد مع د , ط1, السظة 1411 ىا- 1221م – الطظص رة . References The beginning of the diligent and the end of the frugal: Abu al-Walid Muhammad bin Ahmed bin Mohammed bin Ahmed bin Rashid Al- Qurtubi famous Ibn Rushd grandson,(595 A.H). 9. Bada'id al-Sanayeh in the Arrangement of the laws: Alaa al-Din, Abu Bakr bin Masood bin Ahmad al-Kasani Hanafi,(587 A.H), E2, 1406 A.H. 10. Sunan Abi Dawood: Abu Dawud Sulayman ibn al-Ash'ath ibn Ishaq, investigation: Muhammad Mohieddin Abdul Hamid,(275 A.H) - Beirut. 11. Sunan al-Daraqutni: Abu al-Hasan Ali bin Omar bin Ahmed bin Mahdi bin Masoud bin al-Nu'man bin Dinar al-Baghdadi Al-Darkutni (385 A.H), investigation: Shuaib Al-Arnaout, Hassan Abdel-Moneim Shalaby, E1, 1424A.H – 2004A.D - Lebanon. 222 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 12. Explanation of Saheeh al-Bukhaari of Ibn Battal: Ibn Battal Abu Al- Hasan Ali bin Khalaf bin Abdul-Malik,(449 A.H), investigation: Abu Tamim Yasser bin Ibrahim,E2, 1423A.H - 2003 A.D - Riyadh. 12. Explanation of Saheeh al-Bukhaari of Ibn Battal: Ibn Battal Abu Al- Hasan Ali bin Khalaf bin Abdul-Malik,(449 A.H), investigation: Abu Tamim Yasser bin Ibrahim,E2, 1423A.H - 2003 A.D - Riyadh. 12. Explanation of Saheeh al-Bukhaari of Ibn Battal: Ibn Battal Abu Al- Hasan Ali bin Khalaf bin Abdul-Malik,(449 A.H), investigation: Abu Tamim Yasser bin Ibrahim,E2, 1423A.H - 2003 A.D - Riyadh. 13. Explanation of the meanings of the effects: Abu Jaafar Ahmed bin Mohammed bin Salama bin Abdul Malik bin Salama Al-Azdi Al-Hajri Egyptian known Tahawi,( 321 A.H), achieved and presented to him: (Mohammad Zuhri Najjar - Mohammed Sayed Jad Al Haq), E1, 1414 A.H, 1994A.D,a scholar of Al-Azhar Al-Sharif. 13. Explanation of the meanings of the effects: Abu Jaafar Ahmed bin Mohammed bin Salama bin Abdul Malik bin Salama Al-Azdi Al-Hajri Egyptian known Tahawi,( 321 A.H), achieved and presented to him: (Mohammad Zuhri Najjar - Mohammed Sayed Jad Al Haq), E1, 1414 A.H, 1994A.D,a scholar of Al-Azhar Al-Sharif. 14. Saheeh Muslim: Muslim Ibn al-Hajjaj Abul Hassan al-Qushiri al- Nisaburi,(261A.H), investigation: Mohamed Fouad Abdel Baqi, Beirut. 15. Mayor of the Continental Explanation of Sahih al-Bukhaari: Abu Muhammad Mahmud bin Ahmad bin Musa bin Ahmed bin Hussein al- Ghaitabi al-Hanafi Badr al-Din al-Aini,(855 A.H), Beirut. 16. The keys to the explanation of lamp niches: Abulhasan Obaidullah bin Mohammed Abdul Salam bin Khan Mohammed bin Aman Allah bin Hossam Eddin Rahmani Mubarakfuri,(1414 A.H), E3, 1404 A.H, 1984 A.D. 17. References Issues of Imam Ahmad narrated by his son Abi al-Fadl Saleh: Abu Abdullah Ahmad ibn Muhammad ibn Hanbal bin Hilal bin Asad al- Shaibani,( 241A.H)-India. 18. Mahaweb al-Jaleel in explaining Khalil's brief: Shams al-Din Abu Abdullah Muhammad ibn Muhammad ibn Abd al-Rahman al-Trabulsi al- Mughrabi, known as al-Hattab al-Ra'ayni al-Maliki,(954 A.H), E3, 1412 A.H – 1992A.D. 19. Naseb Al-Rayah: Jamal al-Din Abu Muhammad Abdullah bin Yusuf bin Mohammed al-Zaili,(762A.H), investigation: Mohammed Awama,E1, 1418 A.H / 1997A.D, Saudi Arabia. 20. Al-Badr Al-Tamam Explanation of Bologh Al-Maram: Abu Al-Hussain Bin Muhammed bin Saeed Al-Magribi,(1119 A.H), investigation: Abdel Rahman Hassan Mahmoud - Egypt. 21. Saheeh al-Bukhari: Muhammad ibn Isma'il Abu Abdullah al-Bukhaari al- Jaafi, investigation: Muhammad Zuhair bin Nasser al-Nasser, E1, 1422 A.H, Damascus. 22. The jurisprudence of worship on the Maliki doctrine: Kawkab Abeed,E1, 1406 A.H - 1986 A.D - Syria. 23. Musnad Abi Dawood Al-Tialesi: Abu Daoud Sulaiman bin Dawood bin Jaroud Al-Taylasi Al-Basri,(204 A.H), investigation: Dr. Mohammed bin Abdul Mohsen al-Turki, E1, 1419 A.H – 1999 A.D - Egypt. gyp 24. Musnad Ahmed: Ahmed bin Hanbal, investigation: Shoaib Arnaout and others E2, 1420 A.H, 1999 A.D. 25. Al-Faruq Musnad: Abu al-Fidaa Isma'il ibn 'Umar ibn Qatheer al-Qurashi al-Basri and then al-Dimashqi,( 774 A.H), investigation: Abd al-Muti Qalaji, E1, 1411 A.H-1991A.D. 227 ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 26. Neil Al-Awtar: Muhammad bin Ali bin Mohammed bin Abdullah Al- Shawkani, Yemeni (1250A.H), investigation: Essam al-Din Sabati,E1, 1413 A.H – 1993A.D - Egypt. ISLAMIC SCIENCES JOURNAL (2019) Vol 10 (4) :285-303 26. Neil Al-Awtar: Muhammad bin Ali bin Mohammed bin Abdullah Al- Shawkani, Yemeni (1250A.H), investigation: Essam al-Din Sabati,E1, 1413 A.H – 1993A.D - Egypt. 222

https://openalex.org/W4244566152

https://bmcpsychiatry.biomedcentral.com/track/pdf/10.1186/s12888-020-02602-3

English

Mental distress, perceived need and barriers to receive professional mental health care among university students

Research Square (Research Square)

cc-by

Open Access Mental distress, perceived need, and barriers to receive professional mental health care among university students in Ethiopia segid Negash*, Matloob Ahmed Khan, Girmay Medhin, Dawit Wondimagegn and Mesfin Araya Assegid Negash*, Matloob Ahmed Khan, Girmay Medhin, Dawit Wondimagegn and Mesfin Ar © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Abstract 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. * Correspondence: [email protected] Department of Psychiatry, College of Health Sciences, School of Medicine, Addis Ababa University, Addis Ababa, Ethiopia © The Author(s). 2020 Open Access This article is licensed under a Creative C which permits use, sharing, adaptation, distribution and reproduction in any appropriate credit to the original author(s) and the source, provide a link to t changes were made. The images or other third party material in this article a licence, unless indicated otherwise in a credit line to the material. If material licence and your intended use is not permitted by statutory regulation or exc permission directly from the copyright holder. To view a copy of this licence, The Creative Commons Public Domain Dedication waiver (http://creativecom data made available in this article, unless otherwise stated in a credit line to t * Correspondence: [email protected] Department of Psychiatry, College of Health Sciences, School of Medicine, Addis Ababa University, Addis Ababa, Ethiopia * Correspondence: [email protected] Department of Psychiatry, College of Health Sciences, School of Medicine, Addis Ababa University, Addis Ababa, Ethiopia Negash et al. BMC Psychiatry (2020) 20:187 https://doi.org/10.1186/s12888-020-02602-3 Negash et al. BMC Psychiatry (2020) 20:187 https://doi.org/10.1186/s12888-020-02602-3 * Correspondence: [email protected] Department of Psychiatry, College of Health Sciences, School of Medicine, Addis Ababa University, Addis Ababa, Ethiopia Abstract Background: There is limited evidence on the extent of the perceived need and barriers to professional mental health service delivery to university students with mental distress in low- and middle-income countries (LMICs). This study was designed to assess the prevalence of mental distress, perceived need for professional mental health care and barriers to the delivery of services to affected undergraduate university students in Ethiopia. Methods: A multi-stage sampling technique was used to recruit 1135 undergraduate university students. Symptoms of mental distress were evaluated using the Self-Reported Questionnaire (SRQ-20) and a score of above seven was used to identify positive cases. The perceived need for professional mental health care was assessed using a single ‘yes or no’ response item and barriers to mental health care were assessed using Barriers to Access to Care Evaluation (BACE- 30) tool. Percentage, frequency, mean, and standard deviation were employed to summarize demographic characteristics of the participants and to identify common barriers to mental health care service. Moreover, the association of demographic variables with total mean scores of BACE-III sub-scales was modeled using multiple linear regression. Results: The prevalence of mental distress symptoms was 34.6% and the perceived need for professional mental health care was 70.5% of those with mental distress. The top five barriers to receiving professional mental health service were (a) thinking the problem would get better with no intervention, (b) being unsure where to go to get professional help, (c) wanting to solve the problem without intervention, (d) denying a mental health problem existed, and (e) preferring to get alternative forms of mental care. Coming from a rural background, being a second and fourth- year student, and a family history of mental illness were significantly associated with barriers to receive professional mental health service. Conclusion: The high prevalence of mental distress, the paucity of mental health care, and the report of barriers to access what professional mental health care there is among Ethiopian undergraduate students is a call to address the disparity. Keywords: Mental distress, Perceived need, Barrier, Professional mental health care, Ethiopia, Undergraduate students © The Author(s). Background On the other hand, having high social support and enough pocket money are protect- ive factors from mental distress [32]. Mental distress has a negative impact on university students’ academic performance [29]. Evidence shows that mentally distressed students scored poor examin- ation result compared with non-distressed students [33]. Although mental distress has such impact, the treatment gap remains large ranging from 37 to 84% [34]. This treatment gap is also high among Ethiopian university students, where majority of the students receive treat- ment from informal sources such as family, friends, rela- tives, and religious leaders [35]. There are several barriers that hinder students with mental distress from receiving mental health services. Among these: (i) receiv- ing help from friends or family; (ii) preferring to manage mental illness by self; (iii) normalizing mental illness; (iv) thinking that mental illness would get better by itself [36]; (v) lack of perceived need; (vi) being unaware of the existence of professional mental health services; (vii) fear of stigma, concerns about privacy; (viii) skepticism about treatment effectiveness; (iv) socio-economic prob- lem [34]; and (x) denying mental illness [37]. As in any other LMICs, the prevalence and burden of anxiety and depression in Ethiopia is high. For example, a systematic review and meta-analysis study reported that the pooled prevalence of these disorders is 22% [14], which is associated with risk factors such as food insecurity [15], poverty, violence, migration, and sub- stance use [16]. The burden of depression alone contrib- utes to about 6.5% of the burden of diseases [16], however, as noted above few people are able to receive formal mental health services. Evidence showed that the pooled prevalence of help-seeking behaviors of people with depression is 38% [17]. In Ethiopia, most people with mental illness first contact non-professional care providers such as religious leaders and herbalists [18]. But, if the patient remains affected, he/she will go to western trained psychiatric care providers [18]. The psy- chiatric services are mainly concentrated in the capital city of Ethiopia, Addis Ababa [19]. To scale-up the limited mental health services across the country, the government of Ethiopia has planned to expand 100% of mental health care by 2020 [20]. The National Mental Health Strategy was developed in 2012 by the Federal Ministry of Health aimed to decentralize and integrate mental health services at the primary health care level [19]. Background 2017. These students are adolescents, economically dependent on their family and full-time learner, they came from rural-urban backgrounds with diversified cul- tures, languages, and ethnicity. g Mental distress is among the most common type of ex- perience that accompanies mental health problem char- acterized by a mixture of different complaints such as feeling sad, worried, tense or angry [1]. Common mental disorders are a collective noun for anxiety, depression, and somatoform disorders that can adversely affect indi- viduals across the world [2, 3]. According to the World Health Organization [4] 2015 report, over 300 million (4.4%) and 264 million (3.6%) of the total word popula- tion are estimated to suffer from anxiety and depression, respectively [5]. In particular, the contribution of these disorders to the global mental health burden from low- and middle-income countries (LMICs) is high [6]. How- ever, the accessibility of mental health service is still very low, which accounts for the 76–85% treatment gap [7]. This gap has been linked to the lack of skilled human re- sources, lack of mental health policies, lack of access to mental health services, poverty, the preference for infor- mal treatments, a lack of mental health literacy, the fear of stigma, and a low commitment from funders to access the services [8–12]. As a result, the majority of people living with anxiety and depression in LMICs do not re- ceive professional mental health care [13]. The prevalence of mental distress is high among univer- sity students [21]. A cross-cultural web-based survey of 17,348 university students from 23 high-middle-and low- income countries reported that the average depression prevalence is 20% [22]. Another study also reported that the prevalence of depression and anxiety is 68.5 and 54.4%, respectively [23]. Similarly, the prevalence of these disorders ranges from 21.6–49% among Ethiopian univer- sity students [24, 25]. This high prevalence is associated with several factors including: (i) vulnerability of adoles- cence age for early onset of mental distress [26]; (ii) new identity formation [27]; (iii) challenges of being away from home for the first time [28]; and (iv) academic pressure, substance use [29], and financial difficulties [30]. More- over, family histories of mental illness, conflicts with friends, not attending religious services, and being fresh- man are risk factors for mental distress among university students in Ethiopia [31, 32]. © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Page 2 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 Study design, objectives and study period Study design, objectives and study period An institution-based cross-sectional survey was con- ducted among WSU undergraduate students from De- cember 2017 to January 2018. The objective was to estimate the prevalence of mental health problems, per- ceived need, and to identify barriers and demographic predictors to receive professional mental health care. Most studies conducted in Ethiopian universities are primarily focused on assessing the prevalence of mental distress rather than looking at the possible barriers to re- ceiving professional help. Besides of this, there is a litera- ture gap with regard to identifying the perceived need for mental health service and demographic factors asso- ciated with barriers to receive mental health care among university students in Ethiopia. Therefore, this current study is aimed to assess the prevalence of mental dis- tress, perceived need, and identify common barriers to receive professional mental health care among under- graduate students in Wolaita Sodo University (WSU). Our study also investigated the demographic predictors of the barriers to mental health care. The current study findings will fill the literature gap on professional help- Background At the university level, mental health services have been established to support students with mental health problems, although the quality of the service provided is under question. Currently, Ethiopia has 45 public universities, where 392,788 (255,657 male and 137,131female) undergraduate students enrolled in There are associations between demographic variables and perceived need for mental health care. Female stu- dents have more positive attitudes to the utilization of mental health services compared with male students; the possible explanations could be that women experience more mental distress and they give more value for sup- port received from professionals [38]. Conversely, male students are more likely to seek mental health care com- pared with female students; this might be caused by the interaction effect of other demographic variables in the Negash et al. BMC Psychiatry (2020) 20:187 Page 3 of 15 Page 3 of 15 analysis model [39]. Despite this, there is a finding reporting gender is not a predictor for seeking mental health care; possibly this is caused by insignificance gen- der difference in mean scores of depression and self- esteem [40]. Likewise, there is no gender difference in reporting barriers to receive mental health care [41]. Older students are more likely to have positive attitudes toward seeking mental health care than younger stu- dents; possibly caused by the past mental health care re- ceived [42]. Adolescents have a more positive attitude to seek mental health care than adults, because adolescents had more confidence in and better experience of using modern mental health care [39]. There are also insignifi- cant difference in reporting barriers to access mental health care based on age [41]. seeking intention, predictors, and barriers to receive mental health service among undergraduate students in LMICs. Besides this, our findings inform to adapt and study feasibility of psychological intervention for stu- dents with mental distress within Ethiopian universities with potential implication for other LMICs universities. Study setting The current study is conducted at WSU, a public univer- sity located in the Sodo town of Wolaita Sodo Zone, Southern Nations, Nationalities, and Peoples’ Regional State (SNNPR) of Ethiopia. Sodo town is located 320 km south of Addis Ababa. WSU was established in 2007 as a result of the rapid expansion of higher education in Ethiopia. The university began with an intake of 801 stu- dents (609 males and 192 females) in four faculties and sixteen departments. Currently, the university runs undergraduate and graduate programs in six colleges and five schools. During the study period, a total of 12, 028 (7321males and 4707 females) undergraduate stu- dents were registered. These students qualify to join the university by taking the Grade 12 national entrance examination prepared by Ministry of Education. WSU has two counseling offices and two health centers. There are three psychologists in the counseling offices that provide counseling services for students with mental health problems. The two health care centers are the Ottona hospital and the students’ clinic, both of which provide health care services. Ottona hospital is a referral hospital that provides health care services for the com- munity and for the students with severe mental health problem by providing medication. Class years, family history of mental illness, and sub- stance use are also reported as predictors to receive mental health care. For instance, first and fourth-years students are less likely to use mental health services compared with second and third-years students [43], al- though there is no change based on rural-urban back- grounds [44]. However, students who had personal contact with someone with a history of mental illness were significantly associated with decreased help-seeking intention; this could be possibly due to the negative ex- periences students had with a person who they know to have a mental illness [45]. Moreover, students with men- tal distress use substances to manage feeling of discom- fort, which might hinder their interest or preparedness in seeking mental health care [46]. Mental distress, pre- dictors, and barriers to receiving mental health care among university students occur globally, however, there might be higher prevalence, more complex stressors, lower help-seeking behaviors, and a higher treatment gap in LMICs compared with developed countries [47]. For example, even if the prevalence of mental distress is high among university students in LMICs, the majority of them do not receive professional mental health care [48, 49]. Measurements The survey questionnaire consisted of four parts: Demo- graphic Characteristic Questionnaire: used to document variables including participants’ sex, age, religion, ethni- city, marital status, current place of living, area where they grew up, level of the study years, family history of mental illness, and substance use. Self-Reported Questionnaire (SRQ-20): It is a screen- ing tool for mental distress developed by WHO [50]. SRQ-20 is a self-report instrument with 20 binary re- sponses (yes/no) questions. It has the potential of detect- ing cases and non-cases with sensitivity ranging from 63 to 90 and specificity ranging from 44 to 95 [51]. WHO recommends SRQ-20 as a reliable and valid instrument to detect general Common Mental Disorders [51]. It was developed specifically for use in LMICs [50]. SRQ-20 has been previously translated into Amharic language in Ethiopia, locally validated [52, 53], and used in different community [54–56] and institution-based surveys [24, 25, 32, 57] with cut-off points ≥4 [57], ≥7 [25], ≥8 [32] and ≥11 [24]. SRQ-20 has good psychometric properties (i.e. sensitivity 86% and specificity 84%) for detecting in- dividuals with mental distress in the Ethiopian popula- tion with an optimal cut-off point at ≥8 [58]. To identify cases in the current study, a cut-off point of > 7 was used based on a previous validation study of SRQ-20 in Ethiopia that resulted in good sensitivity and specificity using a cut-off point of 8 [58]. The pilot data collected from 38 undergraduate students in a similar population but in a different setting to the current study showed that the internal consistency of SRQ-20 was 0.77. For the current study, BACE-III was translated into the Amharic language by two Amharic language experts whose first language is Amharic and their second lan- guage is English. One expert who knows the subject matter translated the instrument based on the BACE-III translation guide. The masked back-translation was made by two English language experts and one mental health expert. The research team compared the back- translated instrument with the original version of BACE- III and agreed upon the consistency of the translation. The translated BACE-III was piloted on 40 undergradu- ate students in a similar population but in a different area of the current study setting. Its internal consistency was 0.85. Sample size A sample size of 1135 was estimated with an assumed prevalence rate of mental distress 40.9% [32], precision of ±3, 95% confidence interval, and 10% non-response considered. For the other two objectives (perceived need and barriers to receive mental health care) separate sam- ple sizes were not estimated. All the participants who were screened positive for mental distress (> 7) were used as the denominator to estimate the proportion of students having a perceived need for professional mental health care. Those participants who had mental distress symptoms and who did not receive mental health Page 4 of 15 Page 4 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 services from professionals in the past 3 months were eligible to be part of the study. services from professionals in the past 3 months were eligible to be part of the study. The Perceived Need for Professional Mental Health Care Questionnaire: Used to assess the perceived need for professional mental health services in the past 3 months. It has been used in the previous studies [59, 60]. The question is phrased as follows: ‘Was there a time when you thought you should see a doctor, counselor or other health professionals for your mental distress, but you did not go in the past three months?’ with the response options of Yes/No. “Yes” response im- plies the perceived need for mental health care but not received in the past 3 months, whereas “No” response implies no need for mental health care for mental dis- tress. Therefore, the perceived need for professional mental health care in this study implies the number of students who reported “Yes” option. Sampling and procedures A stratified multi-stage sampling technique was used to recruit study participants. A list of students’ names from all first to fifth years was obtained from the registrar of- fice of WSU. The first participant’s name was selected randomly; the remaining participants were selected using systematic random sampling. To accomplish, the first step was stratifying undergraduate students by their schools/colleges (six colleges and five schools). For the second step, the total sample size was allocated into the 11 strata using probability proportional to the number of the students as a measure of size. The third step was selecting participants from each school and college based on the proportion of the size of each department. The fourth step was selecting participants from first to fifth-years based on the proportion to each year. Then, the final step was randomly selecting the first participant and systematically selecting the rest participants from each level and section of the study year. p p Barriers to Access to Care Evaluation (BACE-III): BACE was originally developed to identify barriers to re- ceive professional mental health service for people with mental health problems [61]. It has 30 items to be com- pleted by the participant (self-complete measure). This instrument has good psychometric properties (i.e. valid- ity, reliability, and acceptability) [61]. BACE-III has three dimensions of potential barriers of stigma (12 items), at- titudinal (10 items) and instrumental (8 items) related. This instrument asks about a range of issues that have ever stopped, delayed or discouraged an individual from receiving professional care for a mental health problem in the past 3 months. The response scale ranges from 0 (not at all) to 3 (a lot); the higher score indicating a greater barrier. Five of the thirty items contain a fifth option: “Not applicable”. Findings for each barrier are presented in three ways: mean score for the item, barrier to any degree (the percentage of answering 1, 2 or 3) or major barrier (the percentage of answering 3) based on BACE-III manual for researchers. Measurements Those participants who answered “No” for the perceived need for mental health care measuring questionnaire were asked ‘In the past three months, did you receive help from a psychologist, doctors, friends, family, reli- gious leaders or traditional healers?’ by skipping the BACE-III questionnaire. See supplementary file 1. Ethical considerations h l l f Ethical clearance for the conduct of the study was ob- tained from the Institutional Review Board (IRB) of Addis Ababa University College of Health Sciences. In- formation sheet containing details of the research and rights of the participants was attached to the question- naire. Oral informed consent was obtained from the par- ticipants after we explained to them the purpose of the study, the participation was voluntary, and personal identifiers were not included in the questionnaires. Fi- nally, the obtained data were kept anonymous and confi- dential during all stages of the research. Data analysis Data cleaning and cross-checking were done before ana- lysis using Statistical Packages for the Social Sciences (SPSS version 20). Descriptive statistical measures (i.e. percentage, frequency, mean, and standard deviation) were employed to summarize demographic characteris- tics of the participants and to identify barriers to mental health care services. Pearson chi-square test was used to examine the association between demographic variables with mental health care seeking intention and with the five most commonly reported barriers to receive mental health services. Furthermore, multiple linear regression was also used to model the association between demo- graphic variables with a mean score of BACE-III sub- scales. Univariate regression analysis was used to identify potential candidate variables for multivariable linear re- gression with a p-value of < 0.2 by referring previous published articles [62, 63]. Then, further analysis was carried out using multivariable linear regression. The re- sult was reported as being statistically significant when- ever the p-value is less than 0.05. Training of data collectors and data collection procedures Classroom representatives served as data collectors. A half day training was given by the principal investiga- tor to data collectors about the aim of the research, the contents of data collection tools, how to approach participants, ethical issues, and responsibility of con- trolling missing data. The classroom representatives both males and females were contacted by the re- searcher through the help of their department heads, because they had cell phone numbers of each class- room representative. Then, with the assistance of the classroom representatives, the student participants came to the selected lecture halls and classrooms and the data collectors explained the aim of the study. Fi- nally, after verbal agreement was received, the data collectors started to collect the data by explaining the instructions of all questionnaires with the close super- vision of the principal investigator. To protect the confidentiality of the participants, personal identifiers were not included in the questionnaires; instead, a code was applied. Measurements After the pilot study, the authors examined the applic- ability of each question in the university set-up and no- ticed that item number 27 and 28 need some modifications. Discussion was made with a mental health expert who has experience of adapting mental Page 5 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 immediately checked the existence of incomplete and missed information before the participants left the room. health instruments. Then, question number 27 which says ‘difficulty taking time off work’ was modified as ‘dif- ficulty taking time off education’ and question number 28 which says ‘concern about what people at work might think, say or do’, was modified as ‘concern about what students might think, say or do’. The final version of the instrument was administered to students who scored > 7 on the SRQ-20 and who had a need (those who an- swered “Yes”) to receive professional mental care in the past 3 months of the study period. The internal consistency of the overall BACE-III scale after the revi- sion was 0.85, whereas for stigma sub-scale = 0.83; attitu- dinal sub-scale = 0.67 and instrumental sub-scale = 0.60. Those participants who answered “No” for the perceived need for mental health care measuring questionnaire were asked ‘In the past three months, did you receive help from a psychologist, doctors, friends, family, reli- gious leaders or traditional healers?’ by skipping the BACE-III questionnaire. See supplementary file 1. health instruments. Then, question number 27 which says ‘difficulty taking time off work’ was modified as ‘dif- ficulty taking time off education’ and question number 28 which says ‘concern about what people at work might think, say or do’, was modified as ‘concern about what students might think, say or do’. The final version of the instrument was administered to students who scored > 7 on the SRQ-20 and who had a need (those who an- swered “Yes”) to receive professional mental care in the past 3 months of the study period. The internal consistency of the overall BACE-III scale after the revi- sion was 0.85, whereas for stigma sub-scale = 0.83; attitu- dinal sub-scale = 0.67 and instrumental sub-scale = 0.60. Demographic characteristics A total of 980 undergraduate students completed the screening phase survey from the sample of 1135 students approached, yielding 86.34% response rate. One third (34.6%) of the participants had scored > 7 on SRQ-20. The majority (60.5%) were male. The age of the partici- pants ranged from 17 to 38 years with a mean age of 21.53 years (SD = 2.42). The participants were from di- verse ethnic groups, the majority were from Amhara (34.6%) and Wolaita (20.9%) ethnic groups. Regarding marital status, 82.8% were single and 95.3% were living in the campus. Over half (54.7%) were from urban back- grounds. First-year, second-year, and third-year under- graduate students took 27.7, 26.6, and 25.9% of the sample, respectively (Table 1). The data collection was carried out before the stu- dents’ final examination to control for an inflation of the prevalence of mental distress. Those who scored > 7 on SRQ-20 were asked to answer the questions about the perceived need for professional mental health care and then answer questions in the BACE-III questionnaire. Participants who answered “No” the question about the perceived need for professional mental health care skipped the BACE-III and answered why they did not need mental health treatment in the past 3 months. Fi- nally, after the participants completed the self- administered questionnaires, the data collectors Negash et al. Mental distress The prevalence of mental distress was 34.6%, indicated by 339 participants with SRQ-20 scored higher than 7. It was slightly higher (51.9%) among male students. The item-based response of the study participants to the SRQ-20 is summarized in Fig. 1. The top three fre- quently reported symptoms were: loss of interest in things (37.60%), being tired (36.90), and thought of end- ing one’s life (36.80%). The least reported symptom was handshaking/hand trembling (19.7%). Demographic characteristics BMC Psychiatry (2020) 20:187 Table 1 Demographic characteristics of the study sample (Continued) Variables Total Sample (n) % N (980) Screened positive for mental distress (n) % N (339) Participants with mental distress who have not received formal Mental Care % N (239) Fifth-year 98 (10.0) 25 (7.4) 15 (6.3) Family history of mental illness Yes 67 (6.8) 34 (10.0) 22 (9.2) No 913 (93.2) 305 (90.0) 217 (90.8) Substance use Yes 58 (5.9) 39 (11.5) 31 (13.0) No 922 (94.1) 300 (88.5) 208 (87) Table 1 Demographic characteristics of the study sample (Continued) received the service from informal sources (25.5% from religious leaders, friends, family, and traditional healers) and formal sources (4% from doctors and psychologists). There was no significant gender difference in seeking mental health service, χ2 (1) = 0.48, p = 0.49. Likewise, there were no significant differences among the remaining demographic variables in those seeking men- tal health services (Table 2). Demographic characteristics BMC Psychiatry (2020) 20:187 Page 6 of 15 Table 1 Demographic characteristics of the study sample Variables Total Sample (n) % N (980) Screened positive for mental distress (n) % N (339) Participants with not received for Sex Male 593 (60.5) 176 (51.9) 127 (53.1) Female 387 (39.5) 163 (48.1) 112 (46.9) Age Mean 21.53 21.21 21.22 SD 2.42 1.95 1.82 Minimum 17 18 18 Maximum 38 30 28 Religion Christian Orthodox 543 (55.4) 241 (71.1) 164 (68.6) Christian Protestant 330 (33.7) 50 (17.7) 46 (19.2) Islam 80 (8.2) 30 (8.8) 22 (9.2) Christian Catholic 8 (0.8) 2 (0.6) 2 (0.8) No religion 8 (0.8) 4 (1.2) 3 (1.3) Others 11 (1.1) 2 (0.6) 2 (0.8) Ethnicity Amhara 339 (34.6) 164 (48.4) 110 (46.0) Oromo 155 (15.8) 58 (17.1) 41 (17.2) Wolaita 205 (20.9) 44 (13) 35 (14.6) Gurage 80 (8.2) 24 (7.1) 15 (6.3) Tigre 24 (2.4) 11 (3.2) 8 (3.3) Sidama 58 (5.9) 9 (2.7) 7 (2.9) Hadiya 32 (3.3) 8 (2.4) 6 (2.5) Gamogofa 30 (3.1) 8 (2.4) 6 (2.5) Others 57 (5.7) 13 (3.9) 11 (4.6) Marital status Single 811 (82.8) 268 (79.1) 194 (81.2) In a relation 114 (11.6) 50 (14.7) 31 (13.0) Married but not living together 35 (3.6) 11 (3.2) 9 (3.8) Divorced 14 (1.4) 7 (2.1) 3 (1.3) Married and living together 6 (0.6) 3 (0.9) 2 (0.2) Residence In Campus 934 (95.3) 320 (94.4) 223 (93.3) Off Campus 20 (2) 8 (2.4) 7 (2.9) Both 26 (2.7) 11 (3.2) 9 (3.8) Area of growing Urban 536 (54.7) 176 (51.9) 127 (53.1) Rural 444 (45.3) 163 (48.1) 112 (46.9) Level of study year First-year 271 (27.7) 117 (34.5) 81 (33.9) Second-year 261 (26.6) 85 (25.1) 58 (24.3) Third-year 254 (25.9) 84 (24.8 67 (28.0) Fourth-year 96 (9.8) 28 (8.3) 18 (7.5) Table 1 Demographic characteristics of the study sample Variables Total Sample (n) % N (980) Screened positive for mental distress (n) % N (339) Participants with mental distress who have not received formal Mental Care % N (239) Sex Male 593 (60.5) 176 (51.9) 127 (53.1) Female 387 (39.5) 163 (48.1) 112 (46.9) Age Mean 21.53 21.21 21.22 SD 2.42 1.95 1.82 Minimum 17 18 18 Maximum 38 30 28 Religion Christian Orthodox 543 (55.4) 241 (71.1) 164 (68.6) Christian Protestant 330 (33.7) 50 (17.7) 46 (19.2) Islam 80 (8.2) 30 (8.8) 22 (9.2) Christian Catholic 8 (0.8) 2 (0.6) 2 (0.8) No religion 8 (0.8) 4 (1.2) 3 (1.3) Others 11 (1.1) 2 (0.6) 2 (0.8) Ethnicity Amhara 339 (34.6) 164 (48.4) 110 (46.0) Oromo 155 (15.8) 58 (17.1) 41 (17.2) Wolaita 205 (20.9) 44 (13) 35 (14.6) Gurage 80 (8.2) 24 (7.1) 15 (6.3) Tigre 24 (2.4) 11 (3.2) 8 (3.3) Sidama 58 (5.9) 9 (2.7) 7 (2.9) Hadiya 32 (3.3) 8 (2.4) 6 (2.5) Gamogofa 30 (3.1) 8 (2.4) 6 (2.5) Others 57 (5.7) 13 (3.9) 11 (4.6) Marital status Single 811 (82.8) 268 (79.1) 194 (81.2) In a relation 114 (11.6) 50 (14.7) 31 (13.0) Married but not living together 35 (3.6) 11 (3.2) 9 (3.8) Divorced 14 (1.4) 7 (2.1) 3 (1.3) Married and living together 6 (0.6) 3 (0.9) 2 (0.2) Residence In Campus 934 (95.3) 320 (94.4) 223 (93.3) Off Campus 20 (2) 8 (2.4) 7 (2.9) Both 26 (2.7) 11 (3.2) 9 (3.8) Area of growing Urban 536 (54.7) 176 (51.9) 127 (53.1) Rural 444 (45.3) 163 (48.1) 112 (46.9) L l f d Page 7 of 15 Negash et al. Perceived need for professional mental health care Of the 339 participants who screened positive for mental distress, 239 (127 male and 112 female) had not received mental health services in the past 3 months, because of the barriers to receive the treatment, although they de- sired mental health care as indicated in Table 3. This table shows mean scores of an individual item, standard Perceived need for professional mental health care Of 339 participants with elevated symptoms of mental distress, 70.5% (n = 239) reported a perceived need for professional mental health care in the past 3 months. The remaining 29.5% did not report a need for profes- sional mental health treatment, because they have Fig. 1 Prevalence of mental distress and its distribution of specific symptom Page 8 of 15 Negash et al. BMC Psychiatry (2020) 20:187 deviation, percentage to any degree, and major barriers to receive mental health care. There were top five bar- riers to receiving mental health care the reported per- centage was greater than 60% to any degree (sum of responses of a little, quite a lot, and a lot). The first barrier to seeking mental health care was ‘thinking the problem would get better by itself’ reported by 74.4% to any degree and 37% thought that it would act as a major barrier to receive mental health services. The second was ‘being unsure where to go to get profes- sional care’ which accounted for 71.6% to any degree and 21% reported as a major barrier. The third was ‘wanting to solve the problem on their own’, whereby 71% of the participants reported this as a barrier to any degree and 28% thought that it would act as a major bar- rier to receiving mental health care. The fourth was ‘denying a mental health problem’, where 67.4% of the participants reported this as a barrier to any degree and 38% reported it as a major barrier. The fifth was ‘prefer- ring to get alternative forms of care’ reported as any de- gree of the barrier by 67%, while 34% of the participants reported it as a major barrier to receiving mental health service. Of all the demographic variables, only a family history of mental illness had a significant association, X2 [3] = 14.48 = p = 0.01 with ‘denying mental health prob- lem’ of the top five barriers. See supplementary file 2. Perceived need for professional mental health care Of the top five barriers, the top four were attitudinal related barriers to receiving professional mental health services. The fifth, ‘being unsure of where to get profes- sional care’ was an instrumental-related sub-scale of BACE-III. As a result of, the mean score of attitudinal related barriers sub-scale of BACE-III (M = 1.26, SD = 0.68) was the highest when compared with instrumental related barriers sub-scale (M = 0.78, SD = 0.43) and stigma related barriers sub-scale (M = 0.61, SD = 0.65) of BACE-III. Predictors of attitudinal related barriers to receive professional mental health care In univariate regression analysis (Table 4), fourth-year students perceived significantly more attitudinal related barriers (β = 0.27; 95%CI = 0.24, 1.16; p = 0.003) than the fifth-year students. Multivariable analysis also showed that only fourth-year students perceived significantly more attitudinal related barriers (β = 0.27; 95% CI = 0.21, 1.14; p = 0.01) than the fifth-year students. Predictors of instrumental related barriers to receive professional mental health care In univariate regression analysis (Table 5), female stu- dents perceived significantly fewer instrumental related barriers (β = −.15; 95%CI = −.24, −.02; p = 0.02) than male students. Students from rural background per- ceived significantly more instrumental related barriers Table 2 The association between demographic variables and perceived need for professional mental health care Variables Need for professional mental health care χ 2 P- value Yes (n) No (n) Sex Female 112 51 0.48 0.49 Male 127 49 Religion Orthodox 164 77 3.02 0.39 Protestant 46 14 Muslim 22 8 Others 7 1 Ethnicity Amhara 111 53 3.77 0.88 Oromo 41 17 Wolaita 35 9 Gurage 15 9 Hadiya 6 2 Tigre 8 3 Sidama 7 2 Gamogofa 6 2 Others 10 3 Marital status Single 193 75 6.83 0.15 In a relationship 31 19 Married but no living together 2 1 Divorced 10 1 Married and living together 3 4 Residence In campus 223 97 1.89 0.39 Off campus 7 1 Both 9 2 Area of growing Rural 112 51 0.48 0.49 Urban 127 49 Level of study year First-year 81 36 5.61 0.23 Second-year 58 27 Third-year 67 17 Fourth-year 18 10 Fifth-year 15 10 Family history of mental illness No 217 88 0.61 0.44 Yes 22 12 Substance use No 208 92 1.71 0.19 Yes 31 8 deviation, percentage to any degree, and major barriers to receive mental health care. Perceived need for professional mental health care Students who reported a family history of mental illness perceived significantly more instrumental related barriers (β = 0.16; 95%CI = 0.05, 0.43; p = 0.01) than stu- dents who reported no family history of mental illness. Students who reported substance use perceived signifi- cantly more instrumental related barriers (β = 0.15; 95%CI = 0.03, 0.36; p = 0.02) than students who reported no substance use. A 1 year increase in age was associ- ated with 0.19 unit increased in instrumental related barriers to receiving mental health services (β = 0.19; 95%CI = 0.02, 0.07; p = 0.004). In multivariable analysis, students from rural background perceived significantly more instrumental related barriers (β = 0.16; 95%CI = 0.03, 0.25; p = 0.01) than students from urban back- ground. Besides this, second-year (β = 0.27; 95%CI = 0.02, 0.52; p = 0.03) and fourth-year students (β = 0.29; 95%CI = 0.19, 0.77; p = 0.001) perceived significantly more instrumental related barriers than the fifth-year students. second and fourth-year student, and a family history of mental illness were significantly associated with barriers to receive professional mental health service. The prevalence of mental distress which is reported in the present study is higher than what has been reported in the meta-analysis of the general population studies in Ethiopia [14]. Perhaps our finding is not surprising, be- cause university students are more likely than the gen- eral population to be exposed to mental stress [21]. The possible difference between individual studies reviewed in the meta-analysis [14] and the present study could be partly attributed to the discrepancy in data collection in- strument with cut-off points used, age group, and set- ting. The data collection tools used in the most individual article within the meta-analysis study were ICD-10, PHQ-9, EPDS, K10, HADS etc. with different cut-off points, but in our study we have used SRQ-20 that might be one cause for the discrepancy. The other was a difference in age group and setting, where young person experience higher mental distress compared with adults in the general population. The current prevalence of mental distress is higher than that reported in previous studies conducted among university students [24, 64]. One possible reason for the discrepancy is the difference in the cut-off values used to define mental illness [24, 64]. The other explanation for the difference is other studies did not use locally vali- dated instrument [64]. Perceived need for professional mental health care On the other hand, the present finding is lower than what was reported in previous studies in Ethiopian universities [35, 65]. The first pos- sible justification for the difference might be data collec- tion tool being used to screen mental distress [35, 65]. The second possible reason for the difference could be the timing of the data collection, where our data col- lected prior to the approaching final examinations. The present finding is comparable with a study report con- ducted in Jima University [66]. This might be resulted from similarity of the data collection tool and the cut-off points used to define mental distress. Predictors of stigma related barriers to receive professional mental health care Univariate regression analysis showed that students from rural background perceived significantly more stigma re- lated barriers (β = 0.13; 95%CI = 0.00, 0.33; p = 0.05) than students from urban background. Students who reported a history of mental illness in the family perceived signifi- cantly more stigma related barriers (β = 0.13; 95%CI = 0.01, 0.57; p = 0.05) than students who reported no fam- ily history of mental illness. Students who reported sub- stance use perceived significantly more stigma related barriers (β = 0.13; 95%CI = .00, 0.49; p = 0.05) than stu- dents who reported no substance use. A 1 year increase in age was associated 0.17 unit increased in stigma- related barriers to mental health services (β = 0.17; 95%CI = 0.02, 0.10; p = 0.01). In multivariable analysis, only fourth-year students perceived significantly more stigma related barriers (β = 0.24; 95%CI = 0.14, 1.01; p = 0.01) than the fifth-year students (Table 6). The high prevalence of perceived need for professional mental health services in the current study suggests that most students with mental distress in Wolaita Sodo University remain untreated. This may not be surprising, because most universities in LMICs are ill-equipped to provide services for students’ mental health issues [47]. Previous study also reported only a few university stu- dents receive mental health services for their mental health problems as a result of lack of appropriate ser- vices [67]. Our finding is higher than previously reported in the general population of Ethiopia. For instance, a meta-analysis study reported that the pooled prevalence of the help-seeking intention of people with depression in Ethiopia is 42% [17], which is much lower than the current finding. The possible explanations for the Perceived need for professional mental health care There were top five bar- riers to receiving mental health care the reported per- centage was greater than 60% to any degree (sum of responses of a little, quite a lot, and a lot). The first barrier to seeking mental health care was ‘thinking the problem would get better by itself’ reported by 74.4% to any degree and 37% thought that it would act as a major barrier to receive mental health services. The second was ‘being unsure where to go to get profes- sional care’ which accounted for 71.6% to any degree and 21% reported as a major barrier. The third was ‘wanting to solve the problem on their own’, whereby 71% of the participants reported this as a barrier to any degree and 28% thought that it would act as a major bar- rier to receiving mental health care. The fourth was ‘denying a mental health problem’, where 67.4% of the participants reported this as a barrier to any degree and 38% reported it as a major barrier. The fifth was ‘prefer- ring to get alternative forms of care’ reported as any de- gree of the barrier by 67%, while 34% of the participants reported it as a major barrier to receiving mental health service. Of all the demographic variables, only a family history of mental illness had a significant association, X2 [3] = 14.48 = p = 0.01 with ‘denying mental health prob- lem’ of the top five barriers. See supplementary file 2. Of the top five barriers, the top four were attitudinal related barriers to receiving professional mental health services. The fifth, ‘being unsure of where to get profes- sional care’ was an instrumental-related sub-scale of BACE-III. As a result of, the mean score of attitudinal related barriers sub-scale of BACE-III (M = 1.26, SD = 0.68) was the highest when compared with instrumental related barriers sub-scale (M = 0.78, SD = 0.43) and stigma related barriers sub-scale (M = 0.61, SD = 0.65) of BACE-III. Page 9 of 15 Page 9 of 15 Negash et al. BMC Psychiatry (2020) 20:187 (β = 0.18; 95%CI = 0.04, 0.26; p = 0.01) than students from urban background. Fourth-year students perceived significantly more instrumental related barriers (β = 0.28; 95%CI = 0.17, 0.76; p = 0.002) than the fifth-year stu- dents. Discussion medication side effects) 36.8 (88) 9.6 (23) 239 0.65 (1.01) Thinking that professional care probably would not help 30.9 (74) 7.5 (18) 239 0.54 (0.93) Fear of being put in hospital against my will 21.4 (51) 7.1 (17) 239 0.41 (0.89) Having had previous bad experiences with professional care for mental health 16.7 (40) 4.6 (11) 239 0.30 (0.76) Instrumental-related barriers Not being able to afford the financial costs involved 56.0 (134) 25.5 (61) 239 1.23 (1.25) Having no one who could help me get professional care 59.8 (143) 24.7 (59) 239 1.26 (1.22) Being unsure where to go to get professional care 71.6 (171) 21.0 (51) 239 1.36 (1.11) Difficulty taking time off education 55.2 (132) 17.6 (42) 239 1.10 (1.16) Problems with transport or travelling to appointments 44.4 (106) 17.6 (42) 239 0.92 (1.18) Being too unwell to ask for help 51.9 (124) 14.6 (35) 239 0.96 (1.11) Professionals from my own ethnic or cultural group not being available 26.4 (63) 6.7 (16) 239 0.47 (0.89) Note: Question number 5, 14, 24, and 29 in the BACE-III were not included in this table, because more than 97% of the participants responded “Not applicable” option for each item. Barriers reported percentage greater than 60% to any degree were indicated in bold color Table 3 Barriers to receiving professional mental health care among students with mental distress who have not received mental care in the past three months (n = 239) Barriers to Mental Health Care Mental Distress who did not receive professional mental health treatment (N = 239) Total (N) Item Mean and (SD) Barrier to any Major barrier Table 3 Barriers to receiving professional mental health care among students with mental distress who have not received mental care in the past three months (n = 239) Barriers to Mental Health Care Mental Distress who did not receive Total Item Mean Table 3 Barriers to receiving professional mental health care among students with mental distress who have not received mental i th t th th ( 239) Among the top five reported barriers to receive profes- sional mental health service by the students who recog- nized a need for care, the first was thinking mental distress would get better by itself. Discussion In this study, there is high prevalence of mental distress and perceived need for professional mental health care services among university students. The top five fre- quently reported barriers to receive professional mental health service were: thinking the problem would get bet- ter by itself, being unsure where to go to get professional care, wanting to solve the problem by oneself, denying a mental health problem, and preferring to get alternative forms of care. Having a rural background, being a Page 10 of 15 Negash et al. Discussion BMC Psychiatry (2020) 20:187 Table 3 Barriers to receiving professional mental health care among students with mental distress who have not received mental care in the past three months (n = 239) Barriers to Mental Health Care Mental Distress who did not receive professional mental health treatment (N = 239) Total (N) Item Mean and (SD) Barrier to any degree % (n) Major barrier % (n) Stigma-related barriers Concern about what my family might think, say, do or feel 48.1 (115) 18.4 (44) 239 0.98 (1.18) Concern that I might be seen as weak for having a mental health problem 38.9 (93) 14.2 (34) 239 0.76 (1.10) Feeling embarrassed or ashamed 29.8 (71) 10.9 (26) 239 0.58 (1.01) Concern that I might be seen as ‘crazy’ 31.9 (76) 10.5 (25) 239 0.61 (1.02) Not wanting a mental health problem to be on my medical records 26.0 (62) 8.4 (20) 239 0.50 (0.95) Concern that people might not take me seriously if they found out I was having professional care 28.0 (67) 7.1 (17) 239 0.51 (0.93) Concern that people I know might find out 28.4 (68) 6.7 (16) 239 0.47 (0.87) Concern about what my friends might think, say or do 33.5 (80) 6.7 (16) 239 0.56 (0.92) Concern about what students might think, say or do 31.4 (75) 6.3 (15) 239 0.53 (0.90) Attitudinal-related barriers Thinking I did not have a problem 67.4 (161) 38.1 (91) 239 1.59 (1.29) Thinking the problem would get better by itself 74.4 (178) 36.8 (88) 239 1.65 (1.22) Preferring to get alternative forms of care 66.5 (159) 34.3 (82) 239 1.51 (1.27) Wanting to solve the problem on my own 71.1 (170) 28.0 (67) 239 1.50 (1.18) Preferring to get help from family or friends 58.6 (140) 22.2 (53) 239 1.20 (1.20) Dislike of talking about my feelings, emotions or thoughts 38.0 (91) 9.6 (23) 239 0.69 (1.02) Concerns about the treatments available (e.g. Discussion This indicates that students perceive mental distress would get better with- out receiving any treatment, which may be associated with considering mental health problems as less serious so they are reluctant to use available mental health difference could be a difference in mental health literacy, study population and the data collection instruments be- ing used. Moreover, our study supports the previous web-based survey reporting that 37 to 84% of university students screened positive to mental distress did not re- ceive any professional mental health service [34]. The similarity of the result may be due to using similar data collection tool and similar age group of participants. difference could be a difference in mental health literacy, study population and the data collection instruments be- ing used. Moreover, our study supports the previous web-based survey reporting that 37 to 84% of university students screened positive to mental distress did not re- ceive any professional mental health service [34]. The similarity of the result may be due to using similar data collection tool and similar age group of participants. Page 11 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Table 4 Predictors of attitudinal related barriers to receiving professional mental health care in univariate and multivariable linear regression (n = 239) Variables Attitudinal related barriers Univariate Multivariable Beta 95% CI P-value Beta 95% CI P-value Age .09 −.02, .08 .18 .01 −.05, .06 .91 Level of study years First year .01 −.36, .38 .97 01 −.41, .42 .97 Second year .03 −.33, .43 .80 .03 −.35, .46 .81 Third Year .11 −.20, .54 .37 .12 −.21, .57 .37 Fourth Year .27 .24, 1.16 .003 .26 .21, 1.14 .004 Fifth Year (Ref.) – .80, 1.48 < 0.01 Family mental illness history No (Ref.) – 1.14, 1.32 < 0.01 Yes .12 −.01, .59 .06 .11 −.04, .56 .09 R2 0.08 Note. Reference category results for multivariable were: β = 1.05; 95% CI: −.29, 2.40; P = 0.12. Ref. Discussion = Reference category for univariate regression analysis and CI Confidence Interval for β Table 6 Predictors of stigma related barriers to receiving professional mental health care in univariate and multivariable linear regression (n = 239) Variables Stigma related barriers Univariate Multivariable Beta 95% CI P-value Beta 95% CI P-value Age .17 .01, .10 .01 .10 −.02, .09 .19 Area of growing Urban (Ref.) .42, .65 < 0.01 Rural .13 .00, .33 .05 .12 −.02, .32 .08 Level of study years First year .02 −.32, .37 .89 .11 −.25, .54 .47 Second year −.002 −.36, .36 .99 .08 −.26, .50 .54 Third Year .13 −.18, .54 .32 .18 −.11, .63 .16 Fourth Year .24 .14, 1.01 .01 .24 .14, 1.02 .01 Fifth Year (Ref.) – .19, .83 < 0.01 Family mental illness history No (Ref.) – .50, .67 < 0.01 Yes .13 .01, .57 .05 .09 −.09, −.48 .19 Substance use No (Ref.) .49, .67 < 0.01 Yes .13 .00, .49 .05 .07 −.12, .39 .31 R2 0.10 Note. Reference category for multivariable: β: −.45; 95% CI = −1.72 to .83; p = 0.49. Ref. refers to reference category for univariate regression analysis and CI Confidence Interval for β Note. Reference category results for multivariable were: β = 1.05; 95% CI: −.29, 2.40; P = 0.12. Ref. = Reference category for univariate regression analysis and CI Confidence Interval for β Table 5 Predictors of instrumental related barriers to receiving professional mental health care in univariate and multivariable linear regression (n = 239) Variables Instrumental related barriers Univariate Multivariable Beta 95% CI P-value Beta 95% CI P-value Sex Male (Ref.) – .77, .92 < 0.01 Female −.15 −.24, −.02 .02 −.10 −.20, .02 .12 Age .19 .02, .07 .004 .13 −.01, .07 .09 Area of growing Urban (Ref.) .64, .78 < 0.01 . Rural .18 .04, .26 .01 .16 .03, .25 .01 Level of study years First year .11 −.13, .34 .39 .23 −.05, .47 .11 Second year .14 −.10, .39 .25 .27 .02, .52 .03 Third Year .14 −.10, .37 .27 .23 −.02, .47 .07 Fourth Year .28 .17, .76 .002 .29 .19, .77 .001 Fifth Year (Ref.) .42, .86 < 0.01 Family mental illness history No (Ref.) .70, .82 < 0.01 Yes .16 .05, .43 .01 .09 −.05, .32 .15 Substance use No (Ref.) .70, .82 < 0.01 Yes .15 .03, .36 .02 .09 −.05, .28 .17 R2 0.14 Note. Discussion The present finding supports past studies reporting a major barrier to receiv- ing formal mental health service among university stu- dents with mild to moderate depression and anxiety is preferring to self-medicate [74, 75]. Our study also found that students from rural back- grounds are more likely than students from urban back- grounds to face instrumental related barriers to access professional mental health care. This might be because of young person from rural areas may not have a know- ledge of mental illnesses so that they may not have suffi- cient information about the availability of free mental health services in the university and they may not be psychologically open toward professional mental health services [80]. Our finding compare positively with a study conducted in Australia reporting that adolescents from rural areas have more instrumental related chal- lenges for receiving formal mental health care than ado- lescents from urban areas [81]. Our study also shows that second-year students reported more instrumental- related barriers compared with fifth-year students. This was probably due to the interaction effects of other con- trolled variables in the adjusted model. This may need further study in the future. The fourth barrier identified in the present study is denying mental health problems. Students may not want to recognize their mental health problems due to lack of knowledge about mental illness [69] or they may deny their mental health problem as a coping strategy by rejecting reality and not taking appropriate action to treat their problem [73]. Our finding support a prior study finding reporting that the majority of university students deny mental health problems which hindered them from receiving mental health care [37]. Further- more, the present study interestingly showed that a fam- ily history of mental illness significantly associated with a student denying mental distress. This could be due to students had negative experiences by being with individ- ual with mental illness previously so that they could deny their illness as a coping mechanism [45]. The present study implies that mental distress is prevalent among undergraduate students. Likewise, the need for mental health services is increasing, even though the students are not be able to receive the ser- vice provided in the university. This was because of attitudinal, instrumental and stigma-related barriers. Discussion Particularly, fourth-year students and students from rural background were more likely to report these bar- riers compared with fifth-year and students from urban backgrounds, respectively. This indicates to a need for designing practical mental health interventions to treat students’ mental distress, to alleviate their psychological suffering and minimize the effect on their academic work, general and social functioning [29]. Therefore, the present findings provide useful information and direc- tions for university mental health service providers to create awareness about mental health problems and their service, the benefits of receiving mental health care, and when and where to seek mental health services by distributing flyers, preparing training, and mental health day. All these together enhance to develop active university-based mental health intervention to reduce the prevalence of mental distress and to satisfy the need for receiving mental health service by minimizing the re- ported major barriers. The fifth commonly reported barrier for using mental health service is preferring to get mental health service from informal sources. This suggests that majority of the students receive mental health help from friends, family, relatives, religious leaders, and traditional healers [35], which is also common practice in the general population of Ethiopia [76]. The present finding is also supported by previous studies, where informal sources of mental health care reported by college students was cited as a reason for not receiving mental health services at their university [69, 77]. Interestingly, the current study found that fourth-year students with mental distress are more likely than fifth- year students to report attitudinal, instrumental, and stigma related treatment barriers. Since, the majority of the fourth- and fifth-years students in the present study were from the engineering department, the possible dif- ference could be resulted from as the level of study year/ age increases, students become more stable and have better mental health literacy [78, 79]. In this sense, fourth-year students might encounter more barriers to receiving mental health care compared to fifth-year stu- dents. However, our finding contradicts a study finding first-year students are more likely than their third-year and senior students to perceive a greater number of bar- riers to receiving mental health care [37]. The result dif- ference with the present study possibly due to the difference in the data collection tool, study setting, and sample size in each level of the study year. Discussion Reference category for multivariable: β = −.16; 95% CI = −1.02, 0.70 p = .72. Ref. refers to reference category for univariate regression analysis and CI Confidence Interval for β Table 5 Predictors of instrumental related barriers to receiving professional mental health care in univariate and multivariable linear regression (n = 239) Note. Reference category for multivariable: β: −.45; 95% CI = −1.72 to .83; p = 0.49. Ref. refers to reference category for univariate regression analysis and CI Confidence Interval for β services [68] and it may also be associated with having poor mental health literacy [69]. The current finding supports a prior study reporting that the majority of col- lege students believed that time by itself would solve their mental health problem [59]. Lack of information where to go to get professional care was reported as the second barrier to receiving mental health service in the University. However, WSU has two counseling offices and a teaching referral hos- pital that aim to provide mental health services for stu- dents with mental health problems. This information gap is probably caused by a lack of awareness creation of these services by the mental health service providers. Our finding is supported by prior studies, where the ma- jority of university students had no information about the availability of mental health service in their univer- sity [70, 71]. Wanting to solve mental health problems by oneself is reported as the third common barrier to receive mental health service in the present study. This suggests that most students may not want to share their mental health problems with professionals preferring to handle the problem by themselves. This is possibly due to perceiv- ing their problem as not serious or transitory, being skeptical about the effectiveness of professional mental Page 12 of 15 Page 12 of 15 Page 12 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 the present study, the number of the fourth-year stu- dents was small. health service, fear of stigma, and privacy issue [34]. As a result, they might prefer to manage their mental health problem alone perhaps utilizing both positive and nega- tive strategies such as problem-solving [72], substance uses, and isolation [73] as examples. Funding This research is funded by Addis Ababa University and Wolaita Sodo University. These funders had no role in the study design; data collection, analysis and interpretation of data; in writing the manuscript; and in the decision to submit the paper for publication. Conclusions There is high prevalence of mental distress and per- ceived need for professional mental health service among undergraduate students at Wolaita Sodo Univer- sity that identify the need for professional mental health interventions. Mental health providers in the university should make their services accessible to the students and promote the service for better utilization. Besides the in- terventions, developing preventive mental health educa- tion strategies is essential to address the prevalence of mental distress with the creation of conducive environ- ments that promote and sustain positive mental health for every student. Moreover, preparing for the celebra- tion of mental health day in the university can play a role in changing the attitude of students toward mental health care and improving mental health literacy, be- cause out of the five major barriers, four of them were attitudinal related. In this celebration day, creating Abbreviations BACE: Barriers to Access to Care Evaluation; LMICs: Low- and Middle- Income Countries; SNNPR: Southern Nations, Nationalities, and Peoples’ Region; SRQ: Self-Reported Questionnaire; SPSS: Statistical Packages for Social Sciences; M: Mean; SD: Standard Deviation; WHO: World Health Organization; WSU: Wolaita Sodo University; ICD-10: International Classification of Diseases 10th edition; PHQ-9: Patient Health Questionnaire; EPDS: Edinburgh Postnatal Depression Scale; K10: Kessler Psychological/Mental Distress Scale; HADS: Hospital Anxiety and Depression Scale Supplementary information Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12888-020-02602-3. Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12888-020-02602-3. Further, future research is needed to study barriers to receiving professional mental health care among stu- dents with mental distress who do not wish to receive mental health care from professionals as a result of re- ceiving treatment from informal sources. Additionally, the present study has also investigated some demo- graphic predictors of barriers to receive mental health care, but further study is necessary to examine the asso- ciations of other variables such as mental health literacy and academic results with barriers to receive mental health services. Despite the limitations mentioned above, the present study has some strengths. First, a large num- ber of students participated in the prevalence study. Sec- ond, the study used a locally adapted instrument, SRQ- 20. Third, the research contains findings of the preva- lence of mental distress, perceived needs, predictors, and barriers to receive professional mental health services to- gether; all this information taken together can serve as input for future feasibility studies of mental health inter- ventions for mental distress among university students. Additional file 2. The association between demographic variables and common barriers to receive professional mental health care. Authors’ contributions AN led to conceiving the study, supervision of data collection, developed study design, data analysis, interpretation of the findings, drafted manuscript and revised the manuscript for submission in consultation with co-authors. MAK contributed to data analysis, commenting on all drafts of the manu- scripts, interpretation of the findings, and feedback. GM contributed to data analysis, commenting on all drafts of the manuscripts, interpretation of the findings, and feedback. DW was involved in data analysis, commenting on all drafts of the manuscripts, interpretation of the findings, and feedback. MA led to conceiving the study, data analysis, commenting on all drafts of the manuscripts, interpretation of the findings, and feedback. All co-authors have approved the final version of the manuscript. Acknowledgements We want to thank research participants for their active participation and time. We would also like to acknowledge Addis Ababa University and Wolaita Sodo University for funding and material support. Lastly, we would like to thank Sisay Abayneh, Aklilu Abera, Dagim Degissew, and Dr. Clare Pain for editing the manuscript. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Discussion Caution in Any research has its own limitations; similarly, the present study is not limitation free. First, data were col- lected using self-reported questionnaires so that recalling bias may occur for mental distress symptoms that hap- pened in the past 1 month and rating the degree of bar- riers to mental health care may also difficult to remember. Second, the barriers to the provision of men- tal health care measuring instrument was not locally Page 13 of 15 Page 13 of 15 Page 13 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 adapted, although it was properly translated and piloted for the present research. Third, a screening tool was used to identify participants positive for mental distress; it would have been better to use a diagnosis tool. Fourth, data collectors were classroom representatives, so that they were in a position to know the participants’ re- sponse to each item while checking missing on the ques- tionnaires. Fifth, since the participants were recruited from a single public university, it is difficult to generalize the result to all public universities and private colleges that are found in Ethiopia. Finally, students who received treatment from the informal sources should have to complete BACE-III to understand whether they were aware of their needs for professional care or not, but we did not do that. awareness about the treatment of mental distress like any other physical illness can be emphasized, as well as the benefits of receiving mental health care from profes- sionals, recognizing mental distress in the early stages, and educating students to seek mental health care from professionals in parallel with receiving treatment from alternative sources. Therefore, this paper is a call for ac- tion from university administrations, university mental health care providers, and the Ministry of Science and Higher Education for helping undergraduate students with mental distress and to work to minimizing mental health distress on campus. References Demyttenaere K, Bruffaerts R, Posada-Villa J, Gasquet I, Kovess V, Lepine J, Angermeyer MC, Bernert SD, Morosini P, Polidori G, Kikkawa T. Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization world mental health surveys. Jama. 2004;291(21):2581–90. 33. Andrews B, Wilding JM. The relation of depression and anxiety to life-stress and achievement in students. Br J Psychol. 2004;95(4):509–21. 34. Eisenberg D, Golberstein E, Gollust SE. Help-seeking and access to mental health care in a university student population. Med Care. 2007;45(7):594–601. g y 8. Patel V. Mental health in low-and middle-income countries. Br Med Bull. 2007;81(1):81–96. 35. Gebreegziabher Y, Girma E, Tesfaye M. Help-seeking behavior of Jimma university students with common mental disorders: a cross-sectional study. PLoS One. 2019;14(2):e0212657. 9. Sorsdahl KR, Flisher AJ, Wilson Z, Stein DJ. Explanatory models of mental disorders and treatment practices among traditional healers in Mpumulanga, South Africa. Afr J Psychiatry. 2010;13(4):284–90. 36. Hinderaker DJ. College student mental health and use of counseling center services; 2013. p g y y 10. Saxena S, Thornicroft G, Knapp M, Whiteford H. Resources for mental health: scarcity, inequity, and inefficiency. Lancet. 2007 Sep 8;370(9590):878–89. 37. Vidourek RA, King KA, Nabors LA, Merianos AL. Students’ benefits and barriers to mental health help-seeking. Health Psychol Behav Med. 2014;2(1):1009–22. 11. Hanlon C, Alem A, Lund C, Hailemariam D, Assefa E, Giorgis TW, Chisholm D. Moving towards universal health coverage for mental disorders in Ethiopia. Int J Ment Heal Syst. 2019;13(1):11. 38. Heath PJ, Vogel DL, Al-Darmaki FR. Help-seeking attitudes of United Arab Emirates students: examining loss of face, stigma, and self-disclosure. Couns Psychol. 2016;44(3):331–52. 12. Girma E, Tesfaye M. Patterns of treatment seeking behavior for mental illnesses in Southwest Ethiopia: a hospital based study. BMC psychiatry. 2011;11(1):138. 39. van Zoonen K, Kleiboer A, Cuijpers P, Smit J, Penninx B, Verhaak P, Beekman A. Determinants of attitudes towards professional mental health care, informal help and self-reliance in people with subclinical depression. Int J Soc Psychiatry. 2016;62(1):8493. 13. Lund C, Tomlinson M, De Silva M, Fekadu A, Shidhaye R, Jordans M, Petersen I, Bhana A, Kigozi F, Prince M, Thornicroft G. PRIME: a programme to reduce the treatment gap for mental disorders in five low-andmiddle- income countries. PLoS Med. 2012;9(12):e1001359. 40. Al-Darmaki FR. Attitudes towards seeking professional psychological help: what really counts for United Arab Emirates university students? Soc Behav Personal Int J. 2003;31(5):497–508. 14. References 1. Patel V, Hanlon C. Where there is no psychiatrist, vol. 3. London: RCPsych Publications; 2018. 1. Patel V, Hanlon C. Where there is no psychiatrist, vol. 3. London: RCPsych Publications; 2018. 28. Gress-Smith JL, Roubinov DS, Andreotti C, Compas BE, Luecken LJ. Prevalence, severity and risk factors for depressive symptoms and insomnia in college undergraduates. Stress Health. 2015;31(1):63–70. 2. Mirowsky J, Ross CE. Measurement for a human science. J Health Soc Behav. 2002;43(2):152–70. 2. Mirowsky J, Ross CE. Measurement for a human science. J Health Soc Behav. 2002;43(2):152–70. 3. Drapeau A, Marchand A, Beaulieu-Prévost D. Epidemiology of psychological distress. Mental illnesses understanding, prediction and control; 2012. https://doi.org/10.5772/30872. 29. Field T, Diego M, Pelaez M, Deeds O, Delgado J. Depression and related problems in university students. Coll Stud J. 2012;46(1):193–203. 30. Ryan ML, Shochet IM, Stallman HM. Universal online interventions might engage psychologically distressed university students who are unlikely to seek formal help. Adv Ment Health. 2010;9(1):73–83. 4. Auerbach RP, Mortier P, Bruffaerts R, Alonso J, Benjet C, Cuijpers P, WHO WMH-ICS Collaborators. Mental disorder risk profiles in the World Health Organization World Mental Health Surveys International College Student Project. Int J Methods Psychiatr Res. 2018;10:e1752. 4. Auerbach RP, Mortier P, Bruffaerts R, Alonso J, Benjet C, Cuijpers P, WHO WMH-ICS Collaborators. Mental disorder risk profiles in the World Health Organization World Mental Health Surveys International College Student Project. Int J Methods Psychiatr Res. 2018;10:e1752. 31. Tesfahunegn TB, Gebremariam EH. Mental distress and associated factors among Aksum University students, Ethiopia: a cross-sectional study. BMC psychiatry. 2019;19(1):71. Project. Int J Methods Psychiatr Res. 2018;10:e1752. 5. World Health Organization. Depression and other common mental disorders, global health estimates. 2017. 32. Dachew BA, Bisetegn TA, Gebremariam RB. Prevalence of mental distress and associated factors among undergraduate students of University of Gondar, Northwest Ethiopia: a cross-sectional institutional based study. PLoS One. 2015;10(3):e0119464. 6. Lund C, Breen A, Flisher AJ, Kakuma R, Corrigall J, Joska JA, Swartz L, Patel V. Poverty and common mental disorders in low and middle income countries: a systematic review. Soc Sci Med. 2010;71(3):517–28. 6. Lund C, Breen A, Flisher AJ, Kakuma R, Corrigall J, Joska JA, Swartz L, Patel V. Poverty and common mental disorders in low and middle income countries: a systematic review. Soc Sci Med. 2010;71(3):517–28. 7. Competing interests 25. Tesfaye A. Prevalence and correlates of mental distress among regular undergraduate students of Hawassa University: a cross sectional survey. East Afr J Public Health. 2009;6(1):85–94. Received: 12 July 2019 Accepted: 14 April 2020 Received: 12 July 2019 Accepted: 14 April 2020 Consent for publication N l bl 24. Dessie Y, Ebrahim J, Awoke T. Mental distress among university students in Ethiopia: a cross sectional survey. Pan Afr Med J. 2013;15(1):95. Received: 12 July 2019 Accepted: 14 April 2020 26. MacKean G. Mental health and well-being in post-secondary education settings. Canada: Citeseer; 2011. 27. Kaya M, Genc M, Kaya B, Pehlivan E. Prevalence of depressive symptoms, ways of coping, and related factors among medical school and health services higher education students. Turk Psikiyatri Dergisi. 2007;18(2):137. Ethics approval and consent to participate h l l l b d f h Ethical clearance approval obtained from the Institutional Review Board (IRB) of Addis Ababa University College of Health Sciences. Respondents took part in the study after providing oral consent. Data kept anonymous and Page 14 of 15 Page 14 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 confidential during all stages of the research process. Protocol number: 045/ 17/Psych. Oral consent was approved by the IRB. 22. Steptoe A, Tsuda A, Tanaka Y. Depressive symptoms, socio-economic background, sense of control, and cultural factors in university students from 23 countries. Int J Behav Med. 2007;14(2):97–107. 23. Lun K, Chan CK, Ip P, Ma S, Tsai WW, Wong CS, Wong CH, Wong TW, Yan D. Depression and anxiety among university students in Hong Kong. Hong Kong Med J. 2018;24(5):466–72. References Kassa GM, Abajobir AA. Prevalence of common mental illnesses in Ethiopia: a systematic review and meta-analysis. Neurol Psychiatry Brain Res. 2018;30: 74–85. 41. Luitel NP, Jordans MJ, Kohrt BA, Rathod SD, Komproe IH. Treatment gap and barriers for mental health care: a cross sectional community survey in Nepal. PloS one. 2017;12(8):e0183223. 15. Jebena MG, Lindstrom D, Belachew T, Hadley C, Lachat C, Verstraeten R, De Cock N, Kolsteren P. Food insecurity and common mental disorders among Ethiopian youth: structural equation modeling. PLoS One. 2016;11(11): e0165931. 42. Masuda A, Anderson PL, Edmonds J. Help-seeking attitudes, mental health stigma, and self-concealment among African American college students. J Black Stud. 2012;43(7):773–86. 42. Masuda A, Anderson PL, Edmonds J. Help-seeking attitudes, mental health stigma, and self-concealment among African American college students. J Black Stud. 2012;43(7):773–86. 43. Rodriguez ML, Corse AK, Rosen LD. Mental health services use among medical students: perceived stigma and barriers to care. Med Sci Educ. 2017;27(2):267–72. 43. Rodriguez ML, Corse AK, Rosen LD. Mental health services use among medical students: perceived stigma and barriers to care. Med Sci Educ. 2017;27(2):267–72. 16. Bitew T. Prevalence and risk factors of depression in Ethiopia: a review. Ethiop J Health Sci. 2014;24(2):161–9. 17. Bifftu BB, Takele WW, Guracho YD, Yehualashet FA. Depression and its help seeking behaviors: a systematic review and meta-analysis of community survey in Ethiopia. Depress Res Treat. 2018;2018:1–11. 44. Sareth K, Dang HM, Weiss B. Predictors of Mental Health Help Seeking Among Cambodian Adolescents; 2019. 45. Lally J, ó Conghaile A, Quigley S, Bainbridge E, McDonald C. Stigma of mental illness and help-seeking intention in university students. Psychiatrist. 2013;37(8):253–60. 18. Bekele YY, Flisher AJ, Alem A, Baheretebeb Y. Pathways to psychiatric care in Ethiopia. Psychol Med. 2009;39(3):475–83. 19. Federal Democratic Republic of Ethiopia Ministry of Health. National mental health strategy, 2012/13–2015/16. Addis Ababa: Federal Ministry of Health; 2012. 46. Ruiz P. In: Sadock BJ, Sadock VA, editors. Comprehensive textbook of psychiatry. Philadelphia: lippincott Williams & wilkins; 2000. gy y 20. Federal Democratic Republic of Ethiopia Ministry of Health. Health sector transformation plan: 2015/2016–2019/2020. Addis Ababa: FMOH; 2015. 20. Federal Democratic Republic of Ethiopia Ministry of Health. Health sector transformation plan: 2015/2016–2019/2020. Addis Ababa: FMOH; 2015. 47. Evans-Lacko S, Thornicroft G. WHO world mental health surveys international college student initiative: implementation issues in low-and middle-income countries. Int J Methods Psychiatr Res. 2019;28(2):e1756. 47. Evans-Lacko S, Thornicroft G. References Help-seeking attitudes and behaviors regarding mental health among Turkish college students. Int J Ment Health. 2013;42(2–3):43–59. 78. Bayram N, Bilgel N. The prevalence and socio-demographic correlations of depression, anxiety and stress among a group of university students. Soc Psychiatry Psychiatr Epidemiol. 2008;43(8):667–72. 54. Alem A, Kebede D, Woldesemiat G, Jacobsson L, Kullgren G. The prevalence and socio-demographic correlates of mental distress in Butajira, Ethiopia. Acta Psychiatr Scand. 1999;100:48–55. 79. Furnham A, Cook R, Martin N, Batey M. Mental health literacy among university students. J Public Mental Health. 2011;10(4):198–210. 55. Yimam K, Kebede Y, Azale T. Prevalence of common mental disorders and associated factors among adults in Kombolcha town, Northeast Ethiopia. J Depress Anxiety S. 2014;1:2167–1044. 79. Furnham A, Cook R, Martin N, Batey M. Mental health literacy among university students. J Public Mental Health. 2011;10(4):198–210. 80. Braun B. Barriers to mental health access for rural residents. In: Maryland Family Policy Impact Seminar, vol. 18; 2003. p. 2008. 80. Braun B. Barriers to mental health access for rural residents. In: Maryland Family Policy Impact Seminar, vol. 18; 2003. p. 2008. 56. Mekonnen E, Esayas S. Correlates of mental distress in Jimma town, Ethiopia. Ethiop J Health Sci. 2003;13(1). 81. Quine S, Bernard D, Booth M, Kang M, Usherwood T, Alperstein G, Bennett D. Health and access issues among Australian adolescents: a rural-urban comparison. Rural Remote Health. 2003;3(3):245. Ethiopia. Ethiop J Health Sci. 2003;13(1). 57. Alem A, Araya M, Melaku Z, Wendimagegn D, Abdulahi A. Mental distress in medical students of Addis Ababa University. Ethiop Med J. 2005;43(3):159–66. 57. Alem A, Araya M, Melaku Z, Wendimagegn D, Abdulahi A. Mental distress in medical students of Addis Ababa University. Ethiop Med J. 2005;43(3):159–66. 58. Youngmann R, Zilber N, Workneh F, Giel R. Adapting the SRQ for Ethiopian populations: a culturally-sensitive psychiatric screening instrument. Transcult Psychiatry. 2008;45(4):566–89. References WHO world mental health surveys international college student initiative: implementation issues in low-and middle-income countries. Int J Methods Psychiatr Res. 2019;28(2):e1756. 21. Stallman HM. Psychological distress in university students: a comparison with general population data. Aust Psychol. 2010;45(4):249–57. 21. Stallman HM. Psychological distress in university students: a comparison with general population data. Aust Psychol. 2010;45(4):249–57. Page 15 of 15 Page 15 of 15 Page 15 of 15 Negash et al. BMC Psychiatry (2020) 20:187 Negash et al. BMC Psychiatry (2020) 20:187 72. Ajibade BL, Olabisi OO, Fabiyi B, Ajao OO, Ayeni AR. Stress, types of stressors and coping strategies amongst selected nursing schools students in south- west, Nigeria. Eur J Biol Med Sci Res. 2016;4(3):1–5. 48. AsanteKO,Andoh-ArthurJ.Prevalenceanddeterminantsofdepressivesymptomsamonguniversity studentsinGhana.JAffectDisord.2015;171:161–6. 49. Shamsuddin K, Fadzil F, Ismail WS, Shah SA, Omar K, Muhammad NA, Jaffar A, Ismail A, Mahadevan R. Correlates of depression, anxiety and stress among Malaysian university students. Asian J Psychiatr. 2013;6(4):318–23. 73. Deasy C, Coughlan B, Pironom J, Jourdan D, Mannix-McNamara P. Psychological distress and coping amongst higher education students: a mixed method enquiry. PLoS One. 2014;9(12):e115193. 50. van der Westhuizen C, Wyatt G, Williams JK, Stein DJ, Sorsdahl K. Validation of the self reporting questionnaire 20-item (SRQ-20) for use in a low-and middle-income country emergency Centre setting. Int J Ment Heal Addict. 2016;14(1):37–48. 74. Arria AM, Winick ER, Garnier-Dykstra LM, Vincent KB, Caldeira KM, Wilcox HC, O’Grady KE. Help seeking and mental health service utilization among college students with a history of suicide ideation. Psychiatr Serv. 2011; 62(12):1510–3. 51. Beusenberg M, Orley JH, World Health Organization. A User’s guide to the self reporting questionnaire (SRQ). Geneva: World Health Organization; 1994. 75. Sheppard R, Deane FP, Ciarrochi J. Unmet need for professional mental health care among adolescents with high psychological distress. Austr N Z J Psychiatry. 2018;52(1):59–67. 52. Hanlon C, Medhin G, Alem A, Araya M, Abdulahi A, Hughes M, Tesfaye M, Wondimagegn D, Patel V, Prince M. Detecting perinatal common mental disorders in Ethiopia: validation of the self-reporting questionnaire and Edinburgh postnatal depression scale. J Affect Disord. 2008;108(3):251–62. 76. Kerebih H, Abera M, Soboka M. Pattern of help seeking behavior for common mental disorders among urban residents in Southwest Ethiopia. Qual Prim Care. 2017;25(4):208–16. 53. Kortmann F, Ten Horn S. Comprehension and motivation in responses to a psychiatric screening instrument validity of the SRQ in Ethiopia. Br J Psychiatry. 1988;153(1):95–101. 77. Bilican FI. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 59. Czyz EK, Horwitz AG, Eisenberg D, Kramer A, King CA. Self-reported barriers to professional help seeking among college students at elevated risk for suicide. J Am Coll Heal. 2013;61(7):398–406. 60. Eisenberg D, Speer N, Hunt JB. Attitudes and beliefs about treatment among college students with untreated mental health problems. Psychiatr Serv. 2012;63(7):711–3. 61. Clement S, Brohan E, Jeffery D, Henderson C, Hatch SL, Thornicroft G. Development and psychometric properties the barriers to access to care evaluation scale (BACE) related to people with mental ill health. BMC Psychiatry. 2012;12(1):36. 62. Shumet S, Azale T, Ayano G, Abebaw D, Amare T, Getnet W. Intention to seek help for depression and associated factors among residents of Aykel town, Northwest Ethiopia: cross-sectional study. Int J Ment Heal Syst. 2019; 13(1):18. 63. Choueiry G, Salameh P. Automating data analysis in epidemiology. J Data Sci. 2019;17(1):55–80. 64. Hersi L, Tesfay K, Gesesew H, Krahl W, Ereg D, Tesfaye M. Mental distress and associated factors among undergraduate students at the University of Hargeisa, Somaliland: a cross-sectional study. Int J Ment Heal Syst. 2017; 11(1):39. 65. Haile YG, Alemu SM, Habtewold TD. Common mental disorder and its association with academic performance among Debre Berhan University students, Ethiopia. Int J Ment Heal Syst. 2017;11(1):34. 66. Kerebih H, Ajaeb M, Hailesilassie H. Common mental disorders among medical students in Jimma University, Southwest Ethiopia. Afr Health Sci. 2017;17(3):844–51. 67. Kim JE, Saw A, Zane N. The influence of psychological symptoms on mental health literacy of college students. Am J Orthopsychiatry. 2015;85(6):620. 68. Li W. University students’ mental health help-seeking: intention and service use (doctoral dissertation); 2016. 69. Amarasuriya SD, Jorm AF, Reavley NJ. Depression literacy of undergraduates in a non-western developing context: the case of Sri Lanka. BMC research notes. 2015;8(1):593. 70. Menon V, Sarkar S, Kumar S. Barriers to healthcare seeking among medical students: a cross sectional study from South India. Postgrad Med J. 2015; 91(1079):477–82. 71. Afolabi MO, Daropale VO, Irinoye AI, Adegoke AA. Health-seeking behaviour and student perception of health care services in a university community in Nigeria. Health. 2013;5(05):817.

https://openalex.org/W4382491018

https://zenodo.org/records/8093857/files/V08-N01-39-47.pdf

English

Mechanical and Wear Performance Enhancement of a 3D-Printer Nozzle Fabricated from Millscale-Modified Copper-Zinc Alloy

Zenodo (CERN European Organization for Nuclear Research)

cc-by

*Sekunowo, O.I., Agboola, J.B. and Oserei, L.F. http://doi.org/10.5281/zenodo.8093857 The challenge of dysfunctional engineering components produced via additive manufacturing is a current global concern. This appears to stem from the relatively poor mechanical performances often exhibited by 3D-printers nozzles produced from brass. The current study investigated the feasibility of enhancing the wear and mechanical properties of 3D-printers nozzles using iron-millscale modified copper-zinc alloy as a viable alternative. The methodology entailed addition of varying weight percent of iron-millscale (IMS) into copper-zinc alloy aimed at significantly modifying its microstructure for desirable impact on the mechanical properties. The charges were melted in an induction furnace, cast in metal moulds and characterised for wear and mechanical properties. Results show that the formulation containing 6 wt. % IMS addition exhibited superior mechanical characteristics in terms of hardness (201.4 VHN), wear rate (0.69 cm2 x 10-6), wear resistance (934.9 cm- 2 x 103) and modest impact energy (30.3 J). These results compare well with desirable wear and mechanical properties of high quality 3D printer nozzles. Contributions to this level of performances may have emanated from the refined microstructure showing homogeneously dispersed fine CuFe3Zn2 crystals. It is concluded that the developed alloy is a viable candidate material for 3D printer nozzle production. © 2023 RJEES. All rights reserved. 39 39 Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 p ISSN: 2635-3342; e ISSN: 2635-3350 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 for producing 3D printer nozzle is limited to the printing of only relatively soft materials such as polylactide acid (PLA) and acrylonitrile butadiene styrene (ABS) (Melo et al., 2022; Dizdar and Krishna, 2022). These materials are free from any hard additives such as glass, carbon fibres or metallic particles (Weng et al., 2016; Zhang et al., 2017). Discrimination against these materials is premised on the likelihood of forming grooves and ridges on the inner walls of the nozzle. It also predisposes the nozzle to clogging of its orifices rendering it unusable (Li et al., 2017). In spite of the utility value of copper-zinc alloy, there has not been any significant work done to modify its microstructure in a manner that enhances its wear characteristics and other desirable functional properties. It was reported that the mechanical properties of copper can be improved by alloying resulting in awesome thermal and wear characteristics (Chen et al., 2021; Tezel and Kovan, 2022). However, the limitation imposed on the mechanical property enhancement of Cu-Zn by alloying necessitates additional alloy material which possesses the potential for property improvement that is devoid of debilitating structure compromise (Gibson et al., 2021). As reported by Igelegbai et al., (2017), the feasibility of the foregoing occurring depends on the addition of a material that has the propensity for structure modification in a manner that induces fine grain and coherent crystals within the copper matrix. It has also been shown that the mechanical characteristics of 3D printers’ nozzles impact significantly the quality of the printed products in terms of functionality and aesthetics. Excessive wear of the nozzle is often a precursor for irregular product dimensions and poor surface finish (Tezel and Kovan, 2022). Furthermore, inherent structure defects contribute majorly to deleterious mechanical properties of printed components hence, the need to ensure that 3D printer nozzles exhibit desirable level of hardness and wear resistance behaviours (Igelegbai et al., 2017). In the recent technical review on zinc alloys microstructure and mechanical properties carried out by Pola et al., (2020), it was projected that the use of zinc as alloy in brasses will continue to increase owing to its preferential applications in foundry technologies. O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 Addition of zinc in copper is known to confer good mechanical property and better wear behaviour than conventional copper alloy (Igelegbai and Alo, 2015)). It was also reported that the mechanical properties of Cu-Zn alloy can be improved by adding small amounts of certain elements such as chromium, iron and nickel (Okayasu et al., 2017). Ironmillscale (IMS) is the flaky generic iron oxides (FeO, Fe2O3 and Fe3O4) resulting from oxidation of roll stocks during hot rolling of steel. The accumulation of IMS constitutes a huge waste to steel rolling mills with the attendant disposal challenges. Consequently, some innovative methods for a safe disposal of IMS are being developed. In addition, it has been established that IMS possesses a high propensity to nucleate ferritic structures in some non-ferrous alloys which enhances their mechanical properties (Bugdayci, et al., 2018; Wang, et al., 2021). Thus, the current study intends to employ iron-millscale to refine the parent copper-zinc alloy microstructure for enhanced mechanical properties in terms of wear, impact and hardness 1. INTRODUCTION Amongst other materials such as stainless steel and hardened steel, most three-dimension (3D) printer nozzles are made from brass owing to its great thermal conductive properties (Sukindar et al., 2017; Shakor et al., 2019; Carolo, 2020; Memar et al., 2023). However, due to the relatively low wear characteristics of brass, its usage 40 2.1. Materials The materials used for this study are copper and zinc slabs sourced commercially at the metals section of Owode-Onirin market, Lagos while iron-millscale (IMS) was obtained from African Foundries Limited, Lagos, Nigeria (Figure 1a-c). The preparation of the materials involved oven drying of the IMS using an oven-dryer (model DHG-9030, Japan), ball millingwith the aid of a milling machine (model, JC-QM-2, China) and sieving to 150 µm while both slabs of copper and zinc were mechanically sliced into pieces to render them chargeable into the furnace before melting. 41 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 (a) (b) (c) Figure 1: Research materials used (a) copper slabs (b) zinc slabs (c) pulverised and sieved IMS M th d (c) (a) (b) (b) (b) (c) (a) Figure 1: Research materials used (a) copper slabs (b) zinc slabs (c) pulverised and sieved IMS thods Figure 1: Research materials used (a) copper slabs (b) zinc slabs (c) pulverised and sieved IMS hods Figure 1: Research materials used (a) copper slabs (b) zinc slabs (c) pulverised and sieved IMS 2.3. Characterisation 2.3. Characterisation 2.3.2. Microstructural analysis All the test specimens were prepared according to relevant ASTM E407 standards (2016). One sample from each composition was subjected to both optical and scanning electron microscopy. Optical microscopy (OM) test specimens were prepared by grinding using emery papers of 1000 µm grade and polished using polishing powders to obtain mirror-like surfaces. Ferric chloride (FeCl3) solution mixed in 200ml and 50 ml distilled water was applied as etchant. The specimens’ microstructural features were viewed under OM at 400 magnifications (x400). Scanning electron microscope (SEM) model, EVO LS10 manufactured by Carl Zeiss, Germany was also employed to reveal detailed features of the specimens. 2.3.1. Composition analysis The materials used for the study were subjected to composition analysis using a metal analyser (Model, ARL 3460, Switzerland) while the IMS being compounds of oxides was analysed using X-ray fluorescence (XRF). 2.2.1. Melting and casting The materials were blended according to formulation shown in Table 1 while the IMS was added at varied amounts (6-12 wt. %) and stirred for 45 seconds on each addition. With reference to Ayeni (2018), after melting, the molten alloy was cast in metal moulds and kept for an hour at room temperature to age-harden artificially. Then, the cast samples were stripped from the moulds, fettled and trimmed. Table 1: Material formulation ID Materials, wt. % Cu Zn Millscale Ac 70 30 0 B 70 18 12 C 70 20 10 D 70 22 8 E 70 24 6 2.3. Characterisation 2.3.3. Mechanical characterisation The samples impact energy behaviour was evaluated using Charpy V-notched specimens prepared according to ASTM E23 standard (ASTM 2023). An Avery Denison impact tester, (model 6705, UK) was employed to obtain the specimens impact energy values. Hardness evaluation was conducted on 10 x 10 mm specimens shown in Figure 2a while indentation of the specimen surfaces was done using a hardness machine (Model, Matsuzawa MMTX, Japan) (Figure 2b). The specimens polished surfaces were indented using a quadrangular diamond-shaped pyramid indenter under a load of 100 gf for 10 seconds dwell time. The Vickers indented area and hardness number were calculated using Equations 1 and 2. A= (1) 42 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Science 8(1) 2023 pp. 39-47 d is the average length of the diagonal left by the indentation VHN = ( / ) (2) d is the average length of the diagonal left by the indentation VHN = ( / ) (2) (2) Where F and A are the applied load over the indented area respectively Where F and A are the applied load over the indented area respectively A are the applied load over the indented area respectively (a) (b) Figure 2: Pictures of (a) prepared hardness test specimens and (b) hardness tester p y (b) (a) (a) (b) ( ) ( ) Figure 2: Pictures of (a) prepared hardness test specimens and (b) hardness tester 3. RESULTS AND DISCUSSION 3. RESULTS AND DISCUSSION Wear volume Wear volume (3) Wear volume (3) Where the alloy density was computed to be 8.2 gcm-3 Wear rate (4) Where sliding distance (s) and R (Radius of abrasive wheel = 7.25 cm and N (rpm) = 300) Resistance (5) 3. RESULTS AND DISCUSSION 3.1. Chemical Composition Where the alloy density was computed to be 8.2 gcm-3 Wear rate (4) (5) 2.3.4. Wear characterisation Wear test samples of 10 x 15 mm dimension were used to evaluate the wear characteristics of the developed alloy. The test was conducted on a pin-on-disc machine on which the sample was held stationary against a rotating wheel fixed with a 60-grit emery paper. The additional test parameters include a constant applied load of 11.38 N at 250 rpm rotating speed. The mass-loss in the test samples weight before and after applying the load at intervals of 30 s, 60 s and 90 s were obtained using a digital weigh scale and the data tabulated. Thereafter, the alloys’ wear characteristics were computed using Equations 3 - 5. Wear volume 3.1. Chemical Composition 3.1. Chemical Composition The results of the materials composition analysis are shown in Tables 2-3 respectively for copper (Cu), zinc (Zn) and IMS. Table 2a shows Cu as the main element with silicon (Si), iron (Fe) and Zn relatively substantial. Similarly, Zn concentration in Table 2b as aluminium (Al), Si, Fe and Cu presence are relatively significant. As shown in Table 3, the three generic iron oxides carry a large chunk of the IMS constituents. Generally, given the relative substantial presence of iron, silicon and tin in the research materials composition, the potential for structure modification resulting in mechanical property enhancement appears highly feasible (Huang and Hu, 2022) ) Table 2a: Elemental composition of copper sample Element Cu Si Fe P S Zn Mo Pb Ti Sn Wt. % 71.89 2.95 0.46 0.06 0.05 0.56 0.01 0.03 0.43 23.56 Table 2b: Elemental composition of zinc sample Element Zn Fe Cu Sn Pb Ni Cr Al Mn Si P S Wt. % 79.42 0.63 2.87 0.02 0.51 0.03 0.02 6.56 0.09 9.72 0.06 0.07 43 43 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 Table 3: Chemical composition of iron-millscale by XRF Compounds Weight (%) FeO 67.59 Fe2O3 24.16 Fe3O4 5.09 SiO2 0.02 MgO .0.05 CaO 0.18 MnO 0.03 Al2O3 0.01 ⃰L.O.I 2.87 ⃰L.O.I. = Loss on ignition tructure 3.2. Microstructure 3.2. Microstructure The optical micrographs of the developed alloys are presented in Figures 3(a-e). Starting with the monolithic copper micrograph (Figure 3a), the matrix is overwhelmingly filled with islands of dross-like zinc crystals in α-copper matrix. On addition of 12 wt. %IMS (Figure 3b), the CuFe3Zn2 crystals induced appears to be saturated with IMS. Coupled with the presence of 18 wt. %Zn, most of the crystals became coarse and dispersed in a segregated form. This feature is known to be a precursor to poor mechanical properties (Igelegbai et al., 2017). Similarly, the addition of 10 wt. %IMS (Figure 3c) gave rise to the emergence of two phases with apparent thick grain boundaries which seem to partition the structure. This type of microstructural feature has been shown to induce weak inter-crystal cohesion giving rise to impaired mechanical properties (Igelegbai and Alo, 2015). eak inter crystal cohesion giving rise to impaired mechanical properties (Igelegbai and Alo, 2015). Figure 3a: Optical micrographs of the developed alloy at varied wt. %IMS addition (a) 0 (b) 12 (c) 10 (d) 8 and (e) 6 micrograph shown in Figure 3d contains 8 wt. % IMS, which feature a rather subdued level of grain us, the crystals appear relatively coarse but uniformly dispersed within the matrix. Figure 3e shows raph developed on addition of 6 wt. %IMS. The crystals appear to be relatively fine and usly dispersed in the matrix. It appears the IMS propensity to modify the Cu-Zn alloy microstructure Figure 3a: Optical micrographs of the developed alloy at varied wt. %IMS addition (a) 0 (b) 12 (c) 10 (d) 8 and (e) 6 Figure 3a: Optical micrographs of the developed alloy at varied wt. %IMS addition (a) 0 (b) 12 (c) 10 (d) 8 and (e) 6 Figure 3a: Optical micrographs of the developed alloy at varied wt. %IMS addition (a) 0 (b) 12 (c) 10 (d) 8 and (e) 6 The optical micrograph shown in Figure 3d contains 8 wt. % IMS, which feature a rather subdued level of grain growth. Thus, the crystals appear relatively coarse but uniformly dispersed within the matrix. Figure 3e shows the micrograph developed on addition of 6 wt. %IMS. The crystals appear to be relatively fine and homogeneously dispersed in the matrix. It appears the IMS propensity to modify the Cu-Zn alloy microstructure The optical micrograph shown in Figure 3d contains 8 wt. O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 has been most effective at 6 wt. %IMS addition. This invariably may be taken as the optimum addition of IMS in the Cu-Zn binary alloy system. Thus, the property enhancement potential is expected to be optimum at this level. In order to further explore the alloy microstructural features, both the control sample (Figure 4a) and the 6 wt. %IMS (Figure 4b) were subjected to SEM analysis. has been most effective at 6 wt. %IMS addition. This invariably may be taken as the optimum addition of IMS in the Cu-Zn binary alloy system. Thus, the property enhancement potential is expected to be optimum at this level. In order to further explore the alloy microstructural features, both the control sample (Figure 4a) and the 6 wt. %IMS (Figure 4b) were subjected to SEM analysis. Figure 4: SEM micrographs of the developed alloy at varied wt. %IMS (a) 0 b) 6 Figure 4: SEM micrographs of the developed alloy at varied wt. %IMS (a) 0 b) 6 3.3. Hardness The hardness variation induced in the developed alloys is illustrated in Figure 5 demonstrating increase in hardness values in tandem with increasing IMS addition. The volume fractions of the fine CuFe3Zn2 crystals in the alloys microstructure may have been responsible for the observed trend. The highest hardness (201.7 VHN) occurred at 6 wt. %IMS addition occasioned by the plethora of grain boundaries induced in the alloy which may have enhanced its surface hardness. This level of hardness is adjudged to be sufficient in application such as 3D printer nozzle (Chen et al., 2021). Figure 5: Hardness variation of IMS-modified Cu-Zn alloy 3 4 I t E 0 50 100 150 200 250 Ac B C D E Hardness (VHN) Varied wt. % IMS addition Figure 5: Hardness variation of IMS-modified Cu-Zn alloy 3.4. Impact Energy Figure 6 depicts the impact energy variation of the IMS-modified Cu-Zn alloy. At lower IMS addition, impact energy first increased noticeably from 12.3 J to 22.5 J; after which, it tended to grow marginally as IMS addition increased. It is noteworthy that the microstructure of the alloy contains evenly distributed CuFe3Zn2 particles which enhanced proper matrix bonding giving rise to improved impact energy at low IMS addition. This corroborates the submission by (Pola et al., 2020). The modest rise in the impact energy values of the other compositions can be attributed to poor bonding caused by a seemingly excess IMS concentration. 0 50 100 150 200 250 Ac B C D E Hardness (VHN) Varied wt. % IMS addition Figure 5: Hardness variation of IMS-modified Cu-Zn alloy Varied wt. % IMS addition Figure 5: Hardness variation of IMS-modified Cu-Zn alloy 3.2. Microstructure % IMS, which feature a rather subdued level of grain growth. Thus, the crystals appear relatively coarse but uniformly dispersed within the matrix. Figure 3e shows the micrograph developed on addition of 6 wt. %IMS. The crystals appear to be relatively fine and homogeneously dispersed in the matrix. It appears the IMS propensity to modify the Cu-Zn alloy microstructure 44 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 3.4. Impact Energy Figure 6 depicts the impact energy variation of the IMS-modified Cu-Zn alloy. At lower IMS addition, impact energy first increased noticeably from 12.3 J to 22.5 J; after which, it tended to grow marginally as IMS addition increased. It is noteworthy that the microstructure of the alloy contains evenly distributed CuFe3Zn2 particles which enhanced proper matrix bonding giving rise to improved impact energy at low IMS addition. This corroborates the submission by (Pola et al., 2020). The modest rise in the impact energy values of the other compositions can be attributed to poor bonding caused by a seemingly excess IMS concentration. Figure 6 depicts the impact energy variation of the IMS-modified Cu-Zn alloy. At lower IMS addition, impact energy first increased noticeably from 12.3 J to 22.5 J; after which, it tended to grow marginally as IMS addition increased. It is noteworthy that the microstructure of the alloy contains evenly distributed CuFe3Zn2 particles which enhanced proper matrix bonding giving rise to improved impact energy at low IMS addition. This corroborates the submission by (Pola et al., 2020). The modest rise in the impact energy values of the other compositions can be attributed to poor bonding caused by a seemingly excess IMS concentration. Figure 6 depicts the impact energy variation of the IMS-modified Cu-Zn alloy. At lower IMS addition, impact energy first increased noticeably from 12.3 J to 22.5 J; after which, it tended to grow marginally as IMS addition increased. It is noteworthy that the microstructure of the alloy contains evenly distributed CuFe3Zn2 particles which enhanced proper matrix bonding giving rise to improved impact energy at low IMS addition. This corroborates the submission by (Pola et al., 2020). The modest rise in the impact energy values of the other compositions can be attributed to poor bonding caused by a seemingly excess IMS concentration. 45 45 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 Figure 6: Impact energy variation of IMS-modified Cu-Zn alloy 0 10 20 30 40 Ac B C D E Impact energy (J) Varied wt. % IMS addition Varied wt. % IMS addition Figure 6: Impact energy variation of IMS-modified Cu-Zn alloy 3.5. Wear Characteristics The wear characteristics of the developed alloy containing four different IMS addition is presented in Table 4 and illustrated in Figure 7. At shorter sliding time of 30 s, the rise in temperature between the sliding surfaces presumably may not be high enough to generate full oxide films, hence, the wear rate (1.05-1.88 cm2 x 10-6) at this stage is transient. As the sliding time is prolonged (60 s), the temperature between the sliding surfaces is sufficient to generate oxide films which tends to aid the recovery of the worn-out surface resulting in wear rate decline to 1.06-0.096 cm 2 x 10 -6. decline to 1.06-0.096 cm x 10 . Table 4: Wear characteristics data S N Sliding time (s) 30 60 90 V (cm3. 10-2) WRa (cm2.10-6) WRe (cm-2. 103) V (cm3.10-2) WRa (cm2.10-6) WRe (cm-2. 103) V (cm3. 10-2) WRa (cm2.10-6) WRe (cm-2. 103) Ac 2.35 3.51 284.9 2.67 1.95 512.8 5.93 2.89 345.8 B 1.43 1.05 952.4 1.70 1.24 806.5 2.04 2.99 334.5 C 1.32 1.93 518.1 1.45 1.06 943.4 2.10 1.02 977.2 D 1.27 1.85 539..3 1.34 0.098 1016.1 2.78 1.36 736.4 E 1.28 1.88 532.7 1.31 0.096 1044..3 1.67 0.081 1227.6 Figure 7: Wear rate of IMS-modified Cu-Zn alloy 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Ac B C D E Wear rate, cm2 x10-6 Varied wt. % IMS addition decline to 1.06 0.096 cm x 10 . Table 4: Wear characteristics data S N Sliding time (s) 30 60 90 V (cm3. 10-2) WRa (cm2.10-6) WRe (cm-2. 103) V (cm3.10-2) WRa (cm2.10-6) WRe (cm-2. 103) V (cm3. 10-2) WRa (cm2.10-6) WRe (cm-2. 103) Ac 2.35 3.51 284.9 2.67 1.95 512.8 5.93 2.89 345.8 B 1.43 1.05 952.4 1.70 1.24 806.5 2.04 2.99 334.5 C 1.32 1.93 518.1 1.45 1.06 943.4 2.10 1.02 977.2 D 1.27 1.85 539..3 1.34 0.098 1016.1 2.78 1.36 736.4 E 1.28 1.88 532.7 1.31 0.096 1044..3 1.67 0.081 1227.6 Figure 7: Wear rate of IMS-modified Cu-Zn alloy 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Ac B C D E Wear rate, cm2 x10-6 Varied wt. % IMS addition Table 4: Wear characteristics data S N Sliding time (s) 30 60 90 V (cm3. 10-2) WRa (cm2.10-6) WRe (cm-2. 103) V (cm3.10-2) WRa (cm2.10-6) WRe (cm-2. 103) V (cm3. 10-2) WRa (cm2.10-6) WRe (cm-2. 4. CONCLUSION Iron mill scale modified copper-zinc alloy was successfully produced. Generally, the addition of varied amount of IMS as structure modifier improved the mechanical properties and enhanced the wear characteristics. The formulation containing 6 wt. % IMS addition exhibited superior characteristics in terms of hardness (201.4 VHN), wear rate (0.69 cm2 x 10-6), wear resistance (934.9 cm-2 x 103) and modest impact energy (30.3 J). These results compare well with desirable wear and mechanical properties of high-quality 3D printer nozzles. Contributions to these levels of performances are attributed to the development of homogeneous and coherent fine CuFe3Zn2 crystals within the alloy matrix. It is concluded that the developed alloy is a viable candidate material for 3D printer nozzle production. O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 However, at the long run (90 s sliding time), the oxide layer formed tends to reach a critical thickness, which becomes too weak to withstand the applied load resulting in a rather precarious high wear rate of 1.36-2.99 cm 2 x10 -6. These results indicate that sliding time, surface finish, applied load and microstructural integrity are critical factors that impact wear behaviour of engineering components (Memar et al., 2023). According to Tezel and Kovan (2022), the microstructural integrity has overriding influence on wear behaviour. 6. CONFLICT OF INTEREST There is no conflict of interest associated with this work. There is no conflict of interest associated with this work. 5. ACKNOWLEDGMENT The authors wish to acknowledge the assistance and contributions of the laboratory staff of Department of Physics University of Lagos, Akoka, Lagos where the wear tests were carried out. 3.5. Wear Characteristics 103) Ac 2.35 3.51 284.9 2.67 1.95 512.8 5.93 2.89 345.8 B 1.43 1.05 952.4 1.70 1.24 806.5 2.04 2.99 334.5 C 1.32 1.93 518.1 1.45 1.06 943.4 2.10 1.02 977.2 D 1.27 1.85 539..3 1.34 0.098 1016.1 2.78 1.36 736.4 E 1.28 1.88 532.7 1.31 0.096 1044..3 1.67 0.081 1227.6 Figure 7: Wear rate of IMS-modified Cu-Zn alloy 46 REFERENCES ASTM E407 (2016). Standards Practice for Microetching of Metals and Alloys. ASTM E23 (2023). Standard Text Methods for Notched Bar Impact of Metallic Materials. Bugdayci, M., Alkan, M., Turan, A. and Yuceil, K. (2018). Production of Iron Based Alloys from Mill Scale through Metallothermic Reduction. High Temperature Materials and Processes, 37(9-10), pp. 88-98. Ayeni, O. I. (2018). Sintering and Characterisation of 3D Printed Bronze Metal Filament. Master Thesis, Indi Indiana: Advances Purdue University, USA. Carolo, F. (2020). Best 3D Printer Nozzle-Buyer’s Guide. Available at https://all3dp.com.Accessed on March Chen, X., Bao R. Yi, J. Tao, J. and Li, F. (2021). Enhancing Mechanical Properties of Pure Copper-based Materials with CrxOy Nanoparticles and CNT Hybrid Reinforcement. Journal of Material Science, 56, pp. 3062-3077. Dizdar, S. and Krishna, A. V. (2022). Microstructural and Mechanical Properties of Polylactic Acid/Tin Bronze Tensile Strength Bars Additive Manufactured by Fused Deposition Modeling. Advances in Multidisciplinary Engineering, 21, pp. 566-579. Gibson, I., Rosen, D and Stucker, B. (2021). Additive Manufacturing Technologies: Springer International Publishing; XXIII, 675. Huang, S. and Hu, Q. (2022). Medium-Range Structure Motifs of Complex Iron Oxides. Journal of Applied Physics, 131(7), pp. 902-913. Igelegbai, E. and Alo, O. (2015). Investigation of Mechanical Properties and Microstructure of Brass Alloys Obtained from Recycled Copper and Zinc Metals. International Journal of Scientific and Engineering Research, 6(9), pp. 799-813. Igelegbai, E., Alo, O., Adeodu, A. and Daniyan, I. (2017). Evaluation of Mechanical and Microstructural Properties of α- Brass Alloy Produced from Scrap Copper and Zinc Metal through Sand Casting Process. Journal of Minerals and Materials Characterization and Engineering, 5, pp. 18-28. Li, C; Cheng, W. and Hu, J. (2017). Nozzle Problem Analysis and Optimisation of 3D Printers. Advances in Computer Science Research, 75, pp. 270-273. Melo, J; Santana, L; Idogava, H; Pais, A. and Alves, J. (2022). Effects of Nozzle Material and its Lifespan on the Quality of PLA Parts Manufactured by FFF 3D Printing. Engineering Manufacturing Letters, 11, pp. 20-27. of PLA Parts Manufactured by FFF 3D Printing. Engineering Manufacturing Letters, 11, pp. 20-27. 47 O.I. Sekunowo et al. / Nigerian Research Journal of Engineering and Environmental Sciences 8(1) 2023 pp. 39-47 Memar, S; Azadi, M. and Abdoos, H. (2023). An Evaluation on Microstructure, Wear and Compression Behaviour of Al2O3 / Brass Matrix Nanocomposites Fabricated by Stir Casting Method. Materials today Communications, 34, pp. 105- 123. Okayasu, M; Muranaga, T. and Endo, A. (2017). REFERENCES Analysis of Microstructural Effects on Mechanical Properties of Copper Alloys.Journal of Science: Advanced Materials and Devices,2, pp.128-139. Pola, A., Tocci, M. and Goodwin, F. (2020). Review of Microstructure and Properties of Zinc Alloys.Metals, 10(2), pp. 252-267. Shakor, P; Nejadi, S. and Paul, G. (2019). A Study into the Effect of Different Nozzles Shapes and Fibre-Reinforcement in 3D Printed Mortar. Materials, 12 (10), pp. 1708-1730. Sukindar, N; Ariffin, M; Baharudin, M; Jaafar, C. and Ismail, M. (2017). Analysis on the Impact of Process Parameters on Tensile-Strength using 3D Printer Repetier-Host Software.Journal of Engineering and Applied Science, 12 (10), pp. 416-427. Tezel, T. and Kovan, V. (2022). Determination of Optimum Production Parameters for 3D Printers Based on Nozzle Diameter. Rapid Prototyping Journal, 28(1), pp. 184-194. Wang, X; Jiang, J; Li, G; Shao, W. and Zhen, L. (2021). Particle-Stimulated Nucleation and Recrystallization Texture Initiated by Coarsened Al2CuLi Phase in Al-Cu-Li Alloy. Journal of Materials Research and Technology, 10, pp. 643- 650. Weng, Z; Wang, J; Senthil, T. and Wu, L. (2016). Mechanical and Thermal Properties of ABS/Montimorillonite Nano- Composite for Fused Deposition Modelling 3D Printing. Materials Design, 102, pp. 276-283. Zhang, F., Ma, G. and Tan, T. (2017). The Nozzle Structure Design and Analysis for Continuous Carbon Fiber Composite 3D Printing. Advances in Engineering Research, 136, pp. 193-199.

https://openalex.org/W4309502601

https://www.researchsquare.com/article/rs-2276116/latest.pdf

English

A Novel Synthesis of Rod-shape BaNiSn-Graphene Decorated TiO 2 Composite as a Ternary Photocatalyst to Improve Visible-Light Driven H 2 Evolution with Lactic acid and TEA

Research Square (Research Square)

cc-by

A Novel Synthesis of Rod-shape BaNiSn-Graphene Decorated TiO 2 Composite as a Ternary Photocatalyst to Improve Visible-Light Driven H 2 Evolution with Lactic acid and TEA Yonrapach Areerob  King Mongkut’s Institute of Technology Ladkrabang Md Nazmodduha Rafat  Hanseo University Kefayat Ullah  University of Swat Won-Chun Oh  (  [email protected] ) Anhui University of Science & Technology Research Article License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Additional Declarations: No competing interests reported. Version of Record: A version of this preprint was published at Journal of Inorganic and Organometallic Polymers and Materials on January 12th, 2023. See the published version at https://doi.org/10.1007/s10904-023-02533-y. Page 1/19 Abstract A novel rod-shape BaNiSn-Graphene oxide decorated TiO2 composite (BaNiSn-GT) has beeen synthesized using a simple ultrasonic method to enhance the visible-light-driven H2 evolution. The unique structure between the interfaces of BaNiSn-Graphene and TiO2 provides graphene oxide of contact and excellent electron transfer for H2 evolution activity. The BaNiSn-GT ternary photocatalyst showed that the BaNiSn- GT displayed the highest H2-production rate of 1012 µmol h-1 g-1, which was about 4 times that of rod- shape BaNiSn and BaNiSn-G under visible light irradiation with scavenger media. Moreover, the higher photocurrent density of BaNiSn-GT is correlated with electron-hole recombination, providing evidence for its inhibition, which leads to a longer lifetime of carriers produced by photoelectrons. The mechanism of the photocatalytic H2 evolution of BaNiSn-GT based on a full physicochemical characterization was proposed. This study provides new insight into the efficient hydrogen-evolution of graphene-based photocatalysts. 1. Introduction The evolution of photocatalytic H2 is an interesting way to generate H2 and a graphene oxide on way to overcome the threat of an energy crisis. Despite the many efforts that have been devoted to the development of a highly efficient photocatalyst, the practical application of the H2 evolution rate is still unsatisfactory, this is mainly due to the fast recombination of photogenerated carriers. In one of the newest approaches, heterojunction construction is widely adopted due to its efficiency in facilitating the separation of photogenerated electron-hole pairs and flexibility in material selection [1]. The development of efficient semiconductor composites utilizing visible light for renewable energy production is highly desirable. In many studies to date, many semiconductor materials are used to synthesize photocatalysts that are most suitable and efficient for H2 production under solar irradiation, such as metal oxides, metal selenides, metal sulfides, metal-organic frameworks, and carbon-based materials commonly used in water photolysis [2–5]. Graphene oxide (GO) has recently attracted considerable attention in the field of energy conversion and environmental improvement due to its distinct advantages such as high conductivity, environmental friendliness, excellent solubility, and chemical inertness [6]. In particular, the formation of band gap energy of graphene oxide can be tuned in a wide energy range by controlling the size or functionality of the energy region, so it can be applied as a semiconductor with narrow gap energy for photocatalytic application. In addition, it has been reported that the combination of GO and TiO2 shows improved photocatalytic performance in the field of H2 evolution and decomposition of pollutants [7, 8]. GO, a photocatalyst in the graphene oxide-TiO2 heterojunction, generally acts as a photoabsorbent or photo-sensitive agent to maximize photoelectron emission, and acts as an electron acceptor to receive photoelectrons generated from TiO2 to induce efficient charge separation. However, despite its practical superiority, photocatalytic activity for graphene oxide-TiO2 binary photocatalysts under visible light irradiation was not significantly Page 2/19 Page 2/19 improved for these effects due to the limited ability of graphene oxide to absorb visible light. [9, 10]. To date, many semiconductors with wide visible-light utilization have been explored for hydrogen generation including CdS, ZrO2, BaTi4O9, ZnNb2O6 ternary ZnIn2S4 materials, and polymer composite. However, their wide application is restricted by many factors, including toxicity, photo-corrosion behavior, and poor photocatalytic efficiency. The construction of a non-toxic, inexpensive, and visible-light responsive photocatalyst for hydrogen evolution remains an urgent and challenging topic. 1. Introduction In attempting to overcome these limitations, we try to synthesize multi-component metal sulfide with a tunable band gap and band gap alignment by controlling the composition. Recently, Mariia et al. also reports on the integration of CsPbBr3 on g-C3N4 nanosheets to construct composite photocatalysts [11]. These works have revealed that a well-designed chemical bonded interface in the heterojunction is beneficial for the photocatalytic process. They found that the chemical bonded interface with extensive areas facilitates the migration of carriers across the junction. Our previous work, BaCuZnS-G-TiO2 (BCZS-G-T) showed that Ba,and Zn containing compound with graphene oxide has high hydrogen evolution efficiency under visible-light irradiation for H2 production reaction [12]. For a single conductive or semiconductor material, it is hard to fulfill all demands like cost effective, low bandgap effect and efficient photocatalytic hydrogen evolution. Thus, our study point is to make efficient short-rod type catalyst to for hydrogen evolution [10, 13, 14]. Other issue is implementing graphene oxide with other TiO2, due to the low band gap of graphene oxide and good catalytic activity of TiO2. The preparations of graphene oxide based composite nano- composites are convenient and the obtained products are usually more stable than isolated particles. Most importantly, the intimately bonded graphene oxide/TiO2 surface promotes the separation of the photogenerated electrons and holes, and absorbs a wide range of incident light, thus enhances the photocatalytic performance [15, 16]. In this study, an ultrasonication method was used to synthesize BaNiSn-graphene oxide with a mesoporous TiO2 (BaNiSn-GT) serving as a co-catalyst for H2 evolution. We also investigate the photoelectrochemical properties of pure BaNiSn, BaNiSn-G, and BaNiSn-GT, as well as the effects of the most relevant sacrificial hole scavenger on the photocatalytic H2 evolution performance. Compared to pure BaNiSn and BaNiSn loaded with graphene oxide, the BaNiSn-GT composite exhibits remarkably improved performance in the evolution of photocatalytic H2. As the loading amount of BaNiSn-graphene oxide with TiO2 increased, the H2 generation rate under visible light irradiation (θ > 420nm) increased by about 2.5 times compared to the pure BaNiSn sample. Improved photocatalytic activity can be attributed to the synergistic effect of BaNiSn and graphene oxide-TiO2, where graphene oxide acts as an electron acceptor and medium, while BaNiSn provides an active site for H2 production. We hope that this work can contribute to the design and construction of highly efficient visible light-responsive photocatalysts for sustainable energy harvesting and transition. 2.2 Synthesis BaNiSn-Graphene-TiO2 (BaNiSn-GT) photocatalyst 2.2 Synthesis BaNiSn-Graphene-TiO2 (BaNiSn-GT) photocatalyst The BaNiSn-GT photocatalyst was synthesized using the ultrasonic method. A three-step combined method was used to synthesize this photocatalyst. First, the BaNiSn (BNS) photocatalyst had to be synthesized. In this preparation, 3.75 g and 4.45 g of Barium chloride (Solution A) and Nickel (II) nitrate hexahydrate (Solution B) were individually dissolved in 50 ml water. Then, the A and B solutions were consistently mixed under vigorous stirring for 1 hour, which we named the AB solution. We took 3.50 g Tin (IV) Chloride Pentahydrate (Solution C) contained in 50 ml Hydrochloric Acid and followed the stirring process for 30 min at 60°C. After that, Solution C was mixed dropwise (2 drops/sec) into the previously prepared AB solution. Then, the solution (ABC solution) was sealed with aluminum foil and stirred overnight at 80°C. Finally, the ultrasonic process was followed for 5 hours with 70 Hz amplitude at 90°C. Then, the precipitation had to be washed several times with DI water and dried at 100°C for 12 hours, followed by calcination at 550°C. This collected sample was named BaNiSn. Empirical results were obtained for the quantification of metal compounds from previous studies [12, 17], and the mixing ratio was determined based on these results. Next, we followed the same process to synthesize BaNiSn-G. To synthesize graphene by the oxidation of graphite powder, we used the Hummer–Offeman's method [18]. First, 15 g of natural graphite and 450 ml of sulfuric acid (H2SO4) were mixed with de-ionized water and agitated for an hour at 0°C in an ice bath. Then, the solution was removed from the ice bath, and potassium permanganate (45 g, KMnO4) was slowly added into the solution, which was then continuously agitated at 35°C. After turning to a dim brownish color, the container was sealed and kept at 1°C with agitation for 30 min. Then hydrogen peroxide (H2O2) was added dropwise within 5 min. The precipitation was required to wash with acetone and hydrochloric acid (HCl, 10%) several times, and it was heat-treated at 90°C for 12 h until it formed into a graphite oxide powder. At this point, graphite oxide powder was dissolved in 300 mL DI water, agitated for 30 min, and sonicated for 2 h (using Ultrasonic Processor, VCX 750). Finally, the prepared solution was purified and washed with hot water several times. 2. Materials And Methods Page 3/19 Barium chloride (BaCl2, 99.9% trace metals basis), Nickel (II) nitrate hexahydrate (Ni(NO3)2⋅6H2O, 94.5- 105.5%), and Tin (IV) Chloride Pentahydrate (SnCl4⋅5H2O, 98%) were purchased from commercial suppliers (Merck, Germany), while Graphene powder was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Deionized water (DI) was produced using a Direct-Q3 water purification system. All reagents are of analytical grade and used without further purification. Barium chloride (BaCl2, 99.9% trace metals basis), Nickel (II) nitrate hexahydrate (Ni(NO3)2⋅6H2O, 94.5- 105.5%), and Tin (IV) Chloride Pentahydrate (SnCl4⋅5H2O, 98%) were purchased from commercial suppliers (Merck, Germany), while Graphene powder was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Deionized water (DI) was produced using a Direct-Q3 water purification system. All reagents are of analytical grade and used without further purification. 2.3 Photocatalytic H2 production and Scavengers Experiments In the first step, 50 mg of each of the three types of samples (BaNiSn, BaNiSn-G, and BaNiSn-G) was dispersed in 90 ml of water, and then 10 ml of triethanolamine (TEOA) and an aqueous chloroplatinic acid solution (H2PtCl6⋅6H2O, 0.04 g/L) were added. The solution was stirred for 0.5 hours while maintaining a vacuum state. 30 W of LED lamps (Fawoo, Lumidas-H, Korea, λ ≥ 420 nm) were used as a light source, and the reaction system was maintained at 6 ℃ for the entire process. The production of H2 was determined by the handmade instrumental setting. For hydrogen evolution, we made a handmade set up through U-shape tube with ml/cm scaling. And volume to mol conversion calculation process has been used to calculate the H2 and O2 amount. The schematic diagram of the Hydrogen evolution measurement process is shown in Fig. 10. Scavenger's experiments were performed using aqueous solutions containing 0.1 M Na2SO4 and one of the following hole scavengers: 20 vol.% methanol (MeOH) (pH = 7), 20 vol.% isopropanol (i-PrOH) (pH = 7), Lactic acid (LA), and Triethanolamine (TEA) were used as a reference under the same conditions. 2.4 Instrumental characterization The morphology of the samples was examined by Transmission Electron Microscopy (TEM) (JSM5600 JEOL, Japan) (Hitachi HT7700, operated at 100 kV). The crystal structure, binding energy, and chemical composition of the composite sample were performed using X-ray diffraction patterns (XRD, Mini Flex600 diffractometer) and X-ray photoelectron spectroscopy (XPS, K-Alpha, Thermo Fisher Scientific) as AlKa X- ray radiation sources. We used a UV-vis spectrophotometer (SIMADZU UV-2600) to obtain UV-vis diffuse reflectance spectra in the range of 200 to 800 nm. Photoluminescence spectra (PL) were examined using a Fluorescence Spectrophotometer (Hitachi F-7000) with an emission wavelength region of (280–500) nm at RT. 2.2 Synthesis BaNiSn-Graphene-TiO2 (BaNiSn-GT) photocatalyst Then, the collected amount of solution was put in an oven for 6 h to obtain graphene oxide powder. The 1.0 g graphene oxide was dissolved in 60 ml of water and 40 ml ethanol, 94.0%, and sonicated for 2 hours. After that, the solution was supposed to be combined with the previously made ABC solution and stirred for 1 hour at 100°C, then sonicated for 5 hours with a 70 Hz amplitude at 90°C. The collected mixture was washed with DI water several times and dried at 160°C for 8 hours. It was then calcinated at 600°C for 3 hours. The collected sample was named BaNiSn-G. Page 4/19 At the final step, 1.8 ml of titanium (IV) isopropoxide, 97.0%, was treated with 4 ml of acetic acid and 30 ml of ethanol, 94.0%. The BaNiSn-G solution was prepared following the same process. Then, the TiO2 solution was added dropwise into the BaNiSn-G solution and stirred for 1 hour. One more time of sonication was conducted for 3 hours under the same conditions. Then, the precipitation was washed several times and dried at 160°C for 12 hours. Finally, we calcinated the sample at 600°C for 3 hours. The collected sample was names BaNiSn-GT. 3.1 Structural and Surface morphology The X-ray diffraction (XRD) patterns of BaNiSn, BaNiSn-G, and BaNiSn-GT co-catalyst are displayed in Fig. 1 (a). According to a previous report [19, 20], the graphene exhibits a strong diffraction peak at a 2θ value of 26.3º. The XRD pattern of the BaNiSn sample showed characteristic diffraction peaks of (101), Page 5/19 Page 5/19 (510), (110), (004), and (220) at 15.68 º, 22.45º, 23.78 º, 31.78 º, and 34.83 º, respectively. In addition, the peaks were positioned at the angles (2θ) of 25.4◦, 37.9◦, 48.02◦, 53.9◦, and 55.4◦, which correspond to the anatase-phase TiO2 planes (101), (004), (200), (105), and (211), respectively. The reflective peaks matched with JCPDS No. 48–0928, 15–0780 and 31–0157 corresponding to BaNiO2, BaSnO3 and BaNiSn3 very well, respectively. Furthermore, compared to the XRD pattern of BaNiSn, In the case of the BaNiSn-G sample, some BaNiO2 and BaSnO3 disappeared or showed weak diffraction line peaks, and the BaNiSn3 peak of the metal component was mainly strong. It is believed that some unstable BaNiO2 and BaSnO3 compounds disappeared during the graphene-ultrasonic-hydrothermal reaction. The formation of these crystals showed a similar phenomenon in BaNiSn-GT. In BaNiSn-GT, TiO2 and BaNiSn3 mainly showed strong peaks. Even in this sample, BaNiO2 and BaSnO3 disappeared or showed weak diffraction line peaks. After loading graphene oxide and TiO2, these peaks become less pronounced, especially at higher angle, as a result of the reduced diffraction contrast between walls and graphene oxide layer after the incorporation of TiO2 particles into the layers. EDX element analysis was performed on the BaNiSn-GT sample. The analysis results for this are shown in (b) and (c) of Fig. 1. From this result, it can be seen that C occupies the largest content. In addition, the content is shown in the order of O and Ba. In addition to these, Ti, Ni, and Sn were additionally found. Moreover, the surface morphologies of the synthesized BaNiSn, BaNiSn-G, and BaNiSn-GT composites were measured by SEM and TEM images, as shown in Fig. 2 (a-f). Figure 2 (a-c) depicts the overall surface morphologies of BaNiSn, BaNiSn-G, and BaNiSn-GT composites, which were analyzed via SEM investigation. Figure 2 (d-f) shows the morphology of the samples, which presents as rod shaped with highly interconnected regions with transparent outlets, indicating a few nanometers in size rod like structure. Further, after the addition of graphene oxide, multiple layers of graphene flakes were formed in the BaNiSn-G composite. 3.1 Structural and Surface morphology High temperature condition has been used in the synthesis process for the formation of rod like structure. Meanwhile, the interconnected architecture of the BaNiSn-GT showed a larger surface area of TiO2, which can improve the active size. Figure 2 (d) shows the formation of BaNiSn nanorods with a size of almost 500 nm in aggregated form. Moreover, graphene thin layers could not be clearly observed the BaNiSn surface. Figure 2 (e-f) depicts the TEM image of BaNiSn-GT, which reveals the formation of nanorods that is nicely dispersed on the graphene oxide sheets. The low loading of graphene oxide and TiO2 can be reasonable for unchanged rod shape structure. The high temperature treatment produced framework shrinkage and the samples-maintained rod like structure. 3.3 Raman Analysis Spectra of BaNiSn, BaNiSn-G and BaNiSn-GT measured by Raman spectroscopy are shown in 4, and the main Raman shift peaks at 394 cm-1 (B1g), 517 cm-1 (A1g), and 638 cm-1 (Eg) are due to the characteristic peaks of titanium dioxide nanotubes [25]. The two weak peaks at positions 1350 and 1600 cm-1 are due to the D-band and G-band derived from graphene, respectively. According to graphene's double resonance theory, 1387 cm-1 with a high D-band frequency results from disturbance of the connected hexagonal ring base plane [17]. In addition, the general E12g and A1g vibration modes of BaNiSn showed active modes of 425 and 3425 cm-1, respectively, which is believed to be due to metal compounds. In our experiments, the blue shift of D-band and G-band (approximately 30 cm-1) is likely due to the combined action of quantum confinement effects and the strong interaction between the three compounds of BaNiSn, TiO2 and GO. 3.2 Chemical composition Figure 3 (e) shows two binding energy peaks of C1s at 285.5 and 289.5 eV confirming C = C and C = O or O-C = C, respectively. As can be seen from Fig. 3 (f), the Ti4+-O bond of the TiO2 lattice is typical of the photoelectron peaks of Ti2p3/2 and Ti2p1/2 around 458.7 and 464.4 eV, respectively [22–24]. The results can confirm the structure of the BaNiSn-GT composite. In the overall 3.2 Chemical composition By analyzing X-ray photoelectron spectroscopy (XPS), the interfacial interaction among BaNiSn, graphene oxide, and TiO2 compounds was studied, and the surface composition and chemical state of BaNiSn-GT were analyzed as shown in Fig. 3 (a-f). Figure 3 (a) shows high-resolution XPS spectrum irradiation of Ba3d, Ni2p, Sn3d, O1s, C1s, and Ti2p. The main peak of Ni2p is divided into six sub-peaks by deconvolution operation [21]. Two spin split peaks located at 853.3 and 870.2 eV demonstrate the Page 6/19 presence of Ni2+. Furthermore, the double peaks located at 861.4 and 879.6 eV imply the presence of corresponding satellite peaks (Ni2p3/2 and Ni2p1/2) (Fig. 3 (b)). Peaks of 780.3 and 795.6 eV having a division value of 15.3 eV are all assigned to Ba2+ (Fig. 3 (c)) Also, in Fig. 3 (d), peaks at 486.8 and 495.4 eV are all assigned to Sn4+. Figure 3 (e) shows two binding energy peaks of C1s at 285.5 and 289.5 eV confirming C = C and C = O or O-C = C, respectively. As can be seen from Fig. 3 (f), the Ti4+-O bond of the TiO2 lattice is typical of the photoelectron peaks of Ti2p3/2 and Ti2p1/2 around 458.7 and 464.4 eV, respectively [22–24]. The results can confirm the structure of the BaNiSn-GT composite. In the overall XPS survey results, the intensity of the peak was significantly different, and the quantitative relationship of the element can be seen from this intensity. The intensity of C1s was the largest, and after that, the intensity of the peak gradually decreases in the order of O1s and Ni2p. Also, it can be found that Ti2p, Ba3d, and Sn3d show small peak intensity and occupy a small amount quantitatively. The peaks of O1s consisted of four peaks mainly occurring at 530.50 eV of O-Me, 532.55 eV of O = C, 533.95 eV of O–C = O, and 540.50 eV of O–OH (Fig. 3 (g)). presence of Ni2+. Furthermore, the double peaks located at 861.4 and 879.6 eV imply the presence of corresponding satellite peaks (Ni2p3/2 and Ni2p1/2) (Fig. 3 (b)). Peaks of 780.3 and 795.6 eV having a division value of 15.3 eV are all assigned to Ba2+ (Fig. 3 (c)) Also, in Fig. 3 (d), peaks at 486.8 and 495.4 eV are all assigned to Sn4+. 3.5 Possible photocatalytic mechanisms (Mott-Schottky plots) Theoretical analysis and experimental tests were used simultaneously to verify the successful formation of the Z-system in the composite heterojunction structure. Figure 6 shows a Mott-Shotky plot of BaNiSn, BaNiSn-G, and BaNiSn-GT using a saturated calomel electrode as a reference electrode, and may measure a flat band potential of a material from the plot. In the n-type semiconductor, the flat band potential may be approximated by the position of the energy region of the conductive band CB. From the measured results, the CBs of BaNiSn, BaNiSn-G and BaNiSn-GT are located at -0.6, -0.42, -0.31 V (vs NHE, pH = 7), respectively. Based on the band gap data obtained from the UV-Vis diffusion reflectance spectrum, the valence band positions (VB) of GO and TiO2 are located at + 2.4 and + 3.2V, respectively. 3.6 Photonic spectroscopic analysis (PL, EIS spectra, and Photocurrent) 3.4 Optical performance Light absorbance spectroscopy has been analyzed to confirm the light response of the samples. From the Fig. 5a, it is apparent that the pure BaNiSn has less light absorption ability. After implemented the graphene oxide and TiO2, light response increased. Thus BaNiSn-G, BaNiSn-GT have shown stable peak in the more visible light region. Furthermore, to study the optical absorbances of the BaNiSn, BaNiSn-G, and BaNiSn-GT composites, ultraviolet-visible diffuse reflectance spectroscopy was conducted (Fig. 5b). The bandgaps of the samples were evaluated by the Kubelkae-Munk function [17] αhv = C1 (hv-Eg)1/2 (1) Where α, h, v, Eg, and C1 are absorption coefficients, plank constants, optical frequencies, bandgap energy, and constants, respectively. Reflectivity was converted into absorbance data, and for this transformation, the Kubelkaie-Munk theory was used to obtain it. The photoabsorption increased significantly in the Page 7/19 Page 7/19 range of 2.3 to 1.1 eV. Pure BaNiSn showed 2.05 eV, and the BaNiSn-GT composite presents a band gap of 1.10 eV, which was lower than pure BaNiSn. Low band gap energy values are advantageous for the production of H2 evolution under visible light irradiation. 3.6 Photonic spectroscopic analysis (PL, EIS spectra, and Photocurrent) A higher photocurrent value corresponds to a higher photogenerative charge density, resulting in a more necessary charge for the photooxidation-reduction reaction at the BaNiSn-GT The photocurrent density of the composite is the key to enhancing the performance of the photocatalyst. The instantaneous transient photocurrent measurement of BaNiSn-GT under visible light irradiation is shown in Fig. 7(c). A higher photocurrent value corresponds to a higher photogenerative charge density, resulting in a more necessary charge for the photooxidation-reduction reaction at the BaNiSn-GT heterojunction. BaNiSn-GT obtained the strongest reaction value of 13800 µA cm-2. After four cycles of intermittent on/off irradiation, the stability of the graphene oxide rod photoelectric response of BaNiSn- GT was demonstrated through a nearly decay-free photocurrent curve. 3.6 Photonic spectroscopic analysis (PL, EIS spectra, and Photocurrent) Close interface contact between Junction bonded semiconductors is advantageous for the separation of electrons and holes produced by irradiated light that can be determined by photo-electrochemical measurements. The PL spectrum is described as some combination of photoelectrons and holes received by the semiconductor. The higher the PL strength of the photocatalytic semiconductor composite compound, the slower the recombination and the shorter the electron lifetime. Figure 7(a) shows PL spectra of TiO2, GO, BaNiSn, BaNiSn-G, and BaNiSn-G at an excited wavelength of 550 nm. The emission peak of the BaNiSn-G shows strength similar to that of BaNiSn-GT in the range of 450 to 650 nm. However, BaNiSn-GT showed the lowest peak intensity, and the PL intensity of BaNiSn showed the strongest curve. When BaNiSn-GT was incorporated, the PL strength of BaNiSn decreased dramatically, suggesting a longer lifetime of photo-generated electron-hole pairs [26]. As is well known, TiO2 and GO are basic substances. The results of PL analysis on these can be seen from many documents [12, 27]. Our previous work [12] shows that TiO2 exhibits a high PL curve due to the large bandgap energy value. This shows the effect of the photo-reaction effect late due to the late electron transfer reaction. However, the graphene complex compound presented in this study shows a low PL curve because the electron transfer reaction proceeds easily and quickly. Junction-bonded composites have been studied by electrochemical impedance spectroscopy (EIS). From the Nyquist plot (Fig. 7(b)) it can be clearly observed that the circle radius of BaNiSn-GT is small in the BaNiSn and BaNiSn-G samples, indicating an increase in conductivity and improved charge separation ability after addition of graphene oxide. These results further confirm the higher efficiency of the BaNiSn- GT composite, as shown by photocurrent and photocatalytic experiments, which exhibit the smallest electron transfer resistance at the semiconductor/electrolyte interface. GO is a good conductive material Page 8/19 in complex compound studies [12, 28, 29]. When the metal compound forms a complex with GO, the metallic complex has good photo-electron transfer effect due to the synergy effect. In the results of previous studies [29], many results can be seen. As shown in Fig. 7(b), the GO composite shows the effect of improving the conductivity. The photocurrent density of the composite is the key to enhancing the performance of the photocatalyst. The instantaneous transient photocurrent measurement of BaNiSn-GT under visible light irradiation is shown in Fig. 7(c). 3.6 H2 evolution analysis and Scavenger Effect The activity of the prepared photocatalyst for hydrogen production was conducted under visible light irradiation for 4 hours and was investigated in the presence of visible light of the aqueous methanol solution as a sacrificial agent. These results are shown in Fig. 8 (a), which shows that BaNiSn-GT exhibits relatively high photocatalytic activity with a hydrogen evolution rate of 1012 µmol/h. On the other hand, BaNiSn-G and BaNiSn composites exhibited significantly lower hydrogen evolution rates of 910 µmol/h and 835 µmol/h. The beneficial effect of the complex photocatalyst is the combination of wide band gap and narrow band gap semiconductors, which can significantly improve the interface reaction speed due to the expansion of the life of the charge carrier. In addition, this combination leads to the result of photocatalysts with improved absorption efficiency in the range of visible light by solar irradiation. Radical capture experiments using complex photocatalysts and scavengers were also conducted. Na2SO4, methanol, ethanol, lactic acid (LA), and triethanolamine (TEA) were adopted as scavengers, respectively. As shown in Fig. 8 (b), the addition of TEA has the highest effect on the degradation efficiency, whereas the addition of Na2SO4, methanol, and ethanol having the same concentration and volume leads to almost 50% inhibition effect of the degradation efficiency. Thus, h+ and •OH are essential for the evolution of photocatalytic hydrogen, and in this reaction, •OH has a greater effect than h+. Also, as shown in Fig. 9, the VB position of TiO2 is much closer to the CB position of BaNiSn-G at the hetero interface. The distance between VB and CB position reduced due to zero band gao of graphene oxide. Addition of graphene oxide and TiO2 produced a mid bandgap and hinder the recombination (electron to hole conversion). Thus, the electron lifetime increase and enhance the photocatalytic activity. During light excitation, the photoinduced electrons on the CB of BaNiSn-G can be spatially described as their movement between CB and VB to create a photo-generating hole on the VB. This is an irrefutable solid Z- system charge transfer process over heterogeneous interfaces. The role of BaNiSn in this photoelectron transfer reaction is high hydrogen production by controlling band gap energy. In the photocatalytic experiment, BaNiSn produced an oxide radical to produce a metal-electron conversion. Hydrogen Page 9/19 Page 9/19 molecules can be dissociated through adsorption on the surface of the metal catalyst. 3.6 H2 evolution analysis and Scavenger Effect In the next step, the hydrogen atom moves from the metal catalyst to the non-metal support or adsorbent. Here, the sacrificial scavenger acted as an electron donor that played an intermediate role. In the experiment, the sacrificial scavenger released electrons, which, as shown in Fig. 9, atoms covered holes in the band and increased hole-electron conversion, which is an active site for efficient photocatalytic activity of the photocatalyst. The addition of an auxiliary sacrificial agent also increases the amount of hydrogen evolution. Table 1 Some examples for H2 evolution amounts with other catalysts Catalyst H2 evolution amounts (with the scavenger) H2 evolution amounts References CoMoSx   3.85 µmol/g for 1 min 30 FeMoSx   1.57 µmol/g for 1 min 30 NiMoSx   4.93 µmol/g for 1 min 31 BaCuZnS- graphene- TiO2 5541.04 µmol/g/4 h 2715.60 µmol/g for 4 h 12 LaCdSe‑GO‑ TiO2 443.28 µmol/g. 324.15µmol/g for 4h (sonophotocatalytic) 32 ZnS:Eu quantum dots 11100 µmol/g/h 9000 µmol/g for 1h 33 Ni6(SCH2Ph)12- TiO2   5600 µmol/g for 1h 34 WSe2-graphene- TiO2 2.004 mmol/ 11 h 1.718 mmol/g for 11 h 35 The results of previous studies on hydrogen production studies with several types of photocatalysts are shown in Table 1. Since they have different research conditions such as reaction time and the presence or absence of scavengers, quantitative comparison is problematic. However, from these studies, the accessibility of reactor design and commercialization can be considered in the future. 4. Conclusions In summary, the unique structure of the BaNiSn-GT co-catalyst was successfully prepared using an ultrasonic method. Photoelectrochemical measurements proved that BaNiSn, graphene oxide, and TiO2 act as photo-sensitive agents to react to visible light reactions. The Junction-bonded BaNiSn-GT complex exhibits enhanced photocatalytic H2 production activity and stability under visible light irradiation due to the combination of extended visible light absorption capacity, long life of photogenerated electron-hole Page 10/19 Page 10/19 pairs, and fast interfacial charge transfer rates. The ability of the selected sacrificial agent to clean the photogenic holes was evident in the order of Na2SO4, methanol, ethanol, LA, and TEA solutions. The presence of the sacrificial agent also results in a negative change in the starting potential of the photocurrent, which can be largely explained by the different redox potentials of the additive. This work not only provides an innovative strategy for synthesizing heterojunction ternary hybrid nanocomposites, but also provides new insights to understand the essential role of BaNiSn and TiO2 in graphene oxide- based nanohybrids. Conflicts of Interest: Declare conflicts of interest or state “The authors declare no conflict of interest.” Authors must identify and declare any personal circumstances or interest that may be perceived as inappropriately influencing the representation or interpretation of reported research results. Any role of the funders in the design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results must be declared in this section. If there is no role, please state “The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results”. References 1. Dong Y., Hu Q., Li B., Li X., Chen M., Zhang M., Feng Y., Ding Y., Aminated silicon dioxide enriching iron-containing polyoxometalate catalyst confined in CdS for efficient H2 evolution, Applied Catalysis B: Environmental, 2022, 304, 120998. 2. Zhou H., Chen R., Han C., Wang P., Tong Z., Tan B., Huang Y., Liu Z., Copper phosphide decorated g- C3N4 catalysts for highly efficient photocatalytic H2 evolution, Journal of Colloid and Interface Science, 2022, 610, 126-135. 3. Li X., Hu J., Yang T., Yang X., Qu J., Li C., Efficient photocatalytic H2-evolution coupled with valuable furfural-production on exquisite 2D/2D LaVO4/g-C3N4 heterostructure, Nano Energy, 2022 , 92, 106714. 4. Cao B., Wan S., Wang Y., Guo H., Ou M., Zhong Q., Highly efficient visible-light-driven photocatalytic H2 evolution integrated with microplastic degradation over MXene/ZnxCd1-xS photocatalyst, Journal of Colloid and Interface Science, 2022, 605, 311-319. 5. Cao H., Zhang D., Ning X., Geng J., Constructing magnetically recoverable CdS/CaFe2O4 P–N heterojunctions for enhanced photoelectrochemical properties towards H2 evolution reaction, Ceramics International, 2021, 47, 14, 19668-19677. 5. Cao H., Zhang D., Ning X., Geng J., Constructing magnetically recoverable CdS/CaFe2O4 P–N heterojunctions for enhanced photoelectrochemical properties towards H2 evolution reaction, Ceramics International, 2021, 47, 14, 19668-19677. 6. Choi J.S., Park H.B., Yoon O.J., Kim H.J, Investigation on the role of graphene oxide sheet-platinum composite counter electrode in dye-sensitized solar cell, Thin Solid Films, 2022,  745, 139098. 6. Choi J.S., Park H.B., Yoon O.J., Kim H.J, Investigation on the role of graphene oxide sheet-platinum composite counter electrode in dye-sensitized solar cell, Thin Solid Films, 2022,  745, 139098. Page 11/19 7. Wang P., Deng P., Cao Y., Edge-sulfonated graphene-decorated TiO2 photocatalyst with high H2- evolution performance, International Journal of Hydrogen Energy, 2022, 47, 2, 1006-1015. 8. Altin I., CuO-TiO2/graphene ternary nanocomposite for highly efficient visible-light-driven photocatalytic degradation of bisphenol A, Journal of Molecular Structure, 2022, 1252, 132199. 9. Lv X., Zhang G., Fu W., Highly Efficient Hydrogen Evolution Using TiO2/Graphene Composite Photocatalysts, Procedia Engineering, 2012, 27, 570-576. 10. Gong, W., Meng, X., Tang, X. and Ji, P., 2017. Core-shell MnO2-SiO2 nanorods for catalyzing the removal of dyes from water. Catalysts, 7(1), p.19. 10. Gong, W., Meng, X., Tang, X. and Ji, P., 2017. Core-shell MnO2-SiO2 nanorods for catalyzing the removal of dyes from water. Catalysts, 7(1), p.19. 11. References Pavliuk M.V., Abdellah M., Sá J., Hydrogen evolution with CsPbBr3 perovskite nanocrystals under visible light in solution, Materials Today Communications, 2018, 16, 90-96, 11. Pavliuk M.V., Abdellah M., Sá J., Hydrogen evolution with CsPbBr3 perovskite nanocrystals under visible light in solution, Materials Today Communications, 2018, 16, 90-96, 12. Rafat, M.N., Cho, K.Y., Jung, C.H. and Oh, W.C., 2020, New modeling of 3D quaternary type BaCuZn graphene-TiO2 (BCZS-GT) composite for photosonocatalytic hydrogen evolution with scavenger effect. Photochemical & Photobiological Sciences, 19(12), pp.1765-1775. 13. Chen, H., Chen, Z., Yang, H., Wen, L., Yi, Z., Zhou, Z., Dai, B., Zhang, J., Wu, X. and Wu, P., Multi-mode surface plasmon resonance absorber based on dart-type single-layer graphene. RSC advances, 2022, 12(13), 7821-7829. 14. Zhang, Z., Cai, R., Long, F. and Wang, J., Development and application of tetrabromobisphenol A imprinted electrochemical sensor based on graphene/carbon nanotubes three-dimensional nanocomposites modified carbon electrode. Talanta, 2015, 134, pp.435-442. 15. Zaaba, N.I., Foo, K.L., Hashim, U., Tan, S.J., Liu, W.W. and Voon, C.H., 2017. Synthesis of graphene oxide using modified hummers method: solvent influence. Procedia engineering, 2017, 184, pp.469- 477. 16. Shangguan, Q., Chen, Z., Yang, H., Cheng, S., Yang, W., Yi, Z., Wu, X., Wang, S., Yi, Y. and Wu, P., Design of ultra-narrow band graphene refractive index sensor. Sensors, 2022, 22(17), p.6483. 17. Eskandari M., Shahbazi N., Marcos A.V., Malekfar R., Taboada P., Facile MOF-derived NiCo2O4/r- GRAPHENE OXIDE nanocomposites for electrochemical energy storage applications, Journal of Molecular Liquids, 2022,  348, 118428. 18. Wang X., Liu X., Liu X., Su Y., Wang J., Fan T., Ning X., Seeram Ramakrishn, Yun-Ze Long, Ultralight and multifunctional PVDF/SiO2@GRAPHENE OXIDE nanofibrous aerogel for efficient harsh environmental oil-water separation and crude oil absorption, Carbon, 2022, 193, 77-87. 19. Zhang J., Jia X., Liu T., Yang J., Wang S., Li Y., Shao D., Feng L., Song H., Facile strategy to synthesize porous GRAPHENE OXIDE/ZnO heterostructure for enhanced acetone gas sensing properties, Sensors and Actuators B: Chemical, 2022, 359, 131601. 20. Zhang C., Guo H., Gao Y., Graphene oxideng Y., Jin C., He J., Controllable synthesis of Co3W3N supporting on N-doped GRAPHENE OXIDE as electrocatalysts for oxygen reduction reaction, Chemical Physics Letters, 2022, 793, 139429, 20. Zhang C., Guo H., Gao Y., Graphene oxideng Y., Jin C., He J., Controllable synthesis of Co3W3N supporting on N-doped GRAPHENE OXIDE as electrocatalysts for oxygen reduction reaction, Chemical Physics Letters, 2022, 793, 139429, Page 12/19 Page 12/19 21. References Wang Y., Li J., Yuan R., Gao H., Lv J., Xue D., Hao X., Yang Y., Lookman T., Dkhil B., Lou X., Enhanced electrocaloric effect in BaSn/TiO3 ceramics by addition of CuO, Journal of Alloys and Compounds, 2021, 851, 2021, 156772. 22. Guechi A., Chegaar M., Bouhemadou A., Electronic and optical properties of polar intermetallic compound BaSn3: First principles study, Optik, 2020, 219, 165028, 23. Chen D., Yu W., Wei L., Ni J., Li H, Chen Y., Tian Y., Yan S., Mei L., Jiao J., High sensitive room temperature NO2 gas sensor based on the avalanche breakdown induced by Schottky junction in TiO2-Sn3O4 nanoheterojunctions, Journal of Alloys and Compounds, 2022, 912, 165079. 24. Sugikumar M.V., Sakthivel V.V., Mandarasalam G., Kalayanasundram M., Subramanian S., Photocatalytic degradation of phenylene diamine using TiO2-rGRAPHENE OXIDE under UV radiation, Materials Today: Proceedings, 2022, 57, Part 4, 1694-1699. 25. Wahyuningsih S., Ramelan A.H., Pramono E., Nuryana H.P., Mujahidin M.M.A., Munawaroh H., Hidayat R., Fadillah G., Transformation growth of nanoflower-like GRAPHENE OXIDE-ZnO as an active site platform for H2S sensors, Chemical Physics Letters, 2022, 790, 139351. 26. Zhou H., Chen R., Han C., Wang P., Tong Z., Tan B., Huang Y., Liu Z., Copper phosphide decorated g- C3N4 catalysts for highly efficient photocatalytic H2 evolution, Journal of Colloid and Interface Science, 2022, 610, 126-135. 27. Thao, L.T.S., Dang, T.T.T., Khanitchaidecha, W., Channei, D. and Nakaruk, A., Photocatalytic degradation of organic dye under UV-A irradiation using TiO2-vetiver multifunctional nano particles. Materials, 2017, 10(2), p.122. 28. Hsu, C.P., Lee, K.M., Huang, J.T.W., Lin, C.Y., Lee, C.H., Wang, L.P., Tsai, S.Y. and Ho, K.C., EIS analysis on low temperature fabrication of TiO2 porous films for dye-sensitized solar cells. Electrochimica Acta, 2008, 53(25), pp.7514-7522. 28. Hsu, C.P., Lee, K.M., Huang, J.T.W., Lin, C.Y., Lee, C.H., Wang, L.P., Tsai, S.Y. and Ho, K.C., EIS analysis on low temperature fabrication of TiO2 porous films for dye-sensitized solar cells. Electrochimica Acta, 2008, 53(25), pp.7514-7522. 29. Liu Q., Shi Q., Wang H., and Zhang Q., Laser irradiated self-supporting and flexible 3-dimentional graphene-based film electrode with promising electrochemical properties, RSC Advances 2015, 5, 58, DOI:10.1039/C5RA08431H 29. Liu Q., Shi Q., Wang H., and Zhang Q., Laser irradiated self-supporting and flexible 3-dimentional graphene-based film electrode with promising electrochemical properties, RSC Advances 2015, 5, 58, DOI:10.1039/C5RA08431H 30. Fu, G.; Lee, J. M. Ternary metal sulfides for electrocatalytic energy conversion. Journal of Materials Chemistry A, 2019, 7(16): 9386-9405. 30. References Fu, G.; Lee, J. M. Ternary metal sulfides for electrocatalytic energy conversion. Journal of Materials Chemistry A, 2019, 7(16): 9386-9405. 31. Wang, W.; Yang, L.; Qu, F.; Liu, Z.; Du, G.; Asiri, A. M . A self-supported NiMoS4 nanoarray as an efficient 3D cathode for the alkaline hydrogen evolution reaction. Journal of Materials Chemistry A, 2017, 5(32): 16585-16589. 32. Kovtun A, Jones D, Dell’Elce S, Treossi E, Liscio A, Palermo V. Accurate chemical analysis of graphene-based materials using X-ray photoelectron spectroscopy. Carbon, 2019, 143: 268-275 33. Poornaprakash, B.;  Vattikuti, S. P.; Subramanyam, K.; Cheruku, R.; Devarayapalli, K.;  Kim, Y. L.;  Park, H.; Reddy, M. S. P. Photoluminescence and hydrogen evolution properties of ZnS: Eu quantum dots. Ceramics International, 2021, 47(20): 28976-28984. Page 13/19 Page 13/19 34. Tian, F.;  Chen, J.; Chen, F.; Liu, Y.; Chen, R. Boosting hydrogen evolution over Ni6(SCH2Ph)12 nanocluster modified TiO2 via pseudo-Z-scheme interfacial charge transfer. Applied Catalysis B: Environmental, 2021, 292: 120158. 34. Tian, F.;  Chen, J.; Chen, F.; Liu, Y.; Chen, R. Boosting hydrogen evolution over Ni6(SCH2Ph)12 nanocluster modified TiO2 via pseudo-Z-scheme interfacial charge transfer. Applied Catalysis B: Environmental, 2021, 292: 120158. 35. Zhu, L.; Nguyen, D. C. T.; Woo, J-H; Zhang, Q. F.; Cho, K. Y.; Oh, W. C. An eco-friendly synthesized mesoporous-silica particle combined with WSe2-graphene- TiO2 by self-assembled method for photocatalytic dye decomposition and hydrogen production. Scientific reports, 2018, 8(1): 1-14. 35. Zhu, L.; Nguyen, D. C. T.; Woo, J-H; Zhang, Q. F.; Cho, K. Y.; Oh, W. C. An eco-friendly synthesized mesoporous-silica particle combined with WSe2-graphene- TiO2 by self-assembled method for photocatalytic dye decomposition and hydrogen production. Scientific reports, 2018, 8(1): 1-14. Scheme Scheme 1 is available in Supplementary Files section. Scheme 1 is available in Supplementary Files section. Scheme 1 is available in Supplementary Files section. Figures Page 14/19 igure 1 Figure 1 Page 14/19 Page 14/19 Crystal structure and elemental analysis; (a) XRD patterns of BaNiSn, BaNiSn-G and BaNiSn-GT, (b) and (c) EDX analysis of BaNiSn-GT. (c) gure 2 EM and TEM images of (a, d) BaNiSn, (b, e) BaNiSn-G and (c, f) BaNiSn-GT. Figure 2 SEM and TEM images of (a, d) BaNiSn, (b, e) BaNiSn-G and (c, f) BaNiSn-GT. Figure 2 SEM and TEM images of (a, d) BaNiSn, (b, e) BaNiSn-G and (c, f) BaNiSn-GT. Page 15/19 Figure 3 XPS spectra (a) full XPS survey, (b) Ni 2p, (c) Ba 3d, (d) Sn 3d, (e) C1s, (f) Ti 2p and (g) O1s. Figure 3 XPS spectra (a) full XPS survey, (b) Ni 2p, (c) Ba 3d, (d) Sn 3d, (e) C1s, (f) Ti 2p and (g) O1s. XPS spectra (a) full XPS survey, (b) Ni 2p, (c) Ba 3d, (d) Sn 3d, (e) C1s, (f) Ti 2p and (g) O1s. Page 16/19 Figure 4 Raman Spectra of BaNiSn, BaNiSn-G and BaNiSn-GT. Figure 4 Figure 4 Raman Spectra of BaNiSn, BaNiSn-G and BaNiSn-GT. Page 16/19 Page 16/19 Figure 5 UV-Vis DRS spectra (a) light absorption spectrum of BaNiSn, (b) light absorption spectra of BaNiSn-G and BaNiSn-GT and (c) the bandgap energy of BaNiSn, BaNiSn-G and BaNiSn-GT corresponding Kubelkae- Munk function spectra. Figure 7 Photonic spectroscopic analysis (a) PL Spectra, (b) EIS Spectra and (c) Photocurrent density of BaNiSn, BaNiSn-G and BaNiSn-GT. Photonic spectroscopic analysis (a) PL Spectra, (b) EIS Spectra and (c) Photocurrent density of BaNiSn, BaNiSn-G and BaNiSn-GT. Photonic spectroscopic analysis (a) PL Spectra, (b) EIS Spectra and (c) Photocurrent density of BaNiSn, BaNiSn-G and BaNiSn-GT. Figure 8 Hydrogen evolution amounts ; (a) TiO2, BaNiSn, BaNiSn-G and BaNiSn-GT photocatalysts under visible light and (b) BaNiSn-GT photocatalyst under visible light and scavenger media (LA; Lactic acid and TEA; Triethanolamine) during 1 hour. Figure 5 Figure 5 UV-Vis DRS spectra (a) light absorption spectrum of BaNiSn, (b) light absorption spectra of BaNiSn-G and BaNiSn-GT and (c) the bandgap energy of BaNiSn, BaNiSn-G and BaNiSn-GT corresponding Kubelkae- Munk function spectra. UV-Vis DRS spectra (a) light absorption spectrum of BaNiSn, (b) light absorption spectra of BaNiSn-G and BaNiSn-GT and (c) the bandgap energy of BaNiSn, BaNiSn-G and BaNiSn-GT corresponding Kubelkae- Munk function spectra. Page 17/19 Figure 6 Figure 6 Page 17/19 Page 17/19 Mott-Schottky analysis for calculation of donor density and flat band potential. Mott-Schottky analysis for calculation of donor density and flat band potential. Figure 7 Figure 9 Schematic diagram showing the energy band structure and electrons transfer in the BaNiSn-GT heterojunction. Schematic diagram showing the energy band structure and electrons transfer in the BaNiSn-GT heterojunction. Figure 8 Hydrogen evolution amounts ; (a) TiO2, BaNiSn, BaNiSn-G and BaNiSn-GT photocatalysts under visible light and (b) BaNiSn-GT photocatalyst under visible light and scavenger media (LA; Lactic acid and TEA; Triethanolamine) during 1 hour. Page 18/19 Scheme1.jpg Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. Scheme1.jpg Page 19/19 Page 19/19 Page 19/19

https://openalex.org/W2105032435

http://sceco.ub.ro/index.php/SCECO/article/download/241/227

English

SUBSEQUENT EVENTS AND THEIR IMPORTANCE IN DRAWING UP ANNUAL FINANCIAL STATEMENTS

Studies and Scientific Researches. Economics Edition

cc-by

Studies and Scientific Researches. Economics Edition, No 19, 2014 Studies and Scientific Researches. Economics Edition, No 19, 2014 http://sceco.ub.ro SUBSEQUENT EVENTS AND THEIR IMPORTANCE IN DRAWING UP ANNUAL FINANCIAL STATEMENTS Daniel Botez “Vasile Alecsandri” University of Bacău [email protected] Abstract Subsequent events, called so because they are recorded after the closing date of financial year, must be recorded and taken into account in drawing up the financial statements. Some of them can fundamentally changing financial position of the entity or provide information to users absolutely necessary for their decisions. Accounting treatment in this situation is provided for by International Standard of Accounting ISA 10 “Events after the Reporting Period”, whose provisions are taken to a large extent in national accounting standards. For financial auditor, investigating how the subsequent events are presented in the financial statements shall constitute a procedural obligation. Auditor`s procedure is based on the provisions of an specific International Standard on Auditing, ISA 560 “Subsequent events”. He shall be required to satisfy itself that in the financial statements shall be presented all treatment and appropriate information concerning subsequent events. JEL Classification M40; M42 JEL Classification M40; M42 Subsequent events through the prism of accountants Keywords subsequent events; financial statements; financial audit q g p From the point of view of financial auditor, such intervention on subsequent events is provided for by International Standard on Auditing ISA 560 “Subsequent events”. Auditor`s objective are: j A) Obtaining sufficient appropriate audit evidence on the extent to which events occurring between the date of financial statements and auditor`s report date which provide for an adjustment or the presentation in the financial statements are reflected, as appropriate, in those financial statements, in accordance with the general financial reporting applicable framework; and B) The provisions of answers appropriate to the deeds of which the auditor becomes aware after the auditor`s report date, which, if it had been known to the auditor at that time, could have been determined the auditor to modify auditor1s report. The auditor also should perform audit procedures designed to obtain sufficient appropriate audit evidence, according to which events occurring between the date of financial statements and auditor1s report date which provide for an adjustment or the presentation in the financial statements have been identified. In accordance with the assessment of the risk by auditor, audit procedures may include procedures necessary to obtain sufficient appropriate audit evidence, which involves a review or test accounting records or transactions which take place between the date of financial statements and auditor`s report date. The auditor shall carry out the procedures designed to obtain sufficient evidence as appropriate so that they cover the period between the date of financial statements and auditor`s report date, as far as possible, as close as possible to the data. The auditor also has to take into account the risk assessment by auditor in determination of the nature and extent of such audit procedures, which should include the following: y nature and extent of such audit procedures, which should include the following: A) Obtain an understanding of any proceeding on which the management has established to ensure that the subsequent events are identified; B) Inquiry management, and where appropriate, persons charged with governance, with regard to the production any further event that could affect financial statements; C) The reading minutes, if any, of entity owners meetings, management and persons charged with governance which took place after the date of financial statements query with regard to the matters discussed at such meetings for which there are not yet available minutes. D) The reading of most recent interim financial statements for entity, if there is one. Subsequent events through the prism of accountants From an accounting point of view, the treatment of subsequent events is provided for by International Accounting Standard IAS 10 “Events after the Reporting Period”. From an accounting point of view, the treatment of subsequent events is provided for by International Accounting Standard IAS 10 “Events after the Reporting Period”. The objective of standard is to prescribe when an entity should adjust its financial statements for subsequent balance sheet date events and presentation of information which should be an entity with regard to the date on which it has been authorized publication of financial statements with respect to the events after the balance sheet date. Direct consequence is that an entity should not be to draw up financial statements on the basis of assumptions concerning business continuity if subsequent events indicates that this assumption in not suitable. Subsequent events are those events, both favorable and unfavorable, which occur between balance sheet date and the date on which it has authorized its publication financial statements. May be identifies two types of events: May be identifies two types of events: 1. Those which are proof of the conditions that existed to the balance sheet date ( b t t hi h l d t dj t fi i l t t t ) d 1. Those which are proof of the conditions that existed to the balance sheet date (subsequent events which lead to adjust financial statements); and 2. Those which provide information on conditions that may arise after the date of balance sheet (subsequent events that does not lead to an adjustment of financial statements). Subsequent events shall include all events up to the date on which it is necessary to authorize publication of financial statements, even if those events taking place after public declaration of profit or other financial information selected. 60 Subsequent events through the prism of financial auditor Subsequent events through the prism of financial auditor From the point of view of financial auditor, such intervention on subsequent events is provided for by International Standard on Auditing ISA 560 “Subsequent events”. Auditor`s objective are: Botez An entity is to adjust the amounts recognized in its financial statements to reflect subsequent events when they bring new information relating to situation existing at the time of closing financial year. For example, settlement of a dispute on court judgment. Also, bankruptcy a customer for whom the entity had an adjustment for depreciation led directly to its registration. An entity shall not be required to adjust the amounts recognized in its financial statements to reflect subsequent events that do not lead to adjustment financial statements. In this situation subsequent events must be presented in note. For example, events which call into question a problem concerning business continuity. q g p From the point of view of financial auditor, such intervention on subsequent events is provided for by International Standard on Auditing ISA 560 “Subsequent events”. Auditor`s objective are: If, as result of procedures carried out the auditor identifies events that provide for an adjustment or presentation in the financial statements, the auditor must be able to determine so far as such an event is reflected, as appropriate, in those financial statements, in accordance with the general financial reporting applicable framework. 61 http://sceco.ub.ro Studies and Scientific Researches. Economics Edition, No 19, 2014 401 Suppliers 2. On 20 December 2013, a customer has initiated legal proceedings against the company in relation to a breach of contract. On 29 December 2013, the lawyers of entity has announced that is unlikely that it will be found responsible; therefore they have not been adjusted financial statements, but the situation was presented in notes. On 29 January 2014, the court found the company responsible for breaking a contractual clause and established payment of damages amounting to 130.000 RON. Products supplied, worth 200.000 RON, have been returned. pp Sale of the products has been registered under normal conditions 4111 = % 238.000 Customers 701 200.000 Revenue from sale 4427 38.000 Output VAT 3. On 19 February 2014, a customer has ceased its activity by financial difficulties. He has a debt to the company of 2.500 RON. The board of directors considered that the amount is not relevant and established that is not necessary to adjust. Debt instruments has been registered as normal commercial operation Debt instruments has been registered as normal commercial operation 4111 = % 2.500 Customers 701 2.101 Revenue from sale 4427 399 Output VAT 4111 = % 2.500 Customers 701 2.101 Revenue from sale 4427 399 Output VAT Case study y The Company ABC closed financial year to 31 December 2013. General meeting of shareholders has been convened for 28 February 2014. The threshold of materiality for financial statements as a whole was fixed by auditor at 100.000 RON. For the first time on 20 February it is found that: 1. On 14 January 2014, has been identified a significant fraud. Debts in the amount of 125.000 RON, recorded in the accounts of fictitious suppliers have been paid in bank accounts established by an employee of the company. Events have occurred within the last seven months. The employee was dismissed immediately; they have been initiated legal proceedings against him. The employee has recorded in placeholder different services on behalf of expenditure in services from a third party. Fictitious bills did not include VAT. 628 = 401 125.000 Other expenses with third party Suppliers After he operated fictitious payment in the accounts, where he raised money. 401 = 5121 125.000 Suppliers Bank account 6812 Expenses Provisions for litigations 3. Cessation of a client to a recent date from the date of closure of the exercise indicates a high probability of unpaid and represents, therefore, a further event. As a result, had to be recorded an adjustment for doubtful debts of 2.500 RON, but the entity considered amount irrelevant and did not change their accounts. 3. Cessation of a client to a recent date from the date of closure of the exercise indicates a high probability of unpaid and represents, therefore, a further event. As a result, had to be recorded an adjustment for doubtful debts of 2.500 RON, but the entity considered amount irrelevant and did not change their accounts. Procedures carried out by the management sults of these findings, the management company has taken following steps: As results of these findings, the management company has taken following steps: 1. Fraud was committed by the employee during the course of the financial year. Fraud, errors and irregularities that appear before the date of end-of-year but that are discovered only beyond the year closed suppose adjusting elements and, therefore, 62 Botez financial statements would need to be modified according to the fraudulent activity to the end of the year. financial statements would need to be modified according to the fraudulent activity to the end of the year. To highlight this fact the entity registered a claim on former employee with the value of 125.000 RON. 461 Sundry debtors Debtors income Debtors income Sundry debtors by correcting such result for the year. 2. At the end of the year, the company of such entry in notes about the debt was lacking. However, after the end of the financial year, the court found the company liable for breach of contract. Judicial procedure was initiated on 20 December 2013 (with 10 days before the end of financial year). They were, therefore, evidence about the conditions existing at the end of the year. IAS 10 requires that these conditions are to be taken into account in order to determine whether or not you have recognized a debt. In this case, financial statements require an adjustment for that has been fulfilled criteria for the recognition of a provision for litigation. As a result, the entity has amended financial statements by recording a provision for litigation in value of 130.000 RON and corrected information properly in the explanatory notes. 130.000 Measures, as determined by auditor on fraud issues y • Additional adjustment of the financial statements by the recording of additional claims on an employee for fraud of 50.000 RON • Additional adjustment of the financial statements by the recording of additional claims on an employee for fraud of 50.000 RON 461 = 754 50.000 Sundry debtors Debtors income 461 = 754 50.000 Sundry debtors Debtors income 461 Sundry debtors • Inclusion in your letter to those charged with governance weaknesses have been observed with regard to internal control. • Inclusion in your letter to those charged with governance weaknesses have been observed with regard to internal control. Procedures carried out by auditor on judicial proceedings • Obtaining a copy of the judgment and other correspondence confirming that the entity has been found guilty and must pay for damage customer or verification of confirmation of payment to the customer; • Obtaining a copy of the judgment and other correspondence confirming that the entity has been found guilty and must pay for damage customer or verification of confirmation of payment to the customer; • To ensure that a provision has already been admitted to the original decision of the company; • To ensure that the value of provision is reasonable in relation to its judgment; • Obtaining a written statement from management by confirming treatment of provision. Completing letter concerning statements made in writing by management in accordance with International Standard on Auditing ISA 580 “Written representation” with his opinion on his guilt and the confirmation of lodging entity provision. Procedures carried out by the financial auditor Economics Edition, No 19, 2014 http://sceco.ub.ro http://sceco.ub.ro • Confirm directly with the suppliers of the balances and transactions for the period covered by audit as a result of carrying out these procedures have not been found errors; • Review invoices for the purchase and increase professional skepticism and investigation with respect to their authenticity; • Review files relating to human resources to find evidence of disciplinary proceedings taken against employee. This will also confirm compliance with laws and regulations, in particular in relation to the present legislation employment and retention of funds; • Testing of controls to identify any other weaknesses which may indicate management employees or fraud; • Testing of controls to identify any other weaknesses which may indicate management employees or fraud; g p y • Obtaining written statements from management on fraud issues; • Discussions with the lawyer’s entity with regard to the possibility of recover funds diverted. • Discussions with the lawyer’s entity with regard to the possibility of recover funds diverted. Procedures carried out by the financial auditor Fraud, risk factors of fraud are referred to into International Standard on Auditing ISA 240 “Auditor`s responsibility for fraud in an audit of financial statements”. That fraud has taken place within entity increases the risk of material misstatement due to fraud. As a result, the auditor must apply supplementary procedures. Audit procedures which must be carried out in order to ensure that fraud has been correctly identified and accounted in the financial statements include: • Recalculation of sums involved; In doing the recalculation values laid down by the entity, the auditor also carried out investigations and found out that the amount of 50.000 RON representing fictitious payment has not been included in the accounts drawn up by that entity. • Recalculation of sums involved; • Discussions with the management of the way in which such a fraud took place and why it took seven months to discover fraud (controls should be able to prevent, detect and correct material distortions in a short period of time); • Discussions with the management of the way in which such a fraud took place and why it took seven months to discover fraud (controls should be able to prevent, detect and correct material distortions in a short period of time); By reviewing and applying supplementary procedures for the assessment of internal control, the auditor finds that there is a procedure to provide that payments must be approved by at least two persons and that any new supplier must be assessed by an assessment record. The establishment in circumstances in which the accounting officer has discovered fraud and what working procedures and control (if equipped) contain weaknesses that have allowed employee to commit fraud by this procedure the auditor also found what are the weaknesses of the system of internal control which permitted that the fraud is discovered so late. • Perform procedures by the fund on operations in the vicinity of closing the long entries (in particular those close together, or at the end of the year); • Perform procedures by the fund on operations in the vicinity of closing the long entries (in particular those close together, or at the end of the year); 63 Studies and Scientific Researches. Economics Edition, No 19, 2014 Studies and Scientific Researches. Botez References Accounting Reglementations in accordance with the European directives, approved by OMFP 3055/2009, with completions and modifications. CECCAR (2010), Guide for application OMFP 3055/2009, Bucharest. International Accounting Standard IAS 10 “Events after the Reporting Period”. International Standard on Auditing ISA 560 “Subsequent events”. References Accounting Reglementations in accordance with the European directives, approved by OMFP 3055/2009, with completions and modifications. CECCAR (2010), Guide for application OMFP 3055/2009, Bucharest. International Accounting Standard IAS 10 “Events after the Reporting Period”. International Standard on Auditing ISA 560 “Subsequent events”. Procedures carried out by auditor on uncertain • Discuss customer management reason for which an adjustment has not been registered; • Discuss customer management reason for which an adjustment has not been registered; • Finds that the amount in question in immaterial for financial statements and, under the circumstances in which this amount remains immaterial and at the stage of completion, to express an unqualified opinion; • Finds that the amount in question in immaterial for financial statements and, under the circumstances in which this amount remains immaterial and at the stage of completion, to express an unqualified opinion; g p p q p • Adjusting financial statements by the formation of a provision for doubtful debts customer for 2.500 RON. • Adjusting financial statements by the formation of a provision for doubtful debts customer for 2.500 RON. 6814 = 491 2.500 Expenses Adjustments for debtors 64 References 65

https://openalex.org/W4210584026

https://ieeexplore.ieee.org/ielx7/6287639/9668973/09694597.pdf

English

A Blockchain-Based Framework for Distributed Agile Software Development

IEEE access

cc-by

Received October 5, 2021, accepted January 7, 2022, date of publication January 27, 2022, date of current version February 17, 2022. Digital Object Identifier 10.1109/ACCESS.2022.3146953 Digital Object Identifier 10.1109/ACCESS.2022.3146953 MUHAMMAD SHOAIB FAROOQ 1, ZAREEN KALIM 1, JUNAID NASIR QURESHI2, SAIM RASHEED3, AND ADNAN ABID 1, (Member, IEEE) 1Department of Computer Science, University of Management and Technology, Lahore 54000, Pakistan 2Department of Computer Science, Bahria University Lahore Campus, Lahore 54000, Pakistan 3Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 80200, Saudi Arabia Corresponding author: Muhammad Shoaib Farooq ([email protected]) ABSTRACT Distributed Agile Software Development (DASD) is the most important approach for the modern software industry that allows geographically independent software development. In the past, different tools and frameworks were proposed to solve communication and collaboration issues in DASD but they lacked transparency, trust, traceability, and security. These shortcomings resulted in project failure or overdue, customer dissatisfaction, project deal cancellations, and payment clashes between the customers and development teams. This paper addresses and overcomes the major issues of transparency, trust, security, traceability, coordination, and communication in DASD by embedding blockchain technology. We have proposed a novel blockchain-based framework named as AgilePlus which executes the smart contracts on a private ethereum blockchain for acceptance testing, secure payment, verification of developer’s payment requirements, and automatic payment distribution into the digital wallets of development teams. The execution of these smart contracts automatically assign penalties to the customers for late or non- payments and also to the developers for overdue tasks. Moreover, we have also solved the blockchain’s scalability challenge in AgilePlus by utilizing Interplanetary File System (IPFS) as off-chain storage. Lastly, experimental results prove that the proposed framework enhances transparency, communication, coordination, traceability, security and solves trust issues of both customers and developers in DASD. INDEX TERMS Distributed agile software development, blockchain technology, ethereum blockchain, smart contracts, interplanetary file system. INDEX TERMS Distributed agile software development, blockchain technology, ethereum blockchain, smart contracts, interplanetary file system. is work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ A Blockchain-Based Framework for Distributed Agile Software Development MUHAMMAD SHOAIB FAROOQ 1, ZAREEN KALIM 1, JUNAID NASIR QURESHI2, SAIM RASHEED3, AND ADNAN ABID 1, (Member, IEEE) 1Department of Computer Science, University of Management and Technology, Lahore 54000, Pakistan 2Department of Computer Science, Bahria University Lahore Campus, Lahore 54000, Pakistan 3Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 80200, Saudi Arabia Corresponding author: Muhammad Shoaib Farooq ([email protected]) I. INTRODUCTION AgilePlus connect its users by providing a user-friendly interface and six virtual walls through Decentralized Applications (DApps). AgilePlus consists of six layers named as 1. Agreement, 2. Requirements Elicitation, 3. Prioritization, 4. Design and Development 5. Testing and 6. Payment layer. Each layer has its separate virtual wall which is activated when the process of previous layer is finished by the customer and development team. This framework also executes the smart contracts for acceptance testing, secure payment, verification of developer’s payment requirements, payment distribution, and assigns penalty to its users. All these transactional steps are stored in a private ethereum blockchain. The features of blockchain technology as shown in Fig. 2, ensure trust by storing data in a way that reduces the risk of hacking or tampering the system [18]. Customers and developers face major challenges in dis- tributed environments that need to be addressed for meeting the intended objectives of a project. Customers at geographi- cally distributed locations are concerned about the software development process according to their requirements and satisfaction. However, developers do not trust customers for timely payments, project cancellation cost, and their coordi- nation with the development team. Thus, all these issues can be solved by combining blockchain technology with DASD as blockchain plays a vital role in creating trust between geographically dispersed customers and developers around the globe. g Blockchain technology can have significant impacts on DASD because of its decentralized nature, immutability, dis- tributed database, and distributed ledger [13]. Blockchain is a sequence of blocks that holds the complete record of transactions [14] verified and validated by peer-to-peer nodes in the network without using any central authority [15]. Recent innovations in blockchain technology have enabled the creation of smart contracts that automatically stores and executes code on the blockchain [16]. It also deals with the transferring of ether cryptocurrency in digital wallets without involving any trusted third party [17]. Most of the emerging systems have been shifted on a peer-to-peer, decentralized network, and using transparent distributed ledger systems with a diverse data structure [19]. Therefore, an efficient blockchain embedded framework is a major requirement to overcome all the challenges in DASD. We have also solved the scalability challenge in AgilePlus by storing the huge volume of records about customers, developers, and their communication on off-chain storage IPFS. It reduces the data load on blockchain and enhances the performance of blockchain transactions. I. INTRODUCTION successful software development as shown in Fig.1. However, DASD is the combination of both agile software develop- ment and distributed software development or global software engineering [6]. The DASD is based on geographically dis- persed agile development teams which offer various advan- tages such as increased quality, reduced development time, and cost [7]. Over the last few decades, Distributed agile software devel- opment (DASD) has been used as a common practice with the increase in global software development requirements and most of the software companies have adopted this approach [1]. Software development teams working collab- oratively at different geographical locations is known as Dis- tributed software development [2]. Agile is a set of principles and values for software development based on incremental and iterative development, where requirements and solutions are evolved by the collaboration between self-organizing and cross-functional teams [3], [4]. Agile software development promotes early delivery, adaptive planning, testing for contin- uous improvement, and quick response to changes proposed by the customer [5]. The agile life cycle follows six phases for Many challenges have been faced by development teams when the software development process is based on both agile and distributed software development [8]. The major chal- lenge in DASD is to achieve frequent communication and col- laboration [9]. Most of the risks and challenges in DASD have been associated with geographical distance which involves, trust, traceability, transparency, security, and communica- tion [11], [12], [23], [7], [10]. Recent studies [21], [23] have suggested that current tools for communication can support DASD practices by overcoming the issues of com- munication and collaboration but these researches lacked The associate editor coordinating the review of this manuscript and approving it for publication was Hailong Sun . 17977 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 1. Agile software development life cycle. t t t t bilit d it Th f ll FIGURE 2 Bl k h i f t i A il Pl FIGURE 2. Blockchain features in AgilePlus. FIGURE 1. Agile software development life cycle. FIGURE 1. Agile software development life cycle. FIGURE 1. Agile software development life cycle. FIGURE 2. Blockchain features in AgilePlus. trust, transparency, traceability, and security. Therefore, all these deficiencies in state-of-the-art studies resulted in over- due or project failure, project deal cancellations, customer dissatisfaction, and payment clashes between customers and developers. tions and protects the system against 51% attacks. I. INTRODUCTION Younas et al. [23] proposed and evaluated the Agile Devel- opment Cloud Computing framework (ADCC) that unifies agile development and cloud computing for local and dis- tributed agile development environments. Although evalua- tion results proved that cloud computing in ADCC framework overcomes the problems of face-to-face communication, transparency, scalability, and resource management, yet it has several challenges of data security risks, privacy, interoper- ability, and high cost. Moreover, ADCC lacks blockchain technology to overcome traceability, security, and trust issues. g The novelty of this research is that we have embedded blockchain technology in a user-friendly framework as the social website for solving trust issues of both customers and developers. To the best of our knowledge, a blockchain embedded framework for DASD has not been proposed yet in previous studies. We have executed the smart contracts for acceptance testing, secure payment, verification of devel- oper’s payment requirements, and automatic payment distri- bution between development teams by transferring ethers into their digital wallets. These ethers can be converted into fiat money by using an exchange/ currency conversion service. The execution of smart contracts also assign penalties to both customers and developers, if they have not satisfied the terms and conditions within due date. Moreover, we have solved the blockchain’s scalability issue by utilizing IPFS and also implemented JSON objects. Stray et al. [24] discussed the usage of slack tool for coor- dination and communication in DASD. Slack acts as a forum for customers and development teams to share files, arrange meetings, and improve coordination between customers and developers. This tool is useful in solving collaboration and communication issues but it does not overcome the traceabil- ity, trust, transparency, and security challenges. Qureshi et al. [25] proposed a novel Communication and Coordination (C&C) framework using scrum methodology. It involves distributed agile teams located at multiple loca- tions having a scrum master who acts as technical support person and communication coordinator. Although proposed framework achieved goals of decision making, consensus on sprint backlog, problem-solving, communication, and coor- dination, yet it lacks blockchain technology to solve the traceability, security, trust, and transparency issues. The remaining paper is organized as follows: Section II presents the related work and Section III describes the pre- liminaries used in our framework. Our proposed framework AgilePlus has been presented in Section IV and Section V presents the performance evaluation of AgilePlus. Then, Section VI presents the discussion of this research. I. INTRODUCTION However, all the data within six layers of AgilePlus is stored in blockchain to track the work progress of agile teams. The major objective of this study is to address the key chal- lenges that are associated with the geographical distance in DASD by using blockchain technology. Therefore, AgilePlus framework deploy and execute smart contracts to overcome the issues of trust, transparency, traceability, security, com- munication, and coordination. The key contributions of this research are as follows: We have proposed a blockchain-based framework named as AgilePlus in order to address all the issues faced by customers and developers. This framework embeds a private ethereum blockchain which has been considered promising for improving security as it promotes secure payment transac- We have proposed the AgilePlus framework for DASD which integrates private ethereum blockchain to prevent 51% attacks. The proposed framework uses DApp to provide its users with a user-friendly interface for communication and collaboration. AgilePlus deploy and execute smart contracts 17978 VOLUME 10, 2022 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development ogy to address security, traceability, transparency, and trust problems. for acceptance testing, secure payment, verification of devel- oper’s payment requirements, automatic payment distribu- tion between development teams, and assign penalties to its users if they have not satisfied the terms and conditions on time. AgilePlus allows the users to send and receive ether cryptocurrency in digital wallets for payments which can be converted into fiat currency by using a currency conversion service/ exchange. AgilePlus stores all the records of six lay- ers in blockchain for tracking the work progress of distributed agile teams. Moreover, we have utilized Interplanetary File System (IPFS) as off-chain secondary storage to solve the scalability challenge of blockchain and also implemented JSON objects. Lastly, we have evaluated the performance of AgilePlus based on the increase in chain size over the addition of block’s new transactions and latency for retrieving the longest chain. Lampropoulos and Siakas [22] investigated the communi- cation challenges in DASD and the usage of social media as a collaboration and communication tool. They also explored that effective communication, cultural difference, time zone difference, coordination, and trust are the main challenges in DASD. However, the use of social media only overcomes coordination and communication issues. Furthermore, this research does not overcome the major problems in DASD such as traceability, security, trans- parency, and trust. I. INTRODUCTION Finally, Section VII draws the conclusion and presents the future directions of this research work. Although many researches have been conducted to over- come the communication and collaboration challenges in DASD, yet several studies have discussed the usage of blockchain technology to ensure security, transparency, trace- ability, and trust between users. II. RELATED WORK Demi [26] proposed a blockchain-enabled framework for requirements traceability and management throughout the software development life cycle. The proposed framework mitigates the need to involve tracing tools or third-party requirements management. It also provides a distributed and collaborative environment for increasing the trustworthiness, transparency, and accuracy of the requirements traceability process. However, it is not suggested for DASD as it only solves the issues of requirements engineering process. Many researchers have proposed different tools and frame- works for DASD to overcome the challenges faced by cus- tomers and development teams. Most of them solved the com- munication, collaboration, and coordination issues but little has been explored to solve the trust, traceability, transparency, and security issues by using blockchain technology. Kukreja [20], [21] proposed Winbook framework similar to the social website that integrates user-friendly features such as virtual wall, posts, comments like Facebook, and con- tent organization using color-coded labels, similar to Gmail. These Winbook features overcome the communication and collaboration issues. However, Winbook does not have the capability of prioritization and it lacks blockchain technol- Rene et al. [27] proposed ChainSoft, an efficient platform for outsourcing the development of software and automatic payments by using blockchain technology. This system has been implemented using Github/Travis CI and ethereum 17979 17979 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development contracts, and store data. It uses a peer-to-peer (P2P) network that makes it distributed. smart contracts for solving the challenges of collaboration, security, trust, and testing in a distributed software develop- ment environment. However, this platform does not solve the challenges of transparency and traceability. B. SMART CONTRACTS Marchesi et al. [28] presented the applications of blockchain technology in agile development. They suggested the approach for Scrum or Lean- Kanban processes to deploy smart contracts on ethereum blockchain for accep- tance testing and payment. This approach solves the trust, coordination, and transparency issues for Scrum or Lean- Kanban processes but it is not suggested for the distributed environments. Moreover, the issues of security and traceabil- ity have not been addressed by using blockchain technology. Smart contracts [30] are the lines of code or computer pro- grams that are stored in the ethereum blockchain and exe- cute automatically when certain terms and conditions are satisfied. These smart contracts or transaction protocols for ethereum are written in solidity language and executed on the blockchain in order to document events according to the conditions of an agreement or contract. The objective of using smart contracts is to reduce the involvement of trusted third parties and malicious exceptions. All the above discussions show that still there is a need for an efficient framework to overcome all the major issues of traceability, security, trust, and transparency in DASD by using blockchain technology. However, there is not an adequate framework proposed yet to address all these prob- lems which are faced by customers and agile teams having geographical differences. In this research, we have proposed a novel AgilePlus framework to fulfill the gaps in previous frameworks and researches. III. PRELIMINARIES This section presents the preliminaries used in our proposed framework. It describes the major components of blockchain that have been used in AgilePlus which include Ethereum Blockchain, Smart Contracts, Decentralized Applications (DApps), and Interplanetary File System (IPFS). C. DECENTRALIZED APPLICATIONS (DApps) DApp [31] is one of the main features of ethereum which is also known as a digital program or an application that runs on a peer-to-peer network. DApps are comprised of smart con- tracts which have the user-defined code in order to perform the defined tasks. Ethereum Virtual Machine (EVM) is used to deploy and execute this code. Therefore, DApps consist of a frontend or interface to communicate with users and a smart contract or backend code that runs on a decentralized P2P network. The novelty of our work is that we have used blockchain technology in a user-friendly framework to improve DASD. We have solved the customer’s trust issues by executing smart contracts for acceptance testing. We have also solved the development team’s trust issues for secure payment, veri- fication of their payment requirements, and payment dis- tribution between them by executing smart contracts. The execution of smart contracts automatically assign penalties to the customers and developers for not satisfying the require- ments within due date. AgilePlus also provide development teams with the option of using an exchange/ currency conver- sion service for converting ethers into fiat currency. Further- more, we have used IPFS as off-chain storage to solve the blockchain’s scalability issue and also implemented JSON objects. IV. THE PROPOSED FRAMEWORK: AGILEPLUS This section presents a transparent, scalable, and secure framework to manage DASD for geographically scattered customers and development teams by integrating blockchain. We have proposed AgilePlus framework which follows the agile software development life cycle and embeds private ethereum blockchain to manage the transactions of Agile- Plus efficiently for posts, customer’s quick feedback, accep- tance testing, verification of developer’s payment require- ments as per business logic, and transferring of payments securely in the digitals wallets of development teams. The blockchain technology embedded in AgilePlus framework automatically assigns penalty to the customers for late or non-payments and also to the developers for overdue tasks. All the characteristics of blockchain make it suitable technol- ogy for agile software development in a distributed environ- ment with distributed ledger, distributed database, and strong security properties. The ecosystem of AgilePlus is presented in Fig. 3 which covers all the major features of blockchain. D. INTERPLANETARY FILE SYSTEM (IPFS) IPFS [32], [33] refers to the protocol that uses a P2P net- work for sharing and storing data in a content-addressable distributed file system. The data stored on IPFS is secure from modification as it uses the cryptographic identifier. There- fore, the cryptographic identifier is changed if data needs to be modified. A unique hash value is generated cryptographi- cally for the identification of data files stored on IPFS which does not allow the duplication of files. This cryptographic hash is stored in the blockchain to solve its scalability issue by reducing the data load on it. A. ETHEREUM BLOCKCHAIN Ethereum [29] is an open-source, and decentralized blockchain that consists of the smart contract functionality. It is also known as a public ledger for recording and verify- ing transactions. Ethereum software platform has its digital cryptocurrency known as ether (ETH) which can be shared between accounts that are connected with the ethereum blockchain network. Users interact with their ethereum accounts through ethereum wallets or applications. This dis- tributed blockchain network is also used to run DApps, smart VOLUME 10, 2022 VOLUME 10, 2022 17980 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 3. AgilePlus blockchain ecosystem. The main components of AgilePlus blockchain ecosystem are infrastructure, DApps, identity and privacy, money trans- actions, exchanges, wallets, miners, consensus mechanisms, privacy of user data. The blockchain provides infrastructure for building DApps in order to facilitate AgilePlus users with a user-friendly interface. Customers and developers can FIGURE 3. AgilePlus blockchain ecosystem. privacy of user data. The blockchain provides infrastructure for building DApps in order to facilitate AgilePlus users with a user-friendly interface. Customers and developers can connect with AgilePlus framework via DApp and access the blockchain network. All the stakeholders can use six The main components of AgilePlus blockchain ecosystem are infrastructure, DApps, identity and privacy, money trans- actions, exchanges, wallets, miners, consensus mechanisms, distributed ledger, and distributed storage. AgilePlus frame- work deals with secure payment transactions, payments, and 17981 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 4. Layers in AgilePlus framework. ing several credentials which include Name, Email, Contact Number, and Project Domain to complete the registration process. When a new customer registers AgilePlus, the admin sets up a new project by defining its specific name so that if any user is assigned to multiple projects, he can easily switch between these projects. Admin assigns the customer’s project to the project manager who assign tasks to the expert devel- opers. Admin also assigns a private key to all the stakeholders so that they can log in to the specific project by selecting its name and entering the private key or passcode which allows them to initiate the first layer of agreement. This private key is recorded in the blockchain database of AgilePlus. We have also solved the blockchain’s scalability problem as its design does not allow to store the huge volumes of records on it. A. AGILEPLUS REGISTRATION PROCESS The proposed framework follows blockchain architectural style and the seven-layered blockchain architecture for Agile- Plus has been presented in Fig.6. All the stakeholders register AgilePlus framework by provid- ing their credentials. The project managers and developers register themselves as a candidate by providing their Name, Email, Contact Number, Qualification, Skills, and Experi- ence. A formal interview of the shortlisted candidate has been scheduled by admin of AgilePlus in front of selection board using the English language as a medium of communication. The selected candidates as developers have been recruited within the software project according to their area of exper- tise. However, the customer registers AgilePlus by provid- A. ETHEREUM BLOCKCHAIN Therefore, AgilePlus utilize IPFS as off-chain secondary storage in order to store all the data of regis- tered customers, developers, messages, group chat, and video conferences for enhancing the performance of blockchain transactions by reducing data load on it. This data has a unique IPFS hash which is stored in the blockchain to view records. However, the data of six layers in AgilePlus is stored in ethereum blockchain. FIGURE 4. Layers in AgilePlus framework. B. SCALABILITY virtual walls for effective communication and coordination. The components of AgilePlus blockchain ecosystem provide more control over information to the geographically dis- persed users and promote data transparency in the framework. AgilePlus consists of six layers as shown in Fig.4. It involves the participation of customers in all these layers to track the work progress of development teams by using blockchain traceability. Scalability refers to the property of a system to manage a large amount of data or storage volume and perform efficiently by adding resources to the system. It is one of the major issues in blockchain as it manages a large amount of transaction data in a short time. Therefore, AgilePlus uses IPFS as off-chain distributed storage system to overcome the scalability chal- lenge in blockchain. IPFS uses cryptographic hash which is stored by using P2P network in a decentralized manner. The large volume of AgilePlus user record has been stored on IPFS instead of blockchain. Customers data stored on IPFS involves their registration details such as Name, Email, Contact Number, and Project Domain while the developers data stored on IPFS includes their registration details such as Name, Contact Number, Email, Qualification, Skills, and Experience Duration in the software industry. Moreover, all the messages, group chat, and video conference meetings on AgilePlus have also been stored on IPFS. This decrease in the large amounts of AgilePlus data on blockchain results in the faster performance of transactions. The AgilePlus framework has been shown in Fig. 5 which execute smart contracts in the testing layer for acceptance testing and in the payment layer for verification of devel- oper’s payment requirements and payment distribution in the form of ethers between development teams across the globe. AgilePlus also verifies all the terms and conditions agreed by customers and developers by executing smart contracts. It assigns automatic penalty to the developers if user story has not been completed within due date. Moreover, AgilePlus assigns penalty to the customers in case of late or non- payments. All these steps are recorded and stored in the private ethereum blockchain. 1) INTERFACE LAYER The interface layer consists of AgilePlus user-facing app, decentralized applications, and a web portal that connects the customers and developers to the AgilePlus system. All the users having digital wallets in AgilePlus interact with the user interface to initiate the agile software development process. 17982 17982 VOLUME 10, 2022 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 5 AgilePlus framework FIGURE 5. AgilePlus framework. FIGURE 5. AgilePlus framework. FIGURE 5. AgilePlus framework. 4) TRUST LAYER The trust layer deals with the consensus algorithms such as byzantine fault tolerance and proof-of-work. It also deals with the smart contract’s security analysis, verifica- tion of newly added blocks, and consensus protocols for transactions. The execution results are stored in blockchain layer. The customers and development teams having digital wal- lets in AgilePlus interact with a user interface that consists of web portal and DApp. Web portal allows the customers and development teams to post, comment, chat, send and receive project payments in the form of ether cryptocurrency. The development teams can also use a currency conversion service in order to convert these ethers into fiat currency. Smart contracts are set of rules which define business logic in AgilePlus and the protocols for transactions are defined by these rules. The chain’s credibility is maintained when the rules are defined by smart contracts and the trust layer 7) SECURITY LAYER The security layer is essential for the network of blockchain as it is vulnerable to several security attacks such as 51% attacks. It connects and works parallel to the system and is composed of security algorithms and protocols in order to maintain the security of AgilePlus framework. 2) APPLICATION LAYER 6) INFRASTRUCTURE LAYER The application layer involves metadata of transactions, payments, and various records in AgilePlus such as posts, prototypes, and project agreements between customers and development teams. This layer is also composed of ether cryptocurrency. It connects the interface layer with the busi- ness logic in the form of smart contracts. The infrastructure layer consists of peer to peer network which distributes, verifies, and forwards the transaction mapped on ethereum blockchain within AgilePlus. More- over, it involves verification, communication, and distributed networking mechanisms. The transaction in AgilePlus gets broadcasted to the entire network. Then, each node verifies it according to the predefined parameters and the verified transaction gets stored in blockchain. 3) BUSINESS LOGIC LAYER The business logic layer is comprised of smart contracts within AgilePlus and it deals with the terms and conditions. For smart contracts, the business logic layer acts as an active database which includes invocation, execution of contracts, and rules of communication. 5) TRANSACTION LAYER In transaction layer, the transactions are triggered by Agile- Plus users or smart contracts. It also deals with mining, transaction validation, and block validation. 17983 VOLUME 10, 2022 1) LAYER 1: AGREEMENT LAYER In the first layer, customers and developers set their ini- tial terms and conditions which are recorded as smart con- tracts in the blockchain. These terms and conditions must be achieved in order to clear the acceptance test and enable payment option in the AgilePlus framework. An agreement is achieved when the customer and development team reach a consensus. Customer agrees with the terms and conditions VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 6. AgilePlus blockchain architecture. FIGURE 6. AgilePlus blockchain architecture. a blockchain network. Customers and development teams initiate their software projects by following the process in layers of AgilePlus framework as shown in Fig. 8. integrates consensus algorithms. AgilePlus embeds ethereum blockchain along with proof-of-work implementation for ensuring the chain’s temporal property. Although hashes are stored in blockchain, the data of six layers and private keys to log in to the software project are stored on peer-to-peer net- works and in database storage. The security layer is integrated with all other layers to ensure the authorized access of nodes. FIGURE 10. Modes of communication for requirement elicitation within AgilePlus. blockchain network which enables the requirement elicitation layer. Then, status of the project changes to ‘‘Requirement Elicitation’’ and it has been stored in blockchain notifying all the members in a network. Ethereum in AgilePlus framework involves the deploy- ment of smart contracts in blockchain i.e the data of trans- actions and code implementing the logic for transactions, amongst actors. So, the logic and data have been stored irreversibly which composes of smart contracts. According to the immutability principle of blockchain, the deployment of a smart contract does not allow its code to be changed [34]. The developer has to develop another smart contract in case of bugs or errors. He corrects the errors and this new smart contract will be deployed on the blockchain transferring all the existing data to the new smart contract. FIGURE 10. Modes of communication for requirement elicitation within AgilePlus. In AgilePlus framework, smart contracts have been com- piled into an executable ethereum virtual machine (EVM) bytecode or low-level programming language, and applica- tion binary interface (ABI). This byte code has been used to deploy the smart contract on ethereum which has been compiled by the compiler while ABI provides the interface between two modules of the program. The specification of contract’s function also uses ABI which has to be invoked. A smart contract’s ABI and contract address created during its deployment have been used to interact with the smart contract. These smart contracts must be compiled before deployment so that EVM can interpret and store the contract. Deployment of the smart contract as shown in Fig.9 has been recorded as a transaction in blockchain. to ensure that all the terms and conditions fixed by them in the agreement layer have been satisfied. D. THE LAYERED STRUCTURE OF AGILEPLUS AgilePlus framework is based on a six-layered structure which has been presented in Fig.7. The transactions of each layer are stored in blockchain notifying all the members of VOLUME 10, 2022 17984 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 7. AgilePlus layered structure. set by development team while development team accepts from both customers and developers to initiate the project. FIGURE 7. AgilePlus layered structure. FIGURE 7. AgilePlus layered structure. from both customers and developers to initiate the project. This agreement between customer and development team has been recorded as a smart contract and stored in the from both customers and developers to initiate the project. This agreement between customer and development team has been recorded as a smart contract and stored in the from both customers and developers to initiate the project. This agreement between customer and development team has been recorded as a smart contract and stored in the set by development team while development team accepts the terms and conditions set by customer. Therefore, agree- ment layer involves the acceptance of terms and conditions 17985 VOLUME 10, 2022 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 8. Process in layers of AgilePlus framework. FIGURE 8. Process in layers of AgilePlus framework. 17986 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 9. Deployment of smart contracts within AgilePlus framework. TABLE 2. JSON objects for developer’s agreement in agreement layer. TABLE 2. JSON objects for developer’s agreement in agreement layer. FIGURE 9. Deployment of smart contracts within AgilePlus framework. FIGURE 9. Deployment of smart contracts within AgilePlus framework. TABLE 1. JSON objects for customer’s agreement in agreement layer. 2) LAYER 2: REQUIREMENT ELICITATION LAYER In AgilePlus, customers post user stories on the virtual wall to elicitate requirements. Users can also use five different modes of communication in this framework as shown in Fig.10. All the user stories posted in requirement elicitation layer have been stored as transactions in blockchain. Then, develop- ment team interacts with the customer by replying with the comments on their posts or scheduling a video conference in case of having an issue with any user story. Customer answers all the questions and provides other solutions to the development team. All the posts along with comments have been stored in blockchain while group chat, messages, and video conferences have been stored on IPFS. When customer JSON objects as shown in Table.1 and Table.2 presents the contractual information of the customers and developers deployed on ethereum which are recorded as smart contracts 17987 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 11. Backlog prioritization factors in AgilePlus framework. and development team reach a consensus on user stories, the backlog list has been recorded as a smart contract which enables the prioritization layer The status of this approved FIGURE 12. Design and development in AgilePlus framework. FIGURE 11. Backlog prioritization factors in AgilePlus framework. FIGURE 11. Backlog prioritization factors in AgilePlus framework. and development team reach a consensus on user stories, the backlog list has been recorded as a smart contract which enables the prioritization layer. The status of this approved backlog list changes to ‘‘Prioritization’’ and it moves to the prioritization layer. This transfer has been registered in blockchain notifying all the members of blockchain network. However, if the development team and customer have failed to agree on a backlog list, then customer adds a new list of user stories in the requirement elicitation layer. This step has also been registered in the blockchain network and authorized when both customer and development team agrees with all the posted user stories. The customer can also change or add user stories at any stage of the project. All the members of blockchain network get notified of these changes. FIGURE 12. Design and development in AgilePlus framework. prototypes. Both customers and developers can also schedule a meeting on video conference to communicate their ideas and solutions to each other. 2) LAYER 2: REQUIREMENT ELICITATION LAYER If customer’s feedback is posi- tive for the posted prototypes, development team implements the specific user story while in case of customer’s negative feedback, the development team redesigns the prototypes for high priority task. When development team implements the user story with positive feedback, status of the high priority user story changes to ‘‘Implemented’’ and is stored in the blockchain. Then, the next layer of testing has been enabled for the implemented user story. The status of this implemented user story changes to ‘‘Testing’’ and stored in blockchain network notifying all the members. Fig. 12 shows the design and development of user stories in the AgilePlus framework. 3) LAYER 3: PRIORITIZATION LAYER The approved backlog list has been added to the prioritization layer where customer interacts with the development team by posting comments, posts, scheduling a video conference or group chat for selecting the important user story that must be completed initially by the development team. A user story in AgilePlus is prioritized keeping in view the following factors as shown in Fig.11. Finally, customer and development team prioritize user story from the backlog list according to its relevance and importance to the system which enables the next layer of design and development. The status of this prioritized user story changes to ‘‘Design and Development’’ and the high priority user story moves to the design and development layer. This transfer has been registered in the blockchain network notifying all the members. 5) LAYER 5: TESTING LAYER Testing layer involves testing of each implemented user story. The proposed framework executes the smart contracts for acceptance testing in order to validate implemented user stories. This layer verifies whether all the terms and condi- tions fixed by the customer in agreement and requirement elicitation layer have been successfully achieved. When the implemented user story satisfies all the terms and conditions present in smart contracts, the acceptance test is considered as ‘‘Pass’’. This step enables the next layer of payment and status 4) LAYER 4: DESIGN AND DEVELOPMENT LAYER The high priority user story moves to the design and devel- opment layer where development team designs and posts the prototypes on virtual wall of this layer. Customer view pro- totypes and then give feedback by rating and reviewing these 17988 VOLUME 10, 2022 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 13. Acceptance testing in AgilePlus framework. TABLE 3. JSON objects for acceptance test 1 in testing layer. FIGURE 13. Acceptance testing in AgilePlus framework. TABLE 3. JSON objects for acceptance test 1 in testing layer. TABLE 4. JSON objects for acceptance test 2 in testing layer. FIGURE 14. 51% Attack. network notifying all the members. When the development team fails to pass the acceptance test within user story’s due date, the execution of smart contract marks the status of a user story as ‘‘Overdue’’ and automatically assigns penalty to them. The status change and penalty for an overdue user story have been stored in the blockchain network notifying all the members. JSON objects as shown in Table.3 and Table.4 present the acceptance test results to verify whether all the terms and conditions fixed by customer in the agreement layer have been satisfied. FIGURE 13. Acceptance testing in AgilePlus framework. TABLE 4. JSON objects for acceptance test 2 in testing layer. FIGURE 14. 51% Attack. FIGURE 13. Acceptance testing in AgilePlus framework. TABLE 4. JSON objects for acceptance test 2 in testing layer. TABLE 4. JSON objects for acceptance test 2 in testing layer. FIGURE 14. 51% Attack. FIGURE 14. 51% Attack. FIGURE 14. 51% Attack. FIGURE 14. 51% Attack. FIGURE 13. Acceptance testing in AgilePlus framework. FIGURE 13. Acceptance testing in AgilePlus framework. TABLE 3. JSON objects for acceptance test 1 in testing layer. network notifying all the members. When the development team fails to pass the acceptance test within user story’s due date, the execution of smart contract marks the status of a user story as ‘‘Overdue’’ and automatically assigns penalty to them. The status change and penalty for an overdue user story have been stored in the blockchain network notifying all the members. ABLE 3. JSON objects for acceptance test 1 in testing layer. JSON objects as shown in Table.3 and Table.4 present the acceptance test results to verify whether all the terms and conditions fixed by customer in the agreement layer have been satisfied. 6) LAYER 6: PAYMENT LAYER In this framework, a private ethereum blockchain has been used to protect the system against 51% attacks. Miners or attackers use 51% attack as shown in Fig.14 to control more than 51% network’s hash rate for preventing the confirma- tion of new transactions, halting payments between all the users of AgilePlus, and causing a double-spending problem. A private blockchain provides an extra layer of security as the listed transactions are private. Moreover, it also restricts the participation of nodes to deploy consensus protocols and view transactions. Therefore, private blockchain plays a vital role in preventing 51% attacks and improving payment security in AgilePlus framework. of tested user story changes to ‘‘Payment’’. All these steps have been stored in blockchain notifying all the members of network. However, the acceptance test is considered as ‘‘Fail’’ if all the terms and conditions have not been success- fully satisfied by the development team. In case of acceptance test failure, the development team gets notified which allows them to move implemented user story back to the design and development layer where they fix issues, complete the missing requirements of customer, and implement it again if required. Then, smart contracts execute again for the accep- tance testing by moving user story to the testing layer. There- fore in both cases, AgilePlus assigns acceptance test status ‘‘Pass/Fail’’ as shown in Fig. 13, and stored in blockchain When the implemented high priority user story clears the acceptance test, the payment layer has been automatically 17989 17989 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development TABLE 7. JSON objects for customer’s penalty (no payment) in payment layer. TABLE 8. JSON objects for developer’s penalty in payment layer. TABLE 7. JSON objects for customer’s penalty (no payment) in payment layer. FIGURE 15. Payment and ether cryptocurrency conversion in AgilePlus framework. V. PERFORMANCE In this section, we have evaluated the performance and effec- tiveness of our proposed framework, AgilePlus, on real-case scenarios. TABLE 5. JSON objects for customer’s payment in payment layer. TABLE 5. JSON objects for customer’s payment in payment layer. requirements. If customer successfully satisfies all the pay- ment conditions of development team, a digital payment receipt of customer is generated and the payment is automat- ically distributed among the team members by transferring ethers into their digital wallets. Then, customer receives an email of the completed task from the development team. All the payment transactions have been registered in blockchain network notifying all the members. This step automatically enables the prioritization layer and changes the project status to ‘‘Prioritization’’. However, if customer does not satisfy the payment requirements of developers or pays late for the com- pleted task, the execution of smart contract assigns penalty to him. If customer does not pay for the completed user story within n weeks, his IP address automatically gets blocked from the AgilePlus and stored in blockchain notifying all the members. Here ‘‘n’’ is the number of weeks decided by the development team through consensus. TABLE 6. JSON objects for customer’s penalty (late payment) in payment layer. JSON objects as shown in Table.5, 6, 7, and 8 present the contractual details for payment and penalties assigned to the AgilePlus users to ensure trust. enabled. This layer generates a bill for the tested user story and prompts customer to pay for their completed task oth- erwise, the next user story will not be implemented by the development team. Customer must pay to the development team by sending ethers to their digital wallets in AgilePlus. These ethers can be converted using any currency conversion service into fiat currency such as PKR, AUD, EUR, CAD, and USD, etc. The payment and ether cryptocurrency conversion process in AgilePlus has been shown in Fig.15. FIGURE 15. Payment and ether cryptocurrency conversion in AgilePlus framework. C. PERFORMANCE EVALUATION This section presents the experimental evaluation of Agile- Plus on real-case scenarios in terms of performance. We have implemented the AgilePlus blockchain network where enti- ties transfer data and make transactions in real-time. HTTP requests of ‘‘GET’’ and ‘‘POST’’ have been set up by using the postman tool in order to interact with APIs. We have used the six functions as use cases to validate our proposed framework: 1. get_chain 2. connect_node 3. mine_block 4. add_transaction 5. is_chain_valid 6. replace_chain VOLUME 10, 2022 1. get_chain B. EVALUATION METRICS The metrics used for evaluating AgilePlus framework include block size evaluation and latency. These are briefly explained as follows: A. PERFORMANCE ASSESSMENT In order to test the efficiency of AgilePlus, we have used Spyder IDE Version 4.2.5 for implementing the AgilePlus blockchain network in python. For performance testing, The customer’s payment for the tested user story executes smart contract for the verification of developer’s payment 17990 VOLUME 10, 2022 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development FIGURE 16. The average increase in file size. FIGURE 17. Latency for retrieving the longest chain. FIGURE 17. Latency for retrieving the longest chain. FIGURE 16. The average increase in file size FIGURE 17 Latency for retrieving the longest chain FIGURE 16. The average increase in file size. FIGURE 17. Latency for retrieving the longest chain. Postman tool’s Version 8.5.1 has been used to interact with APIs. Moreover, we have also used matplotlib library in python for the data visualization of chain size and latency. A random transaction has been used to send HTTP requests for 700 repetitions in postman. For 700 blocks, the size of chain having a range between 0.448 KB to 500 KB, increases by adding new blocks with transactions in the blockchain. Fig. 16 shows that the increase in number of blocks increases the blockchain size. The average increase in size is 248 KB. Fig.16 also shows that the newly mined block having a large number of transactions affect the size of files by increasing it from 100 to 110 blocks and 590 to 600 blocks in a linear fashion. The number of transactions in a block is not defined at the development time for this particular use case which results in a slight increase in the size of newly mined blocks. 1) BLOCK SIZE EVALUATION The size of block refers to the amount of data stored in a block. A single block consists of transaction data in chain. The amount of data or block size has been examined to analyze the average increase in size of file/ blockchain. We have examined all the current chains in blockchain network of AgilePlus and replaced them with the network’s longest existing chain against that specific timestamp by following all the steps required for processing the HTTP request. From this process, we have analyzed the time taken for executing the replace_chain request of HTTP in order to evaluate the latency for achieving the current longest chain. Fig. 17 shows random delays in the time taken for execut- ing the HTTP requests. The reason behind these random increments and decrements in time for executing a request is the blockchain’s decentralized nature which involves multi- ple nodes and does not allow any central system to control the data. Furthermore, these multiple servers in the Agile- Plus blockchain network use their system capacities which depend on their local machine’s response time against multi- ple requests, speed of their machines, and internet bandwidth. Such factors cause latency to execute the HTTP requests on different servers. 2) LATENCY Latency is defined as the delay that occurs when the com- ponent of a system waits for another system component to respond to an action. In terms of blockchain network, it refers to the time between the submission of transaction to a network and the first acceptance confirmation by the network. D. COMPARISON OF AGILEPLUS FRAMEWORK WITH RELATED WORK 2. connect_node 3. mine_block We have also discussed some parameters that are present in our proposed framework and these parameters have been used for comparison with the related work in this domain. The 4. add_transaction 5. is_chain_valid 6. replace_chain 17991 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development TABLE 9. Comparison between AgilePlus framework and related work. TABLE 9. Comparison between AgilePlus framework and related work. TABLE 9. Comparison between AgilePlus framework and related work. p g execution of smart contracts verify whether all the terms and conditions of the customers and developers have been fulfilled. comparison between AgilePlus framework and related work has been shown in Table 9. Traceability: AgilePlus tracks the work progress of dis- tributed agile teams by storing each step in blockchain and notifying all the members in network. 2) SCALABILITY • It prevents 51% attacks IPFS has been used as off-chain storage in AgilePlus to solve the scalability problem of blockchain for storing the records of customers, developers, and their communication. This decrease in data load of blockchain results in the faster performance of AgilePlus transactions. • It allows secure payment and distribution between vir- tual agile teams. • A private key is required for all the stakeholders to log in to the specific project in order to restrict the participation of unauthorized nodes. 1) BLOCKCHAIN-BASED FRAMEWORK AgilePlus embeds private ethereum blockchain to ensure security, transparency, trust, and traceability. Security: The integration of private ethereum blockchain in AgilePlus is useful for the following reasons: 3) COMMUNICATION Transparency: The software development process is transpar- ent in AgilePlus as data is always available at any time to all the dispersed stakeholders and they get notified with each update. AgilePlus uses DApp for providing a user-friendly interface to its users. The stakeholders can use five communication modes in all six layers which include posts, comments, group chat, messages, and video conference to ensure effective communication. Trust: AgilePlus creates trust by executing smart con- tracts to achieve customers and developers satisfaction. The VOLUME 10, 2022 17992 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development 4) COORDINATION • Blockchain technology is useful for tracking the work progress of scattered agile teams by recording their per- formance in blockchain. • Blockchain technology is useful for tracking the work progress of scattered agile teams by recording their per- formance in blockchain. Our proposed framework involves consensus in agreement and requirement elicitation layers for coordination between customers and developers. The significant findings show that the integration of blockchain technology in AgilePlus successfully overcomes the issues of trust, security, transparency, traceability, communication, and coordination. These characteristics of blockchain avoid project deal cancellations, payment clashes between customers and developers, customer dissatisfaction, and project overdue or failure. Moreover, the utilization of IPFS as off-chain secondary storage greatly reduces the data load on blockchain which results in the faster performance of AgilePlus transactions. 5) TESTING AgilePlus execute smart contracts for acceptance testing to verify whether all the terms and conditions of customer have been successfully fulfilled by the development team. VI. DISCUSSION This section formally discusses the experimental results of our scalable, secure and transparent framework, AgilePlus, which is efficient in terms of performance. The proposed framework is useful in the successful development of agile- based projects, providing its users with a collaborative and distributed environment. AgilePlus addresses and overcomes the major challenges of trust, traceability, security, trans- parency, communication, and coordination in DASD by using blockchain technology due to its decentralized nature, con- sensus mechanism, distributed storage, and strong security properties. However, the lack of such fundamental factors in distributed environments may result in overdue or project fail- ures, payment clashes, project deal cancellations, customer dissatisfaction, and lack of trust. Our experimental results prove that blockchain technology is the major strength of this research which greatly enhances the software development process in DASD. However, the previous studies have not proposed an efficient, transparent, secure, and scalable framework by using blockchain technol- ogy to solve the major problems of trust, traceability, secu- rity, and transparency in DASD. These existing researches lacked blockchain technology and only solved the issues of coordination and communication in DASD. Therefore, our proposed framework, AgilePlus, outperforms all the previ- ous frameworks and researches for DASD by incorporating blockchain technology. The limitation of incorporating ethereum blockchain in AgilePlus framework involves high energy consumption as mining of block requires a lot of power. Moreover, data modification is hard, once it is stored in the blockchain. All the performance evaluation results show that the emerging technology of blockchain has a great potential to transform the future of DASD and can drive major changes in offshore software companies. The performance evaluation results of our proposed frame- work indicate that blockchain technology in AgilePlus plays a vital to eliminate all these barriers for the successful devel- opment of software projects in DASD. AgilePlus integrates a private ethereum blockchain and provides the users with six layers each having a separate virtual wall to follow the agile development life cycle. We consider that integration of blockchain technology in AgilePlus framework is quite beneficial for the following reasons: • The decentralized nature of blockchain supports DASD. • The decentralized nature of blockchain supports DASD. VII. CONCLUSION AND FUTURE WORK REFERENCES 2020. [7] M. Singh, N. Chauhan, and R. Popli, ‘‘A framework for transitioning of traditional software development method to distributed agile software development,’’ in Proc. Int. Conf. Issues Challenges Intell. Comput. Techn. (ICICT), vol. 1, Sep. 2019, pp. 1–4. [31] K. Wu, Y. Ma, G. Huang, and X. Liu, ‘‘A first look at blockchain- based decentralized applications,’’ Softw., Pract. Exp., vol. 51, no. 10, pp. 2033–2050, Oct. 2021. [8] K. B. Awar, M. S. I. Sameem, and Y. Hafeez, ‘‘A model for applying agile practices in distributed environment: A case of local software industry,’’ in Proc. Int. Conf. Commun., Comput. Digit. Syst. (C-CODE), Mar. 2017, pp. 228–232. [32] R. Kumar and R. Tripathi, ‘‘Blockchain-based framework for data storage in peer-to-peer scheme using interplanetary file system,’’ in Handbook of Research on Blockchain Technology. New York, NY, USA: Academic, 2020, pp. 35–59. [9] Y. I. Alzoubi and A. Q. Gill, ‘‘An empirical investigation of geographi- cally distributed agile development: The agile enterprise architecture is a communication enabler,’’ IEEE Access, vol. 8, pp. 80269–80289, 2020. [33] M. U. Javed, M. Rehman, N. Javaid, A. Aldegheishem, N. Alrajeh, and M. Tahir, ‘‘Blockchain-based secure data storage for distributed vehicular networks,’’ Appl. Sci., vol. 10, no. 6, p. 2011, Mar. 2020. [10] G. Ammad, U. Iqbal Janjua, T. M. Madni, M. F. Cheema, and A. R. Shahid, ‘‘An empirical study to investigate the impact of communication issues in GSD in Pakistan’s IT industry,’’ IEEE Access, vol. 7, pp. 171648–171672, 2019. [34] M. S. Farooq, M. Khan, and A. Abid, ‘‘A framework to make charity col- lection transparent and auditable using blockchain technology,’’ Comput. Electr. Eng., vol. 83, May 2020, Art. no. 106588. [11] M. A. B. Khalid, A. Farooq, and W. Mahmood, ‘‘Communication chal- lenges for distributed teams,’’ Int. J. Eng. Manuf., vol. 11, no. 1, pp. 19–28, Feb. 2021. [12] S. Maro, J.-P. Steghofer, E. Knauss, J. Horkoff, R. Kasauli, R. Wohlrab, J. L. Korsgaard, F. Wartenberg, N. J. Strom, and R. Alexandersson, ‘‘Man- aging traceability information models: Not such a simple task after all?’’ IEEE Softw., vol. 38, no. 5, pp. 101–109, Sep. 2021. [13] K. Salah, M. Rehman, N. Nizamuddin, and A. Al-Fuqaha, ‘‘Blockchain for AI: Review and open research challenges,’’ IEEE Access, vol. 7, pp. 10127–10149, 2019. [14] Z. Zheng, S. Xie, H. Dai, X. Chen, and H. Wang, ‘‘An overview of blockchain technology: Architecture, consensus, and future trends,’’ in Proc. VII. CONCLUSION AND FUTURE WORK The current state-of-the-art tools and frameworks for DASD are inefficient as they lack security, traceability, transparency, and trust. Thus, there is a great need for an efficient frame- work to solve all these issues. In this paper, we have pro- posed a novel blockchain embedded framework, AgilePlus, to overcome the major challenges of traceability, trans- parency, trust, security, communication, and coordination in DASD. AgilePlus uses DApp to provide its users with a user-friendly interface and stores the transaction records in blockchain for tracking the work progress of virtual agile teams. The proposed framework executes the smart contracts for acceptance testing, secure payment, and verification of developer’s payment requirements along with the payment distribution between team members by transferring ethers into their digital wallets. The execution of smart contracts automatically assign penalties to its customers for late or non-payments and also to the developers for overdue tasks. We have used a private ethereum blockchain in AgilePlus to provide security against 51% attacks. Moreover, we have also used IPFS as off-chain storage to solve the scalability issue of • Blockchain provides a user-friendly interface for six virtual walls through DApps to ensure effective com- munication and collaboration between distributed users across the world. • Blockchain uses consensus mechanism to improve coor- dination in DASD. • The ethereum blockchain in AgilePlus is useful for achieving automatic acceptance testing, verification of developer’s payment requirements, and payment distri- bution to the digital wallets of development team by executing smart contracts. • Blockchain technology facilitates the distributed users with digital wallets and cryptocurrency for payments such as ether (ETH). • Customer satisfaction is achieved as blockchain ensures transparency. • Blockchain technology ensures security as it provides secure payment transactions. Furthermore, AgilePlus uses a private ethereum blockchain to restrict the par- ticipation of nodes for preventing 51% attacks. 17993 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development blockchain. Finally, our experimental results have proved that integration of blockchain technology in AgilePlus success- fully overcomes all the above-mentioned major challenges which greatly enhances DASD. blockchain. Finally, our experimental results have proved that integration of blockchain technology in AgilePlus success- fully overcomes all the above-mentioned major challenges which greatly enhances DASD. [19] A. Khelifi, O. Aziz, M. S. Farooq, A. Abid, and F. REFERENCES [24] V. Stray, N. B. Moe, and M. Noroozi, ‘‘Slack me if you can! Using enter- prise social networking tools in virtual agile teams,’’ in Proc. ACM/IEEE 14th Int. Conf. Global Softw. Eng. (ICGSE), May 2019, pp. 111–121. [1] M. D. Kahya and Ç. Seneler, ‘‘Geographical distance challenges in dis- tributed agile software development: Case study of a global company,’’ in Proc. 3rd Int. Conf. Comput. Sci. Eng. (UBMK), Sep. 2018, pp. 78–83. [25] R. Qureshi, M. Basheri, and A. A. Alzahrani, ‘‘Novel framework to improve communication and coordination among distributed agile teams,’’ Int. J. Inf. Eng. Electron. Bus., vol. 10, no. 4, pp. 16–24, Jul. 2018. p pp [2] W. Aslam and F. Ijaz, ‘‘A quantitative framework for task alloca- tion in distributed agile software development,’’ IEEE Access, vol. 6, pp. 15380–15390, 2018. Int. J. Inf. Eng. Electron. Bus., vol. 10, no. 4, pp. 16–24, Jul. 2 [26] S. Demi, ‘‘Blockchain-oriented requirements engineering: A framework,’’ in Proc. IEEE 28th Int. Requirements Eng. Conf. (RE), Aug. 2020, pp. 428–433. [3] S. Alsaqqa, S. Sawalha, and H. Abdel-Nabi, ‘‘Agile software development: Methodologies and trends,’’ Int. J. Interact. Mobile Technol., vol. 14, no. 11, p. 246, Jul. 2020. [27] M. Król, S. Reñé, O. Ascigil, and I. Psaras, ‘‘ChainSoft: Collaborative software development using smart contracts,’’ in Proc. 1st Workshop Cryp- tocurrencies Blockchains Distrib. Syst., Jun. 2018, pp. 1–6. [4] A. M. Gheorghe, I. D. Gheorghe, and I. L. Iatan, ‘‘Agile software develop- ment,’’ Inform. Economica, vol. 24, no. 2, pp. 90–100, 2020. [28] V. Lenarduzzi, M. I. Lunesu, M. Marchesi, and R. Tonelli, ‘‘Blockchain applications for agile methodologies,’’ in Proc. 19th Int. Conf. Agile Softw. Develop., Companion, May 2018, pp. 1–3. [5] D. Kunda, M. Mulenga, M. Sinyinda, and V. Chama, ‘‘Challenges of agile development and implementation in a developing country: A Zambia case study,’’ J. Comput. Sci., vol. 14, no. 5, pp. 585–600, May 2018. [29] S. Ferretti and G. D’Angelo, ‘‘On the ethereum blockchain structure: A complex networks theory perspective,’’ Concurrency Comput., Pract. Exp., vol. 32, no. 12, pp. 1–11, Jun. 2020. [6] P. T. Robinson, ‘‘Communication network in an agile distributed software development team,’’ in Proc. ACM/IEEE 14th Int. Conf. Global Softw. Eng. (ICGSE), May 2019, pp. 100–104. [30] G. A. Oliva, A. E. Hassan, and Z. M. J. Jiang, ‘‘An exploratory study of smart contracts in the Ethereum blockchain platform,’’ Empirical Softw. Eng., vol. 25, pp. 1864–1904, Mar. VII. CONCLUSION AND FUTURE WORK Bukhari, ‘‘Social and economic contribution of 5G and blockchain with green comput- ing: Taxonomy, challenges, and opportunities,’’ IEEE Access, vol. 9, pp. 69082–69099, 2021. [20] N. Kukreja and B. Boehm, ‘‘Process implications of social networking- based requirements negotiation tools,’’ in Proc. Int. Conf. Softw. Syst. Process (ICSSP), Jun. 2012, pp. 68–72. For future work, we have planned to provide users with the option to give ratings and reviews for AgilePlus at the end of project. We have also planned to integrate a translator in AgilePlus to support different languages. Furthermore, our future work also involves the execution of smart contracts for solving task distribution issues between development teams in AgilePlus framework. [21] N. Kukreja, ‘‘Winbook: A social networking based framework for collab- orative requirements elicitation and WinWin negotiations,’’ in Proc. 34th Int. Conf. Softw. Eng. (ICSE), Jun. 2012, pp. 1610–1612. [22] G. Lampropoulos and K. Siakas, ‘‘Communication in distributed agile software development: Impact of social media-social networking,’’ in Proc. SQM, vol. 26, 2018, p. 43. [23] M. Younas, D. N. A. Jawawi, A. K. Mahmood, M. N. Ahmad, M. U. Sarwar, and M. Y. Idris, ‘‘Agile software development using cloud computing: A case study,’’ IEEE Access, vol. 8, pp. 4475–4484, 2020. REFERENCES IEEE Int. Congr. Big Data, Jun. 2017, pp. 557–564. [15] M. Niranjanamurthy, B. N. Nithya, and S. Jagannatha, ‘‘Analysis of blockchain technology: Pros, cons and SWOT,’’ Cluster Comput., vol. 22, no. S6, pp. 14743–14757, Nov. 2019. MUHAMMAD SHOAIB FAROOQ is currently working as a Professor in computer science with the University of Management and Technology, Lahore. He is also an Affiliate Member of George Mason University, USA. He possesses more than 26 years of teaching experience in the field of computer science. He has published many peer-reviewed international journals and confer- ence papers. His research interests include theory of programming languages, blockchain IoT, inter- net of vehicles, machine learning, distributed systems, and education. MUHAMMAD SHOAIB FAROOQ MUHAMMAD SHOAIB FAROOQ is currently working as a Professor in computer science with the University of Management and Technology, Lahore. He is also an Affiliate Member of George Mason University, USA. He possesses more than 26 years of teaching experience in the field of computer science. He has published many peer-reviewed international journals and confer- ence papers. His research interests include theory of programming languages, blockchain IoT, inter- learning, distributed systems, and education. [16] O. R. Kabi and V. N. Franqueira, ‘‘Blockchain-based distributed market- place,’’ in Proc. Int. Conf. Bus. Inf. Syst. Cham, Switzerland: Springer, Jul. 2018, pp. 197–210. pp [17] I. Acharjamayum, R. Patgiri, and D. Devi, ‘‘Blockchain: A tale of peer to peer security,’’ in Proc. IEEE Symp. Ser. Comput. Intell. (SSCI), Nov. 2018, pp. 609–617. [18] P. D. Kate and P. S. Powar, ‘‘Review of blockchain architecture implemen- tation in various fields of technologies,’’ Int. J., vol. 6, no. 3, pp. 23–26, 2021. 17994 VOLUME 10, 2022 M. S. Farooq et al.: Blockchain-Based Framework for Distributed Agile Software Development ZAREEN KALIM was born in Lahore, Pakistan. She received the B.S. and M.S. degrees in software engineering from the University of Management and Technology, Pakistan. Her research interests include agile software development, requirements engineering, and blockchain. SAIM RASHEED was born in Pakistan. He received the Ph.D. degree in the area of brain–computer interaction from the University of Milan, Italy, in 2011. He has almost 20 years of experience in academia and research. He has been actively participating and contributing in many international conferences and journals. His cur- rent research interests include human–computer interaction, gamification, and natural language processing. ADNAN ABID (Member, IEEE) received the B.S. degree from the National University of Computer and Emerging Science, Pakistan, in 2001, the M.S. degree in information technology from the National University of Science and Technology, Pakistan, in 2007, and the Ph.D. degree in com- puter science from the Politecnico di Milano, Italy, in 2012. He spent one year at EPFL, Switzer- land, to complete his M.S. thesis. He is currently working as a Professor with the Department of Computer Science, University of Management and Technology, Pakistan. He has almost 40 publications in different international journals and con- ferences. His research interests include computer science education, infor- mation retrieval, and data management. He has served as a reviewer in many international conferences. JUNAID NASIR QURESHI received the M.S. MUHAMMAD SHOAIB FAROOQ degree in computer science from the University of Central Punjab, Lahore. He is currently pur- suing the Ph.D. degree in computer science with the University of Management and Technology, Lahore, Pakistan. He is currently working as a Senior Lecturer with the Department of Com- puter Science, Bahria University Lahore Campus (BULC). He have 12 years of professional experi- ence in education and industry. His research inter- ests include the Internet of Things (IoT), blockchain, distributed systems, human–computer interaction, software engineering, agile project develop- ment, and education. 17995 VOLUME 10, 2022

W1537019044.txt

https://www.scielo.br/j/rbccv/a/tP5f7XXnRVjdXznv7BrpRNS/?lang=pt&format=pdf

pt

A RBCCV está no PubMed/Medline!

Brazilian Journal of Cardiovascular Surgery

cc-by

Editorial A RBCCV está no PubMed/Medline! Domingo M. Braile* dia 31 de outubro de 2007 vai ficar marcado na história da Revista Brasileira de Cirurgia Cardiovascular/Brazilian Journal of Cardiovascular Surgery (RBCCV/BJCVS) e da Sociedade Brasileira de Cirurgia Cardiovascular (SBCCV). Após tantos anos de lutas iniciadas pelo Prof. Dr. Adib Jatene e continuadas pelo Prof. Dr. Fábio Jatene, nessa data recebemos a carta comunicando que finalmente tínhamos sido aceitos no PubMed/ Medline. Concretizou-se, deste modo, a previsão feita no Editorial do volume 22.3. A RBCCV/BJCVS tornou-se, assim, uma das poucas revistas da especialidade naquela importante base de dados. Agradeço a todos que nos ajudaram e a DEUS por ter-nos permitido alcançar o destaque merecido para a cirurgia cardiovascular brasileira e de todo o Hemisfério Sul. Os colegas devem sentir-se todos prestigiados por esta láurea, que reconhece a qualidade do cirurgião brasileiro!!! O Medline é a mais importante base de dados online da literatura internacional da área médica, destinado a revistas científicas da área biomédica, patrocinado pela National Library of Medicine (NLM), dos Estados Unidos da América. A partir de agora, a inserção internacional da revista aumentará muito e, conseqüentemente, receberemos mais manuscritos qualificados do Exterior. Este reconhecimento vem coroar todos os esforços e investimentos empreendidos pela diretoria da SBCCV, nos últimos anos, na área científica, assim como avaliza a qualidade do trabalho e o alto valor da produção científica do cirurgião cardiovascular brasileiro, que tornaram possível esta imensa conquista. Mesmo diante das respostas negativas anteriores, jamais desanimei e, em maio deste ano, estive em Washington com o Administrador do Medline, Dr. Sheldon Kotzin, que forneceu informações importantes para o sucesso na submissão da revista. O Dr. Kotzin, aliás, fez questão de cumprimentar a RBCCV/BJCVS, destacando o nosso esforço. O e-mail dele está O publicado na sessão de Cartas ao Editor, a partir da página 517, juntamente com os cumprimentos recebidos de vários colegas. Também não posso esquecer-me de destacar o trabalho do Corpo Editorial da revista, que se dedicou com entusiasmo para desenvolver as mudanças, refinamento e formatação da revista, em conformidade com os elevados padrões requeridos pela Medline para a obtenção da indexação. Da mesma forma, foi fundamental o apoio da Bireme e da Scielo, para o preenchimento correto do formulário de acordo com os padrões do Medline. A partir de agora, os resumos em inglês dos artigos da RBCCV/BJCVS já estão disponíveis no PubMed/Medline, começando com o primeiro fascículo de 2007. A indexação é tão importante que foi destaque no Boletim de Novembro do CTSNet, site que congrega algumas das principais revistas da área da cirurgia cardiovascular e torácica do mundo. Estou pleiteando junto aos gestores do CTSNet que a RBCCV/BJCVS migre da página “Other Cardiovascular Journals” para a “Journals”, na qual se encontram os principais periódicos. É um processo demorado e trabalhoso, mas já comprovamos que é possível superar barreiras aparentemente intransponíveis. E é bom frisar que a indexação não é o ponto final de nossa jornada, mas sim o início de uma nova fase, com os cuidados sendo redobrados para nos mantermos no nível editorial exigido, tanto na parte científica, com a já costumeira exigência dos revisores na avaliação dos manuscritos, bem como na apresentação gráfica, pois a revista continuará sendo avaliada periodicamente. Mais uma vez, quero agradecer ao apoio de todos! Nosso próximo passo é conseguirmos a indexação na Base de Dados da ISI, dando assim o caráter definitivo à RBCCV/BJCVS de publicação de referência no contexto internacional. Outra boa notícia que gostaria de compartilhar é a I de que após muitas gestões o “String” (contagem de acessos) do nosso site (www.rbccv.org.br) está sendo armazenado na Scielo. Assim, toda vez que alguém acessar o nosso site, haverá um registro na Scielo. De janeiro a outubro de 2007, a Scielo contabilizou mais de 350 mil acessos à RBCCV, média de mais de 1.100 por dia. Para efeito de comparação, o total, nos primeiros 12 meses de contagem (fevereiro de 1998 a janeiro de 1999), foi de 168 artigos acessados. Os números são idênticos no site da revista. Com a indexação no Medline, a tendência é estes índices crescerem sensivelmente. Ainda dentro das boas notícias dos últimos meses, não posso deixar de cumprimentar o editor associado da RBCCV/BJCVS Prof. Dr. Luiz Felipe Moreira, que foi escolhido pela Comissão de Seleção da Sociedade Brasileira de Cardiologia como Editor Chefe dos Arquivos Brasileiros de Cardiologia para o quadriênio 2010-2013. A seleção do nome do Dr. Luiz Felipe faz jus a sua brilhante carreira acadêmica, premiando sua relevante contribuição à cirurgia cardíaca e cardiologia brasileira e mundial. Em outubro, finalmente a verba de R$ 9.500,00 da Capes, solicitada em 2006, foi disponibilizada. Também enviamos o pedido de auxílio referente a 2007 ao CNPq, II esperando que, mais uma vez, sejamos aquinhoados com os recursos, fundamentais para a manutenção da revista. Como já se tornou padrão na RBCCV/BJCVS desde 2005, estamos publicando, na página 522, a relação dos revisores que colaboraram conosco durante este ano. É uma forma de agradecer e homenagear aqueles que fazem um trabalho cuidadoso e anônimo, mas sem o qual nossa revista não teria atingido o patamar atual, tampouco teria sido indexada pelo Medline. O ano de 2007 está chegando ao fim. Aproveito para cumprimentar ao corpo editorial da revista, diretoria e funcionários da SBCCV pelo seu apoio e aos anunciantes, pela confiança. Desejo a todos um Feliz Natal e que 2008 seja repleto de realizações. Recebam o meu abraço, *Editor RBCCV/BJCVS 

https://openalex.org/W3152789875

https://www.researchsquare.com/article/rs-292692/v1.pdf?c=1631893260000

English

Misdiagnosis of Dengue Fever and Co-infection With Malaria and Typhoid Fevers in Rural Areas in Southwest Nigeria

Research Square (Research Square)

cc-by

Misdiagnosis of Dengue Fever and Co-infection With Malaria and Typhoid Fevers in Rural Areas in Southwest Nigeria Lawrence Ehis Okoror  (  [email protected] Federal University Oye-Ekiti Emmanuel Olufemi Bankefa  Federal University Oye-Ekiti Oluchi Mariam Ukhureigbe  Federal University Oye-Ekiti Evelyn Olubunmi Ajayi  Federal University Oye-Ekiti Stephen Kayode Ojo  Federal University Oye-Ekiti Bryan Ogeneh  Federal University Oye-Ekiti Lawrence Ehis Okoror  (  Law Federal University Oye-Ekiti Emmanuel Olufemi Bankefa  Federal University Oye-Ekiti Oluchi Mariam Ukhureigbe  Federal University Oye-Ekiti Evelyn Olubunmi Ajayi  Federal University Oye-Ekiti Stephen Kayode Ojo  Federal University Oye-Ekiti Bryan Ogeneh  Federal University Oye-Ekiti Misdiagnosis of Dengue Fever and Co-infection With Malaria and Typhoid Fevers in Rural Areas in Southwest Nigeria Misdiagnosis of Dengue Fever and Co-infection With Malaria and Typhoid Fevers in Rural Areas in Southwest Nigeria Lawrence Ehis Okoror  (  [email protected] ) Federal University Oye-Ekiti Misdiagnosis of Dengue Fever and Co-infection With Malaria and Typhoid Fevers in Rural Areas in Southwest Nigeria Lawrence Ehis Okoror  (  [email protected] ) Federal University Oye-Ekiti Introduction Dengue and Malaria fevers are the most common arthropod-borne diseases caused by mosquito bite and they also have similar signs and symptoms. Typhoid fever, which is caused by Salmonella typhi and 3 types of Salmonella paratyphi also share similar symptoms with both malaria and dengue fevers. Co-infection of Malaria and Dengue was first reported in 2005 [1], while Orhe et al in 2003 report a co-infection of malaria and typhoid fever [2] and that the few cases that have been reported indicated that, co-infections maybe more severe than singles infection (Malaria or Dengue fever) [3], although there is a paucity of reports for Dengue and typhoid fevers co-infection. In most cases, it is common to first think of malaria diagnosis when feverish syndromes are observed in patients and secondly, typhoid fever will be suspected and even treated empirically. Dengue virus infection is rarely taken into consideration by the clinicians because the disease is not considered as endemic of which this could lead to fatal consequences. Hence, early diagnosis of Dengue infection will not only prevent complications such as Dengue Hemorrhagic fever (DHF) and Dengue Shock syndrome (DSS) but will also curtail unnecessary consumption of antimalaria drugs and antibiotics thereby reducing menace of antimicrobial resistance. Dengue fever, also known as break bone fever, is an infectious tropical disease caused by the Dengue virus, a member of the Flaviviridae family [4]. About 390 million cases of Dengue infections are reported every year [5] and the virus is a cause of serious health problems in many tropical and subtropical areas of the world. Dengue hemorrhagic fever (DHF) first emerged as a public health problem in 1954, when the first epidemics occurred in other regions of the world in the 1980s and 1990s caused by all four serotypes of Dengue virus [6]. DHF and DSS, are major public health concerns because of their severe and often fatal disease in children as approximately 90% of DHF victims are children less than 15years of age [7]. Dengue virus is transmitted to humans by the bite of an infected Aedes mosquito mostly Aedes Egypti [8]. It primarily propagates in skin dendritic cells and replicate in target cells such as the monocytes or macrophages [9]. Symptoms include fever, headache, muscle and joint pains, and a characteristic skin rash that is similar to measles. Conclusion Conclusively, there was significant number of misdiagnosed cases of DV for either malaria or typhoid, hence it is recommended to include DV screening into routine hospital test especially in cases of malaria and typhoid negative by rapid diagnostic testing. Result Of all the samples tested 315 (29.4%) were positive to DV NS1 while 50 (6.7%) and 13 (3.9%) of 714 malaria samples and 333 typhoid samples respectively had Dengue fever co-infection. Co-infection of the three types of fever occurred in 5 (0.5%). A total of 54 (5%) DV cases were wrongly diagnosed for malaria while 14 (1.3%) DV cases was wrongly diagnosed as typhoid. Background Dengue and malaria have similar symptoms and arthropod vector and their mode of transmission coupled with differential diagnosis. Though typhoid fever differs from dengue and malaria by not having arthropod vector and different mode of transmission, it shares differential diagnosis with Dengue and Malaria which make misdiagnosis possible. This misdiagnosis of these three diseases has since been a major concern towards therapeutic administration because of their co-occurrence in many cases. Methods This study focused on the misdiagnosis of dengue fever for malaria or typhoid fever since the three have differential diagnosis and could co- occur. 741 samples were collected from malaria patient and 333 samples for typhoid fever outpatient at the health department facilities in rural communities of South West Nigeria. The samples were tested for dengue virus (DV) NS1 protein, anti DV IgM, anti DV IgG and RT-qPCR. Result Research Article Page 1/10 Page 1/10 Introduction In few cases, the disease progress into life-threatening DHF resulting in bleeding, low levels of blood platelets or blood and plasma leakage, or into dengue shock syndrome where low blood pressure occurs [10] and can lead to death. Malaria remains the deadliest infectious diseases in Africa and its parasites belongs to the genus Plasmodium. In humans, malaria is caused by P. falciparum, P. malariae, P. ovale, P.knowlesi [11]. Among the parasite known to transmit malaria, P. falciparum is the most common species identified (75%) followed P. vivax (20%) [12]. Although P. falciparum traditionally accounts for the majority of deaths while recent studies also suggests that P. vivax malaria can also be life-threatening [13]. Like in Dengue, symptoms of malaria include fever, headache, rash, vomit and joint pain. humans, malaria is caused by P. falciparum, P. malariae, P. ovale, P.knowlesi [11]. Among the parasite known to transmit malaria, P. falciparum is the most common species identified (75%) followed P. vivax (20%) [12]. Although P. falciparum traditionally accounts for the majority of deaths while recent studies also suggests that P. vivax malaria can also be life-threatening [13]. Like in Dengue, symptoms of malaria include fever, headache, rash, vomit and joint pain. Page 2/10 Study Design This study is a cross sectional research conducted in several health facilities in rural areas of South West, Nigeria.A cross section of the patients seeking diagnosis for malaria and typhoid were tested for Dengue Virus (DV) NS1,IgM, IgG, using ELISA and Reverse Transcriptase Polymerase Chain Reaction (RT-qPCR). The DV test was done independent of whether the patients are positive for malaria and typhoid or not. Inclusion Criteria This study involved all consenting out-patients that reported to health facilities for malaria and typhoid fever during the course of the research. Any sample testing positive to the different types of malaria parasite were included in the study and any sample testing positive to different causative agents of typhoid fever was included in the study or both diseases All collected sample must have been tested in the facility and result registered in the facility register. These samples were then tested for dengue NS1 used as a first line marker for dengue virus. All samples testing positive for dengue NSI were again tested for anti DV IgM, IgG and RT-PCR. Test for Malaria Malaria testing was done in the clinics using the Rapid Diagnostics Testing after which samples were immediately shipped to the laboratory and confirmed using the Giemsa staining technique. Test for Typhoid Fever Typhoid fever was tested in theclinical laboratory using the slide agglutination and typhoid RDT, samples were shipped to the laboratory for confirmation using the tube agglutination technique as adapted to the use of microtiter plate, all the patients were made to come back a week after the first test for a second sample collection used for paired sample testing [18]. Sample collection 5 ml of blood was aseptically collected from patients (n=1074) seeking malaria and/or typhoid diagnosis in the health institutions in rural areas of South West Nigeria from October to September in the year sampled. The blood samples were collected into EDTA bottles from each participant by a trained phlebotomist using needle and syringe and were immediately transported in cold chain to the Microbiology Laboratory.Each bottle was labeled indicating their age,sex and location.Blood samples were shared into 2 EDTA bottles and one to be used for DV ELISA and the other for malaria and typhoid. Page 2/10 Typhoid fever also remains an important public health problem in many developing countries of the world including Nigeria [14] about 11 million cases of typhoid fever occur annually with 600,000 deaths. In tropical Africa and some other developing countries enteric fever is rampant because of the low socio-economic and poor hygienic conditions in these regions. During the past few years in Nigeria, there have been a high incidence of typhoid fever creating fears or panic in any febrile illness which has led to drug abuse among populace especially chloramphenicol Dengue, malaria and typhoid are three major public health concerns in tropical settings and developing countries. Few cases of Dengue- Malaria coinfection has been reported in Nigeria [15] not much of Dengue (DV) and typhoid has been reported. Moreover, concurrent infection with the three different infective agents especially DV and malaria lead to an overlap of their clinical features. This can pose a diagnostic challenge to the physician, especially in endemic areas [16]. There has also been reports of misdiagnosis of the of malaria and DV infections [17], which could be due to differential diagnosis of both infections. Although reports of DV and typhoid infection are sparse, it is therefore imperative to investigate the co-occurrence due to their differential diagnosis. This study is not aware of any study which investigated DV, malaria and typhoid co-infection and misdiagnosis in Nigeria. Hence this study investigated the frequency of co-occurrence of dengue, malaria and typhoid fevers in rural communities in South West Nigeria, as well as their possible misdiagnosis of dengue fever for malaria and/or typhoid fever. Results A total of 1074 blood samples were collected from different health facilities in different rural locations in South West Nigeria of which 741 (69%) blood samples were positive for malaria parasites, 333 (31%) were positive to typhoid fever while from the malaria and typhoid positive samples, 315 (29.4%) were positive to dengue using the NS1 protein (Fig. 1). From DV NS1 results, 80 (25.4%) of the total samples were positive to anti DV IgM and 20 (6.3%) positive to anti DV IgG while 287 (91.1%) samples were positive for DV by RT-PCR (Table 1) confirmation. Of the total number of 741 confirmed malaria cases, 50 (6.7%) also had DV NSI (co-infection) {t = 5.8540) and also confirmed by RT-PCR, There was 13 (3.9%) typhoid- DV co-infection (t = 9.3611) from the 333 samples positive to typhoid fever, while malaria/typhoid co-infection with DV had 5 (0.5%) samples positive (p = 0.9296 and p-0.9432 respectively). A total of 54 DV positive samples were wrongly diagnosed as malaria parasites while 14 (1.3%) samples positive for DV weremisdiagnosed as typhoid fever (χ = 86.877, p = 0.0000001), showing an association between all misdiagnosed samples.Age group distribution shows that age groups 21–26, 27–32 and 15–20 have the highest number of all samples positive for all infections (Fig. 1). Age group 39–44 years had the highest number of DV malaria co- infection, however, age group 27–32 and 33–38 were higher for DV typhoid co-infection while 27–32 and 33–38 years bracket has the highest positive cases for the 3 co-infection. These age brackets also has the highest number of DV misdiagnosis with 15 cases respectively after age group 39–44 which had 20 cases. Age group 27–32 and 33–38 had the highest misdiagnosed samples for DV with 6 samples respectively. Figure 2 shows sex distribution to malaria, typhoid, DV and all their co-infections. Males were more infected with malaria parasites 412 (55.6%) while females were 329 (44.4%). For typhoid 218 (65.5%) were females while males 115 (34.5%) were positive. DV NS1 had 200 (63.5%) positive males and 115 (36.5%) positive samples were females. Other DV parameters also followed the same trend. Monthly distribution shows that the months of October, September, August, July, and June had the highest number of positive cases to malaria and DV, samples positive to typhoid fever did not show a particular distribution pattern (Table 3). RNA Extraction RNA was extracted using the Norgen Biotek total RNA extraction kit. 100 ml of non-coagulating whole blood was collected into well-labeled RNAse free microfuge tubes and 350 lysis buffer added to the blood in the microfuge tubes and the extraction procedure carried out as directed by the manufacturer (Norgen Biotek, Ontario Canada). All samples testing positive to DV NS1, IgM and IgG had their RNA extracted RT-PCR. Results November and June has the highest number of co-infection cases with Malaria/DV(t= -0.639, p = 0.537) as compared to others, while September has the highest number of co-infection for the 3 parameters (t = 0.905, p = 0.387) and October and March had the highest number of typhoid and DV co-infection (t =  1.008, p = 0.537) (Fig. 1). The concentration of the NS1 protein detected by ELISA technique as analyzed by the myassays microplate software was moderate andevenly distributed across all samples positive for DV NS1 (mean = 6.25mM) irrespective of sex, age or month of sample collection (Fig. II), the DV anti IgM concentration was also (mean = 41.94) with men having higher concentration and age distribution shows that adults of age groups 33–38, 37–44 and 27–32 having higher concentration (Fig. III).The Anti DV IgG also vary like the IgM with a mean of 51.72 (Fig. IV). RT- PCR procedure RT-qPCR was carried out on all samples testing positive to dengue NS1 by ELISA technique after RNA extraction. A PCR reaction mixture was set up using the MAXIMA SYBR green with ROX RT-PRCR master mix to achieve a total volume of 25ml using the hot start as described by the manufacturer. The primers used were obtained from already published research [19] which are universal primers targeting 3′ untranslated region of all complete genome sequences of dengue virus available in GenBank (n = 3,305)DENV_F- GCATATTGACGCTGGGARAGAC, DENV_R1-3 -TTCTGTGCCTGGAATGATGCTG, DENV_R4- YTCTGTGCCTGGATWGATGTTG) and probe (DENV_P- CAGAGATCCTGCTGTC). Hence all 4 types of DEV will be detected. The primers were sent to Inqaba Biotech, South Africa for synthesis. The PCR mixtures were put into the thermocycler (Biorad icycler, (Biorad USA). The PCR program was as follows- UDG pretreatment 50°C 2 mins for 1 cycle, initial denaturation 95°C for 10 mins 1 cycle, denaturation at 95°C for 15 sec 40 cycle, annealing at 60°C for 30 sec 40 cycle and extension at 72°C for 30 sec 40 cycles. Data acquisition was then done and analyzed with the icycler software for CT values. Enzyme Linked Immunosorbent Assay Blood samples for ELISA was immediately centrifuged at 3000 rpm (Beckman Microfuge centrifuge centrifuge) and sera separated from the whole blood and immediately used for ELISA. Sera stored in the refrigerator were brought out and allowed to attain room temperature as well as all the reagents. Sera were dispensed into the antibody impregnated ELISA microplate and the test carried out as described by the manufacturer of the ELISA test kits (Melsin Medicals, China). ELISA kits used included ELISA NS1, IgM and IgG for research. ELISA plates Page 3/10 were loaded into the microplate reader (Molecular devices, USA) at optical density of 450nm, absorbance were generated and analyzed using the myassays software to generate the concentration of each sample in each of the parameters for analysis. were loaded into the microplate reader (Molecular devices, USA) at optical density of 450nm, absorbance were generated and analyzed using the myassays software to generate the concentration of each sample in each of the parameters for analysis. Discussion This study is a cross-sectional and comparative analysis of DV, malaria and typhoid fever in co-infection in rural settings of South West Nigeria, with misdiagnosis of dengue fever for malaria or typhoid fever. DV fever co-infection with malaria and typhoid fevers was analyzed using 4 different parameters which included NS1, IgM, IgG antibodies and later confirmed with RT-PCR. The NS1 is first protein to be Page 4/10 Page 4/10 produced in the infection, which ensures the first early window is not missed in the DV infection, the IgM is the first antibody to be produced in response to an infection while the IgG will remain even in convalescence. A combination of all these ensures that no case is missed during the period of the study. The RT-PCR is used to validate all the results from the serological analysis and that any DV missed by ELISA technique will be captured by RT-PCR and at the same time quantifying the antigen in each sample. Of the total 315 DV NS1 positive samples, 287 samples were RT-PCR positive, thereby confirming that 28 samples though came in contact with DV but has not yet established enough antigen to be detected by PCR, which thereby confirms the usefulness of the NS1 protein in detecting dengue infection. This study generally reveals a high incidence of Malaria/DV co-infection as well as Malaria typhoid but the trio co-infection was not significant. A study of prevalence of concurrent dengue and malaria was reported by Charrel et al.[1] to have 0% prevalence in Netherlands, which oppose the result from this study. This reason for this may be due to geographical variation and high level of mosquito control in Netherlands, which is lacking in Nigeria. In addition, the mosquito vector for the two diseases have different habitats, malaria mosquito vector has its habitat in the forest [16] while Dengue mosquito vector main habitat is in the city [19]. Hence, overlapping of the habitat may not be easily available in Netherlands as is the case in Nigeria. Significant low Malaria-Dengue concurrent rate can be expected [20]. Dengue-Malaria co-infection has been referred to as uncommon phenomenon in the temperate regions. However, reports showed that the highest dengue-Malaria reported was 23.21% found in Pakistan while as low as 0.01% was found in Senegal [21]. Another study conducted in Brazil recorded 2.8% [7]. Discussion The report from Pakistan was similar to that from this study which is as a result of similar geographic conditions of both regions. The Senegal report also opposed results from this study and a major factor responsible for that is due to the time of the year and city the Senegal sampling was done, since both countries are in sub-Saharan Africa as indicated by the 2.84% Dengue virus and malarial concurrent infection within Ilorin metropolis In Nigeria [22]. Okoror et al. [23] also reported a prevalence of a high prevalence of DV infection (57.5%) in selected rural communities in South West Nigeria since all the hospitals do not officially screen for DV, and the differential diagnois DV, malaria and typhoid despite similar clinical presentations makes misdiagnosis not unusual. Therefore, co-infections may give rise to wrong diagnosis especially in areas where the clinician depends on empirical treatment. Moreover, the treatment regimens for these coinfections are not the same as those for mono- infections. Hence, a delay in implementing the appropriate treatment regimen for these different infections due to poor diagnosis can result in fatal consequences in compromising the patient’s health. Age group distribution revealed that age group 21–26 through to 39–40 have the highest prevalence of Malaria/DV co-infection and they were statistically significant (p = 0.9296: CI = 0.05). These age group also had the highest prevalence for malaria alone as well as for typhoid. The reason for this may be attributed to active (working class) which are found in this age range and their outdoor engagement on a day to day routine, making contact with mosquito unavoidable [24]. The study is also supported by Okoror et al [25] who reported the endemic nature of typhoid in Nigeria. In addition to this, sex distribution shows that more males were more infected with Malaria as compared to females and also Malaria DV co-infection was more in males compared to females, as well as typhoid DV coinfection. The overall infection also show similar pattern of sexual distribution. This was also reported by Dhanya et al. [26], though more males have been shown to have more outdoor activities than their female counterparts. The distribution of Malaria/DV and Typhoid monthly, show that the months of November and June had the highest Malaria/DV co-infection distribution while October and March showed the highest co-infection distribution for typhoid and DV with significant statistical difference from other months. Discussion This coincides with period of high rainfall and progression to dry season for malaria/DV infection and the period of high breed of mosquitoes. For typhoid/DV co-infection the months of highest prevalence coincides with period of high rainfall as these are the period of lowest hygiene due to high flooding which increases incidence of typhoid fever, and high proliferation of mosquitoes leading to high incidence of DV. Some authors have also reported both seasonal and monthly distribution of Malaria/DV co-infection. Dhanya et al. [26] in an India study reported a monthly distribution of Malaria and Dengue co-infection and reported their highest prevalence was between July and December which is in agreement with some of the months reported in this study. Savargaonkail et al.[24] also reported similar distribution in India. Total samples for malaria, typhoid and DV also follow similar pattern of distribution. Other factors that may have influenced this distribution includes rainfall fluctuation, humidity, and temperature [27, 28]. The concentration of NS1 protein, which was evenly distributed across the study population further goes to confirm the high prevalence of DV in the population and their misdiagnosis for malaria and typhoid fever. This is more importantly noticed with the high anti DV IgM and IgG which however was distributed along sex and age. The high concentration noticed in males goes to establish the earlier claims that males are more involved with out-door activities in the study population and therefore more exposed to the vector. Author contribution LEO conceived the study and was involved from writing of proposal through ethical defense and sample collection, laboratory analysis and to writing up of the paper. EOB was involved in supervision of laboratory analysis and writing of the final paper. OMU was involved in sample collection and laboratory analysis, EOA participated in sample collection and laboratory analysis, SKO and BO were involved in final data analysis. Conclusion In conclusion this research showed that lots of dengue positive samples are misdiagnosed for malaria and typhoid fevers and as such dengue testing should be incorporated to national testing scheme especially in areas where high dengue infection has been reported like in Page 5/10 Page 5/10 this study population so as not to compromise the patients’ health. The importance of this paper can not be undermined because of the public health implication of undermining or misdiagnosis of dengue for malaria or typhoid, a situation which could endanger the health of the population because of the like progression of dengue fever to DSS leading to death. this study population so as not to compromise the patients’ health. The importance of this paper can not be undermined because of the public health implication of undermining or misdiagnosis of dengue for malaria or typhoid, a situation which could endanger the health of the population because of the like progression of dengue fever to DSS leading to death. Availability of Data Materials Not applicable Not applicable List Of Abbreviations NS1, DV, DV IgG, DV IgM, RT-QPCR, DHF, DSS, ELISA, EDTA, RDT, RNA, PCR, MAXIMA, SYBR, ROX, RT-PRCR, DENV, UDG, CT Consent for Publication We declare no competing interest. Authors received no funding from anywhere, but study was carried out from their personal purse. Authors received no funding from anywhere, but study was carried out from their personal purse. A k l d t Acknowledgement We acknowledge the contribution of all our project students who were involved in sample collection during this study. Ethical Clearance Ethical approval was granted by Ethical Review committee of the Joseph Ayo Babalola University, Ikeji-Arakeji, Nigeria. All participants in this study agreed to participate through filling of questionnaire. Conflict of Interest We declare no conflict of interest. Author did not receive any form of financial assistance whatsoever. Ethical Clearance We declare no conflict of interest. Author did not receive any form of financial assistance whatsoever. References 1. Charrel RN, Brouqui P, Foucault C, De Lamballerie X. Concurrent dengue and malaria. Emerg Infect Dis.2005;11(7):1153-1154. [PMC free article] [PubMed] 1. Charrel RN, Brouqui P, Foucault C, De Lamballerie X. Concurrent dengue and malaria. Emerg Infect Dis.2005;11(7):1153-1154. [PMC fre article] [PubMed] 2. Orhue, Okoror LE. and IB Enweani. Co-occurrence of malaria and typhoid fever in patients attending clinics in Irrua Specialist Hospital Irrua, Nigeria. Journal of Science Engineering and Technology 2003; (1) 4745-4752. 3. Elpeboin L, Hanf M, Dussart P, Ouar-Epelboin S, Djossou F, Nacher M. Is Dengue and malaria co-infection more severe than single infections? A retrospective matched-pair study in French Guiana. Malar J.2012; 11:142. 3. Elpeboin L, Hanf M, Dussart P, Ouar-Epelboin S, Djossou F, Nacher M. Is Dengue and malaria co-infection more severe than single infections? A retrospective matched-pair study in French Guiana. Malar J.2012; 11:142. 4. https://www.euro.who.int/__data/assets/pdf_file/0006/246165/Fact-sheet-Dengue-Eng.pdf 4. https://www.euro.who.int/__data/assets/pdf_file/0006/246165/Fact-sheet-Dengue-Eng.pdf 5. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL. The global distribution and burden of Dengue. 2013;496:504- 7.doi:10.1038/nature 12060. Page 6/10 Salam N, Mustafa S, Hafiz A, Chaudhary AA, Deeba F, Parveen S. Global Prevalence and distribution of coinfection of malaria, dengue and chikungunya: a systemic review. BMCPub. Health 2018; 18:71010.1186/s12889-018-5626-z. 22. Kolawole OM, Seriki AA, Irekola AA, Bello KE, Adeyemi OO. Dengue virus and malaria concurrent infectionamong febrile subjects within Ilorin metropolis, Nigeria. JMed. Virol.2017; 89(8):1347-1353 23. Okoror, L E, Osanyinlusi S A, Ukhureigbe O M, Udenze, D O.Prevalence of Dengue Virus Antibody among Residents of a Rural Community in Southwestern, Nigeria. J. of Microbiol, 2019 33(2): 4512 – 4517 24. Savargaonkai D, Sinha S, Srivastava B, Nagpal B.N, Sinha A, Shamim A, Das R, Pandi V, Anvikar AR, Valecha N. An Epidemiological Study of Dengue and its Coinfections in Delhi. Inter. J. of Infect. Dis. 74(2018): 41-46. 24. Savargaonkai D, Sinha S, Srivastava B, Nagpal B.N, Sinha A, Shamim A, Das R, Pandi V, Anvikar AR, Valecha N. An Epidemiological Study of Dengue and its Coinfections in Delhi. Inter. J. of Infect. Dis. 74(2018): 41-46. 25. Okoror LE, Osanyinlusi SA, Ukhureigbw OM and Udenze DO. Prevalence of Dengue Virus among residents of a rural community in Southwestern Nigeria. Nigerian Journal of Microbiology 2019; 33 (2) : 4512-4517 25. Okoror LE, Osanyinlusi SA, Ukhureigbw OM and Udenze DO. Prevalence of Dengue Virus among residents of a rural community in Southwestern Nigeria. Nigerian Journal of Microbiology 2019; 33 (2) : 4512-4517 26. Dhanya P. R, Shilpa K, Vivek H. A comparative study on dengue and malaria infections and analysis on the prevalence of co-infection from a rural tertiary care hospital. Int J ResMedSci. 2018; 6(6):2111-2115 26. Dhanya P. R, Shilpa K, Vivek H. A comparative study on dengue and malaria infections and analysis on the prevalence of co-infection from a rural tertiary care hospital. Int J ResMedSci. 2018; 6(6):2111-2115 27. Wongkoon S, Jaroensutasinee M, Jaroensutasinee, K; Distribution, seasonal variation and Dengue transmission prediction in Sisakett, Thailand, Indian J. of Med. Res, 2013;138(3):347-53. 27. Wongkoon S, Jaroensutasinee M, Jaroensutasinee, K; Distribution, seasonal variation and Dengue transmission prediction in Sisakett, Thailand, Indian J. of Med. Res, 2013;138(3):347-53. 28. Pakhare A, Yogesh S, Ankur J, Rashmi J, Arun K, Rajnish J. A Study of Spatial and Meteorological determinants of dengue outbreak in Bhopal City. J Vector Borne Dis 2016;53 (3):225-233 Page 6/10 Page 6/10 6. Nimmannitya S. Dengue haemorrhagic fever: current issues and future research. Asian- Oceanian Journal of Paediatrics and Child Health. 2002; 1: 1-21. 7. Magalhaes BM, Siqueira AM, Alexandra MA. vivax malaria and dengue fever coinfection: a cross- sectional study in the Brazililian Amazon. PLoS NegL Trop Dis.2014; 8:e3239.10.1371/ journal.pntd.0003239. 8. Kumar V, Nagpai, BN, Veena S, Anna G, Sanjeev K, Paul R. Comparison of egypti breeding in localities of different socio-economic group of Delhi; India.; Int. J. of Mosquito Res.2015; 2(3): 83-88. 9. Clyde K, Kyle JL, Harris E. Recent advances in deciphering viral and host determinants of dengue virus replication and pathogenesis. J Virol. 2006;80(23):11418-11431. doi:10.1128/JVI.01257-06 0. Wiwanitkit V “Unusual mode of transmission of dengue”. of infect in Developing Countries (2010); 11. Barber BE, Rajahram GS, Grigg MJ, William T, Anstey NM. World malaria report: time to acknowledge Plasmodium knowlesi malaria. Malar J. 2017;16:135.doi:10.1186/s112936-017-1787-y. 12. Nadjm B, Behrens RH. “Malaria: An update for physicians”. Infect. DisClinics of NorthAmerica2012; 26(2): 243- 59.doi:10.1016/j.idc.2012.03.010. PMID 22632637. 13. Baird JK. “Evidence and implications of mortality associated with acute Plasmodium vivax malaria”. Microbiol. Rev.2013; 26(1):36- 57,doi10.1128/CMR.00074-12. PMC 3553673. PMID 23297258. 14. Alaribe, AAA Ejezie GC and Ezedinachi, NU. The prevalence of Salmonella antibody among malaria patients in Calabar. of Med. Lab. Sci.2013; 7: 34 – 41. 1998. 7. 15. Nasir S, Shoeb M, Abdul H, Anis AC, Farah D and Shama P. Global prevalence and distribution of coinfection of malaria, dengue and chikungunya: a systemic review. BMCPub.Health2018; 18:710 doi:10.1186/s12889-018-018-5626-z. 16. Obsomer V, Defourny P, Coosemans M. The Anopheles dirus complex: spatial distribution and environment drivers. Malar J 2007; 6:26. [PMC free article] [PubMed]. 7. Wilairatana P, Tangpukdee N and Krudsood SMisdiagnosis of Malaria in Malaria-Dengue Endemic DOI: 10.4172/2329-9088.1000e119 17. Wilairatana P, Tangpukdee N and Krudsood SMisdiagnosis of Malaria in Malaria-Dengue Endemic Area. TropMed&Surgery 2014; 2:14. DOI: 10.4172/2329-9088.1000e119 18. Okoror L.E., Umolu P.I., Orhue P, Omigie O.O. and R. Akpe. Widal test as adapted to the use of microtitre plate. Discovery and Innovation 2005; 17 (1/2) 45-49. 19. Alm E, Lesko B, Lindegren G, Ahlm C, So¨derholm S , Falk KI, Lagerqvist N, Universal Single-Probe RT-PCR Assay for Diagnosis of Dengue Virus Infections. PLOs Neglected Trop. Dis: 2014;8(12): e3416. doi:10.1371/journal.pntd.0003416 20. Ward DI. A case of fatal Plasmodium falciprum malaria complicated by acute dengue fever in East Timor. Am J Trop Med2006;75(1):182-185. [PubMed]. 21. Tables Page 7/10 Age Group Number     Positive Malaria Typhoid DV NSI DV IgM DV IgG DV PCR Malaria/ DV Typhoid/DV Malaria/DV/Typhoid Wrong diagnosis Malaria Wrong diagnosis Typhoid 15-20 110 53 48 21 9 46 0 0 0 0 0 21-26 135 49 48 27 5 48 10 5 1 3 1 27-32 115 61 36 10 0 34 9 5 2 15 6 33-38 101 61 37 8 2 34 9 0 1 15 6 39-44 97 44 45 8 3 40 22 0 1 20 1 45-50 86 25 39 6 1 30 0 0 0 1 0 51-56 62 21 41 0 0 36 0 2 0 0 0 57-62 35 19 21 0 0 19 0 1 0 0 0 Total 741 333 315 80 20 287 50 13 5 54 14 of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Table II: Sex distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Gender                                      Number     Positive Malaria Typhoid DV NSI DV IgM DV IgG DV PCR Malaria/ DV Typhoid/DV Malaria/DV/Typhoid Wrong diagnosis Malaria Wrong diagnosis Typhoid Male 412 115 200 54 10 151 35 7 4 37 3 Female 329 218 115 26 10 136 15 8 1 17 3 Total 741 333 315 80 20 287 50 15 5 54 6 Table III: Monthly distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Month Dengue Typhoid Malaria October 41 44 87 November 20 12 44 December 12 10 39 January 12 10 27 February 16 12 27 March 29 29 46 Table III: Monthly distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Month Dengue Typhoid Malaria October 41 44 87 November 20 12 44 December 12 10 39 January 12 10 27 February 16 12 27 March 29 29 46 April 35 28 67 May 41 39 71 June 40 39 90 July 38 38 93 August 28 27 81 September 21 27 69 Total 333 315 741 Table III: Monthly distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Table III: Monthly distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Month Dengue Typhoid Malaria Table III: Monthly distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria distribution of Dengue fever, Malaria and typhoid fever in some rural settlement in South West Nigeria Month Dengue Typhoid Malaria October 41 44 87 November 20 12 44 December 12 10 39 January 12 10 27 February 16 12 27 March 29 29 46 April 35 28 67 May 41 39 71 June 40 39 90 July 38 38 93 August 28 27 81 September 21 27 69 Total 333 315 741 Page 8/10 Figures Figures Figures Figure 1 Monthly Distribution of Co-occurrence of Dengue Fever with Malaria and Tyhpoid Fever in Rura Communities in Southwest Nigeria Figure 1 Figure 1 Monthly Distribution of Co-occurrence of Dengue Fever with Malaria and Tyhpoid Fever in Rura Communities in Southwest Nigeria Monthly Distribution of Co-occurrence of Dengue Fever with Malaria and Tyhpoid Fever in Rura Communities in Southwest Nigeria Page 9/10 Figure 2 DV NS1 concentration of samples tested for malaria, typhoid in rural areas of South West Nigeria Figure 2 DV NS1 concentration of samples tested for malaria, typhoid in rural areas of South West Nigeria Page 9/10 DV NS1 concentration of samples tested for malaria, typhoid in rural areas of South West Nigeria Figure 3 Anti DV IgM concentration of samples tested for malaria, typhoid and testing for positive to Dengue NS1 in rural areas of South West Nigeria Figure 3 Anti DV IgM concentration of samples tested for malaria, typhoid and testing for positive to Dengue NS1 in rural areas of South West Nigeria Page 10/10 Figure 4 Anti DV IgG concentration of samples tested for malaria, typhoid and testing for positive to Dengue NS1 in rural areas of South West Nigeria Figure 4 Anti DV IgG concentration of samples tested for malaria, typhoid and testing for positive to Dengue NS1 in rural areas of South West Nigeria Page 10/10

https://openalex.org/W4386430498

https://www.e3s-conferences.org/articles/e3sconf/pdf/2023/57/e3sconf_ebwff2023_03005.pdf

English

The content of polyphenols in low-growing apple rootstocks depending on the type of soil in Central Russia

E3S web of conferences

cc-by

The content of polyphenols in low-growing apple rootstocks depending on the type of soil in Central Russia Vyacheslav Zakharov1*, Valentina Gulidova1, and Tat'jana Zubkova1 1 Bunin Yelets State University, 399770 Yelets, Russia Abstract. Field studies were carried out in 2017-2021. in horticultural farms of Central Russia. We analyzed 11 varieties of apple trees in four seasons (winter, spring, summer, autumn) for the content of polyphenols in the rootstocks of apple trees - dwarf 62-396, semi-dwarf 54-118 and dwarf PB-9, grown on 7 types of soil: floodplain granular, floodplain layered , podzolized chernozem, leached chernozem, meadow-chernozem, gray forest and sod-podzolic soil. It was found that of all the organs of cultivated varieties of apple trees, the largest amount of polyphenols is concentrated in 1-2-year-old shoots. The highest concentration of flavonols in the shoots of all studied rootstocks and catechins in the shoots of rootstock 54-118 was when growing on the most low-fertile soddy-podzolic soil. It was found that the content of flavonols in apple shoots is less affected by the weather conditions of the year than the content of catechins. The content of catechins in the annual shoots of the apple tree is an unstable variable indicator. Its value may increase or decrease by 5.4-12.0 times during the first two years. Of the three apple tree rootstocks, the largest amount of flavonols was noted in shoots 62-396, and catechins - in shoots 54-118. It was found that with an increase in the proportion of fine dust and physical sand and a decrease in the proportion of medium dust in the soil, the content of flavonols in the shoots of the dwarf apple rootstock 62-396 increases, the less medium dust in the soil, the more flavonols are contained in the shoots of the semi-dwarf apple rootstock 54-118. With an increase in the content of medium dust in the soil, the content of catechins increases in the shoots of the dwarf rootstock of the PB-9 apple tree. It was proposed that the accumulation of flavonols in the annual shoots of apple rootstocks should be considered as an adaptation to a decrease in soil fertility. * Corresponding author: [email protected] E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). 1 Introduction Phenolic compounds of the bark and pith of apple wood have antimicrobial, antioxidant [25], biostimulating, anti-inflammatory, cardioprotective [8], and anticarcinogenic effects [20]. For this reason, more and more attention is being paid to methods for extracting these substances from apple wood in order to enrich food products with them [10]. According to the content of biologically active substances in the leaves and shoots of the apple tree, one * Corresponding author: [email protected] © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 can judge the physiological state of the tree. For example, the content of starch and lignin are important indicators of the degree of ripening of annual shoots of an apple tree, and, consequently, their frost resistance [1], while flavonoids are key modulators of auxin transport, which in turn affects the size of fruits [13]. can judge the physiological state of the tree. For example, the content of starch and lignin are important indicators of the degree of ripening of annual shoots of an apple tree, and, consequently, their frost resistance [1], while flavonoids are key modulators of auxin transport, which in turn affects the size of fruits [13]. Polyphenols in apple shoots increase plant resistance to Waltz's canker (a type of canker of fruit trees) [17]. The most important vitamins, the physiological role of which has been established only in relation to apple fruits, are: L-ascorbic acid, β-carotene, folic acid, inositol, pantothenic acid, thiamine, riboflavin, niacin, pyridoxine, biotin, tocopherol [7]. Phenolic substances are found in plants in high concentrations and their functions are diverse. For example, indole-3-butyric acid promotes the rooting of green cuttings [6]. From 52 to 87% of the phenolic compounds of apple wood (bark, core, roots) is usually occupied by phlorizin, and phenolic acids account for 10% [21]. The content of phlorizin in apple leaves ranges from 290 to 720 µg/kg [9]. The leaves of scab-resistant apple varieties (Topaz, Goldstar, Goldrush) contained significantly more flavonols (rutin, quercetin, phenol 173) than the leaves of resistant apple varieties (Golden Delicious Weinsberg and Golden Delicious Clone B). The content of these flavonols did not depend on any external stress factors, but changed only by the phenophases of the growing season and was also determined by the plant genotype [24]. 1 Introduction A gene was found that is responsible for the concentration of phlorizin in apple leaves [12] and for the biosynthesis of flavonoids in apples [19]. The level of phenolic compounds in apple leaves is more stable during the entire harvest period than in fruits. While the group of polyphenols procyanidins predominates in fruits, dihydrochalcones dominate in leaves [26]. If polyphenolic metabolites (especially chlorogenic and isochlorogenic acids) are isolated from apple leaves and then treated with them on growing apple plants, the effect of scab suppression is observed [18]. When an apple tree is artificially infected with late blight, chlorogenic and coumaric acids are produced in its tissues [22]. An increased content of 3- hydroxyflorizin in apple leaves correlated with plant resistance to scab and late blight [16]. In transgenic plants of the domestic apple tree, an increased production of such polyphenols as catechins and proanthocyanidins in the leaves [15], but a reduced concentration of flavonoids in the shoots [11] was found. When blocking the synthesis of anthocyanins in the leaves of red-leaved apple trees, their viability decreases due to necrotic lesions of the leaves and the amount of epicatechin increases [23]. p The presence of phlorizin in apple buds ensures a normal transition of the juvenile phenological phase into the adult vegetative phase, and in the reproductive phenophase, the disappearance of myricitrin in the bark and the absence of caffeic acid in aboveground tissues were noted [28]. The relationship between the content of phenol in the shoots of apple rootstocks and their growth force was established [27]. When apple leaves were treated with one of the enzyme inhibitors (prohexadione Ca), catechins were formed in the leaves [14]. It is known that the growth regulator gibberellic acid and chlorocholine chloride can be used to change the content of polyphenols and their distribution over organs in woody plants [5]. When apple trees of the Pepin saffronny variety were darkened, the content of polyphenols in the shoots decreased [4]. The question of the influence of external factors on the content of polyphenols in the leaves or shoots of the apple tree remains poorly understood. The aim of the work was to find out how the type of soil affects the content of flavonols and catechins in the shoots of apple rootstocks in the conditions of central Russia. 2 Materials and methods Field studies were carried out in 2017-2021 at the experimental site of Yelets State University named after I.A. Bunin. Laboratory analyzes of the soil and shoots of the apple tree were carried out on the basis of the research agrochemical laboratory of the specified university. 2 2 E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 Agrochemical analyzes of the soil were carried out according to the instructions of the TSINAO [3]. At the beginning of our research, we found out which organ of the apple tree is the most indicator and rich in polyphenols. In the Timiryazevsky farm of the Dolgorukovsky district of the Lipetsk region, 11 varieties of apple trees were studied (Lobo, Sinap Orlovsky, Northern Sinap, Antonovka ordinary, Bogatyr, Spartan, Melba, Pepin saffron, Mantet, Uelsi, Zhigulevskoye). For each variety for 3 years in each season (winter, autumn, spring, summer), the content of polyphenols was determined. The organs of the apple tree for the content of polyphenols were also taken from different soils (podzolized chernozem, leached chernozem, podzolized chernozem-meadow soil). Then all these data were averaged. g A model experiment with the three most common apple rootstocks in production: (dwarf 62-396, semi-dwarf 54-118 and dwarf PB-9) was established in the spring of 2016 at the YEGU training field. The soils were selected from the humus horizons of 7 predominant soils of the Lipetsk region, which are under a long-term fallow. The mass of soil in one vessel is 4 kg. One rootstock was planted in each vessel. The repetition of experience 5-fold. The location of the plots is 3-tier sequential. Planting scheme for rootstocks 75 x 20 cm. The soil was sampled from a layer of 0-30 cm in the following places: floodplain granular - from the floodplain terrace of the Bystraya Sosna River, floodplain layered - from the floodplain of the Bystraya Sosna River, leached chernozem - from the watershed area near the Vorgol River, podzolized chernozem - from the edge of the oak-linden forest on the high floodplain terrace of the Vorgol River, gray forest - from the forest oak-linden massif on the watershed, soddy-podzolic - from the pine massif on the watershed. Shoots were taken in mid-February. Polyphenols were defined as the sum of flavonols and catechins. The content of flavonols and catechins in annual increments was determined by the photometric method [2]. 6.13 46.14 3.81 13.32 18.77 11.83 56.08 43.92 Sod-podzolic 0.5 41.6 22.8 13.6 12.7 8.8 64.9 35.1 According to the content of physical clay, podzolized chernozem and gray forest soil, heavy loamy, turned out to be the heaviest in terms of granulometric composition. The lightest, that is, with less clay, were floodplain granular, floodplain layered and meadow- chernozem soils. The leached chernozem and soddy-podzolic soil occupied an intermediate position - they had a medium loamy granulometric composition. p y y g p The thinnest humus horizon (A1) was noted in the profile of soddy-podzolic soil - only 10-15 cm. In other soils, the thickness of the humus horizon was much greater - from 40 to 80 cm. We found the following content of polyphenols in different organs of apple varieties: small roots of apple varieties (up to 3 mm) - 273.3 mg%, live apple fruits - 309.8 mg%, dead apple fruits - 131.0 mg%, live skeletal apple tree roots - 426.6 mg%, fallen apple tree leaves - 448.4 mg%, live one-year-old apple tree shoots - 503.1 mg%, live 6-12-year-old apple tree branches - 62.5 mg%, live two-year-old apple tree shoots - 818.5 mg%, apple fruits - 54.3 mg%. It has been established that the maximum amount of polyphenols is concentrated in the youngest growths of the apple tree (1-2 years old). Therefore, a further subject of research under the conditions of a model experiment was the annual growth of apple rootstocks. In the model experiment, it was found that the highest amount of flavonols in the shoots of all three studied rootstocks was when growing on soddy-podzolic soil (Table 2). Table 2. The content of flavonols in annual increments of apple rootstocks depending on the typ soil (average for 2020-2021), mg% Soil type Аpple rootstock 62-396 54-118 PB-9 Chernozem leached 735.6 915.5 831.8 Meadow-chernozem 740.3 911.3 825.6 Chernozem podzolized 1076.1 983.5 764.4 Gray forest 1000.5 990.4 770.6 Floodplain granular 695.3 1135.2 783.7 Floodplain layered 946.0 911.5 935.5 Sod-podzolic 1520.6 1186.2 1063.2 NSR05 40.3 38.6 37.1 Accuracy of experience,% 7.2 7.0 6.4 e 2. The content of flavonols in annual increments of apple rootstocks depending on the type of soil (average for 2020-2021), mg% 3 Results Gray forest, podzolized chernozem and sod-podzolic soils had a pH of water extract of 5.1- 5.3, humus content of 3.3-3.4%, total nitrogen 0.17%, mobile phosphorus 6.6-7.2 mg /100 g, exchangeable potassium 6.1-6.9 mg/100 g, other soils - pH - 7.1-7.8; humus - 5.7-6.2%, total nitrogen - 0.3%, mobile phosphorus - 17.1-25.4 mg/100 g, exchangeable potassium - 10.9- 13.0 mg/100 g, respectively. Exchangeable calcium was contained in floodplain soils 2.5- 3.0, in gray forest, soddy-podzolic and podzolized chernozem - 6.6, in leached chernozem - 9.0 mg-ekv/ 100 g. Exchangeable magnesium in the floodplain granular soil contained 0.2, and in other soils - 0.5-0.7 mg- ekv/100 g. The greatest differences between the soils occurred in their granulometric composition (Table 1). Table 1. Granulometric composition of soils in the Yelets region in the 0-30 cm layer, % The content of fractions of different diameters, mm 1-0.25 0.25- 0.05 0.05- 0.01 0.01- 0.005 0.005- 0.001 up to 0.001 sand (more than 0.01) clay (up to 0.01) Floodplain granular 1.91 28.97 14.68 16.73 35.54 4.08 45.55 54.44 Chernozem podzolized 0.23 42.11 16.4 20.18 6.65 14.42 58.75 41.25 Meadow-chernozem 1.63 39.14 5.35 18.13 32.04 3.71 46.12 53.88 Chernozem leached 0.68 37.54 24.89 21.21 12.0 3.67 63.11 36.89 Floodplain layered 3.09 29.78 10.21 19.99 35.71 1.22 43.08 56.92 Gray forest Table 1. Granulometric composition of soils in the Yelets region in the 0-30 cm layer, % 3 E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 4 Discussion According to the smallest amount of flavonols in the shoots of apple rootstocks, no binding to any type of soil has been established. All three rootstocks had a minimum of these substances on different soils: 62-396 - on the floodplain granular, 54-118 - on the floodplain layered, PB-9 - on podzolized chernozem. If we compare the rootstocks with each other, then the largest amount of flavonols was noted in the shoots of the rootstock 62-396, and catechins - in the shoots of the rootstock 54-118. Research has established relationships between individual soil fractions and polyphenols. So, with an increase in the content of fine dust (0.01-0.005 mm) in the soil, the content of flavonols in the shoots of the rootstock increases 62-396 (r=0.51). As the content of physical sand in the soil increased (larger than 0.01 mm), the content of 62-396 flavonols in the rootstock shoots increased (r=0.57). With an increase in the proportion of physical clay in the soil, an opposite trend was observed for these substances (r=-0.57). Inverse relationships were also established: with a decrease in the proportion of medium dust (0.01-0.005) in the 4 E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 soil, the content of flavonols increased in the shoots of the rootstock 62-396 (r=-0.52) and in the shoots of the rootstock 54-118 (r=-0.66 ). soil, the content of flavonols increased in the shoots of the rootstock 62-396 (r=-0.52) and in the shoots of the rootstock 54-118 (r=-0.66 ). It was revealed that the greatest number of catechins in all rootstocks was on soils: 62- 396 - on floodplain granular, 54-118 - on soddy-podzolic, PB-9 - on leached chernozem (Table 3). able 3. The content of catechins in annual increments of apple rootstocks depending on the type of soil (average for 2020-2021), mg% Table 3. 4 Discussion The content of catechins in annual increments of apple rootstocks depending on the type of soil (average for 2020-2021), mg% Soil type Аpple rootstock 62-396 54-118 PB-9 Chernozem leached 80.1 51.8 115.0 Meadow-chernozem 78.5 50.3 109.9 Chernozem podzolized 76.3 61.3 60.1 Gray forest 74.2 62.0 59.8 Floodplain granular 119.4 54.2 51.0 Floodplain layered 53.0 155.0 61.3 Sod-podzolic 70.0 161.3 53.4 NSR05 25.3 20.2 19.3 Accuracy of experience, % 6.2 5.9 5.7 The smallest amount of catechins in the shoots of apple rootstocks was not tied to one type of soil, but was found in all rootstocks on different types of soil: 62-396 on podzolized chernozem, 54-118 on leached chernozem, and PB-9 on floodplain granular soil. It was determined that with an increase in the content of medium dust in the soil in the shoots of the dwarf rootstock of the apple tree PB-9, the content of catechins increases. The content of catechins in the shoots of apple rootstocks had fluctuations in the context of the years. So, in 2020, the variation ranged from 64.8 to 294.9 mg%, and in 2021 - from 5.5 to 54.5 mg%. The difference was 5.4-12.0 times. This trend was found for all rootstocks and soils. According to the content of flavonols, such a trend was not revealed. The content of flavonols in the shoots of apple rootstocks in 2020 ranged from 681.8 to 1971.7 mg%, and in 2021 - from 508.4 to 1227.8 mg%. g Thus, of the considered polyphenols, flavonols are more sensitive to soil changes than catechins. The latter rather vary due to weather conditions. Of the soil indicators that influenced the content of flavonols and catechins is the thickness of the humus horizon, the content of physical clay, physical sand, medium and fine dust. The accumulation of flavonols in the annual shoots of apple rootstocks can be considered as an adaptation to a decrease in soil fertility. 5 Conclusion 1. Of all the organs of the apple tree, the largest amount of polyphenols is concentrated in 1-2-year-old shoots. The highest concentration of flavonols in the shoots of all three studied rootstocks was when growing on the lowest-fertile soddy-podzolic soil. The accumulation of flavonols in these organs of the apple tree can be considered as an adaptation to a decrease in soil fertility. 2. The content of flavonols is less affected by the weather conditions of the year than the content of catechins in apple shoots. Catechins in annual apple tree shoots are an unstable variable. Their content may increase or decrease by 5.4-12.0 times during the first two years. 3. With an increase in the proportion of fine dust and physical sand and a decrease in the proportion of medium dust in the soil, the content of flavonols in the shoots of the dwarf 5 E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 apple rootstock 62-396 increases, but in the shoots of the semi-dwarf apple rootstock 54-118 this indicator increases only with a decrease in the average dust in the soil. apple rootstock 62-396 increases, but in the shoots of the semi-dwarf apple rootstock 54-118 this indicator increases only with a decrease in the average dust in the soil. 4. With an increase in the content of medium dust in the soil, the content of catechins increases in the shoots of the dwarf rootstock of the PB-9 apple tree, and in the shoots of the rootstock 54-118 this indicator was maximum only when growing on the most low-fertile soddy-podzolic soil. 4. With an increase in the content of medium dust in the soil, the content of catechins increases in the shoots of the dwarf rootstock of the PB-9 apple tree, and in the shoots of the rootstock 54-118 this indicator was maximum only when growing on the most low-fertile soddy-podzolic soil. 6 Acknowledgements The authors express their gratitude to the Rector of Bunin Yelets State University for his assistance in conducting this study References 1. E.I. Barskaya, Changes in chloroplasts and maturation of shoots in connection with the frost resistance of woody plants (Nauka, Moscow, 1967) 1. E.I. Barskaya, Changes in chloroplasts and maturation of shoots in connection with the frost resistance of woody plants (Nauka, Moscow, 1967) 2. L.I. Vigorov, A.Ya. Tribunskaya, Methods for determining flavonols and flavones in fruits and berries in Proceedings of the III All-Union Seminar on biologically active (medicinal) substances of fruits and berries, Sverdlovsk (1968) 2. L.I. Vigorov, A.Ya. Tribunskaya, Methods for determining flavonols and flavones in fruits and berries in Proceedings of the III All-Union Seminar on biologically active (medicinal) substances of fruits and berries, Sverdlovsk (1968) 3. TSINAO instructions for conducting mass soil analyzes in zonal agrochemical laboratories (Kolos, Moscow, 1973) 4. I.V. Plotnikova, V.F. Verzilov, V.K. Bakun, V.S. Alexandrova, S.F. Zagursky, Influence of lighting conditions on the content of IAA and phenolic compounds in the tissues of apple shoots and cuttings (Phytohormones - growth regulators, Moscow, 1980) 5. L.V. Runkova, Comparative study of the level of polyphenols in some tree species in connection with the treatment with gibberellin and chlorcholine chloride (Phytohormones - growth regulators, Moscow, 1980) 6. L.V. Runkova, T.V. Khromova, T.I. Suslova, E.R. Safina, About some polyphenols of three-lobed almonds in connection with the rooting of its cuttings treated with indole- 3-butyric acid (Phytohormones – plant growth regulators, Moscow, 1980) 7. L.V. Sadovskoy, L.V. Solovyova, l.V. Shvergunova, Physiology of fruit plants (Kolos Moscow, 1983) 8. M.J. Aliaño-González, J. Gabaston, V. Ortiz-Somovilla, E. Cantos-Villar, Biomolecules 12, 238 (2022) 9. S. Ben-Othman, H. Kaldmäe, R. Rätsep, U. Bleive, A. Aluvee, T. Rinken, Antioxidants 10, 189 (2021) 10. L.C. Da Silva, J. Viganó, L.M. De Souza Mesquita, et.al, Food Chemistry X(12), 100- 133 (2021) 11. А. Dare, R. Hellens, Plant Signaling & Behavior 8, 25033 (2013) 12. А. Dare, S. Tomes, J.M. Cooney, D.R. Greenwood, R.P. Hellens, Plant Physiology and Biochemistry 72, 54-61 (2013) 13. A.P. Dare, S. Tomes, M. Jones, T.K. McGhie, D.E. Stevenson, R.A. Johnson, D.R. Greenwood, R.P. Hellens, The Plant Journal 74(3), 398-410 (2013) 14. T.C. Fischer, H. Halbwirth, S. Roemmelt, et.al, Physiologia Plantarum 128(4), 604-617 (2006) 6 E3S Web of Conferences 420, 03005 (2023) EBWFF 2023 https://doi.org/10.1051/e3sconf/202342003005 15. H. Flachowsky, I. Szankowski, T.C. Fischer et al., Planta 231, 623–635 (2010) 16. O.S. Hutabarat, H. Flachowsky, I. Regos et al. Planta 243, 1213–1224 (2016) 17. Y. Jing, M. Zhan, C. Li, T. References Pei, Q. Wang, P. Li, F. Ma, C. Liu, Molecular Plant Pathology 23(8), 1170-1186 (2022) 18. D.S. Kirkham, Microbiology 17(2), 491 (1957) 19. H. Li, H. Flachowsky, T.C. Fischer et al., Planta 226, 1243–1254 (2007) 20. E. Martino, D.C. Vuoso, S. D’Angelo et al., Sci Rep 9, 13045 (2019) 21. M.M. Moreira, F.M. Barroso, А. Boeykens, H. Withouck, S. Morais, С. Delerue-Matos Industrial Crops and Products 104 210-220 (2017) 22. M. Skłodowska, E. Gajewska, E. Kuźniak, M. Wielanek, A. Mikiciński, P. Sobiczewski, Journal of Phytopathology 159(7-8), 495-504 (2011) 23. I. Szankowski, H. Flachowsky, H. Li et al., Planta 229, 681–692 (2009) 24. V. Usenik, M. Mikulič-Petkovšek, A. Solar, F. Štampar, Journal of Plant Diseases and Protection 2, 137-44 (2004) 25. H. Withouck, А. Boeykens, W. Luyten, R. Lavigne, Journal of Biologically Active Products from Nature 9, 1 (2019) 26. А. Wojdyło, J. Oszmiański, Antioxidants 9(7), 567 (2020) 27. F. Yildirim, A.N. Yildirim, B. San, S. Ercişli, Erwerbs-Obstbau 58(1), 25-29 (2016) 28. X.Z. Zhang, Y.B. Zhao, C.M. Li, D.M. Chen, G.P. Wang, R.F. Chang, H.R. Shu Journal of Plant Physiology 164(5), 574-580 (2007) 7 7 7

https://openalex.org/W2186678929

https://papiro.unizar.es/ojs/index.php/historiografias/article/download/2462/2187

Spanish; Castilian

Historicidad y pseudo-historicidad del universo: un análisis del concepto de simultaneidad”/“Historicity and Pseudo-Historicity of the Universe: an Analysis of the Concept of Simultaneity

Historiografías

cc-by-sa

Abstract This article examines in a critical way the assumption that underpins the new historiographical trend of the Big History or Deep History, that is, the construction of a narrative that, comprising from the alleged supreme historical event of the Big Bang to the birth of human intelligence, makes use of a theory of everything, the so-called Consilience. Facing this conventional image, prevailing because of the prestige of science, the author holds that this vision is no more than the result of a constructed historical narrative which stems from the overlap of a traditional time narrative to some equations only partially valid which are neither completely developed nor proved by the experimental facts. Historicity and Pseudo-Historicity of the Universe: an Analysis of the Concept of Simultaneity José Carlos Bermejo Barrera Universidad de Santiago de Compostela. España [email protected] Resumen El presente artículo examina de modo crítico el supuesto en el que se apoya las nueva corriente historiográfica de la Big History o Deep History, esto es, la construcción de un relato, que abarca desde el supuesto acontecimiento histórico primordial del Big Bang hasta al nacimiento de la inteligencia humana, y que se sirve de una teoría del todo, la llamada Consilience. Frente a esta imagen convencional, que se impone gracias al prestigio de la ciencia, el autor sostiene que dicha visión es más más bien el producto de un relato histórico construido que procede, a su vez, de la superposición de un tiempo narrativo tradicional a unas ecuaciones que sólo son parcialmente válidas y no están plenamente desarrolladas ni comprobadas con hechos experimentales. Historiografías, 5 (Enero-Junio, 2013): pp.107-117. Historiografías, 5 (Enero-Junio, 2013): pp.107-117. Key Words Big History, Consilience, simultaneity, relativistic mechanics, narrative time. Palabras clave Big History, Consilience, simultaneidad, mecánica relativista, tiempo narrativo. 107 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. La idea según la cual el universo tiene una historia y está dotado de una determinada antigüedad, medible por la astrofísica, es hoy en día patrimonio común de la cultura occidental. Y suele considerarse además que todo eso puede ser estudiado mediante un único procedimiento científico, que abarcaría un largo proceso que se habría iniciado con el Big Bang y culminado con el desarrollo de la especie y la inteligencia humanas, como culminación de la evolución de la vida en el cosmos. Subyacen en esa idea numerosos supuestos metafísicos, lógicos y epistemológicos, según los cuales, por ejemplo, la racionalidad es una característica del propio universo, razón por la cual debe ser abarcable y explicable mediante el método científico. Unos supuestos de los que los propios científicos no son conscientes, lo que les lleva a veces a creer que el propio tiempo es el objeto de su estudio: un tiempo que puede ser a la vez cosmológico, biológico e histórico, como se puede ver en el conjunto de estudios publicados en el libro coordinado por Katinka Ridderbos (Ridderbos, 2003). O bien a pensar que el mismo método permite analizar y explicar la física de partículas, la vida y la inteligencia humana – perfectamente reducible a sus bases biológicas – como ocurre en el caso del conocido libro de H. R. Maturana y F. J. Varela (Maturana y Varela, 1998). El supuesto metafísico en el que se asienta la cosmología estándar, que integra la astrofísica con la física de partículas, está siendo asumido cada vez más por los historiadores anglosajones, quienes sostienen, como en el caso de Fred Spier, que debe escribirse una Big History, que abarque desde el Big Bang al presente (Spier, 1996). Una “Gran Historia” que desembocaría en el estudio del mismísimo cerebro, y a la que otros historiadores, como Daniel Lord Smail, llaman también Deep History (Smail, 2008). Sería esa una historia que se enmarcaría en el bloque de lo que el viejo Círculo de Viena llamó la “ciencia unificada”, que se denominó Consilience en la terminología reduccionista de la sociobiología de E. O. Wilson (Wilson, 1988). Una historia que tendería siempre al reduccionismo, intentando analizar las propiedades de los sistemas emergentes, como la vida, las sociedades humanas o el conocimiento, reduciéndolos a los fenómenos de los estratos ontológicos más simples. Palabras clave Hace ya dos siglos dijo Inmanuel Kant que las intuiciones sin conceptos son ciegas – debiendo entenderse como intuiciones en su vocabulario los datos de la observación –, y que los conceptos sin intuiciones son vacíos. Esto, que es válido para el conocimiento en general, lo es más para la historia, en la que ya señaló en su momento Benedetto Croce que es esencial el análisis lógico de los conceptos de los que se sirve (Croce, 1917). Por ello analizaremos, desde un punto de vista epistemológico, un concepto clave para la historia y para la ciencia física: el concepto de simultaneidad. Suele decirse que frente a la física newtoniana, que defendía la noción de simultaneidad, por considerar que la velocidad de la luz era infinita, y por lo tanto que 108 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. cualquier acontecimiento físico es percibido en el espacio simultáneamente a la aparición en el mismo, la mecánica relativista negaría la simultaneidad, a partir del principio de que todas las ondas y partículas no pueden desplazarse a más de 300.000 kilómetros por segundo. Razón por la cual, desde un punto cualquiera del espacio, siempre se percibe un acontecimiento transcurrido en el pasado, debiendo pasarse de la idea de espacio a la de espacio-tiempo. Se admite sin discusión que esta idea del espacio-tiempo sería la clave de la macro- teoría cosmológica del Big Bang. Sin embargo, Einstein no defendió dicha teoría, ni tampoco Kurt Gödel, otro de los grandes matemáticos conocedores de la relatividad, defendiendo ambos más bien un universo estacionario (Yourgrau, 2007). Del mismo modo, desde hace unos años, viene afirmando Julian Barbour (Barbour, 1999) que la idea del tiempo de la llamada cosmología estándar no está justificada matemáticamente, ni tampoco está avalada por los datos de la física de partículas. Personalmente, he señalado que la cosmología estándar, tal y como se presenta globalmente en la teoría del Big Bang, es simplemente un relato, una narración, que pretende sentar las bases de una nueva historia global del universo (Bermejo Barrera, 2010, pp. 61-86). No voy a reiterar ahora los argumentos ya desarrollados, pues de lo que se trata es solo de analizar epistemológicamente la idea de simultaneidad. Sin embargo, convendría volver a dejar claros algunos conceptos acerca de las dos concepciones básicas del tiempo. Palabras clave Así, existen dos nociones de tiempo: a) El tiempo de la física, que es básicamente la medida del movimiento en función de la anterioridad y de la posterioridad, de acuerdo con la vieja definición contenida en la Física aristotélica. Este es el tiempo de la mecánica, ya sea clásica o relativista. b) El tiempo entendido como un flujo, ya sea ese flujo biológico, cognoscitivo – en el que el tiempo se asocia a la memoria – o ficticio, siendo éste el tiempo de las narraciones de todo tipo. b) El tiempo entendido como un flujo, ya sea ese flujo biológico, cognoscitivo – en el que el tiempo se asocia a la memoria – o ficticio, siendo éste el tiempo de las narraciones de todo tipo. Curiosamente, es este tipo de tiempo narrativo el que se esconde en la cosmología estándar bajo un inmenso ropaje de ecuaciones y datos experimentales (Penrose, 2006). Y en ella, como en cualquier narración, se parte de las siguientes bases. Se llama narración a una construcción verbal compleja en que un sujeto o un protagonista, en este caso el universo, desarrolla una acción continua en el tiempo, un tiempo que se divide en pasado, presente y futuro. Ese sujeto narrativo, o protagonista del relato, subordina todos los elementos narrativos, sean del tipo que sean, al desarrollo de una acción, que podrá ser lineal o no, y con un fin cerrado o abierto. Parece bastante evidente que ese protagonista sería el cosmos, o el universo en la narración histórico-científica de la cosmología estándar. Intentaremos a continuación analizar las contradicciones que implica el superponer continuamente el tiempo de la mecánica con el tiempo narrativo, así como poner de manifiesto que ideas tales como la del universo no pertenecen al ámbito de la física, ni de ninguna otra ciencia, sino al viejo ámbito de la metafísica, asumido sin ser plenamente 109 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. conscientes de ello por parte tanto de los físicos como de los historiadores defensores de las Big o Deep Histories. conscientes de ello por parte tanto de los físicos como de los historiadores defensores de las Big o Deep Histories. Prolegómeno En la física clásica, tal y como fue establecida por Isaac Newton, se utilizaron dos conceptos fundamentales: los del espacio y tiempo absolutos. Sostenía Newton que el espacio es independiente de los acontecimientos que transcurren en su seno, que por definición son todos los acontecimientos posibles, y que permanece constante e inmóvil. Ese espacio es llamado sensorium Dei, o lo que es lo mismo, es el órgano mediante el cual Dios conoce el mundo, estando omnipresente en todos y cada uno de sus puntos. El espacio newtoniano no es pues solo el escenario en el que transcurren todos los acontecimientos físicos, sino también el teatro en el que son contemplados mediante el conocimiento. Siendo ese espacio ilimitado, si no infinito, para hacer posible que la existencia de una única fuerza en el cosmos, la gravitación, no consiga provocar un colapso en este modelo de universo estacionario. Lo mismo ocurriría con el tiempo, que fluye uniformemente al margen de los acontecimientos que transcurren en su seno. Y que del mismo modo es también sensorium Dei, es decir, el instrumento por el cual Dios esta siempre presente y gracias al cual conoce lo que puede ocurrir en cualquier momento. Es pues también el tiempo escenario y teatro a la vez. La física relativista y su síntesis mecánico-cuántica superaron estas ideas de espacio y tiempo absolutos, pero vuelven a caer en ellas al pretender desarrollar una teoría del todo y una historia del cosmos, de acuerdo con un modelo narrativo tradicional, es decir, de acuerdo con el eterno modelo de cualquier narración posible. A continuación, intentaremos ponerlo de manifiesto manejando únicamente las ideas de espacio y tiempo mecánico y tiempo como flujo. Por ello, nos limitaremos a hablar solo de espacio, acontecimientos y puntos, pudiendo ser sustituidos estos conceptos por los de partículas, por parte de quien lo desee y quiera introducir las matizaciones pertinentes. Axioma 1: todo espacio es simultáneo consigo mismo. Axioma 1: todo espacio es simultáneo consigo mismo. Puesto que todos sus puntos están presentes en él, en tanto que pertenecen al mismo. Axioma 2: todo tiempo es simultáneo consigo mismo. Axioma 2: todo tiempo es simultáneo consigo mismo. En tanto que todos los instantes posibles están contenidos en su propio flujo. La no formulación clara de estos axiomas es lo que llevó a Newton a concebir las ideas de espacio y tiempo absolutos. Ambos axiomas están igualmente presentes en la cosmología estándar, que en su caso sin embargo pretende presentarlos como hechos. Veamos cómo ello fue así a partir de la relatividad einsteiniana: 110 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. 1- En el universo existe una constante, la velocidad de la luz. Ningún movimiento ni ninguna información puede desplazare a una velocidad mayor a ella. Por eso no es posible en la mecánica relativista la existencia de la simultaneidad. 2- Sea AB una recta en la que A emite una señal o una partícula a B, siendo el tiempo del viaje de esa señal o partícula: t = e/c 2- Sea AB una recta en la que A emite una señal o una partícula a B, siendo el tiempo del viaje de esa señal o partícula: t = e/c 3- La señal enviada a A a B en un tiempo t es percibida en B como un acontecimiento del pasado, ya que no son simultáneas ni su emisión ni su recepción. Lo que obligó a introducir la noción de espacio-tiempo y reformular la mecánica clásica. 3- La señal enviada a A a B en un tiempo t es percibida en B como un acontecimiento del pasado, ya que no son simultáneas ni su emisión ni su recepción. Lo que obligó a introducir la noción de espacio-tiempo y reformular la mecánica clásica. 4- Sea t(a) un reloj que mide el tiempo en A, y t (b) un reloj que mide el tiempo en B. Como esos relojes no son simultáneos, al estar alterado el tiempo por el movimiento, puesto que el tiempo del trayecto será: t = e/c, habría que suponer que se producirían alteraciones en el reloj de B, al llegar la señal, que sería percibida como propia del pasado y no del presente. Por ello se suele decir que en el presente percibimos el pasado del universo, hasta llegar a su instante inicial: Big Bang. 17- Llamamos universo al conjunto de todo lo que existe. 17- Llamamos universo al conjunto de todo lo que existe. Axioma 2: todo tiempo es simultáneo consigo mismo. 5- Esto no es filosóficamente correcto, puesto que, dado que en la mecánica relativista espacio y tiempo forman un todo indisociable, consecuentemente el pasado de B en el tiempo t = e/c, solo sería perceptible en otro punto del espacio C, al que se llegaría en ese mismo tiempo 5- Esto no es filosóficamente correcto, puesto que, dado que en la mecánica relativista espacio y tiempo forman un todo indisociable, consecuentemente el pasado de B en el tiempo t = e/c, solo sería perceptible en otro punto del espacio C, al que se llegaría en ese mismo tiempo Estaría pues ese otro punto C separado en el espacio-tiempo a distancia igual que la de A a B. Lo que quiere decir simplemente esto: no es posible mantener la idea de simultaneidad en la mecánica relativista, aunque ello se hace subrepticiamente. 6- Si consideramos que hay un reloj en A, que marca el tiempo de A y otro en B, que marca el tiempo de B, teniendo que transformar con una ecuación el tiempo de A en B estamos cayendo en una contradicción, puesto que utilizamos a la vez la idea de tiempo de la mecánica – como medida del movimiento –, y la idea biológica del tiempo como flujo continuo. 7- En el universo no pueden darse trayectorias rectilíneas, puesto que la gravedad hace que el espacio sea curvo. Todas las trayectorias posibles se realizan en unas líneas geodésicas, cuya curvatura puede ser variable según la masa, que determina la densidad del espacio-tiempo. 7- En el universo no pueden darse trayectorias rectilíneas, puesto que la gravedad hace que el espacio sea curvo. Todas las trayectorias posibles se realizan en unas líneas geodésicas, cuya curvatura puede ser variable según la masa, que determina la densidad del espacio-tiempo. No obstante, el ejemplo anterior seguiría siendo válido, aunque podría ocurrir que el pasado de B en C coincidiese con A, en una trayectoria cerrada. Ello supondría afirmar que al ser el pasado de A presente en B y el pasado de B presente en A se caería en una tautología, que supondría la anulación del tiempo, puesto que no puede haber movimiento ni consecuentemente tiempo en un punto del espacio, a menos que se configure interiormente como otro espacio confinado en él. 111 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. Axioma 2: todo tiempo es simultáneo consigo mismo. 8- En el espacio-tiempo relativista dos acontecimientos están separados por tiempos iguales si están a la misma distancia (suponiendo que la información o la partícula viajen a la velocidad de la luz), sea cual sea la curvatura del espacio. 8- En el espacio-tiempo relativista dos acontecimientos están separados por tiempos iguales si están a la misma distancia (suponiendo que la información o la partícula viajen a la velocidad de la luz), sea cual sea la curvatura del espacio. 9- En el espacio-tiempo relativista la simultaneidad de todos los acontecimientos en el espacio presupone un universo estacionario; en la cosmología estándar no. 9- En el espacio-tiempo relativista la simultaneidad de todos los acontecimientos en el espacio presupone un universo estacionario; en la cosmología estándar no. 10- En la cosmología estándar se consigue la simultaneidad perfecta cuando se da una singularidad. 11- Una singularidad es simultánea en sí misma, puesto que en ella se anulan el tiempo y el movimiento. Es esa singularidad el instante inicial del universo con un tiempo cero, siendo su curvatura y densidad infinitas. 12- El modelo estándar se basa en correlacionar tiempo, curvatura, densidad y energías mediante la construcción de conjeturas que nunca podrán ser comprobadas ni objeto de experimentación. 13- El modelo estándar anula la idea de observación física, puesto que para observar el instante inicial habría que retroceder en el espacio-tiempo a la misma distancia que a la edad estimada del universo, lo cual es metafísicamente imposible, puesto que supondría reestablecer dos ideas imposibles: a) la de simultaneidad, negada por la relatividad, y b) la del espacio como sensorium hominis, una vez apartado Dios de la teoría física. 14- Conjeturar estadios sucesivos de la evolución del cosmos, desde el Big Bang al presente, volviendo a un eterno retorno con un posible Big Crunch, no es más que la construcción de otra narración cosmológica. 15- El tiempo del Big Bang es un tiempo narrativo complejo con todas sus variables: energía, curvatura, densidad, masa, carga, campos de diferentes tipos, englobados bajo la idea común de espacio. 16- El espacio del Big Bang no es una magnitud física, sino la vieja idea del universo. 16- El espacio del Big Bang no es una magnitud física, sino la vieja idea del universo. 17- Llamamos universo al conjunto de todo lo que existe. 18- La física no puede desarrollar teorías acerca de todo lo que existe, ni descubrir teorías del todo, a menos que pretenda volver a Newton y al espacio y tiempo como sensori Dei. 19- La cosmología estándar se configura como una sucesión de estadios cosmológicos en el tiempo. 20- Cada estadio por sí mismo puede ser reconstruido como tal por la física teórica, pero no su sucesión, aunque se utilicen variables como la pérdida de temperatura, la entropía… ISSN 2174-4289 112 112 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. 21- La sucesión de los estadios de la historia del cosmos se construye a partir no de un tiempo físico, sino de un tiempo narrativo, que se superpone a él. 22- En el tiempo narrativo de esa cosmología todos los estadios y acontecimientos cosmológicos son simultáneos con el instante inicial del universo, dentro de un flujo temporal, ya que en todo flujo temporal existe también una simultaneidad dentro del marco de una sucesión, por ser todos los acontecimientos conexos con ella. 23- Consecuentemente, la teoría narrativa de la cosmología estándar, que además solo sería viable si no tuviésemos conocimiento de la existencia de la materia oscura y la energía oscuras (véase el problema en Penrose, 2006), es conceptualmente una rama de la vieja metafísica. 24- Dado que la materia oscura forma parte del espacio tiempo, suponer su existencia impide, por definición, formular cualquier teoría global acerca del mismo. Y lo mismo ocurre con la energía. 25- En la narración cosmológica no se puede pasar sin solución de continuidad de la física de partículas al desarrollo de la vida y luego de la inteligencia humana, como hacen muchos biólogos e historiadores sin explicar cómo y solo utilizando una mera sucesión narrativa. 26- Ello es así porque aunque la vida solo sea concebible en esa narración en determinadas condiciones cosmológicas, geofísicas o químicas, el flujo del tiempo de la vida posee una lógica propia, al igual que el flujo de la historia humana. 26- Ello es así porque aunque la vida solo sea concebible en esa narración en determinadas condiciones cosmológicas, geofísicas o químicas, el flujo del tiempo de la vida posee una lógica propia, al igual que el flujo de la historia humana. 17- Llamamos universo al conjunto de todo lo que existe. 27- La vida se ha desarrollado en una parte confinada del espacio, y por ello el espacio-tiempo de la vida queda al margen de las leyes de la evolución mecánico-cuántica del cosmos. 28- La existencia de un determinado conjunto de condiciones para que la vida sea posible es una condición necesaria, pero no suficiente, para dar cuenta de su origen y su desarrollo (Ward y Brownlee, 2000). 29- Lo mismo puede afirmarse de la historia humana, considerada como parte de la evolución de la vida biológica, condición necesaria pero no suficiente para dar cuenta de su lógica. 30- No es necesario apelar a poderes sobrenaturales para explicar el origen de la vida o de la especie humana, sino desarrollar modelos explicativos y narrativos diferentes que intenten dar cuenta, en la medida de lo posible, de ese tipo de fenómenos. 31- La cosmología estándar, al confundir sistemáticamente las condiciones necesarias con las suficientes, cae en un reduccionismo totalmente exacerbado. 31- La cosmología estándar, al confundir sistemáticamente las condiciones necesarias con las suficientes, cae en un reduccionismo totalmente exacerbado. 113 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. 32- En la cosmología estándar, dejando a un lado su intento de integrar el mundo de la vida y la historia humanas (obra siempre más de biólogos o historiadores que de físicos, como dijimos), se construye un metarrelato cosmológico. 32- En la cosmología estándar, dejando a un lado su intento de integrar el mundo de la vida y la historia humanas (obra siempre más de biólogos o historiadores que de físicos, como dijimos), se construye un metarrelato cosmológico. 33- En este metarrelato cosmológico se reconstruye de nuevo el tiempo absoluto, gracias a la confusión entre tiempo físico y tiempo como flujo o tiempo narrativo. 34- En ese metarrelato cosmológico todos los acontecimientos físicos vuelven a ser simultáneos en el flujo temporal global, que abarca todos los estadios mecánico-cuánticos sucesivos. 35- Los estadios mecánico-cuánticos no pueden ser sucesivos físicamente hablando, aunque sí metafísicamente, porque dependerían del conocimiento de un conjunto de leyes globales que fuesen diferentes a la evolución de la propia estructura de la idea de tiempo cosmológico, único principio rector. 36- El conocimiento de esas leyes supondría la anulación de la idea de metarrelato cosmológico, ya que traería consigo relevar al tiempo narrativo por otro meramente físico. 17- Llamamos universo al conjunto de todo lo que existe. 37- La propia existencia de la materia y energía oscuras hace que el conocimiento de esas leyes sea inconcebible, pues dichas leyes podrían no serles aplicables aquélla, de la que no es posible conocer nada. 38- La física es un conocimiento positivo y progresivo, pero que nunca puede llegar a clausurarse a sí mismo, razón por la cual no será posible jamás elaborar “teorías del todo”, ni hallar una única ley del universo, como la que Newton creyó haber hallado, por mucho que la física incremente su conocimiento. 39- El conocimiento físico depende tanto de la observación como de la capacidad de construir modelos matemáticos que permitan procesar e integrar la información. 40- No es posible agotar las observaciones posibles. 40- No es posible agotar las observaciones posibles. 41- Nadie puede afirmar que no sea posible concebir modelos matemáticos absolutamente inéditos. La historia de las matemáticas es buena prueba de ello. 42- Creer que los datos experimentables disponibles en un momento dado son todos los posibles carece de sentido. No solo porque siempre se podrán realizar nuevas observaciones, sino porque se pueden crear nuevas tipologías de fenómenos, como ocurrió en la historia de la física con el descubrimiento del electromagnetismo, la radioactividad… 43- Por ello, intentar integrar en un modelo global cerrado observaciones fragmentarias, procesadas con una tecnología en constante cambio, formuladas con conceptos que puede ser renovados o abandonados en cualquier momento (piénsese en la existencia del éter), con instrumentos lógicos y matemáticos siempre mejorables, carece de sentido científico. 114 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. 44- El ansia de integración de todos los conocimientos en un modelo único y cerrado es característica de la mala filosofía o de la mala teología, que cree poder sintetizar ciencia y religión. 44- El ansia de integración de todos los conocimientos en un modelo único y cerrado es característica de la mala filosofía o de la mala teología, que cree poder sintetizar ciencia y religión. 45- Algunos científicos actuales creen poder realizar esa integración en modelos como la Consilience de E. O. Wilson (Wilson, 1998). Por definición, esos intentos han de ser fallidos, ya que no existe un método científico único, sino diferentes métodos para las diferentes ciencias, de acuerdo con los tipos de fenómenos observados, los métodos e instrumentos de observación y las formas de procesar y analizar los datos observados. 17- Llamamos universo al conjunto de todo lo que existe. Siempre y cuando no estemos hablando de ciencias formales, como las matemáticas, en las que ninguna observación es posible. 46- El discurso sobre la ciencia es un discurso que no se refiere a los fenómenos observables. 46- El discurso sobre la ciencia es un discurso que no se refiere a los fenómenos vables. 47- El discurso sobre la ciencia no se refiere al mundo. 48- El discurso sobre la ciencia se refiere a la ciencia misma. 60- Consecuentemente, todo metarrelato cosmológico es básicamente poesía. 61- La física, aunque tiene innegables valores estéticos, no puede ser poesía, pues ha de delimitar los hechos de los que habla, el sentido de los términos que utiliza y los instrumentos abstractos de sus análisis. La física se construye a partir de la negación de la intuición – base de la poesía –, como hace ya muchos años señaló Gaston Bachelard, (Bachelard, 1951). Por ello la física ha de renunciar a la función poética. 62- El abandono de la función poética en la cosmología estándar supondría dejar de lado los intentos de construir un metarrelato y quedarse solo con los datos fragmentarios, y los esfuerzos de procesarlos, integrarlos y construir modelos parciales, transitorios y perecederos, que es en donde reside la trágica grandeza de esa ciencia. 62- El abandono de la función poética en la cosmología estándar supondría dejar de lado los intentos de construir un metarrelato y quedarse solo con los datos fragmentarios, y los esfuerzos de procesarlos, integrarlos y construir modelos parciales, transitorios y perecederos, que es en donde reside la trágica grandeza de esa ciencia. 63- Consecuentemente, cuando los historiadores trasladan al campo de la historiografía el modelo de la cosmología estándar no contribuyen al desarrollo de una nueva ciencia, sino que se limitan a asumir la función poética propia de la física. 63- Consecuentemente, cuando los historiadores trasladan al campo de la historiografía el modelo de la cosmología estándar no contribuyen al desarrollo de una nueva ciencia, sino que se limitan a asumir la función poética propia de la física. 64- No se pueden sustituir las funciones poéticas propias de la historia (Bermejo Barrera, 2005, pp. 182-194), por las de la física, ya que corresponden a necesidades expresivas distintas, dentro de la estructura global del lenguaje (Bermejo Barrera, 2011). 64- No se pueden sustituir las funciones poéticas propias de la historia (Bermejo Barrera, 2005, pp. 182-194), por las de la física, ya que corresponden a necesidades expresivas distintas, dentro de la estructura global del lenguaje (Bermejo Barrera, 2011). 49- La ciencia no es un objeto neutro observable. 59- Todo relato cosmológico global es a su vez evocativo en el tiempo, narrativo y es un intento de dar sentido a la historia del universo, la vida y el hombre. 59- Todo relato cosmológico global es a su vez evocativo en el tiempo, narrativo y es un intento de dar sentido a la historia del universo, la vida y el hombre. 60- Consecuentemente, todo metarrelato cosmológico es básicamente poesía. 60- Consecuentemente, todo metarrelato cosmológico es básicamente poesía. 49- La ciencia no es un objeto neutro observable. 49- La ciencia no es un objeto neutro observable. 49- La ciencia no es un objeto neutro observable. 49- La ciencia no es un objeto neutro observable. 50- Las ciencias son el resultado de las acciones colectivas que llevan a cabo los científicos. 50- Las ciencias son el resultado de las acciones colectivas que llevan a cabo los científicos. 51- El discurso sobre la ciencia, o las ciencias, puede ser a la vez un intento de analizar formalmente lo que esas ciencias son, y eso sería el objeto de la filosofía de las ciencias, o una forma de expresión de los propios científicos. 52- Cuando los científicos hablan de las ciencias sin analizarlas como objetos de conocimiento desde el exterior, básicamente están expresando sus deseos, temores e inquietudes, objetos básicos de la expresión humana. 53- El discurso de los científicos sobre la ciencia cumple, pues, una función expresiva, individual o colectiva, que es una expresión poética. Aunque se presente como el resultado final de las propias ciencias, no es más que wishful thinking. 54- La poética de la ciencia tiene como fin dar sentido a la propia vida de los científicos, y conseguir simultáneamente su reconocimiento social como personas y como colectivos. 55- El mundo poético es un mundo cerrado, perfecto y autorreferente, por su propia naturaleza. 56- En el mundo poético se anula ficticiamente la distancia entre la realidad y el deseo, el hablante y el mundo, al igual que en el lenguaje de la religión. 56- En el mundo poético se anula ficticiamente la distancia entre la realidad y el deseo, el hablante y el mundo, al igual que en el lenguaje de la religión. 115 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. 57- El mundo poético se construye a partir de la metáfora, y consecuentemente de los desplazamientos entre los varios sentidos y significados de cada palabra. 57- El mundo poético se construye a partir de la metáfora, y consecuentemente de los desplazamientos entre los varios sentidos y significados de cada palabra. 58- El mundo poético es básicamente evocativo, narrativo y expresivo. E intenta construir un sentido global de la vida y el cosmos. 58- El mundo poético es básicamente evocativo, narrativo y expresivo. E intenta construir un sentido global de la vida y el cosmos. Bibliografía Spier, Fred, The Structure of Big History. From the Big Bang until Today. Amsterdam Amsterdam University Press, 1996. Ward, Peter D. y Brownlee, Donald, Rare Earth. Why Complex Life is Uncommon in t Universe. New York: Copernicus, 2000. Wilson, Edward O., Consilience: the Unity of Knowledge. New York: Knopf, 1988. ourgrau, Palle, Un mundo sin tiempo. El legado olvidado de Gödel y Einstein. Barcelona: usquets, 2005, 2007. Bibliografía Bachelard, Gaston, Activité rationaliste de la physique contemporaine. París: PUF, 1951. Barbour, Julian, The End of Time. London: Phoenix, 1999. ISSN 2174-4289 116 Bachelard, Gaston, Activité rationaliste de la physique contemporaine. París: PUF, 1951. Barbour, Julian, The End of Time. London: Phoenix, 1999. Bermejo Barrera, José Carlos, “On History Considered as Epic Poetry”, History and Theory, 44, 2 (2005): 182-194. Bermejo Barrera, José Carlos, The Limits of Knowledge and the Limits of Science. Santiago de Compostela: USC Editora, 2010. Bermejo Barrera, José Carlos, Los Límites del lenguaje. Proposiciones y categorías. Madrid: Akal, 2011. Croce, Benedetto, Logic as the science of the pure concept. London: MacMillan, 1917. Maturana, Humberto R. y Varela, Francisco J., The Tree of Knowledge. The Biological Roots of Human Understanding. Boston & Londres: Shambhala, 1998. Penrose, Roger, El camino a la realidad. Una guía completa de las leyes del universo. Barcelona: Debate, 2004, 2006. Ridderbos, Katinka (ed.), El tiempo. Cambridge: Cambridge University Press, 2002, 2003. ermejo Barrera, José Carlos, “On History Considered as Epic Poetry”, History and heory, 44, 2 (2005): 182-194. Bermejo Barrera, José Carlos, The Limits of Knowledge and the Limits of Science. Santia de Compostela: USC Editora, 2010. Bermejo Barrera, José Carlos, Los Límites del lenguaje. Proposiciones y categorías. Madrid: Akal, 2011. Croce, Benedetto, Logic as the science of the pure concept. London: MacMillan, 1917. Maturana, Humberto R. y Varela, Francisco J., The Tree of Knowledge. The Biological Roots of Human Understanding. Boston & Londres: Shambhala, 1998. Penrose, Roger, El camino a la realidad. Una guía completa de las leyes del universo. Barcelona: Debate, 2004, 2006. , , dderbos, Katinka (ed.), El tiempo. Cambridge: Cambridge University Press, 2002, 2003. 116 ISSN 2174-4289 Historiografías, 5 (Enero-Junio, 2013): pp.107-117. Smail, Daniel L., On Deep History and the Brain. Berkeley: University of California Press, 2008. Spier, Fred, The Structure of Big History. From the Big Bang until Today. Amsterdam: Amsterdam University Press, 1996. Ward, Peter D. y Brownlee, Donald, Rare Earth. Why Complex Life is Uncommon in the Universe. New York: Copernicus, 2000. Wilson, Edward O., Consilience: the Unity of Knowledge. New York: Knopf, 1988. Yourgrau, Palle, Un mundo sin tiempo. El legado olvidado de Gödel y Einstein. Barcelona: Tusquets, 2005, 2007. Smail, Daniel L., On Deep History and the Brain. Berkeley: University of California Press, 2008. Spier, Fred, The Structure of Big History. From the Big Bang until Today. Amsterdam: Amsterdam University Press, 1996. Smail, Daniel L., On Deep History and the Brain. Berkeley: University of California Press, 2008. Para citar este artículo: José Carlos Bermejo Barrera, “Historicidad y pseudo-historicidad del universo: un análisis del concepto de simultaneidad,” Historiografías, 5 (enero-diciembre, 2013): pp. 107-117, y f g p Yourgrau, Palle, Un mundo sin tiempo. El legado olvidado de Gödel y Einstein. Barcelona: Tusquets, 2005, 2007. Profile The author is professor of Ancient History at the University of Santiago (Spain), an expert in philosophy of history and historiography, and has studied the Ancient Greece and Rome myths. Among his major works, the series that include the term “Historia teórica” stand out: Ensayos de Historia teórica (Madrid, Akal, 1987), Replanteamiento de la historia: ensayos sobre Historia teórica II (Madrid, Akal, 1989), La fundamentación lógica de la historia: Introducción a la Historia teórica (Madrid, Akal, 1991), Genealogía de la historia: ensayos sobre Historia teórica III (Madrid, Akal, 1999) (written with José Andrés Piedras Monroy), Qué es la Historia teórica (Madrid, Akal, 2004), El fin de la historia: ensayos de historia teórica (Madrid, Akal, 2007), Introducción a la Historia teórica (Madrid, Akal, 2009), La consagración de la mentira. Entre la realidad y el silencio (Madrid, Siglo XXI, 2012). El autor es catedrático de Historia Antigua de la Universidad de Santiago (España), es especialista en filosofía de la historia e historiografía, y estudioso de los mitos de la Antigua Grecia y Roma. Entre sus principales obras destacan la serie que lleva como título el término “Historia teórica”: Ensayos de Historia teórica (Madrid, Akal, 1987); Replanteamiento de la historia: ensayos sobre Historia teórica II (Madrid, Akal, 1989); La fundamentación lógica de la historia: Introducción a la Historia teórica (Madrid, Akal, 1991); Genealogía de la historia: ensayos sobre Historia teórica III (Madrid, Akal, 1999) (escrito con José Andrés Piedras Monroy); Qué es la Historia teórica (Madrid, Akal, 2004); El fin de la historia: ensayos de historia teórica (Madrid, Akal, 2007); Introducción a la Historia teórica (Madrid, Akal, 2009), José Carlos Bermejo, La consagración de la mentira. Entre la realidad y el silencio (Madrid, Siglo XXI, 2012). Fecha de recepción: 29 de agosto de 2012 Fecha de aceptación: 7 de febrero de 2013 Publicado: 15 de junio de 2013 Fecha de recepción: 29 de agosto de 2012 Fecha de aceptación: 7 de febrero de 2013 Fecha de recepción: 29 de agosto de 2012 Fecha de aceptación: 7 de febrero de 2013 Publicado: 15 de junio de 2013 Para citar este artículo: José Carlos Bermejo Barrera, “Historicidad y pseudo-historicidad del universo: un análisis del concepto de simultaneidad,” Historiografías, 5 (enero-diciembre, 2013): pp. 107-117, http://www.unizar.es/historiografias/historiografias/numeros/5/bermejo.pdf http://www.unizar.es/historiografias/historiografias/numeros/5/bermejo.pdf 117 ISSN 2174-4289

https://openalex.org/W3146350445

https://actavet.vfu.cz/media/pdf/avb_1991060020099.pdf

English

The Morphogenesis of Circumvallate Papillae and the Differentiation of Taste Buds in the Pig at 41 to 84 Days of Prenatal Development

Acta veterinaria Brno

cc-by

THE MORPHOGENESIS OF CIRCUMVALLATE PAPILLAE AND THE DIFFERENTIATION OF TASTE BUDS IN THE PIG AT 41 TO M DAYS OF PRENATAL DEVELOPMENT F. TICHY Department of Anatomy, Histology and Embryology University of Veterinary Science, 612 42 Bmo Received Febru4ry 6, 1990 Received Febru4ry 6, 1990 ACTA VET. BRNO,8O, 1991: 99-100 ACTA VET. BRNO,8O, 1991: 99-100 Abstract Tichy, F.: The Morphogenesis of CircumtJallate PapiUae and the Differentiation of Taste Buds in the Pig at 41 to 64 Days of Prenatal DefJe/opment. Acta vet. Bmo, 60, 1991: 99-1()()' Samples of tongues collected from porcine foetuses at 41, 44, 50, 53, 57 and ~ days after fc:rtilization were examined for the appearance and development of surface lingual structures. At each of the stages studied, the morphogenesis of circumvallate papillae and the differentiation of taste buds were recorded. Attention was paid to the formation of gustatory glands· and their ducts and to the development of the furrow encircling the circumvallate papilla. The differentiation of the other types of lingual papillae was also studied. The formation of taste buds in the epithelium coincided with the shape differen- tiation of circumvallate papillae. The buds appeared first in the dorsal surface epithelium and later in the wall epithelium. The number of differentiating buds within one papilla increased with foetal age. Different cell types composing the taste bud could be distinguished soon after the bud was formed. g The circUmvallate papilla was either formed as a unified structure of typical appe- arance or had first a composite base later giving rise, by fusion of its segments, to the typical papilla. yp p p The other, lingual papillae developed at later stages than the circumvallate papillae. ll ll d ff The other, lingual papillae developed at later stages than the Tongue, circumtJallate papilla, taste bud, differentiation, pig Tongue, circumtJallate papilla, taste bud, differentiation, pig The structure, function and distribution of taste buds were studied as early as in the second half of the last century. The first observations were made in fish (Schulze 1863), then in mammal' (Loven 1868; Schwalbe 1868). The general information on the structure of the taste organ in some species of laboratory mammals and in man were obtained by means of light microscopy at the beginning of this century (Kolmer 1910; Retzius 1912; Heidenhain 1914). Pro~s Pro~s in understanding the mechanism of taste perception occurred with the advent of tech- niques ailowing observations at the ultrastructural level. Materials and Methods Samples were collected from the tongues of porcine foetuses at 41, 44, 50, 53, 57 and 64 days after fertilization. The age of foetuses was estimated by their crown-rump lenght according to Evans and Sack (1973). .. ' W Three pigs were sampled in each age category. The material Was excised from the tongues in the following areas: junction between thebody:,~d J;oot, lateral part of the root, apical dorsum linguae. The samples were fixed immediately in 10 % neutral formaldehyde. To prevent artifacts ensuing from a sudden dehydration of embryonic tissues rich in water, the graded alcohol series, starting with 10 % alcohol, increased by 10 % in each subsequent bath. The sections were routinely stained with haematoxylin and eosin. To intensify the visualization of selected structures, the impregnation technique according to Gomori and nuclear red staining were used. Some of the seetions were stained with the greerttrichrome reagent. . ' , Abstract Many of data on the submicroscopic struc- ture of taste buds together with characterization of their cell types (Beidler and Smallman 1965; De Lorenzo 1958; Engstrom and Rytzner 1956; Fahrmannand Schuchard 1967; Farb- man 1965a; Farbman and Vonkers 1971; Fujimoto and Murray 1970; Murray 1969,1971, 1973; Murray and Murray 1960, 1967, 1970; Murray et aI. 1969; Takeda and Hoshino 1975; Trujillo-Cenoz 1957; Uga 1966,1969) have provided the grounds for the development of biochemical and physiological methods for investigation of taste perception (Beidler 1970; De Han and Graziadei 1971,1973; Desgranges 1966; Hirata and Nada 1975; Spoendlin 1970; Takeda 1976; Beauchamp and Cowart 1986). All these studies have contributed signi- ficantly to the understandiDg of processing and transport of taste stimuli within the bud. The most important information and methods of investigation have been reviewed and preSented in text- books (Cormack 1984; Banks 1981; Smolich and Michael 1985; Rooss and Reith 1985). 100 The. de,telopment of the' taste organ· in' the lingual epithelium is well recorded in both man and various mammalian species (Beidler and Smallman 1965; Farbman 1965b, 1971; Takeda 1972). None of the relevant studies, however, has been concerned with the stages of ontogenic development or has paid attention to the changes occurring during development of the gustatory receptor. Our earlier study (Tichy and Cerny 1987) described the differentiation of taste buds ~t selected stages of ontogeny in sheep. This paper will deal with the formation, development and distribution of the taste buds and their functionally and topographically related structures in the prenatal pig~ Results' mucosa at 41 days offoetaLage (Fig. 1, Plate I, at the end of the volume The mucosal surface of the dorsum linguae was uneven. Frequent dome-shaped protrusions:,lthe anlagen. (rudiments) of papillae, could be seen .along·.the lateral margin,s of the tongue.J'hey were more prominent in the caudal regions, parti- cularly at the junction, of the body and root of the tongue. The mucosa was covered with markedly stratified epithelium .. It.consisted of 2 to 3 lower layers, in: which cells . with rounded hyperchromatic nuclei were arranged in a palisad~-like manner, and. an upper layer of poorly-stained cells irregular in shape and size .. The,10wer layer cells, distin(:tly smaller than the upper layer ones, constituted the lower half of the epithelium,. while the other half was made of only' one layer of light, large cells. The upper layer was markedly thinner above the dome-like protrusions; in some instances it was even missing. The linguiu epithelium had a uniform appearance allover the mucosal surface wi~out any changes indicating the onset of taste bud formation. :'A :A distinct basement membrane separated ,the epitQ.elium from the layer of mucosal connective tissue containing numerous bloc)d. ~essels. This extended against the epithelium. forming the stromata of ptimitive papillae. Below these, connectiv~ tissue increased in density and began to organize which was a:,:commencem,ent of the aponeurosis linguae. , .' , , p g Linguiu mucosa at 44 days of foetal age (Fig. 2, Plate I) 1"'h d lik i id h li i f 1"'he: dome-like protrusions evident on the linguiu surface in the previous pe- riod were masked completely w~th proliferating epithelial cells; which. gave the mucosal. surface a mildly undulating appearance. '. Cross-sections through the linguiu mucosa showed distinct anlagen of circum- vallate papillae which were marke4, with compact cell bands preceding the for- mation ofa furrow around each papill~.· No anlagen of the other types of linguiu papillae were laid at this stage. ' ,., g pithelium of tllelingual surface, similarly to the previoll..C1. period, ha The epithelium of tllelingual surface, similarly to the previoll..C1. p 101 two parts: the lower layer consisted of several strata of well-stained cells, the upper layer was made up of one or two strata of light, large, irregular cells. These poorly-stained cells produced the cell bands marking the extent and localization of the future encircling furrow. Lingual mucosa at 50 days of foetal age (Figs 3, 4, Plate II) At the lateral margins of the dorsum linguae, the lingual mucosa showed large disc-shaped elevations markedly protruding above the surface at the junction between the body and root of the tongue. It could be seen in cross-sections that the discs were the anlagen of circumvalla- te papillae; they had already attained a typical shape but had not yet developed the encircling furrows. Their underlying cell bands, however, were found to run deeply into the connective tissue of the lamina propria mucosae. With their bottom part they penetrated under the base of the circumvallate papilla, while their terminal part produced fork-like extensions later developing into ducts of the gustatory glands. The other types of lingual papillae were observed only as mild protrusions of the lingual mucosa. p g The lingual epithelium did not differ in structure from that of the previous age stzge. However, its thickness was greater due to an increased number of layers of poorly-stained polyedric surface cells. y p y p y The dorsal surface of the anlagen of circumvallate papillae showed the epithe- lium with taste buds at different stages of differentiation. These presented most frequently as cell clusters slightly exceeding in height the basal epithelial layer and easily discernible by less intense staining. All the cells involved in the com- mencement of taste buds at this stage were uniform in appearance. Cells of the basal epithelial layer attached to the anlagen of the taste buds attained a spindle- -like shape; this was accompanied by similar morphological changes in their nuclei. The rudiment of the aponeurosis linguae was more marked than in the previous age category, but under the bases of circumvallate papillae the thickening of connective tissue could not be seen. Lingual mucosa at 53 days of foetal age (Fig. 5, 6, Plate III) The epithelium of the dorsum linguae at this stage was characterized by nu- merous protuberances, varying in size and shape, and shallow and narrow de- pressions. The protuberances were the future papillae not yet differentiated in shape, the depressions indicated the separation of the papillae from the surround- ing tissue. The epitheliaum did not differ in structure from that seen in the pre- vious period. p A cross section through the dorsum linguae demonstrated more frequent and larger anlagen of papillae than in the 50-day-old foetus. Results' g In the epithelium of the dorsum linguae in the close vicinity of the cell bands, the lower epithelial layer included cells differing by their elongated shape and low intensity of staining. These cells extended down to the basement membrane and, in cross-sections, were bordered by pairs of cells with dark elongated nuclei. y p g The structure of the musocal connective tissue was markedly denser and more vascularized than in the previous period. The rudiment of the aponeurosis lin- guae was also more distinct. Lingual mucosa at 50 days of foetal age (Figs 3, 4, Plate II) Lingual mucosa at 50 days of foetal age (Figs 3, 4, Plate II) The picture, however, did not yet allow us to distinguish between different types of lingual papillae with the exception of the circumvallate papillae. 102 These papillae had a typical configuration and, because of their size, they became the most conspicuous formations on the lingual epithelium. Although the encircling furrow was not completed, its basis in the form of a cell band ex-· tended deeply into the mucosal connective tissue. In the cell band, a narrow groove began to form from th~ epithelial surface down. At this stage it ran down one third of the total cell band height. Its lower part divided into thin projections directed towards the aponeurosis linguae rudiment. These would later .constitute the gustatory gland ducts. g y g Taste buds in various stages of differentiation were observed in the epithelium of the dorsal surface and walls of the circumvallate papilla as well as in the epi- thelial basis of the encircling furrow. Most frequently they presented as round cell clusters which did not reach as far as the epithelial surface but were covered with one or two layers of cells. The taste bud cells included cells with dark nuclei and those with light nuclei. A clear cut between the sensory. and supporting cells could not be made by light microscopic observations at this stage. y g p g The connective tissue of lamina propria mucosae was' rich in blood vessels. The aponeurosis linguae rudiment had the same appearance as in the previous period. Lingual mucosa at 57 days of foetal age (Figs 7 to 11, Plates IV, V and VI) The tongue surface differed markedly from that seen in the lower age category. The mucosa within the dorsum linguae appeared to consist of numerous big,. 'Separated papillae still undifferentiated in shape. Occasionally the separation was only suggested. The lateral surface of the tongue had a mildly undulating appe- arance. . The most conspicuous structures seen in cross-sections were the rapidly de- veloping circumvallate papillae. They had a typical shape and were bordered by deeply extending epithelial bands. These contained round hollow spaces of different volumes seen both near the surface and in the ba$al Iarts of the cell band. The spaces near the surface were elongated md separated with one or two long epithelial cells. The basal spaces were narrow and markedly smaller. Lingual mucosa at 50 days of foetal age (Figs 3, 4, Plate II) They presented as oval cell aggregations of different size. At the border of the lamina propria mucosae and the muscle tissue, a thick dense layer of the aponeurosis linguae was observed. It was not seen under the bases of the circumvallate papillae. Anlagen of secretory compartments of the gIl. gustatoriae were found in the layer of mucosal connective tissue and also deep !llllong fasciculi of the develop- ing tongue wasculature. They presented as oval cell aggregations of different size. g g y p gg g At the border of the lamina propria mucosae and the muscle tissue, a thick dense layer of the aponeurosis linguae was observed. It was not seen under the bases of the circumvallate papillae. Lingual mucosa at 64 days of foetal age (Figs 12 to 16, Plates VI, VII, VIII). The mucosal surface of the dorsum linguae bore numerous papillae much bigger than those seen in the previous period. Apart from circumvallate papillae, however, they were not differentiated in terms of shape. The lateral surface of the lingual mucosa showed elevations indicating the formation of foliate papillae. g g p p TJ:;te circumvallate papillae presented as large formations separated from the surrounding tissue by well defined furrows. Their dorsal surface was seen as a pattern of depressions and elevations. A cross-section through a depression showed that a thin cell band,. involving a lumen, extended from its bottom down- ward into the lingual musculature to be connected with secretory regions of the gustatory gland. These ducts opening onto the dorsal surface of the circumvallate papillae had lumina of varying width but never as wide as the true orifices of the gustatory gland. g y g The lingual epitheliuni on the dorsal surface of the circumvallate papillae had the appearance of a mature papilla with one or two surface layers of flat cells showing signs of keratinization. The epithelium lining the furrow around the circumvallate papilla grew thinner toward the furrow bottom. The epithelium on the rest of lingual mucosa was characterized by a distinct surface layer of poorly- -staining large polyedric cells, as at the previous stage. g g p y p g The anlagen of foliate papillae present on the lateral surfaces of the tongue were not separated distinctly from the surrounding tissue. When sectioned trans- versally, they presented as cylinders of connective tissue covered with a thick stratified epithelium composed of light polyedric cells. Lingual mucosa at 50 days of foetal age (Figs 3, 4, Plate II) The bases of the epithelial bands were divided into many developing ducts of the gustatQry gland which ran deep into the differentiating muscular tongue base. Their terminal portions separated rounded formations, later presenting as se- cretory components of the gland. In cross-sections the ducts at this stage showed lumina. At this foetal age, another form of the circumvallate papilla was regularly observed. These atypical papillae comp osed of several segments were surrounded with solid cell bands whose lower parts branched into ducts of the gustatory gland. The bigger segments were always found in the centre while their size decreased close to the periphery of th is composite anlagen of the circumvallate papilla. These composite anlagen were observed on the tongue of one and the same foetus together with the usual anlagen of circumvallate papillae described above. Taste buds were generally found on the circumvallate papillae showing the typical differentiation. They occurred in the epithelium of the dorsal surface and of the walls. They had a characteristic appearance and were never found reaching the surface of the epithelium. T hey included nuclei of two types: hyper- chromatic nuclei, smaller in size and oval in shape, and nuclei with diffuse chro- matin, larger in size and also val in shape. 103 The anlagen of the atypical papillae, particularly in the epithelium of the dorsal surfaces of the segments, also showed differentiating taste buds. Their appearance was reminiscent of that observed in the previous age category, where it was not possible to differentiate various cell types on the basis of their nucleus characteristics. Taste buds at the initial stage of differentiation were also recorded on the pri- mitive anlagen of foliate papillae situated on the lateral surface of the tongue. They presented as cell clusters on the top of the papillary anlage. They did not extend beyond the basal epithelial layer and from the connective tissue of the lamina propria mucosae were· separated with a fold of the basement membrane. p p p The cells of the basal epithelial layer which were in the close vicinity of the taste bud were elongated, with rod-shaped nuclei. Changes in shape found in cells attached to the taste bud were more marked with increasing bud differen- tiation. Anlagen of secretory compartments of the gIl. gustatoriae were found in the layer of mucosal connective tissue and also deep !llllong fasciculi of the develop- ing tongue wasculature. Discussion The picture of the lingual mucosa with its structures and components is well documented by both light and electron microscopic studies. Similarly, a lot of data have been accumulated on taste buds, giving, comprehensive information on their shape, size, ultrastructure, the mechanism of development and principles of taste perception. The cell types constituting the taste bud, and their ultrastruc- ture and function have also received attention. Considerably less information has been reported on the processes related to the· commencement of various mucosal structures of the tongue and their further development in the foetuses of domestic animals. Particularly incomplete are data concerning the sequence of events leading to the development, differentiation and distribution of taste buds and their relation to the development of lingual papillae. p p In the porcine foetus, the most significant period for morphological changes in the lingual mucosa seems to be from day 50 to day 64. This stage is marked by a rapid development of various components of the lingual mucosa; the anla- gen of papillae are formed and taste buds are iniciated. g p p In the surface lingual epithelium of the porcine foetus, the first signs of diffe- rentiation are apparent between day 41 and day 44. It is 10 days earlier than in ovine foetuses (Tichy and Cerny 1987). Epithelial cells proliferate at a high rate, which makes the uneven appearance of the dorsal surface at day 41 become smooth at day 44. At the same time growing epithelial bands penetrate into the mucosal connective tissue forming the first, not too distinct outlines of the future circumvallate papillae. At 44 days the basal epithelial layer of the lingual mucosa within the anlages of papillae contains cells with poorly-stained cytoplasm; these are not seen in the ovine foetus of the same age (Tichy and Cerny 1987). The cells are found single or in small clusters and are thought to be implemented in the initial stages of taste bud formation. Their adjoining cells change in morpho- logy; they became rod-shaped and so do their nuclei. The slender cells encompass the light cells; they will presumably give rise to one of the cell types constituting the taste bud. Some authors (Beidler and Smallmann 1965; Farbman 1965a) regard them as a stock of cellular material which serves to supply cells to the de- veloping taste bud. Lingual mucosa at 50 days of foetal age (Figs 3, 4, Plate II) The germinative part of the epithelium was limited to one or two layers of small but well-stained cells with oval hyperchromatic nuclei. yp Taste buds were observed in the epithelium on the dorsal surface and in the walls of circumvallate papillae. Taste buds in the process of differentiation were also present on the dorsal surface of the anlagen of foliate papillae. p g p p The taste buds observed in the epithelium of circumvallate papilla g p p aste buds observed in the epithelium of circumvallate papillae had a ty 104 pical shape and consisted of two well differentiated cell types. Cells with nuclei containing less chromatin were more numerous than those with hyperchromatic nuclei. Epithelial cells adjoining the taste bud were markedly increased in length and made the outer boundary of the bud. The apical end of the bud was overlaid with one or two layers of flat cells preventing communication with the surface. pical shape and consisted of two well differentiated cell types. Cells with nuclei containing less chromatin were more numerous than those with hyperchromatic nuclei. Epithelial cells adjoining the taste bud were markedly increased in length and made the outer boundary of the bud. The apical end of the bud was overlaid with one or two layers of flat cells preventing communication with the surface. y p g The taste buds on foliate papillae, still in the process of differentiation, appeared as cell clusters with the bases extending beyond the lower epithelial margin into the lamina propria mucosae, from which they were separated with basement membrane folds. At this stage of differentiation it was possible to distinguish cells with hyperchromatic nuclei from those with light nuclei. yp g The lamina propria mucosae and the aponeurosis linguae had similar structures to those found in the previous age group. Discussion Similar arrangements of epithelial cells adjoining the developing taste bud were, in more pronounced forms, seen at the other stages of prenatal 105 development, which implies that all cell types are continually supplied (Farb- mann 1971; Whiteside 1927). At 50 days the anlagen of taste buds are distinct- ly laid. They are formed by cluster of light cells which do not extend beyond the basal epithelial layer. In the light microscope they have a uniform appearance. At this stage it was not possible to distinguish cell types reported in electron- -microscopic studies (De Lorenzo 195~; Engstrom and Rytzner 1956; Fahrman and Schuchard 1967; Farbman 1965, 1971; Graziadei and De Han 1971; Murray and Murray 1960, 1967, 1970 and others). The buds at this stage are localized to the dorsal surface of the anlagen of circumvallate papillae. It is suggested that this localization is also typical for the early stages of development of the lingual mucosa in other domestic animals (Tichy and Cerny 1987). In the wall epithelium of the circumvallate papillae, the taste buds appear later, at 53 days of foetal age, but at greater numbers. As suggested by some authors, the taste bud differentiation is initiated by a contact of epithelial cells with a nerve fibre (Desgranges 1966; Farbman 1965a, b; Fujimoto and Murray 1970; Kurosumi and Kurosumi 1969; Munger 1965; Spoendlin 1970; Takeda and Hoshino 1975). The site of contact determines the location of the future bud (Takeda 1976). The presence of taste buds in the epithelium is, in our opinion, directly related to t1!e development and growth of the anlagen of circumvallate papillae (Tichy and Cerny 1987). From day 57 on, the circum- vallate papilla anlagen undergo marked differentiation, while the occurrence of taste buds on their dorsal surface is being reduced, so that, from day 64 on, they occur only occasionally. It is of interest that taste buds begin to from in the lingual epithelium at the time when circumvallate papillae start to differentiate in shape. In contrast to the findings published earlier (Tichy and Cerny 1987) in porcine foetuses the commencement of differentiation of both the papilla and the bud occur at the same time. Taste buds on the anlagen of foliate papillae differentiate later than those on circumvallate papillae, i. e. from day 57. They appear on the dorsal surface of the foliate papilla at the time of its commencement. Discussion If the view that the differen- tiation of taste buds is initiated by contact of epithelial cells with a nerve fibre (Farbman 1965; Murray and Murray 1967; Takeda 1976) is correct then the area of developing circumvallate papillae is supplied with nerves earlier than the lateral parts bearing the anlagen of foliate papillae. This is also evidenced by our finding that in each period the stage of differentiation of the epithelium was higher on the dorsum linguae than on the lateral surfaces. g g The differentiation of cell types inside the taste bud can be seen from day 53, when the occurrence of two cell types differing in the appearance of nuclei was first recorded. This seems to be the early period of development of sensory and supportive cells. The ultrastructure of this and other types of cells composing the taste bud was documented by many electron microscopic studies (D e L 0- renzo 1958; Farbman 1965a, b; Gray and Watkins 1965; Graziadei 1969; Murray 1971, 1973; Murray and Murray 1969; Takeda 1972; Takeda and Hoshino 1975; Trujillo-Cenoz 1957; Uga 1966, 1969). j g In the period between the 50th and 64th days of prenatal development, the epithelium of the foundations of circumvallate papillae shows taste buds at various stages of differentiation in terms of shape, size and structure. First they look like cell clusters located within the basal epithelial layer. In some instances they penetrate below the epithelial base into the layer of mucosal connective tissue, from which they are separated by a fold of the basement membrane. 106 On day 57 the developing taste buds reach as far as the surface layer of epithelium but on day 64 they still do not communicate with the epithelial surface. This implies that in this period the bud differentiates only in structure while maturation leading to the definite size and localization occurs at the following stages. On day 57 the developing taste buds reach as far as the surface layer of epithelium but on day 64 they still do not communicate with the epithelial surface. This implies that in this period the bud differentiates only in structure while maturation leading to the definite size and localization occurs at the following stages. g g g An unexpected finding was the detection of composite anlagen of circumvallate papillae on day 57. Their occurrence was regular along with that of typical circum- vallate papillae. Discussion At 64 days of prenatal development, however, the anlagen of circumvallate papillae presented as typical uniform structures. It can be speculated whether the composite anlagen will have any effect on the appearance of circum- vallate papillae in the adult animal whether they occur only as an insignificant anomaly in the course of development. We favour the concept of involvement of composite papillary anlagen in the final appearance of the circumvallate papillae because in the pig these papillae are know to have a variety of shapes and sizes. In the spaces between segments of the composite anlage of the circumvallate papilla, the gll. gustatoriae ducts are formed; they persist even after the segments fuse into a unified anlage of the papilla at 64 days. Our observations show that the ducts open at the bottom of depressions on the dorsal surface of the circum- vallate papilla. These cannot be considered the definite ducts because their lumina are noticcebly smaller and they, themselves, disappear ot the later stages. y y pp g The finding of hollow spaces among the cells of the epithelial band, later the furrow encircling the papilla, was also unusual. They are first seen on day 57 and are accounted for by changes in the compact band, which leads to the for- mation of the encircling furrow. From day 50 on, it was observed that the continuity of the aponeurosis linguae under the anlagen of circumvallate papillae was broken. The absence of the apo- neurosis linguae in these regions may be explained as enabling the developing ducts of the gustatory gland to grow downward. Since no similar report on the development and maturation of the lingual mucosa of the prenatal period reported in this paper has been found in the lite- rature, the observations described here can be considered an original contribution to this field investigation. Conclusions The composite anlagen of circumvallate papillae are first seen on day 57~ but by day 64 the circumvallate papillae is a uni- fied structure. S. The ducts of gustatory glands are commenced at 53 days and the first lumi- na can be observed on day 57. At that time the lamina propria mucosae and lingual musculature show the first signs of secretory regions of the glands. The ducts are formed between segments of the composite anlagen of circumvallate papillae at 57 days and persist even after the segments disappear in the unified bases of the papillae at 64 days. S. The ducts of gustatory glands are commenced at 53 days and the first lumi- na can be observed on day 57. At that time the lamina propria mucosae and lingual musculature show the first signs of secretory regions of the glands. The ducts are formed between segments of the composite anlagen of circumvallate papillae at 57 days and persist even after the segments disappear in the unified bases of the papillae at 64 days. p p y 9. The furrow encircling the circumvallate papillae is started when cells of the epithelial band begin to move apart at 53 days. On day 57 ~ spaces among cells appear and the cleft is completed at 64 days. p p y 9. The furrow encircling the circumvallate papillae is started when cells of the epithelial band begin to move apart at 53 days. On day 57 ~ spaces among cells appear and the cleft is completed at 64 days. pp p y 10. Foliate papillae begin to differentiate at 57 days. pp p y 10. Foliate papillae begin to differentiate at 57 days. 11. The rudiment of the aponeurosis linguae is indicated at 41 days and at 50 days it presents as a compact layer of connective tissue reduced under the bases of circumvallate papillae. 11. The rudiment of the aponeurosis linguae is indicated at 41 days and at 50 days it presents as a compact layer of connective tissue reduced under the bases of circumvallate papillae. Morfogeneze hrazene papUy a diferenciace chut'ovych pohark1i fetu prasete v obdo1>i 41.-64. dne Conclusions This paper describes the differentiation of certain structures of the lingual mucosa in the pig at 41 to 64 days of foetal age. Attention was focused on the morphogenesis of circumvallate papillae, the differentiation and localization of taste buds in the lingual epithelium and the development of some other related structures. The following conclusions have been drawn: . 1. Taste buds appear in the lingual epithelium first on the dorsal surface of developing circumvallate papillae at 50 days of prenatal development. However, the first signs of their differentiation can be recognized as early as on day 44. 1. Taste buds appear in the lingual epithelium first on the dorsal surface of developing circumvallate papillae at 50 days of prenatal development. However, the first signs of their differentiation can be recognized as early as on day 44. 2. Taste buds begin to form on the anlagen of foliate papillae on day 57. g g y 2. Taste buds begin to form on the anlagen of foliate papillae on 3 A id bl i i h f b d i h 3. A considerable increase in the amount of taste buds in the epithelium of circumvallate papillae is recorded from day 53. 4. Taste buds are at first limited to the dorsal epithelial surface of the circum- vallate papillae. At 53 days they appear in the wall epithelium involved in the basis of the encircling furrow. From day 64 on, the occurrence of taste buds on the dorsal surface of circumvallate papillae is a rare finding. 107 5. The differentiation of ceU types inside the taste bud can be recorded from the 53rd day of foetal development. 5. The differentiation of ceU types inside the taste bud can be recorded from the 53rd day of foetal development. y p 6. Taste buds do not reach up to the surface of epithelium during the period described in this paper. p p 7. The anlagen of circumvallate papillae are laid down at 44 days of foetal age and attain their appearance at 64 days. The composite anlagen of circumvallate papillae are first seen on day 57~ but by day 64 the circumvallate papillae is a uni- fied structure. p p 7. The anlagen of circumvallate papillae are laid down at 44 days of foetal age and attain their appearance at 64 days. Morfogeneze hrazene papUy a diferenciace chut'ovych pohark1i fetu prasete v obdo1>i 41.-64. dne V praci je venovana pozornost vzniku~ vYvoji a morfogenezi hrazenych papil a diferenciaci chuiovych poharkli v epitelu jazyka prasete od 41. do 64. dne intrauterinniho vjvoje. j j Odebrany byly vzorky jazyka praseCich fern ve stari 41~ 44~ 50~ 53~ 57 a 64 dnli po koncepci. Byl sledovan vznik~ vjvoj a vzhled povrchovjch struktur sliznice jazyka v jednotlivjch etapach vYvoje se zretelem na morfogenezi hrazenych papil a diferenciaci chuiovych pohark'li v jejich epitelu. Soucasne byla pozornost za- merena na vYvoj a utvareni gU. gustatoriae a jejich vYvodli~ vznik brazdy ohrani- cujfci hrazenou pa:pilu a diferenciaci dalsfch typli jazykovjch papil. j p p yp j y j p p Chuiove poharky se v epitelu zakladaji soucasne s tvarovou diferenciaci hrazene papily. Nejprve jsou lokalizovany v epitelu dorsalni plochy hrazene papily~ v epi- telu sten se objevuji teprve pozdeji. Postupne se zvysuje pocet diferencujicich se pohark'li v rozsahu jedne hrazene papily. K diferenciaci bunek chuioveho po- harku na jednotlive typy dochazi kratce po jeho vzniku. j ypy p j Hrazene papily se zakladaji bud jako jednotny utvar typickeho vzhledu~ nebo je zaklad mnohocetny~ a papila pak vznika postupnjrn splynutim jednotlivjch dilCich zakladli. Vjvody gU. gustatoriae luminizuji souCasne se vznikem obkru- zujici brazdy hrazene papily. j y p p y Ostatni typy jazykovych papil vznikaji pozdeji nez papila hrazena. 108 Mopcj:)oreHe3 JKenOOOB8Toro COCO'lK8 M AMCIXtlepeHQM8QMSI BKYCOBblX • COCOliKOB nnoAa CBMHbM B nepMoA 41- 64 CyTOK B pa60Te YAeneHo BHMMaHMe B03HMKHOBeHMIO, pa3BMTMIO M Mopcpore- He3Y )l(en060BaTblx COCOI.JKOB M AMcpcpepeHlIeallMM BKYCOBblX COCOI.JKOB B 3nMTenMM Sl3blKa nopocSlT, Hal.JMHaSi C 41 M KOHl.JaSi 64 AHeM BHYTPM- YTp06HOH )l(M3HM. p ) ( OT6MpanM np06bl Sl3blKa nnOAOB nopOCSIT B B03pacTe 41, 44, 50, 53, 57 M 64 CYTOK nocne 3al.JaTMSI. II1ccneAoBanM B03HMKHOBeHMe, pa3BMTMe M BHeWHMH BMA nOBepxHocTHblX CTPYKTYP cnM3McToH Sl3blKa Ha OTAenb- HblX 3Tanax pa3BMTMSI C Yl.JeTOM MopcporeHe3a )l(en060BaTblx COCOI.JKOB M AMcpcpepeHlIMallMM BKYCOBblX COCOI.JKOB B ero 3nMTenMM. BHMMaHMe 6bl- no oAHoBpeMeHHo HanpaBneHo Ha pa3BMTMe M CPOpMMpoBaHMe gIl. gusta- toriae M MX npOTOKoB, B03HMKHOBeHMIO 60P03Abl, OrpaHM'IMBalOU\eH )l(e- n060BaTblH COCOl.JeK, M AMcpcpepeHlIMallMM APyrMx TMnOB COCOI.JKOB Sl3b1Ka. BKycoBble COCOI.JKM B 3nMTenMM SlBnSlIOTCSI oAHoBpeMeHHo C AMcpcpepeH- lIMallMeH CPOPMbl OCHOBOH )l(en060BaToro COCOI.JKa. CnepBa OHM HaXOASlTCSI B 3nMTenMM AopcanbHOH nOBepXHOCTM )l(en060BaToro COCOI.JKa, B 3nMTenMM CTeHOK OHM nOSlBnSlIOTCSI HeCKonbKO n03)1(e. nOCTeneHHO YBenMI.JMBaeTCSI I.JMCneHHOCTb AMcpcpepeHlIMpYIOU\MXCSI COCOI.JKOB B npeAenax OHoro )l(eno- 60BaToro COCOI.JKa. ,lJ.McpcpepeHlIMallMSI KneTOK BKYCOBoro COCOI.JKa Ha OT- AenbHble TMnbl npOMCXOAMT BCKope nocne ero B03HMKHOBeHMSI. Morfogeneze hrazene papUy a diferenciace chut'ovych pohark1i fetu prasete v obdo1>i 41.-64. dne OT6MpanM np06bl Sl3blKa nnOAOB nopOCSIT B B03pacTe 41, 44, 50, 53, 57 M 64 CYTOK nocne 3al.JaTMSI. II1ccneAoBanM B03HMKHOBeHMe, pa3BMTMe M BHeWHMH BMA nOBepxHocTHblX CTPYKTYP cnM3McToH Sl3blKa Ha OTAenb- HblX 3Tanax pa3BMTMSI C Yl.JeTOM MopcporeHe3a )l(en060BaTblx COCOI.JKOB M AMcpcpepeHlIMallMM BKYCOBblX COCOI.JKOB B ero 3nMTenMM. BHMMaHMe 6bl- no oAHoBpeMeHHo HanpaBneHo Ha pa3BMTMe M CPOpMMpoBaHMe gIl. gusta- toriae M MX npOTOKoB, B03HMKHOBeHMIO 60P03Abl, OrpaHM'IMBalOU\eH )l(e- n060BaTblH COCOl.JeK, M AMcpcpepeHlIMallMM APyrMx TMnOB COCOI.JKOB Sl3b1Ka. , p p p y BKycoBble COCOI.JKM B 3nMTenMM SlBnSlIOTCSI oAHoBpeMeHHo C AMcpcpepeH- lIMallMeH CPOPMbl OCHOBOH )l(en060BaToro COCOI.JKa. CnepBa OHM HaXOASlTCSI B 3nMTenMM AopcanbHOH nOBepXHOCTM )l(en060BaToro COCOI.JKa, B 3nMTenMM CTeHOK OHM nOSlBnSlIOTCSI HeCKonbKO n03)1(e. nOCTeneHHO YBenMI.JMBaeTCSI I.JMCneHHOCTb AMcpcpepeHlIMpYIOU\MXCSI COCOI.JKOB B npeAenax OHoro )l(eno- 60BaToro COCOI.JKa. ,lJ.McpcpepeHlIMallMSI KneTOK BKYCOBoro COCOI.JKa Ha OT- AenbHble TMnbl npOMCXOAMT BCKope nocne ero B03HMKHOBeHMSI. OCHOBOH )l(en060BaTblX COCOI.JKOB SlBnSlIOTCSI eAMHble xapaKTepHoro BMAa CPOpMMpOBaHMSI MnM MHOrOI.JMCneHHble cpopMallMM; nanMnna BnocneACTBMM B03HMKaeT nOCTeneHHblM cnMSlHMeM OTAenbHblX eAMHMI.JHblX OCHOB. npo- TOKM gIl. gustatoriae BblXOASIT oAHoBpeMeHHo C B03HHKHOBeHHeM OKPY- )l(alOU\eH 60P03Abl )l(en060BaToro COCOI.JKa. OCTanbHble THnbl COCOI.JKOB Sl3blKa B03HHKalOT n03)1(e )l(eno60BaToro co- COI.JKa. References I.: Electron microscope study of the developing taste bud in rat fungiform pa- FARBMAN, A. I.: Fine structure ~f~e tas~e bud. J . Ultrastruct. Res., 12, 1965a: 328 350 FARBMAN, A. I.: Electron microscope study of the developing taste bud in rat fungiform pa- pilla. Develop. Biol. 11, 1965b: 110-135 . :" .. ' p p , FARBMAN, A. I.: Development of the taste bud. In: L. M. Beidler (Ed.) Handbook of sensory PhYllil>logy, Vol. IV" Chemical ,Sens~lb P~art 2, Springer, New York, ·1971, pp. 51-62 FARBMAN A I "'"YONKERS J D Fi f h b d i th d N p p FARBMAN, A. I.: Development of the taste bud. In: L. M. Beidler (Ed.) Handbook of sensory PhYllil>logy, Vol. IV" Chemical ,Sens~lb P~art 2, Springer, New York, ·1971, pp. 51-62 FARBMAN, A. I.:"'" YONKERS, J. D.: Fine structure of the taste bud in the mud puppy, Nectu- l~US maculosus. Amer.J. Anat.,l~l. '1971.:,.353-370 , . PhYllil>logy, Vol. IV Chemical ,Sens~lb P~art 2, Springer, New York, ·1971, pp. 51-62 FARBMAN, A. I.:"'" YONKERS, J. D.: Fine structure of the taste bud in the mud puppy, Nectu- l~US maculosus. Amer.J. Anat.,l~l. '1971.:,.353-370 , . gy p FARBMAN, A. I.:"'" YONKERS, J. D.: Fine structure of l~US maculosus. Amer.J. Anat.,l~l. '1971.:,.353-370 , , FUJIMOTO, S.-MURRAY, R. G.: Fine structure of degeneration and regeneration in dener- '~vated rabbitvallate taste buds . .l).pat. Rec.,l68, 1970; 393-:-414. . : . '.' ) p , , ; GRAY, E. G.-WATKINS, K. C.: Electron microscopy oftaste buds of the rat. Z. Zellforsch., 68, 1965: 583-595 P. P. C.: The ultrastructure of vertebrate taste buds. In: (ed. by) C. Pfaffmann: d taste, New York, Rockefeller University, 1969, pp. 315-330 , GRAZIADEI, P. P. C.: The ultrastructure of vertebrate taste buds. In: (ed. by) C Olfaction and taste, New York, Rockefeller University, 1969, pp. 315-330 , , y, , pp GRAZIADEI, P. P. C. - DE HAN, R. S.: The ultrastructure of frog's taste organs. Acta anat., SO, 1971: 563-603 , HEIDENHAIN, M.: tl'ber die Sinnesfelder und die Geschmacksknospen der papilla foliata des Kaninchens. Arch. mikr. Anat., 85, 1914: 365-479 HIRATA, K.-NADA, 0.: A monoamine in the gustatory cell of the frog's taste organ. Cell Tiss. Res., 159 1975: 101-108 , KOLMER, W.: bber Strukturen im Epithel des Sinnesorgane. Anat. References Anz., 36, 1910: 281-299 KUROSUMI, K.-KUROSUMI, U.-SUZUKI, H.: Fine structures of Merkel cells and asso- ciated nerve fibers in the epidermis of certain mammalian species. Arch. histol. jap., 30, 1969: 295-313 D.: Beitriige zur Kenntniss von Bau der Geschmackswiirzchen der Zunge. Arch. mikr 1868: 96-154 . MUNGER, B. L.: The intraepidermal innervation of the snout skin of the opossum. A light and electron microscope study, with observations on the nature of Merkel's Tastzellen. J. Cell BioI., 26, 1965: 79-97 MURRAY, R. G.: Cell tapes in rabbit taste buds. In: Qlfaction and taste III, Pfaffm York, Rockefeller Univ. Press, 1969, pp. 331-344 pp MURRAY, R. G.: Ultrastructure of taste receptors.' In: Handbook of sensory physiology, vol. IV Chemical senses, part 2 Taste, Beidler, ed. Berlin-Heidelberg-New York: Springer, 1971, pp.31-50 pp MURRAY, R. G.: The structure of taste buds. In: Ultrastucture of sensory organ, v man, ed. Amsterdam: North Holland, 1973, pp. 1-81 pp MURRAY, R. G.-MURRAY, A.: The fine structure of the taste buds of Rhesus and Cynomol- gus monkeys. Ant. Rec, 138, 1960: 211-233 g y MURRAY, R. G.-MURRAY, A.: Fine structure of taste buds of rabbit foliate papillae. J. Ul- trastruct. Res., 19, 1967: 327-353 MURRAY, R. G.-MURRAY, A.: The anatomy and ultrastructure oftaste endings. In: (ed. by) G. E. W. Wolstenholme and J. Knight: Taste and smell in vertebrates. London, J. and A. Chur- chil, 1970, pp. 3-30 , , pp MURRAY, R. G.-MURRAY, A.-FUJIMOTO, S.: Fine structure of gustatory cells in rabbit taste buds. J. Ultrastruct. Res., 27, 1969: 444-461 RETZIUS, G.: Zur Kenntniss des Geschmackorgans beim Kaninchen. BioI. Untersuch., n. f. 17, 1912: 72-113 . ROOSS, M. H. - REITH, E. J.: Histology a Text and Atlas. Harper & Row, Publishers, J. B. Lip- pincott company, New York, Cambridge, Philadelphia, San Francisco, London, 1985, pp. 400-403 SCHULZE, F. E.: tl'ber die becherf6rmigen Organe der Fische. Z. wissensch. Zool., 12, 1863: 218-292 . SCHWALBE, G.: Ober die Geschmacksorgane der Siiugetiere und des Menschen. Arch. mikr. Ant., 4., 1868: 154-210 SMOLICH, A.-MICHEL, G.: Mikroskopische Anatomie der Haustiere. VEB Gustav Fischer Verlag :ena, 1985, pp. 91-94, pp. 363-364 g , , pp , pp SPOENDLIN, H.: Taste buds. In: (ed. by) A. Bischoff: Ultrastructure of the periph system and sense organs. Stuttgart, Georg Thieme Verlag, 1970, pp. 326-337 SPOENDLIN, H.: Taste buds. In: (ed. by) A. Bischoff: Ultrastructure of the peripheral nervous system and sense organs. References BANKS, W. J.: Applied veterinary histology. Williams and Wilkins, Baltimore, London, 1981, pp.374-376 J pp BEACHAMP, G. K.-COWART, B. J.: Development of Sweet Taste Sweetness. Ilsi human nutrition Reviews, Symp. on Sweetness, Geneva, Switzerland, May 21-23,1986: 127-130 BEIDLER, J. M.-SMALLMAN, R. L.: Reneva1 of cells within taste buds. J. Cell BioL, 27, 1965 263 2 2 pp BEACHAMP, G. K.-COWART, B. J.: Development of Sweet Taste Sweetness. Ilsi human nutrition Reviews, Symp. on Sweetness, Geneva, Switzerland, May 21-23,1986: 127-130 J J BEACHAMP, G. K.-COWART, B. J.: Development of Sweet Taste Sweetness. Ilsi human nutrition Reviews, Symp. on Sweetness, Geneva, Switzerland, May 21-23,1986: 127-130 BEIDLER, J. M.-SMALLMAN, R. L.: Reneva1 of cells within taste buds. J. Cell BioL, 27, 1965: 263-272 , y p , , , y , BEIDLER, J. M.-SMALLMAN, R. L.: Reneva1 of cells within taste buds. J. Cell BioL, 27, 1965: 263-272 J BEIDLER, J. M.: Physiological properties of mammalian taste receptors. In: Taste and smell in vertebrates. Wolstenholme and Knight, eds. Ciba Found. Sywp. London, J. & A. Churchill, 1970, pp. 51-67 , pp CORMACK, D. H.: Introduction to Histology. J. B. Lippincott Company Philadelphia, London, New York, 1984, pp. 287-289. , , pp DE HAN, R. - GRAZIADEI, P.P. C.: Functional anatomy of frog's taste organs. Experientia, 27, 1971: 823-826. , DE HAN, R.-GRAZIADEI, P.P.C.: The innervation of frog's taste organ. "A histochemical study." Life Sci., 13, 1973: 1435-1449 J J y DE LORENZO, A. J.: Electron microscopic observation on the taste buds of the ra physic. Biochemic. CytoL, 4, 1958: 143-150 p y y , , DESGRANGES, J. C.: Sur 1a double innervation des cellules sensorielles des bourgeons du gout. des barbillons du Poisson-chat. C. r. Acad. Sci. (Paris), 263, 1966: 1103-1106 ( ), , ENGSTROM, H. - RYTZNER, C.: The fine structure of taste buds and taste fibers. Ann. OtoL (St. Louis), 65, 1956: 361-375 109 EVANS, H. E. -;- SNACK, W. 0.: Prenatal development of domestic laboratory. mammals. Growth curves, external features and selected references Zbl. Vet. Med. C 2, 1973: 11-45 , , FAHRMANN.> W.-SCHVCHARDT,E.;Licht- und electronenmikroskopische Befunde an den Geschmacksknospen der Axolodzunge. Experientia, 23, 1967: 657 - 659 ..' p g p , , RBMAN, A. I.: Fine structure ~f~e tas~e bud. J •. Ultrastruct.·Res., 12, 1965a: 328-3 FARBMAN, A. I.: Fine structure ~f~e tas~e bud. J •. Ultrastruct.·Res., 12, 1965a: 328-350 FARBMAN, A. References Stuttgart, Georg Thieme Verlag, 1970, pp. 326-337 TAKEDA M Fi 'It t f d l i t t b d i h f t l i ll t pap~e SPOENDLIN, H.: Taste buds. In: (ed. by) A. Bischoff: Ultrastructure of the peripheral nervous system and sense organs. Stuttgart, Georg Thieme Verlag, 1970, pp. 326-337 TAKEDA, M.: Fine 'Itructure of developing taste buds in human fetal circumvallate pap~e (in Japanese with &glish summary). Acta anat. nippon., 47,1972: 325-337 system and sense organs. Stuttgart, Georg Thieme Verlag, 1970, pp. 326-337 TAKEDA, M.: Fine 'Itructure of developing taste buds in human fetal circumvallate pap~e AKEDA, M.: Fine 'Itructure of developing taste buds in human fetal circumvallat (in Japanese with &glish summary). Acta anat. nippon., 47,1972: 325-337 110 TAKEDA, M.: An electron microsoopis study on the innervation in the taste buds of the mouse circumvallate papillae. Arch. histol. jap., 39,1976: 257-269 p p j p TAKEDA, M.-HOSHINO, T.: Fine structure of taste buds in the rat. Arch. histol. jap., 37, 1975: 395-413 F.-C~, TICHY, F.-C~, H.: The morphogenesis of circumvallate papillae and differentiation of taste buds in shee'p ontogeny. Acta vet. Brno, 56, 1987: 261-274 taste buds in shee p ontogeny. Acta vet. Brno, 56, 1987: 261 274 TRUJILLO-CENOZ, 0.: Electron microscope study of the rabbit gustatory bud. Z. Zellforsch., 48, 1957: 272-280 p g y , TRUJILLO-CENOZ, 0.: Electron microscope study of the rabbit gustatory bud. Z. Zellforsch., 48, 1957: 272-280 UGA, S.: The fine structure of gustatory receptors and their synapses in frog's tongue. Symp. Cell Chem., 18, 1966; 75-86 UGA, S.: The fine structure of gustatory receptors and their synapses in frog's tongue. Symp. Cell Chem., 18, 1966; 75-86 jap., 31, 1969; 59 72 WHITESIDE, B.: The regeneration of the gustatory apparatus in the rat. J. compo Neurol., 40, 1927: 33-45 j p , , ; WHITESIDE, B.: The regeneration of the gustatory apparatus in the rat. J. compo Neurol., 40, 1927: 33-45 Plate I. Plate I. Tichy F.: The Morphogenesis of Circumvallate Papillae ... pp. 99-100. Fig. 1: A part of the lingual surface of a porcine foetus at 41 days. Simple elevations on the dor- sum linguae (1). The surface layer of poorly-stained cells of the epithelium (2). Thickening con- nective tissue as a rudiment of the aponeurosis linguae (3). Haematoxylin-eosin (HE); magnifi- cation: x 250. Tichy F.: The Morphogenesis of Circumvallate Papillae ... pp. 99-100. Fig. 1: A part of the lingual surface of a porcine foetus at 41 days. Simple elevations on the dor- sum linguae (1). The surface layer of poorly-stained cells of the epithelium (2). Thickening con- nective tissue as a rudiment of the aponeurosis linguae (3). Haematoxylin-eosin (HE); magnifi- cation: x 250. Fig. 2: A part of the lingual sun ace of a porcine foet1ls at 44 days. A developing cell band (1) as the basis of the fUrtow encirclm.g the clrcumvalliite papilla. A light cell in the germinative layer of the epithelium (2), bordered by well-stained cells with elongated nuclei. A capillary in the layeI of mucosal connective tissue (3). The rudiment of the aponeurosis linguae (4). HE, magnifica- tion: x 250. - Fig. 2: A part of the lingual sun ace of a porcine foet1ls at 44 days. A developing cell band (1) as the basis of the fUrtow encirclm.g the clrcumvalliite papilla. A light cell in the germinative layer of the epithelium (2), bordered by well-stained cells with elongated nuclei. A capillary in the layeI of mucosal connective tissue (3). The rudiment of the aponeurosis linguae (4). HE, magnifica- tion: x 250. - Fig. 2: A part of the lingual sun ace of a porcine foet1ls at 44 days. A developing cell band (1) as the basis of the fUrtow encirclm.g the clrcumvalliite papilla. A light cell in the germinative layer of the epithelium (2), bordered by well-stained cells with elongated nuclei. A capillary in the layeI of mucosal connective tissue (3). The rudiment of the aponeurosis linguae (4). HE, magnifica- tion: x 250. - Plate II. Fig. 3: A forming circumvallate papilla on the tongue of a porcine foetus at 50 days (2). A taste bud at the early stage of differentiation on the dorsal surface of the papillary anlage (1). The an- lagen of ducts of the gIl. gustatoriae (4). The developing aponeurosis linguae (3). HE, magnifica- tion: X 100. Plate II. Fig. Plate I. 3: A forming circumvallate papilla on the tongue of a porcine foetus at 50 days (2). A taste bud at the early stage of differentiation on the dorsal surface of the papillary anlage (1). The an- lagen of ducts of the gIl. gustatoriae (4). The developing aponeurosis linguae (3). HE, magnifica- tion: X 100. Fig. 3: A forming circumvallate papilla on the tongue of a porcine foetus at 50 days (2). A taste bud at the early stage of differentiation on the dorsal surface of the papillary anlage (1). The an- lagen of ducts of the gIl. gustatoriae (4). The developing aponeurosis linguae (3). HE, magnifica- tion: X 100. Fig. 4: A detail of a differentiating taste bud (1) in the epithelium of the dorsal surface of the anlage of a circumvallate papilla in a 50-day-old porcine foetus. HE, magnification: x 400. Fig. 4: A detail of a differentiating taste bud (1) in the epithelium of the dorsal surface of the anlage of a circumvallate papilla in a 50-day-old porcine foetus. HE, magnification: x 400. Plate III. Plate III. Fig. 5: The anlage of a circumvallate papilla on the tongue of a porcine foetus at 53 days. Taste buds forming on both the dorsal and lateral surfaces of the papilla (1). The rudiment of the apo- neurosis linguae in the layer of mucosal connective tissue (2). Differentiating ducts of the gIl. gustatoriae (3) HE, magnification: x 250. Fig. 5: The anlage of a circumvallate papilla on the tongue of a porcine foetus at 53 days. Taste buds forming on both the dorsal and lateral surfaces of the papilla (1). The rudiment of the apo- neurosis linguae in the layer of mucosal connective tissue (2). Differentiating ducts of the gIl. gustatoriae (3) HE, magnification: x 250. Fig. 5: The anlage of a circumvallate papilla on the tongue of a porcine foetus at 53 days. Taste buds forming on both the dorsal and lateral surfaces of the papilla (1). The rudiment of the apo- neurosis linguae in the layer of mucosal connective tissue (2). Differentiating ducts of the gIl. gustatoriae (3) HE, magnification: x 250. Fig. 6: A part of the anlage of a circumvallate papilla with a developing taste (1) bud in the lingual epithelium of a 53-day-old porcine foetus. An indication of the cleft in the compact cell band (2) in the process of encircling furrow formation. Plate I. HE, magnification: x 400. Fig. 6: A part of the anlage of a circumvallate papilla with a developing taste (1) bud in the lingual epithelium of a 53-day-old porcine foetus. An indication of the cleft in the compact cell band (2) in the process of encircling furrow formation. HE, magnification: x 400. Fig. 6: A part of the anlage of a circumvallate papilla with a developing taste (1) bud in the lingual epithelium of a 53-day-old porcine foetus. An indication of the cleft in the compact cell band (2) in the process of encircling furrow formation. HE, magnification: x 400. Plate IV. Fig.-i: The lingual surface of a porcine foetus at 57 days. A differentiating circumvallate papilla with spaces of different sizes in the epithelium (1), surrounded with undifferentiated anlagen of other lingual papillae (2). Cross-sectioned lumina of gustatory gland ducts (3) and the indication of secretory regions (4). The aponeurosis linguae (5) is missing under the base of the circumvallate papilla. HE, magnification: x 100. Plate IV. Fig.-i: The lingual surface of a porcine foetus at 57 days. A differentiating circumvallate papilla with spaces of different sizes in the epithelium (1), surrounded with undifferentiated anlagen of other lingual papillae (2). Cross-sectioned lumina of gustatory gland ducts (3) and the indication of secretory regions (4). The aponeurosis linguae (5) is missing under the base of the circumvallate papilla. HE, magnification: x 100. Fig.-i: The lingual surface of a porcine foetus at 57 days. A differentiating circumvallate papilla with spaces of different sizes in the epithelium (1), surrounded with undifferentiated anlagen of other lingual papillae (2). Cross-sectioned lumina of gustatory gland ducts (3) and the indication of secretory regions (4). The aponeurosis linguae (5) is missing under the base of the circumvallate papilla. HE, magnification: x 100. Fig. 8: A developing circumvallate papilla with the spaces in. the epithelial basis of the encircling furrow (1) and differentiating taste buds in the wall epithelium (2) in a 57-day-old porcine foetus. In the connective tissue and lingual musculature, regions of the developing gustatory gland (3) can be seen. The aponeurosis linguae (4). HE, magnification: x 100. Fig. 8: A developing circumvallate papilla with the spaces in. the epithelial basis of the encircling furrow (1) and differentiating taste buds in the wall epithelium (2) in a 57-day-old porcine foetus. Plate I. In the connective tissue and lingual musculature, regions of the developing gustatory gland (3) can be seen. The aponeurosis linguae (4). HE, magnification: x 100. Fig. 8: A developing circumvallate papilla with the spaces in. the epithelial basis of the encircling furrow (1) and differentiating taste buds in the wall epithelium (2) in a 57-day-old porcine foetus. In the connective tissue and lingual musculature, regions of the developing gustatory gland (3) can be seen. The aponeurosis linguae (4). HE, magnification: x 100. Plate V. Fig. 9: A detail of Fig. 8. Spaces in the epithelial basis of the encircling furrow (1). A differentia- ting taste bud (2). Two types of nuclei in cells of the bud. Elongated nuclei of pericanalicular cells (-+). HE, magnification: x 400. Plate V. Fig. 9: A detail of Fig. 8. Spaces in the epithelial basis of the encircling furrow (1). A differentia- ting taste bud (2). Two types of nuclei in cells of the bud. Elongated nuclei of pericanalicular cells (-+). HE, magnification: x 400. Fig. 10: The composite anlage of a circumvallate papilla in a porcine foetus at 57 days. Sections through the lumina of gustatory gland ducts (1) in the connective tissue. HE, magnification: x 100. Fig. 10: The composite anlage of a circumvallate papilla in a porcine foetus at 57 days. Sections through the lumina of gustatory gland ducts (1) in the connective tissue. HE, magnification: x 100. Plate VI. 1 Fig. 11: The surface of the lateral part of the tongue in a 57-day-old porcine foetus. The anlage of a foliate papilla (2) with a developing taste bud (1). HE, magnification: X 250. Plate VI. Fig. 11: The surface of the lateral part of the tongue in a 57-day-old porcine foetus. The anlage of a foliate papilla (2) with a developing taste bud (1). HE, magnification: X 250. Fig. 12: A circumvallate papilla in a porcine foetus at 64 days. The dorsal surface shows occa- sional taste buds (1) and depressions (2) varying in depth. Ducts of the gustatory gland (3) open into the depressions. HE, magnification: X 100. Fig. 12: A circumvallate papilla in a porcine foetus at 64 days. The dorsal surface shows occa- sional taste buds (1) and depressions (2) varying in depth. Ducts of the gustatory gland (3) open into the depressions. HE, magnification: X 100. Plate VII. Fig. Plate I. 13: A part of the circumvallate papilla of the 64-day-old porcine foetus. The epithelial surface is covered with a layer of keratinized cells (1). The epithelial bands extend into the connec- tive stroma of the papilla (2) as remnants of the composite anlage. An inconspicuous dense connec- tive tissue layer under the base of the papilla (3). HE. magnification: x 250. Plate VII. Fig. 13: A part of the circumvallate papilla of the 64-day-old porcine foetus. The epithelial surface is covered with a layer of keratinized cells (1). The epithelial bands extend into the connec- tive stroma of the papilla (2) as remnants of the composite anlage. An inconspicuous dense connec- tive tissue layer under the base of the papilla (3). HE. magnification: x 250. Fig. 13: A part of the circumvallate papilla of the 64-day-old porcine foetus. The epithelial surface is covered with a layer of keratinized cells (1). The epithelial bands extend into the connec- tive stroma of the papilla (2) as remnants of the composite anlage. An inconspicuous dense connec- tive tissue layer under the base of the papilla (3). HE. magnification: x 250. Fig. 14: An area of the dorsal surface of the circumvallate papilla at 64 days. A taste bud (1). A layer of keratinized cells on the epithelial surface (2). The connective stroma contains noduli of lymphoreticular tissue (3). HE. magnification: x 400. Fig. 14: An area of the dorsal surface of the circumvallate papilla at 64 days. A taste bud (1). A layer of keratinized cells on the epithelial surface (2). The connective stroma contains noduli of lymphoreticular tissue (3). HE. magnification: x 400. Fig. 14: An area of the dorsal surface of the circumvallate papilla at 64 days. A taste bud (1). A layer of keratinized cells on the epithelial surface (2). The connective stroma contains noduli of lymphoreticular tissue (3). HE. magnification: x 400. Plate VIII. Plate VIII. Fig. 15: A part of the circumvallate papilla at 64 days. The wall epithelium shows numerous taste buds (1). Ducts of the gustatory glands (2) open onto the bottom of the encircling furrow. HE, magnification: x 400. Fig. 15: A part of the circumvallate papilla at 64 days. The wall epithelium shows numerous taste buds (1). Ducts of the gustatory glands (2) open onto the bottom of the encircling furrow. HE, magnification: x 400. Fig. Plate I. 16: The surface of the lateral part of the tongue at 64 days. A developing foliate papilla with a taste bud (1) in the dorsal epithelium. Fasciculi in the process of organization (2). HE, 'magnification: x 250. Fig. 16: The surface of the lateral part of the tongue at 64 days. A developing foliate papilla with a taste bud (1) in the dorsal epithelium. Fasciculi in the process of organization (2). HE, 'magnification: x 250.

https://openalex.org/W3186579135

https://www.nature.com/articles/s41598-021-94107-6.pdf

English

Antagonist effects of grain boundaries between the trapping process and the fast diffusion path in nickel bicrystals

Scientific reports

cc-by

Antagonist effects of grain boundaries between the trapping process and the fast diffusion path in nickel bicrystals OPEN J. Li, A. Hallil, A. Metsue, A. Oudriss, J. Bouhattate & X. Feaugas* Hydrogen-grain-boundaries interactions and their role in intergranular fracture are well accepted as one of the key features in understanding hydrogen embrittlement in a large variety of common engineer situations. These interactions implicate some fundamental processes classified as segregation, trapping and diffusion of the solute which can be studied as a function of grain boundary configuration. In the present study, we carried out an extensive analysis of four grain-boundaries based on the complementary of atomistic calculations and experimental data. We demonstrate that elastic deformation has an important contribution on the segregation energy which cannot be simply reduced to a volume change and need to consider the deviatoric part of strain. Additionally, some significant configurations of the segregation energy depend on the long-range elastic distortion and allows to rationalize the elastic contribution in three terms. By investigating the different energy barriers involved to reach all the segregation sites, the antagonist impact of grain boundaries on hydrogen diffusion and trapping process was elucidated. The segregation energy and migration energy are two fundamental parameters in order to classify the grain-boundaries as a trapping location or short circuit for diffusion. Hydrogen has a strong tendency to segregate or interact with structural defects as point defects, solid solution species, precipitates, dislocations and internal interfaces (inter-phases, grain-boundaries, …). This situation affects the apparent solubility and the mobility of hydrogen and consequently modifies the embrittlement process of metals and their ­alloys1. Many ­studies2–6 support the role of hydrogen state in the control of the properties of hydrogenated materials in a large diversity of cases (hydrogen diffusion, hydrogen induced cracking, electrical properties in semiconductors, catalyze …) and more precisely their mechanical behavior (embrittlement itself). In this context, hydrogen seems to have a larger implication on intergranular ­fracture7–13. While segregation and diffusion of hydrogen at grain boundaries (GBs) have been of great interest for understanding hydrogen embrit- tlement (HE), the difficulty of hydrogen analysis at grain boundary scale is the limited number of studies carried out in this subject for nickel alloys (see for a review on the subject ­references14, 15). The interactions of hydrogen with grains and grain boundaries are often treated numerically and experimentally so as to separate with a little ­confrontation16–18. www.nature.com/scientificreports www.nature.com/scientificreports LaSIE UMR CNRS 7356, La Rochelle Université, Av. Michel Crépeau, 17042 La Rochelle Cedex 1, France. *email: [email protected] www.nature.com/scientificreports/ These values can be significantly modified as a function of the conditions applied to atomistic calculations (size of the box, DFT or EAM potential…) but globally the minimum segregation energy increases with the increases of the CSL index. Hallil et al. suggest that for Σ3 GBs, that the GB character (energy, and excess volume) can be treated by the notion of the inclination angle ϕ between the two symmetrical tilt grain boundaries (STGB): coher- ent twin boundary (CTB) and symmetrical incoherent twin boundary (SITB) ­configurations46, 60. Energy and excess volume expands with ϕ and at the same time the minimum segregation energy of hydrogen ­grows46. Based on MD/MC simulations, larger systems can be investigated. On the other hand, Moody et al.47 have pointed out that the hydrogen concentration is enhanced in tilt Σ9(221) high energy grain boundary in nickel. More recently, Brien and ­Foilles43 have studied the hydrogen segregation in inclined Σ3 〈110〉 nickel GBs using the hybrid MC/MD and an analytic segregation model. The maximum concentration of hydrogen occurs at the boundary at the inclination with the highest enthalpy. This result also gives a correlation between the hydrogen segregation and the GBs energy since the GB energy amplifies with the inclination angle for the nickel GB Σ3 〈110〉. The hydrogen segregation phenomenon is more pronounced for high energy GBs which may be explained by the high excess volume for these GBs. All these outcomes suggest a correlation with the geometric and energetic configuration of grain boundaries and segregation properties, but the physical bases of this relationship stay ambiguous. More recently, the local state was considered in some nickel grain-boundaries46, 56. Some correlation seems to be possible between the local deformation of hydrogen segregation volume defined by polyhedrons using the Voronoi tessellation method. These studies suggest that elastic dilatation and distortion deformation of the site is partially responsible for the segregation energy. Based on these considerations, some authors have used a continuum approach to evaluate the impact of the elastic field associated with GBs on segregation ­processes61. The respective contribution of short and long-range stress continues to be an open question. h p g g p q The roles of grain boundaries (GBs) in hydrogen diffusion processes were determined from density functional theory calculations by some authors in fcc ­metals51, 62, 63. www.nature.com/scientificreports/ Additionally, SIMS ­mapping29, 34–36 and recent Atom Probe Tomography ­observations37 highlight a gradient of hydrogen content with a path length higher than the GBs thickness which suggests that hydrogen diffusivity and segregation processes cannot be only discussed in relation to the local structure of grain boundaries.f g Based on atomistic simulations, several computational efforts have focused on the hydrogen segregation and diffusion properties and embrittlement consequences for some selected grain boundary in ­nickel19, 38–57. Classi- cally, the grain boundaries are characterized by their energy, excess volume and geometric parameters such as Coincidence Site Lattice; easily accessible data using density functional theory (DFT) or molecular dynamics (MD) ­simulations58, 59. These characteristics are determining factors on the interaction properties of solutes with the GB; however, such global values sometimes appear to be far from representative of local behaviour. According to extensive atomistic simulations, the segregation energy is essential to the understanding of dynamic processes of solute evolution in materials. The minimum segregation energy, commonly used to characterize GB capac- ity to interact with hydrogen, vary significantly from −0 04 to −0 37 eV depending on the GB character19,45–47, Based on atomistic simulations, several computational efforts have focused on the hydrogen segregation and diffusion properties and embrittlement consequences for some selected grain boundary in ­nickel19, 38–57. Classi- cally, the grain boundaries are characterized by their energy, excess volume and geometric parameters such as Coincidence Site Lattice; easily accessible data using density functional theory (DFT) or molecular dynamics (MD) ­simulations58, 59. These characteristics are determining factors on the interaction properties of solutes with the GB; however, such global values sometimes appear to be far from representative of local behaviour. According to extensive atomistic simulations, the segregation energy is essential to the understanding of dynamic processes of solute evolution in materials. The minimum segregation energy, commonly used to characterize GB capac- ity to interact with hydrogen, vary significantly from − 0.04 to − 0.37 eV depending on the GB ­character19,45–47, 53. The effects of last one can be evaluated on the base of literature data for Σ3(111) (− 0.04 eV)45, 46, Σ3(110) (− 0.21 eV)46, Σ3(221) (− 0.21 eV)46, Σ3(112) (− 0.24 eV)46, Σ5(012) (− 0.23 to − 0.37 eV)19, 45, 48–51, Σ5(001) (− 0.16 eV)52, Σ5(310) (− 0.32 eV)54, Σ9(221) (− 0.2 eV)46, Σ99(557) (− 0.15 eV)46and Σ17(140) (− 0.34 eV)54. www.nature.com/scientificreports/ to an eased path associated with lower energy barrier. The predominance of one phenomenon over the other depends on the grain boundary energy and the excess of free ­volume28. These results gathered from a correlation between large data of diffusion coefficient and grain boundary character seems to highlight some ­exceptions29 which suggest a considerable diversity of local processes. These mechanisms were improved and discussed an extensive variety of experimental technics with a high spatial resolution which has been reviewed ­recently14, 15. In fcc metals and alloys, the existence of short-circuit diffusion paths of hydrogen was illustrated using the hydrogen microprint technics and Secondary Ion Mass Spectrometry (SIMS) ­mapping14, 15, 29–32. The preferential ingress of hydrogen along grain boundaries was observed by Tanaka et al.32 using Ga-FIB-TOF-SIMS to directly visualize deuterium distribution in fcc steel. Microprint technic shows that not all grain boundaries are gener- ally decorated with Ag crystals, which suggest that hydrogen transport capacity of a boundary depends on its microstructural specificity (character, orientation, …)30, 31. More recently, Tof-SIMS and EBSD were combined to investigate statistically hydrogen distribution around grain boundaries in polycrystalline ­nickel29. Our results sug- gest that grain boundaries can be categorized into two families according to how hydrogen is distributed across the grain boundary. The first family designates random grain boundaries which reveal a sharp gap for hydrogen concentration profile across the grain boundaries. The second one is special Σ3n grain boundaries which presents a smooth gradient of hydrogen concentration cross the grain boundary. Despite these new relevant results, it is clear to conclude that actually it stays difficult to demonstrate that hydrogen distribution results in a heterogene- ous behavior of diffusion and segregation processes or both. Recent in situ SKPFM analyses using for detecting the local hydrogen distribution around GBs, demonstrate that a random GB with a misorientation of 43° does not significantly facilitate hydrogen diffusion, while a coherent Σ3 twin GB provides a fast path for hydrogen ­transport33. This last result seems in opposite with Oudriss ­works28, 29 and questions the simple view based on random and Coincidence Site Lattice character (CSL—Σ). Antagonist effects of grain boundaries between the trapping process and the fast diffusion path in nickel bicrystals OPEN Also, at the atomistic scale, there are many works that are rapidly interested in inter-granular ­decohesion19–23 in relation to segregation without a precise focus on the elementary process which occurs at the grain boundary. In contrast, the short circuit of diffusion within the grain boundaries remains a remarkable complex subject that would require more in-depth analysis.hi p j q p y There is little available experimental data of the hydrogen segregation for properly defined GBs. Indeed, adequately characterizing a GB persists to be an engineering challenge, especially the crystallographic orienta- tion and the misorientation angle need to be suitably controlled. For two decades, a large debate was supported by many studies on the possible conflict between the fact that grain-boundaries can be a either a trapping site and/or a short circuit of ­diffusion24–28 in fcc materials. More recently, based on a larger experimental investiga- tion Oudriss et al.27–29 have reported that the GBs with low misorientation (Σ1) and a category of “special” grain boundaries (Σ3–Σ29) are usually preferential areas for hydrogen trapping in polycrystalline nickel. In fact, considering their ordered structure, this kind of boundary is accommodated by defects (dislocations, vacancies and more complex organization) that represent potential traps of hydrogen. In opposite, the high angle “random” grain boundaries are considered as the “disordered phase” where the hydrogen diffusion is accelerated in relation | https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 www.nature.com/scientificreports/ Results S Some remarkable results from experimental works. Recently, an extensive collection of experimen- tal data was gathered for the polycrystalline nickel to characterize diffusion and trapping processes which occur along grain ­boundaries9, 27–29. This database is first revisited in the present work to introduce some open ques- tions. The effective diffusion coefficient ­Deff was assessed based on classical Fick’s law to describe hydrogen flux across a polycrystalline membrane obtained during electrochemical permeation test. Using a large range of grain sizes it was possible to modulate the fraction of random grain boundary (fR) and special grain bound- ary (fCSL)27, 28. A linear relationship is clearly established between Ln(Deff) and fR the fraction of random GBs, (Fig. 1a) which illustrates the fact that random GBs can act as a short circuit of diffusion. In opposite, it was suggested that CSL GBs are specific locations for deep trapping based on the fact that the number of trapping sites ­NT increases as function with the fraction of CSL grain boundaries (fCSL) (Fig. 11b in Oudriss et al. ­work28). Despite this demonstration of the global impact of both nature of GBs (Special and Random) on diffusion and trapping processes, some questions seem to be open. The separation of Special and Random class of GBs based only on the coincidence lower than Σ29 seems unreasonable considering the energy of grain boundaries (see Fig. 2a as an example). Additionally, the hydrogen concentration ­CH increases with the fraction of random fR which suggests that random GB is also a specific location of trapping. This effect was clearly identified as a consequence of a vacancy cluster formation (SAV) process. A linear relationship with vacancy concentration and hydrogen was identified ­(Cvac = 0.15 × ­CH 28 which illustrates that the fact that the increase of hydrogen concentration is directly a consequence of vacancy formations without clearly establishing that the formation is directly promoted by random grain boundaries. The antagonist properties of trapping sites and short diffusion paths of random GBs illustrate some ambiguities of the interpretation of experimental data. More recently, we had the opportunity to use both TOF-SIMS and EBSD and combinate their analyses to retrieve the statistical information on the location of hydrogen near the GBs as a function of its ­character29. Results S Initially, these data were only analyzed in term CSL Σ3n and random GBs, but in the present work, we show the opportunity to distin- guish the coincident twin boundary CTB (Σ3 {111}) to other CSL boundaries OTG (ATGB, asymmetric tilt grain boundary and SITB symmetric incoherent twin boundary: Σ3 {112}). Figure 1b illustrates the different profiles of hydrogen content observed around GBs after pre-charging and releasing steps which corresponds to a study state. A gap of hydrogen concentration between both adjacent grains is mostly related to random GBs. Consequently, it seems that when hydrogen diffusion occurs along GBs, hydrogen does not cross the GB easily but moves along the GB. In opposite, a constant profile where no significant modification of hydrogen content is mainly related to both grains around GB is observed for coincident twin boundary CTB where no defect and elastic distortion are necessary to accommodate the misorientation between adjacent grains. y j g For the other CSL Σ3n boundaries OTB, a hydrogen gradient around GBs is observed where a large density of defects is necessary to accommodate the misorientation between two adjacent grains. Additionally, it was reported at a micrometer length scale of the gradient of hydrogen concentration significantly higher than GBs thickness (nanometer)29 which suggests the occurrence of the long-range internal stresses near OTB GBs. Con- sidering these statistical results, we note that even though we associate a type of grain boundary type (random or special) with a hydrogen concentration profile, this result is by no means exclusive. This conclusion allows us to develop a work specifically on selected grain boundaries in a large domain of representative GBs. We have followed the hydrogen content as a function of charging time for a specific electrochemical charging condition which corresponds to a thermodynamic system defined by ­PH2 = 800 atm and T = 300 K on two systems single and bicrystal with the same ingress surface. The evolution of hydrogen concentration ­CH as a function of charg- ing time is presented in Fig. 1c for the {110} and {100} single crystals and bicrystal Σ11{332} with a common hydrogen ingress orientation of {110}. In each case, hydrogen concentration increases with time and reaches a saturation plateau corresponding to an apparent solubility. No significant difference is observed between both single crystals. www.nature.com/scientificreports/ www.nature.com/scientificreports/ Despite numerous experimental and numerical studies, short-circuits of diffusion and trapping processes within grain boundaries in fcc metals and alloys remain a complex subject that is still poorly understood. Fur- thermore, the confrontation of experience and numerical works has not been currently used in this subject which reduces the quality of the interpretations.f Despite numerous experimental and numerical studies, short-circuits of diffusion and trapping processes within grain boundaries in fcc metals and alloys remain a complex subject that is still poorly understood. Fur- thermore, the confrontation of experience and numerical works has not been currently used in this subject which reduces the quality of the interpretations.f q y p In the present work, a substantial effort was made to gain further understanding of the key issues of hydrogen segregation and diffusion processes near GBs. The hydrogen/grain-boundaries interactions have been examined for four different configurations of nickel bi-crystal systems to question a considerable variety of grain boundaries energy and excess volume. The hydrogen mobility and trapping process have been investigated based on the electrochemical hydrogen charging technique and on atomistic simulations using an Embedded Atom Method (EAM) potential. The confrontation of both technics allows to elucidate some relevant queries on the contribu- tion of grain-boundary geometry to the mobility and trapping of hydrogen. The segregation process is discussed in relation to the systematic determination of short and long-range elastic distortions and the short-circuit of diffusion process is clarified with a confrontation of the different diffusion paths and the segregation energy of each grain boundaries considered. Both aspects offer new insight to disclose the impact of grain boundaries on some physical properties. www.nature.com/scientificreports/ The energy barriers along the diffusion path towards and within GBs has been related for Σ3 〈110〉 {111} and Σ11 〈110〉 {113) in fcc Fe-γ62, and for Σ3 〈110〉 {111}51, 63, Σ5 〈100〉 {210}51, 63, Σ5 〈100〉 {310}63, Σ11 〈110〉 {113}63, Σ25 〈100〉 {430}63 and Σ41 〈100 〉 {540}63 in fcc Nickel. In fcc Fe-γ, the Σ3 GB repels hydrogen and the Σ11 offers an easy diffusion path parallel to the GB ­plane62. In fcc nickel, the Σ3 and Σ11 present a quite similar diffusion behavior tho the ­bulk63 and Σ5 GBs exhibit low-barrier paths to facilitate hydrogen diffusion along the ­GBs51, 63 while Σ25 and Σ41 exhibit high- barrier regions which suggest a slower diffusion of hydrogen than the ­bulk63. The authors suggest that a trapping model in relation with the dislocation density is sufficient to relate these ­data63. Despite these appreciable results, a minor confrontation was proposed in the literature between hydrogen diffusivity and segregation capability of GBs which doesn’t offer the opportunity to clarify the trapping process inside the GBs. Scientific Reports | (2021) 11:15533 | https://doi.org/10.1038/s41598-021-94107-6 Results S In opposite the hydrogen content is largely lower in the considered bicrystal (5 wppm instead of 9 wppm for single crystals). We provide the same comparison for three other configurations in Fig. 1d for the stationary state. A very low difference is observed between bicrystal with Σ3 {111} (CTB) or Σ11 {311} in com- parison with single crystal with a similar hydrogen ingress orientation. In opposite for bicrystal with Σ5 {310} or Σ11{332} the hydrogen content obtained are lower than the one determined for equivalent single crystals (same hydrogen ingress orientation). To resume, for bicrystals the hydrogen content is lower for bicrystals than single crystal with an intensity defined by ΔCH (difference of hydrogen content between single crystal and bicrystal) https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ 0 2 4 6 8 10 12 0 20 40 60 80 100 120 140 time (h) CH (wppm) single crystals - orientations (110) and (100) bi-crystal 11{332} - orientation {110} (c) (d) 1.E-14 1.E-13 1.E-12 1.E-11 1.E-10 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 fR Deff (m2/s) , CH x10-8 [H/Ni] Deff CH (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Gap Constant Break Gradient Random GBs CSL GBs S3 CTB S3 OTB 3 3 (b) Probability Profiles of hydrogen Time (h) CH (wppm) CH (wppm) Figure 1. Experimental data collected on the diffusion and solubility in nickel single and polycrystalline pure metal. (a) Effective diffusion coefficient ­Deff and hydrogen concentration ­CH as a function with the fraction of random grain-boundaries ­fR (data collected with Oudriss et al. ­works28 electrochemical permeation for the same conditions and for different grain sizes between 10 nm to 200 μm). (b) Probability to obtain different gradient profiles of hydrogen concentration near GBs obtained by ToF–SIMS ­(data29 revisited, CTB Coherent Twin Boundary Σ3, Other Twin Boundary Σ3). (c) Hydrogen concentration ­CH versus charging time for single crystal and Σ11-{332} bi-crystal with the hydrogen ingress orientation along 〈 110 〉 (hydrogen charging at the cathodic current density of 5 mA/cm2 in 0.1 M NaOH at 300 K). (d) Hydrogen solubility ­CH in single crystals and bi-crystals with different hydrogen ingress orientations: two orientations were studied: {110} and {100} (hydrogen charging at the cathodic current density of 5 mA/cm2 in 0.1 M NaOH at 300 K for 3 days, Σ3 (Σ3- {111} CTB), Σ5 (Σ5-{310}), Σ11 (Σ11-{311}) and Σ11 * (Σ11-{332}). Results S 2a, the first group of our selected GBs: Σ3 {111} (CTB) and Σ11 {311} corresponds to CSL GBs and the second group: Σ5 {310} and Σ11{332} corresponds to Random GBs. Moreover, for the GBs selected in the present work a linear relation can be found between ­EGB and ­Vex: EGB ≈3.32 × Vex despite the fact that we observed a large scatter in the Fig. 2a. The possible correlation between the excess volume ­Vex and the difference of solubility ΔCH is evaluated in Fig. 2b and allows suggesting that a higher disorder in grain-boundary (higher excess volume/Random GBs) promotes the hydrogen diffu- sion along grain boundaries which offers an explanation of fact that the hydrogen content is largely lower in the second group: Σ5 {310} and Σ11{332} (Random GBs). Based on these experimental data, the understanding of the competition between hydrogen segregation and diffusion along the grain-boundaries stays a challenge, which is questioned using atomistic calculations in present work. which depends on the grain boundary character and suggests that GBs act as a short-circuit of diffusion. The difference ΔCH between bi-crystals and single crystal are questioned in term of the energy of grain boundary and the excess volume ­Vex. Both values are evaluated using atomistic calculation with EAM potential (see method) and confronted with data base of Olmsted et al.62 (Fig. 2a). Two domains corresponding to Special or CSL GBs and Random GBs are represented in this figure. Additionally, our studied GBs are defined with red dots. In a specific range of energy and excess volume (intersection between two domains), it is impossible to distinguish both kinds of GBs (CSL or Random). According to the Fig. 2a, the first group of our selected GBs: Σ3 {111} (CTB) and Σ11 {311} corresponds to CSL GBs and the second group: Σ5 {310} and Σ11{332} corresponds to Random GBs. Moreover, for the GBs selected in the present work a linear relation can be found between ­EGB and ­Vex: EGB ≈3.32 × Vex despite the fact that we observed a large scatter in the Fig. 2a. The possible correlation between the excess volume ­Vex and the difference of solubility ΔCH is evaluated in Fig. Results S 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Gap Constant Break Gradient Random GBs CSL GBs S3 CTB S3 OTB 3 3 (b) Probability Profiles of hydrogen 1.E-14 1.E-13 1.E-12 1.E-11 1.E-10 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 fR Deff (m2/s) , CH x10-8 [H/Ni] Deff CH (a) (d) CH (wppm) 0 2 4 6 8 10 12 0 20 40 60 80 100 120 140 time (h) CH (wppm) single crystals - orientations (110) and (100) bi-crystal 11{332} - orientation {110} (c) Time (h) CH (wppm) Figure 1. Experimental data collected on the diffusion and solubility in nickel single and polycrystalline pure metal. (a) Effective diffusion coefficient ­Deff and hydrogen concentration ­CH as a function with the fraction of random grain-boundaries ­fR (data collected with Oudriss et al. ­works28 electrochemical permeation for the same conditions and for different grain sizes between 10 nm to 200 μm). (b) Probability to obtain different gradient profiles of hydrogen concentration near GBs obtained by ToF–SIMS ­(data29 revisited, CTB Coherent Twin Boundary Σ3, Other Twin Boundary Σ3). (c) Hydrogen concentration ­CH versus charging time for single crystal and Σ11-{332} bi-crystal with the hydrogen ingress orientation along 〈 110 〉 (hydrogen charging at the cathodic current density of 5 mA/cm2 in 0.1 M NaOH at 300 K). (d) Hydrogen solubility ­CH in single crystals and bi-crystals with different hydrogen ingress orientations: two orientations were studied: {110} and {100} (hydrogen charging at the cathodic current density of 5 mA/cm2 in 0.1 M NaOH at 300 K for 3 days, Σ3 (Σ3- {111} CTB), Σ5 (Σ5-{310}), Σ11 (Σ11-{311}) and Σ11 * (Σ11-{332}). which depends on the grain boundary character and suggests that GBs act as a short-circuit of diffusion. The difference ΔCH between bi-crystals and single crystal are questioned in term of the energy of grain boundary and the excess volume ­Vex. Both values are evaluated using atomistic calculation with EAM potential (see method) and confronted with data base of Olmsted et al.62 (Fig. 2a). Two domains corresponding to Special or CSL GBs and Random GBs are represented in this figure. Additionally, our studied GBs are defined with red dots. In a specific range of energy and excess volume (intersection between two domains), it is impossible to distinguish both kinds of GBs (CSL or Random). According to the Fig. Results S (b) the difference of hydrogen solubility ΔCH between single crystal and bi-crystal with the same hydrogen ingress orientation as a function with the grain boundary excess volume ­Vex (ΔCH are collected by experimental works, ­Vex is determined numerically using atomistic code). Segregation energy versus grain‑boundary character. The segregation energy for the different sites in and near the GBs has been determined for the four GBs studied. According to previous works (see as ­examples46, 54, 65, 66), the segregation energy of hydrogen in a specific location in and near GBs is relatively reported to the octahedral site since it is the most stable one in fcc nickel phase (the insertion energy for octahe- dral site is Eins oct=0.1775 eV while for tetrahedral site it is equal to 0.586 eV). The energy can be represented as a function of the distance from the GB plane (Fig. 3a). Since a large variety of segregation energy can be obtained for some GBs and specifically for large excess volume (see Σ11-{332} as an example) we characterise this distri- bution by the maximum of this energy ­Eseg(max) and the thickness, ­eGB of GBs defined from an energy point of view. ­eGB characterises a length scale of the impact of GB on the segregation of hydrogen. ­Eseg(max) is obtained in the GBs core and depends on the GB character. Both parameters ­(Eseg(max) and ­eGB) are an increasing function of the grain boundary energy, ­EGB (Fig. 3b). According to the linear relation between grain boundary energy and the excess volume ­Vex (Fig. 2a), the last results can be interpreted as a consequence of the disorder impact on the intensity of segregation process and the domain of the occurrence. To question the possible implication of this disorder on the hydrogen segregation site near grain boundaries, we characterise precisely each location of hydrogen segregation firstly by the hydrogen atomic volume, ­VH. This volume defined by neighbouring nickel atoms is calculated using the Voronoi method. The hydrogen atomic Voronoi volume at the bulk octahedral site ­VOct is 5.773 Å3 and presents a cubic morphology. All segregation sites contiguous to the GB region have a higher atomic volume and a more stable segregation state (Fig. 3d). Concerning the lower energy GBs, the segregation energy ­Eseg for the majority of sites is a quite linear function of the hydrogen atomic volume, ­VH (domain I, blue curve). Results S 2b and allows suggesting that a higher disorder in grain-boundary (higher excess volume/Random GBs) promotes the hydrogen diffu- sion along grain boundaries which offers an explanation of fact that the hydrogen content is largely lower in the second group: Σ5 {310} and Σ11{332} (Random GBs). Based on these experimental data, the understanding of the competition between hydrogen segregation and diffusion along the grain-boundaries stays a challenge, which is questioned using atomistic calculations in present work. https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ 0 1 2 3 4 5 6 7 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Excess Volume, Vex (Å) (b) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Excess Volume, Vex (Å) [Olmsted 2019] present work Random GBs CSL GBs Σ3-(111) CTB Σ3-(110) ATGB Σ3-(112) SITB Σ11-(332) Σ5-(310) Σ11-(311) (a) Grain-boundary energy EGB (J/m2) Difference of hydrogen “solubility” ∆CH (wppm) Figure 2. (a) Grain-boundaries energy ­EGB versus the excess volume ­Vex (data from atomistic calculations courtesy to Olmsted et al.64). The grains boundaries studies in present work are represented in the picture with red dots. Our calculation has been performed with the same atomic potential. (b) the difference of hydrogen solubility ΔCH between single crystal and bi-crystal with the same hydrogen ingress orientation as a function with the grain boundary excess volume ­Vex (ΔCH are collected by experimental works, ­Vex is determined numerically using atomistic code). 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Excess Volume, Vex (Å) [Olmsted 2019] present work Random GBs CSL GBs Σ3-(111) CTB Σ3-(110) ATGB Σ3-(112) SITB Σ11-(332) Σ5-(310) Σ11-(311) (a) Grain-boundary energy EGB (J/m2) 0 1 2 3 4 5 6 7 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Excess Volume, Vex (Å) (b) Difference of hydrogen “solubility” ∆CH (wppm) Figure 2. (a) Grain-boundaries energy ­EGB versus the excess volume ­Vex (data from atomistic calculations courtesy to Olmsted et al.64). The grains boundaries studies in present work are represented in the picture with red dots. Our calculation has been performed with the same atomic potential. Results S The lower energy GBs presents a lower excess volume (Fig. 2a) which corresponds to a large number of sites with potential low hydrogen atomic volume. Consequently, the linear part of the evolution of segregation energy versus hydrogen atomic volume (Fig. 3d) is mainly representative to segregation sites of lower energy GBs. However, this linear relationship is not available for high energy GBs with an atomic volume above 6.6 Å3 (domain II) in accordance with our previous work related to Σ3n special ­GBs46. p p In domain II, a quasi-plateau is reached in terms of energy (− 0.22 eV) with a large scatter. Consequently, the hydrogen atomic volume seems to be insufficient to question the elastic energy contribution to the segregation energy which allows in the following to consider the morphology of the different sites. The segregation sites in the GB region have complicated local geometry structures, thus, we will describe the geometry of all the potential segregation sites in GBs core in detail. An illustration of this approach is shown in Fig. 3c where the segregation positions and their volume geometry are presented for the Σ11-{332} GB. The position numbers are ranked from the most to less stable segregation energy. According to the Voronoi tessellation, the hydrogen volume at the octahedral site in nickel bulk is a cubic form with 14 neighbouring atoms. The closer the hydrogen atoms get to the GB core region, the greater the geometry deformation occurs. Consequently, the segregation energy and the hydrogen volume size are extremely dependent on the local environment. However, a direct relationship cannot be found among these factors. The morphology of the different sites highlights the fact that the deformation of the site is not isotropic for most cases. We will discuss this crucial aspect in the next section. After the considera- tion of the local volume deformation, the hydrogen atom insertion at different sites can be discussed in terms of local energy of hydrogen ­EH (the energy of atom i in the EAM method with i = H). This energy is the sum of the kinetic and the potential energy, which differs from the segregation energy ­Eseg only if we have a long-range effect associated with the insertion of hydrogen. Figure 4a shows the relationships between the segregation energy, ­Eseg and the hydrogen atom energy, ­EH. Results S These sites have a similar cubic form with a relatively low hydrogen atomic volume. This linearity disappears when the distortion becomes significant ­(VH > 6.6 Å3) and probably with a long-range impact of this distortion (see the next section). Moreover, instead of domain II, we observed some specific situations as the fact that some sites where hydrogen atomic volume ­VH is lower than the one of the octahedral interstitial sites ­VOct (5.773 Å3). These situations have been recently related near a free surface of nickel single ­crystal67. Fig. 3d. For low hydrogen atom energy, a linear relation with segregation energy is clearly established ­(Eseg = 0.54 × ­EH + 0.0864) which corresponds to a range of low hydrogen atomic volume range ­VH < 6.6 Å3 (domain I). The slope 0.54 of ­Eseg versus ­EH coincides to the ratio between the unstrained atomic volume ΩNi (10.904 Å3) and one of the unstrained octahedral interstitial sites ­VOct (5.773 Å3). Consequently, the hydrostatic deformation induced by the incorporation of hydrogen is a second order term for the relationship between ­Eseg and ­EH. For a segrega- tion energy equal to zero ­(Eseg = 0 eV), the hydrogen atom energy, ­EH is equal to 0.16 eV which is quite similar to the value of the insertion energy for octahedral site Eins oct (0.1775 eV). For a higher value of ­EH above − 0.5 eV (domain II) a large scatter demonstrates that probably long-range displacements of the nickel atoms occur in addition to the short-range displacement of the neighbouring atoms. Concerning Fig. 3d, the picture highlights the fact that hydrogen atom is more stable at a deep segregation site with high atomic volume. Actually, we have noticed that a part of the data has a linear evolution ( Eseg ∼A × [VH −Voct] with a slope A equal to 0.43). These sites have a similar cubic form with a relatively low hydrogen atomic volume. This linearity disappears when the distortion becomes significant ­(VH > 6.6 Å3) and probably with a long-range impact of this distortion (see the next section). Moreover, instead of domain II, we observed some specific situations as the fact that some sites where hydrogen atomic volume ­VH is lower than the one of the octahedral interstitial sites ­VOct (5.773 Å3). These situations have been recently related near a free surface of nickel single ­crystal67. Segregation energy versus elastic energy. Results S Two domains are clearly observed in relation with the ones defined in Scientific Reports | (2021) 11:15533 | https://doi.org/10.1038/s41598-021-94107-6 www.nature.com/scientificreports/ -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 5.5 6 6.5 7 7.5 8 8.5 Segregaon energy, Eseg (eV) Hydrogen atomic volume, VH (Å3) Σ5 - {310} Σ11 - {332} Σ11 - {311} Σ3 - CTB Σ3 - ATGB 35.26° Σ3 - ATGB 54.74° Σ3 - SITB Domain I Domain II (d) (c) -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 -15 -10 -5 0 5 10 15 H distance from GB plane (Å) S11 - {332} S5 - {310} S11 - {311} S3 - {111} Σ11 Σ5 Σ11 Σ3 (a) 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 0.5 1 1.5 2 2.5 3 0 200 400 600 800 1000 1200 GB tickness, e (nm) Grain boundary energy, EGB (mJ/m2) e [present work] e [Hallil et al. 2017] Eseg(max) [present work] Eseg (max) [Hallil et al. 2017] Hydrogen segegraon energy, Eseg(max) (eV) (b) Hydrogen segregation energy, Eseg (eV) Figure 3. (a) Hydrogen segregation energy, ­Eseg as a function of the hydrogen location in the grain boundary core region. (b) Grain boundary thickness, e and maximum hydrogen segregation energy, ­Eseg(max) vs the grain boundary energy ­EGB. (c) Segregation position of hydrogen in GB Σ 11-{311} and their volume geometry defined by Voronoi’s method. (d) Segregation energy as a function of the hydrogen atomic volume ­VH for the different locations in the seven grain boundaries studied. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 0.5 1 1.5 2 2.5 3 0 200 400 600 800 1000 1200 GB tickness, e (nm) Grain boundary energy, EGB (mJ/m2) e [present work] e [Hallil et al. 2017] Eseg(max) [present work] Eseg (max) [Hallil et al. Results S 2017] Hydrogen segegraon energy, Eseg(max) (eV) (b) -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 -15 -10 -5 0 5 10 15 H distance from GB plane (Å) S11 - {332} S5 - {310} S11 - {311} S3 - {111} Σ11 Σ5 Σ11 Σ3 (a) Hydrogen segregation energy, Eseg (eV) (c) -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 5.5 6 6.5 7 7.5 8 8.5 Segregaon energy, Eseg (eV) Hydrogen atomic volume, VH (Å3) Σ5 - {310} Σ11 - {332} Σ11 - {311} Σ3 - CTB Σ3 - ATGB 35.26° Σ3 - ATGB 54.74° Σ3 - SITB Domain I Domain II (d) Figure 3. (a) Hydrogen segregation energy, ­Eseg as a function of the hydrogen location in the grain boundary core region. (b) Grain boundary thickness, e and maximum hydrogen segregation energy, ­Eseg(max) vs the grain boundary energy ­EGB. (c) Segregation position of hydrogen in GB Σ 11-{311} and their volume geometry defined by Voronoi’s method. (d) Segregation energy as a function of the hydrogen atomic volume ­VH for the different locations in the seven grain boundaries studied. Fig. 3d. For low hydrogen atom energy, a linear relation with segregation energy is clearly established ­(Eseg = 0.54 × ­EH + 0.0864) which corresponds to a range of low hydrogen atomic volume range ­VH < 6.6 Å3 (domain I). The slope 0.54 of ­Eseg versus ­EH coincides to the ratio between the unstrained atomic volume ΩNi (10.904 Å3) and one of the unstrained octahedral interstitial sites ­VOct (5.773 Å3). Consequently, the hydrostatic deformation induced by the incorporation of hydrogen is a second order term for the relationship between ­Eseg and ­EH. For a segrega- tion energy equal to zero ­(Eseg = 0 eV), the hydrogen atom energy, ­EH is equal to 0.16 eV which is quite similar to the value of the insertion energy for octahedral site Eins oct (0.1775 eV). For a higher value of ­EH above − 0.5 eV (domain II) a large scatter demonstrates that probably long-range displacements of the nickel atoms occur in addition to the short-range displacement of the neighbouring atoms. Concerning Fig. 3d, the picture highlights the fact that hydrogen atom is more stable at a deep segregation site with high atomic volume. Actually, we have noticed that a part of the data has a linear evolution ( Eseg ∼A × [VH −Voct] with a slope A equal to 0.43). www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 4. (a) Hydrogen segregation energy ­Eseg versus hydrogen energy ­EH. (b) Hydrostatic σm and deviatoric stress σD as a function of hydrogen atomic volume ­VH. (c) Hydrogen energy ­EH vs the elastic strain energy ­Eel (per unit of volume) (d) Contributions to the hydrostatic, ­EH and deviatoric, ­ED parts of energy to the elastic strain energy ­Eel. Figure 4. (a) Hydrogen segregation energy ­Eseg versus hydrogen energy ­EH. (b) Hydrostatic σm and deviatoric stress σD as a function of hydrogen atomic volume ­VH. (c) Hydrogen energy ­EH vs the elastic strain energy ­Eel (per unit of volume) (d) Contributions to the hydrostatic, ­EH and deviatoric, ­ED parts of energy to the elastic strain energy ­Eel. process the elastic dipole tensor ­Pij is used to describe the equilibrium in a continuum solid of a point-force ­distribution46, 73–75. The elastic dipole tensor characterizes the changes for both the volume and the shape of the interstitial site during the relaxation procedure after the incorporation of the solute in the lattice. From lattice point defect distortion theory, the dipole tensor is given by: process the elastic dipole tensor ­Pij is used to describe the equilibrium in a continuum solid of a point-force ­distribution46, 73–75. The elastic dipole tensor characterizes the changes for both the volume and the shape of the interstitial site during the relaxation procedure after the incorporation of the solute in the lattice. From lattice point defect distortion theory, the dipole tensor is given by: (1) Pij =  m f m/sol j rm i (1) where f m/sol j is the force exerted from the solute surrounding atoms m and rm i is the atomic displacement of atoms. In a first approach, the lattice distortion around the solute is supposed to occur mainly on the insertion site closest to the neighboring atoms. Results S As previously highlighted in different configurations, the insertion of a solute in a specific location induces a geometric transformation which cannot be only described as a volume change (see as ­examples16, 46, 68–73). The modification of the geometric configuration of the volume defined by Voronoi’s method results in an equilibrium with the different interactions between the neighboring atoms of the solute. The contribution of the elastic strain energy of this process can be evaluated by considering the respective effect of hydrostatic strain (change of the volume) and shear strain (change of the morphology, deviatoric part of strain). In order to investigate the contribution of the elastic strain energy to the segregation https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | with Ed el = 1 6G∗(σd)2VH.h 6G This equation predicts that for total elastic energy Eel= Em el + Ed el = 0eV that EH = Echem = − 0.653 eV which is higher than the one related in Fig. 4c in domain II (− 0.85 eV). This difference suggests that the contribution of the electronic component of energy increases in domain II when the hydrogen volume increases to a value near VH vac . In the present work, the elastic contribution of the energy of hydrogen ­EH is between 0.286 eV and 0.082 eV in regime I and lower than 0.082 eV in regime II. The main effect is consequently observed in regime I where the hydrostatic strain contribution is predominant and has the same order of magnitude as the chemical contribution.i Both Figs. 3d ( Eseg vs VH ) and 4a ( Eseg vs EH ) clearly demonstrate that a significant number of insertion sites do not verify the relationships previously defined. These situations correspond to sites of the grain-boundaries with higher energy and excess volume (Σ11-{332} and Σ5-{310}). The fact that we cannot find a relation with Eseg and EH or VH , suggesting that a significant part of the segregation energy depends on the long-range elastic distortion. To support this idea, two situations are illustrated in Fig. 5. The first one corresponds to the location (a) where a long-range displacement is observed along the axis 〈110〉 and a short-range displacement in any other directions (Fig. 5a). The profile of the radial deformation is shown in Fig. 5b with a length scale of around ∼ 8 Å and an amplitude of 16%. The first neighbor’s energy (1 to 4) corresponds to a long-range elastic interaction ELR el ∼4.3109Pa (or 0.16 eV) which needs to be integrated in the total strain energy density. With this contri- bution the location (a) follows the linear curve Eseg vs Eel defined in domain I with Eel = Ed el + Em el + ELR el  . The second situation illustrated in Fig. 5d is the location (b) in domain II which verifies the linear curve Eseg vs Eel without any addition of long-range elastic interactions term because quasi-isotropic short-range displacements are only observed. Consequently, the large scattering observed in Fig. 4a results from the long-range elastic energy between 0.08 eV and 0.44 eV. Diffusion paths. www.nature.com/scientificreports/ of about 2 GPa. Below this value a non-linear evolution is established and referred to as domain II. As it is illus- trated by the Fig. 4d, the elastic strain energy ­Eel is dominated by the hydrostatic stress for the first linear regime ( Eel ≈ 1 2B(σm)2 in domain I, ­Eel > 2 GPa, ­EH >  − 0.5 eV and ­VH < 6.6 Å3) while the other regime (domain II, ­Eel < 2 GPa, ­EH <  − 0.5 eV and ­VH > 6.6 Å3) is due to the contribution of the deviatoric and hydrostatic stresses with a similar contribution for the large volumes. Assuming that local energy of hydrogen is dominated by the elastic strain energy induced by the insertion of the solute, the slope of linear behavior observed in domain I between hydrogen energy, ­EH and the elastic strain energy per unit volume, ­Eel (Fig. 4c) is directly associated with an apparent hydrogen volume Vap H equal to 12 Å3. In regime I, the hydrogen volume ­VH stays equal to the hydrogen volume of the bulk octahedral interstitial site (5.773 Å3) with only a difference lower than 10%. The discrepancy observed between Vap H and ­VH is directly associated with the fact that firstly (Eq. 2) we do not consider the impact of hydrogen on the elastic properties. The degradation of elastic properties correlated to the incorporation of hydrogen is known well documented in the ­literature17, 73, 76–81. Based on the linear-elastic theory of the equilibrium between interstitial solute and sur- rounding metal lattice, the apparent bulk modulus B* can be determined considering the Eshelby’s inclusion model in case of isotropic elastic properties of a ­solid76, 77: B∗= B  2(1−2υ) (1+υ)   . Consequently, considering elastic strain energy ­Eel dominated by the hydrostatic stress with the bulk modulus correction we find Vap H = 6.5 Å3 which is in accordance with ­VH obtained in regime I (between 5.7 to 6.4 Å3). In agreement with the Fig. 4c, this result indicates that the local energy of hydrogen ­EH verifies that: ) (3) EH = Em el + Echem EH = Em el + Echem (3) with a dilatational energy equal to: Em el = (1+υ) 4B(1−2ν)(σm)2VH. Th d f h i E (3) Echem i h l with a dilatational energy equal to: Em el = (1+υ) 4B(1−2ν)(σm)2VH. The second term of the equation Eq. www.nature.com/scientificreports/ Thus, in isotropic medium, the stress σij and the elastic strain energy ­Eel by unit volume induced from the insertion of the hydrogen interstitial are related to the dipole ­tensor46: (2) σij = Pij VH = σmδij+σ d ijandE el = Em el + Ed el = 1 2B(σm)2 + 1 6G (σd)2 (2) where σm and σd are respectively the hydrostatic stress ( 1 3[σ11 + σ22+σ 33]) and the deviatoric stress defined as Von Mises equivalent stress ( 2σ 2 d = (σ11 −σ22)2 + (σ11 −σ33)2 + (σ33 −σ22)2 + 6(σ 2 12 + σ 2 23 + σ 2 13)) . B and G are the bulk and shear modulus respectively. Em el and Ed el represent the dilatational and distortion energies. The evolution of hydrostatic and deviatoric part of stress field as a function of hydrogen atomic volume is illustrated in Fig. 4b. Two behaviors can be distinguished: in domain I, for a lower value of ­VH below 6.6 Å3, the hydrostatic component is predominant, whereas in domain II both stresses have the same order of intensity. The direct consequence of this result is observed in the correlation between the local energy of hydrogen ­EH and the elastic strain energy ­Eel (Fig. 4c). This one illustrates a change of behavior between both domains defined previously. In domain I, a linear regime between both energies is observed with a transition around a strain energy density https://doi.org/10.1038/s41598-021-94107-6 https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ www.nature.com/scientificreports/ (3), Echem is the elect el 4B(1 2ν) The second term of the equation Eq. (3), Echem is the electronic contribution or “chemical” term which cor- responds to the embedding energy due to the distortion of chemical bonds near the solute ( Echem= − 0.653 eV). Equation (3) can be extended in regime II, considering the increase of the hydrogen atomic volume and the implication of distortion part of the elastic energy Ed el . For the last component, we also consider the degradation of elastic properties with hydrogen ( G∗= 3B∗ (1−2υ) 2(1+υ)   ). In domain II, ­VH increases with the decrease of the elastic energy to reach a value near 8.2 Å3. This value is very close to the atomic volume of hydrogen localized in a vacancy VH vac ( ∼ 8.66Å3)46 where the contribution of elastic energy is negligible instead of an electronic contribution in accordance with the present ­results82. Equation (3) can be rewritten considering the previous comment as: (4) EH = Em el + Ed el + Echem (4) EH = Em el + Ed el + Echem with Ed el = 1 6G∗(σd)2VH.h The short circuit of diffusion along GBs can be discussed as a function with a reference state defined as a diffusion path of the perfect crystal. The nudged elastic band method (NEB) method has been used to calculate the Minimum Energy Paths (MEP) and their associated energy barriers (Table 1). Several stable segregation sites in GBs were looked upon for three principal path directions: (i) the hydrogen atom moves from the bulk site to the GB site (Fig. 6a), (ii) the hydrogen atom moves between two GB sites along the GB plane (Fig. 6b), (iii) the hydrogen atom moves between two GB sites along the tilt axis (Fig. 6c). The first investigation of the migration energy Em has been carried out in the perfect nickel crystal. For the easier path, the hydrogen atom moves between octahedral sites (O) through a metastable tetrahedral site (T) in nickel bulk and the energy barriers, E0−T m is equal to 0.47 eV in accordance with the DFT works: ∼ 0.46 ­eV83 and ∼ 0.41 ­eV84. This migration energy is defined as the reference state, the hydrogen atom is able to move faster if the energy barrier is lower than this reference energy. The movement of a hydrogen atom in the GB core and cross the GB is illustrated in Fig. 6 where it is given the MEP and the energy barriers of NEB calculations. The energy barrier is calculated https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ 1 2 3 3 1 2 Σ11-{332} <110> 4 -0.4 -0.3 -0.3 -0.2 -0.2 -0.1 -0.1 0.0 0.1 0.E+00 1.E+09 2.E+09 3.E+09 4.E+09 5.E+09 6.E+09 7.E+09 Hydrogen segregaon energy, Eseg (eV) Elasc strain energy, Eel (Pa) S5 - {310} S11 - {332} S11 - {311} S3 - {111} 5 11 3 11 Domain I Domain II (a) (b) Initial state Final state Σ11-{311} (b) (a) (c) (d) Figure 5. The loss of the linear relation between the segregation energy, ­Eseg and the elastic strain energy ­Eel(c) is directly a consequence of the anisotropy of the displacement field and “long range” elastic distortion. (a) An illustration of the anisotropy of the displacement field along 〈110〉 near a Σ11-{332}. (b) The relative radial displacement dr/r along 〈110〉 as a function of the r coordination from H (corresponding to the Σ11-{332} illustrated in (a)). with Ed el = 1 6G∗(σd)2VH.h (d) Moderate quasi-isotropic short range displacement for a position of hydrogen which respects the linear relation of ­Eseg versus ­Eel (see picture (c), GB Σ11-{311}). (b) 1 2 3 3 1 2 Σ11-{332} <110> 4 (a) -0.4 -0.3 -0.3 -0.2 -0.2 -0.1 -0.1 0.0 0.1 0.E+00 1.E+09 2.E+09 3.E+09 4.E+09 5.E+09 6.E+09 7.E+09 Hydrogen segregaon energy, Eseg (eV) Elasc strain energy, Eel (Pa) S5 - {310} S11 - {332} S11 - {311} S3 - {111} 5 11 3 11 Domain I Domain II (a) (b) (c) Initial state Σ11-{311} (d) (d) Final state Initial state Initial state Final state Final state Figure 5. The loss of the linear relation between the segregation energy, ­Eseg and the elastic strain energy ­Eel(c) is directly a consequence of the anisotropy of the displacement field and “long range” elastic distortion. (a) An illustration of the anisotropy of the displacement field along 〈110〉 near a Σ11-{332}. (b) The relative radial displacement dr/r along 〈110〉 as a function of the r coordination from H (corresponding to the Σ11-{332} illustrated in (a)). (d) Moderate quasi-isotropic short range displacement for a position of hydrogen which respects the linear relation of ­Eseg versus ­Eel (see picture (c), GB Σ11-{311}). with the initial position energy as the reference point. We have considered several positions such as the most stable segregation sites (type A) and the highest volume sites (type B). A and B positions are the same only for stable segregation sites (type A) and the highest volume sites (type B). A and B positions are the same only for Table 1. Energy barriers for the most stable segregation position A and the highest volume position B along the different directions X; Y and Z. The forward ( EF m ) and backward ( EB m ) paths are differentiated. with Ed el = 1 6G∗(σd)2VH.h Grain-boundaries Segregation site X, along GB Y, across GB Z, along GB Σ3-{111} CTB A/B EF m = EB m = 0.52 eV EF m = EB m = 0.46 eV – Σ3-{112} SITB A/B EF m = EB m = 0.6 eV EF m = 0.42 eV/EB m = 0.65 eV – Σ11-{311} A/B EF m = EB m = 0.51 eV EF m = 0.47 eV/EB m = 0.72 eV EF m = EB m = 0.6 eV Σ11-{332} A EF m = EB m = 0.61 eV EF m = 0.48 eV/EB m = 0.81 eV EF m = EB m = 0.23 eV B EF m = EB m = 0.3 eV EF m = 0.46 eV/EB m = 0.53 eV EF m = EB m = 0.22 eV Σ5-{310} A EF m = EB m = 0.51 eV EF m = 0.46 eV/EB m = 0.79 eV EF m = EB m = 0.5 eV B EF m = EB m = 0.3 eV EF m = 0.46 eV/EB m = 0.7 eV EF m = EB m = 0.45 eV Table 1. Energy barriers for the most stable segregation position A and the highest volume position B alon the different directions X; Y and Z. The forward ( EF m ) and backward ( EB m ) paths are differentiated. the GBs Σ11-{311} and Σ3-{111} (CTB). In each situation, we have considered the forward ( EF m ) and backward ( EB m ) paths. All the energies are reported in Table 1. the GBs Σ11-{311} and Σ3-{111} (CTB). In each situation, we have considered the forward ( EF m ) and backward ( EB m ) paths. All the energies are reported in Table 1. For the CTB Σ3-{111} according to the migration energy, a hydrogen atom can easily cross the GB ( EGB,Y m = 0.46eV ∼E0−T m = 0.47eV ) but it is slightly slowed down along the GB plane ( EGB,X m = 0.52eV > E0−T m = 0.47eV ). In opposite, for the SITB Σ3-{112} according to the migration energy, a hydrogen atom can easily cross the GB ( EGB,Y m = EF m = 0.42eV ∼E0−T m = 0.47eV ) but the backward path is more difficult ( EGB,Y m = EB m = 0.65eV > E0−T m  ). with Ed el = 1 6G∗(σd)2VH.h Additionally, the migration energy along grain-boundary is https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ entificreports/ -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (eV) Coordination ratio path A path B -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (eV) Coordination ratio path A path B -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (eV) Coordination ratio path A path B (a) (b) (c) Figure 6. Energy barriers and minimum energy path between the known initial and final states for the GB Σ11–{332}. (a) Diffusion paths perpendicular to the grain boundary y =  〈332〉. (b, c) diffusion paths along the grain boundary respectively for x =  〈311〉 and z =  〈110〉. A is the path between the most stable segregation positions and the bulk. B is the path between the highest volume positions and the bulk. -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (eV) Coordination ratio path A path B (a) Energy (eV) (b) -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (eV) Coordination ratio path A path B Energy (eV) (b) (c) -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (eV) Coordination ratio path A path B Energy (eV) Coordination ratio Figure 6. Energy barriers and minimum energy path between the known initial and final states for the GB Σ11–{332}. (a) Diffusion paths perpendicular to the grain boundary y =  〈332〉. (b, c) diffusion paths along the grain boundary respectively for x =  〈311〉 and z =  〈110〉. A is the path between the most stable segregation positions and the bulk. B is the path between the highest volume positions and the bulk. slower in the GB core than in the bulk EGB,X m = EF m = EB m = 0.6eV > E0−T m  ). Consequently, the segregation sites are probably trapping sites in present SITB Σ3-{112}. with Ed el = 1 6G∗(σd)2VH.h For the low energy/excess volume GB Σ11-{311} (CSL GB), the hydrogen atom can move easily from the bulk to the GB core ( EGB,Y m = EF m = 0.47eV = E0−T m  ), but the backward path is more difficult ( EGB,Y m = EB m = 0.72eV > E0−T m  ) as are the other directions ( EGB,X m = 0.51eVandEGB,Z m = 0.6eV) where the dif- fusion gets slower in the GB core. This GB has a few numbers of sites for the hydrogen segregation (four sites) but its segregation sites are probably trapping sites. https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ The Random GB Σ11-{332} is a high energy/excess volume GB (Fig. 6). It has several hydrogen segregations sites in the GB core (14 sites). For the segregation site A, the energy barrier along the X direction is higher than the reference energy in nickel bulk ( EGB,X m = 0.61eV > E0−T m = 0.47eV ), in opposite for the segregation site B where we observed a fast diffusion path with lower migration energy ( EGB,X m = 0.3eV < E0−T m = 0.47eV ). A quite similar value for the migration energy along Z direction is obtained for the segregation sites A and B ( EGB,Z m equal 0.23 eV for A site and 0.22 eV for B site) which allows to define this direction as a fast dif- fusion path. For the cross direction Y, the hydrogen atom can move easily from the bulk to the GB core ( EGB,Y m = EF misequalto0.48eVforAsiteand0.46eVforBsite ), but the backward path is more difficult ( EGB,Y m = EB m is equal to 0.81 eV and 0.53 eV respectively for A and B sites).h q p y The Random GB Σ5-{310} is the highest energy/excess volume GB in our samples, it has less segregation sites than the GB Σ11-{332} in the GB core (8 sites). Along the X direction, it exists a diffusion path for which hydrogen atom moves faster at the highest volume sites B ( EGB,X m = 0.3eV < E0−T m = 0.47eV ). However, the diffusion at the most stable segregation sites A slows down (EGB,X m = 0.51eV > E0−T m = 0.47eV) and can be considered as a trapping site. Discussionh The contribution of shear elastic energy far from any defect stays less common feature related in the literature but a few works show clearly the significance of that ­one69–71, 73, 88. The shear or d ih deviatoric elastic strain energy Ed el can be mostly near the dislocation core and have a large contribution to the trapping ­energy70, 71. In the present work, the respective contribution on elastic strain energy of hydrostatic Em el and deviatoric Ed el strains in grain boundary (Fig. 4b,d) reveals a significant influence of shear strain in domain II for higher hydrogen atomic volume ­(VH > 6.6 Å3). Additionally, no clear correlation can be obtained in this regime between segregation energy and hydrogen atomic volume (Fig. 3d). The second fundamental aspect emerges in a contribution of a long-range elastic distortion ELR el to segregation energy as illustrated in Fig. 5. The importance of this contribution was appropriately illustrated for some specific sites (Fig. 5) where the amplitude of this energy offers the opportunity to explain the discrepancy observed between some experimental data and linear relation between the segregation energy ­Eseg and hydrogen energy ­EH (Fig. 4a).f d g g gy g y g gy g Consequently, the segregation energy can be formulated as a function of the different contributions ( Ed el , Em el  LR el  , Echem, Eins octa) : (5) Eseg = Voct Ni  EH −Eins octa + ELR el with EH = Em el + Ed el + Echem (5) The chemical component of energy, Echem evolves between − 0.653 eV to − 0.85 eV, the short range elastic energy Em el + Ed el is evaluated between 0.082 eV to 0.286 eV, the long range internal stresses can be reach a value of 0.44 eV and the insertion energy for octahedral site Eins oct is equal to 0.1775 eV.hf The chemical component of energy, Echem evolves between − 0.653 eV to − 0.85 eV, the short range elastic energy Em el + Ed el is evaluated between 0.082 eV to 0.286 eV, the long range internal stresses can be reach a value of 0.44 eV and the insertion energy for octahedral site Eins oct is equal to 0.1775 eV.hf The hydrogen enrichment at the different grain boundaries studied in the present work can be described with a modified analytical Langmuir-McLean segregation isotherm (LML). Discussionh The shear or deviatoric elastic strain energy Ed el can be mostly near the dislocation core and have a large contribution to the trapping ­energy70, 71. In the present work, the respective contribution on elastic strain energy of hydrostatic Em el and deviatoric Ed el strains in grain boundary (Fig. 4b,d) reveals a significant influence of shear strain in domain II for higher hydrogen atomic volume ­(VH > 6.6 Å3). Additionally, no clear correlation can be obtained in this regime between segregation energy and hydrogen atomic volume (Fig. 3d). The second fundamental aspect emerges in a contribution of a long-range elastic distortion ELR el to segregation energy as illustrated in Fig. 5. The importance of this contribution was appropriately illustrated for some specific sites (Fig. 5) where the amplitude of this energy offers the opportunity to explain the discrepancy observed between some experimental data and linear relation between the segregation energy ­Eseg and hydrogen energy ­EH (Fig. 4a).f d y gyhii geometry using space tessellation or Voronoi description. Zhou et al.56 recently argued that the local volume of the site can be analyzed using only five polyhedrons. Our present investigations clearly illustrate the limitation of this description where we observed a large diversity of geometric forms (see complementary data) especially for grain boundaries with the highest excess volume. According to a large set of data, the segregation energy cannot be directly correlated with the variation of the volume (Fig. 3d) without considering the asymmetry of some sites (see Fig. 3c, site 1) and their distortion (deviatoric part of local strain tensor) during the incorporation of the solute. Another aspect needs to be integrated into this analysis in relation to the impact of hydrogen on the atomic configuration around the insertion site, which depends on the distribution of the different sites along the GB. The first aspect can be considered as a short-range interaction while the second one which corresponds to a long-range distortion according to the length scale larger than the size of the site and the in inhomogene- ity of the distribution of these sites. According to the thermodynamic approach, Larche et al.86 and Kirchheim et al.87 highlighted the possible contribution of the hydrostatic and the deviatoric strain energy on the chemical potential. More recently, the contribution of defects and elastic anisotropy have been questioned on the solubil- ity and diffusivity in ­nickel16. with Ed el = 1 6G∗(σd)2VH.h Along the Z direction, the energy barrier higher than the reference energy in nickel bulk and cannot be considered as a new path of diffusion (Table 1). For the cross direction Y, the hydrogen atom can move easily from the bulk to the GB core ( EGB,Y m = EF m = 0.46eVforforAandBsites ), but the backward path is more difficult ( EGB,Y m = EB m is equal to 0.79 eV and 0.7 eV respectively for A and B sites) which promotes the stabilization of hydrogen on GB. Discussionh The implications of grain-boundary on hydrogen embrittlement are one key understanding to extend the mate- rial performance in an aggressive environment. The design of new materials and alloys requires to improve the fundamental interaction of hydrogen with structural features of grain-boundaries. Before any analysis of the intergranular fracture is necessary to understand the mobility of hydrogen in relation to the microstructural fea- tures. Commonly as any solutes, the first step is to focus on the segregation process. The general approach of the segregation is based on a differentiation between an insertion site in GB and the one in lattice (octahedral in fcc alloys) in terms of energy and ­volume58, 59, 65, 85. The domain of the insertion is defined as a volume with specific y gyhi pi geometry using space tessellation or Voronoi description. Zhou et al.56 recently argued that the local volume of the site can be analyzed using only five polyhedrons. Our present investigations clearly illustrate the limitation of this description where we observed a large diversity of geometric forms (see complementary data) especially for grain boundaries with the highest excess volume. According to a large set of data, the segregation energy cannot be directly correlated with the variation of the volume (Fig. 3d) without considering the asymmetry of some sites (see Fig. 3c, site 1) and their distortion (deviatoric part of local strain tensor) during the incorporation of the solute. Another aspect needs to be integrated into this analysis in relation to the impact of hydrogen on the atomic configuration around the insertion site, which depends on the distribution of the different sites along the GB. The first aspect can be considered as a short-range interaction while the second one which corresponds to a long-range distortion according to the length scale larger than the size of the site and the in inhomogene- ity of the distribution of these sites. According to the thermodynamic approach, Larche et al.86 and Kirchheim et al.87 highlighted the possible contribution of the hydrostatic and the deviatoric strain energy on the chemical potential. More recently, the contribution of defects and elastic anisotropy have been questioned on the solubil- ity and diffusivity in ­nickel16. The contribution of shear elastic energy far from any defect stays less common feature related in the literature but a few works show clearly the significance of that ­one69–71, 73, 88. www.nature.com/scientificreports/ Σ5-{310}. This order is similar to the one of excess volume of GBs and allows to conclude that Σ5-{310} presents a higher capability to segregation than Σ11-{332} although the number of sites is lower to the one of Σ11-{332}. According to the Langmuir–McLean segregation isotherm model without solutes ­interactions65, 66, it is pos- sible to establish a relationship between the segregation concentration and the bulk concentration in a binary Σ5-{310}. This order is similar to the one of excess volume of GBs and allows to conclude that Σ5-{310} presents a higher capability to segregation than Σ11-{332} although the number of sites is lower to the one of Σ11-{332}. Σ5-{310}. This order is similar to the one of excess volume of GBs and allows to conclude that Σ5-{310} presents a higher capability to segregation than Σ11-{332} although the number of sites is lower to the one of Σ11-{332}. According to the Langmuir McLean segregation isotherm model without solutes interactions65, 66 it is pos According to the Langmuir–McLean segregation isotherm model without solutes ­interactions65, 66, it is pos- sible to establish a relationship between the segregation concentration and the bulk concentration in a binary system. The expression is given in: (6) CGB = n  i=1 Cmax,i 1 1 + 1−Cbulk Cbulk exp  Eseg,i kBT (6) Cseg,i = CHseg,i/CNi is the molar fraction of segregation on site i, Cbulk = CH/CNi is the molar fraction in the nickel bulk. Hydrogen atoms occupy the octahedral site in the nickel bulk, the unit cell of fcc structure gives the maximum saturation Coct/CNi = 1. The saturation of segregation Cmax,i of the boundary on a site i depends on the number of the sites with the same segregation energy ­Eseg,i. As illustrated in Fig. 7a for Σ11-{332}n = 14 sites can be distinguished. Considering the temperature of 300 K, the evolution of hydrogen concentration segregated on GBs is represented as a function of the bulk concentration (Fig. 7c).hih The picture confirms the analysis in terms of the segregation energy density of states. The segregation sites in the high energy GBs such as Σ11-{332} and Σ5-{310} start to be occupied by hydrogen atoms at a very low bulk concentration (Cbulk ~ ­10−7 H/Ni). The GB Σ11-{311} begins the segregation around Cbulk = ­10−6 H/Ni while the GB Σ3-{111} initiates the segregation around Cbulk = ­10−3 H/Ni. www.nature.com/scientificreports/ The segregation is first initiates at the most stable positions, and then the less steady positions commence to segregate hydrogen atoms when the first position reaches saturation (70%–80% occupation). It has been reported that high local concentrations can act as a seed for local hydrides in ­nickel92, 93 because of the attractive H–H interactions. Moreover, EP/TDS experi- mental data indicates that the limit of hydrogen solubility in nickel single crystal is around Cbulk = 5 × ­10−4 H/ Ni (~ 7.5 wppm at 300 K). Consequently, we compared thus the segregation in different GBs within the range of the bulk concentration between ­10–7 and 5 × ­10−4 H/Ni. Hydrogen segregation is stronger in Σ5-{310} than in Σ11-{332}, the segregation concentration is dominated by the most stable segregation position of which ­Eseg,1 Σ5-{310} = − 0.3252 eV ­Eseg,1 Σ11-{332} = − 0.3188 eV. Additionally, the GB Σ3-{111} acts as the nickel bulk with no significant segregation at all. h d l l d d k h h ll i Once hydrogen atom is localized in a GB, it is now appropriated to ask now how the atom will move or remain trapped using the evaluation of the migration energies.h The hydrogen atom in all GBs (Σ3-{111}, Σ11-{311}, Σ5-{310} and Σ11-{332}) can easily move from the nickel bulk to the GB core: the forward energy EGB,Y m = EF m ≈ 0.47 ± 0.01 eV is quite similar to the migration energy in the bulk Ebulk m = E 0−T m = 0.47 eV. GBs such as Σ11-{311}, Σ5-{310} and Σ11-{332} prevent the hydrogen atom from moving from the GB core to the bulk (backward energy EGB,Y m = EB m > 0.47 eV), however Σ3-{111} GB act if the bulk had little influence on this direction ( EGB,Y m = EB m = 0.46 eV). The energy barrier is able to provide an estimation for the ratio of the diffusion coefficient in comparison to the diffusion in the perfect nickel bulk as it was previously proposed by Di Stefano et al.51: (7) DGB Dbulk = exp  −EGB,i m −Ebulk m kBT  with i ∈{X, Y, Z} (7) Two situations can be considered when the backward energy is higher than the one in bulk. www.nature.com/scientificreports/ The first one cor- responds to the segregation of hydrogen to an easy migration path along the grain boundary (directions X and/ or Z) where the EGB,XorZ m is lower than in the bulk and consequently DGB Dbulk > 1. This situation is commonly named a short circuit of diffusion. The second case is a situation where all the migration energies along GB are higher than in the bulk and consequently DGB Dbulk < 1. This situation is defined as a trapping configuration.f Dbulkhii Figure 8 illustrates a distinction (bulk red line) between the accelerated diffusion and the trapping in seg- regation sites type A (most stable segregation sites) and B (highest volume sites) for the different GBs studied. High energy/excess volume GBs (Σ11-{332} and Σ5-{310}) have one or two accelerated diffusion path(s) where the hydrogen atom moves ­103 ~ ­104 faster than the diffusion in the nickel bulk. These results are confirmed by our experimental data where for both GBs the hydrogen content in bi-crystals is lower than in single crystal for the same orientation (Fig. 1c,d). Indeed, Σ11-{332} is a GB that have more segregation sites in the GB core with two accelerated diffusion paths, and none of these segregation sites has the trapping behavior (Fig. 8). While GB Σ5-{310} has only one accelerated diffusion path and the deepest segregation site in this GB is a trapping position for the hydrogen atoms (Fig. 8). Σ11-{311} GB do not present any accelerating diffusion paths but only trapping sites in the GB core. Moreover, no significant difference is observed in terms of hydrogen content in single and by-crystals (Fig. 1d). These results allow to conclude that this GB is not a fast diffusion path as CTB Σ3-{111}. Similar results in GBs Σ11-{311} and Σ3-{111} have been reported by Du et al.62 in γ-Fe based on their DFT calculations. To conclude, based on a combined analysis of the segregation energy and migration energy of the different potential sites of hydrogen in one GB, it is possible to define one GB as a short circuit of diffusion or trapping location for hydrogen. In perspective, we can enlarge this analysis for a large number of GB configura- tions considering the similarity in terms of EGB,i m and EH using atomistic calculation and machine ­learning90. Discussionh Instead of initial ­approach89, it was recently ­answered90 the importance to consider the variability of the segregation energy of one grain ­boundary91. Consequently, for each grain boundary, we have determined the number of sites for each segregation energy for one representative motif (see Fig. 7a as an example for Σ11-{332}). Considering the segregation energy density of states, ρ(Eseg,i) for each GBs, we observed that ρ is maximum for low segregation energy in Σ 3-{111} (Fig. 7b). In opposite, ρ is maximum for the highest segregation energy for Σ5-{310}(Fig. 7b). The segregation energy corresponding to the maximum of ρ follows the relation: ­Eseg Σ3-{111} > ­Eseg Σ11-{311} > ­EsegΣ11-{332} > ­Eseg Scientific Reports | (2021) 11:15533 | https://doi.org/10.1038/s41598-021-94107-6 www.nature.com/scientificreports/ Conclusion d l d We developed a methodology based on the confrontation between experimental hydrogen charging and atomis- tic modeling to elucidate the competition between hydrogen trapping and short circuit of diffusion along grain boundaries. https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | w.nature.com/scientificreports/ Using thermal desorption spectroscopy and electrochemical permeation testing we measured the diffusivity Figure 7. (a) The unit cell for GB Σ11-{332} with position details ((31 Ni and 28 sites for hydrogen segregation, 14 sites with different segregation energies). (b) The segregation energy density of states (ρ) as a function of absolute value of segregation energy ­Eseg for four GBs. (c) the evolution of hydrogen concentration segregated on GBs is represented as a function with the bulk concentration. www.nature.com/scientificreports/ Figure 7. (a) The unit cell for GB Σ11-{332} with position details ((31 Ni and 28 sites for hydrogen segregation, 14 sites with different segregation energies). (b) The segregation energy density of states (ρ) as a function of absolute value of segregation energy ­Eseg for four GBs. (c) the evolution of hydrogen concentration segregated on GBs is represented as a function with the bulk concentration. Figure 7. (a) The unit cell for GB Σ11-{332} with position details ((31 Ni and 28 sites for hydrogen segregation, 14 sites with different segregation energies). (b) The segregation energy density of states (ρ) as a function of absolute value of segregation energy ­Eseg for four GBs. (c) the evolution of hydrogen concentration segregated on GBs is represented as a function with the bulk concentration. Using thermal desorption spectroscopy and electrochemical permeation testing, we measured the diffusivity of hydrogen and trapping energy in single and poly crystals of nickel on a large variety of configurations. Our experimental results are challenged by our atomistic simulations to provide a straightforward understanding of the apparently ambiguous and antagonist effects of grain boundaries between the trapping process and the fast diffusion path. The origin of the acceleration of the diffusivity along grain boundaries is observed when the excess volume increases, which is clarified in terms of migration energies and the distribution of segregation energies. The fast diffusion of hydrogen observed in some grain boundary configurations corresponds to high segregation energy sites and different paths along inter-connected sites of low migration energy. In opposite, https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ Figure 8. Methods M i l d Material design and structural characterization. The different nickel samples investigated were high purity single crystals (99.999% purity), bi-crystals (99.95% purity) and poly-crystals/nanocrystals (99.99% purity). Nickel single crystals with three different crystallographic orientations 〈001〉, 〈110〉 and 〈111〉 were provided by Goodfellow. Samples are cylindrical with an 18 mm diameter for nickel 〈001〉 and 11 mm for nickel 〈110〉 and 〈111〉. They were obtained by Bridgman-Stockbarger method with an accuracy of ± 3°. Nickel bi-crystals were made in Laboratory (provided by professor L. Priester (ICMPE) and developed in Mines St-Etienne School). The material for solidification was electrolytic nickel which was remelted under vacuum. The bicrystals were grown from a seed in an argon atmosphere using the horizontal boat method (Chalmers method) of a final length of 14.4 cm at a rate of about 3 mm per hour. The impurity contents are given lower than 0.05 wppm, the sulfur content could be enough to promote intergranular ­segregation94. All samples were then carried out a desulfurization heat treatment at 550 °C for 10 h in flowing hydrogen. Polycrystalline nickel with micrometric grain sizes between 18 and 200 μm was controlled by thermo-mechanical processing at con- stant annealing temperature for a predefined time in a controlled atmosphere on cold-drawn round nickel rods provided by ­Goodfellow95. The nanometric sizes from 20 to 120 nm were synthesized by electrodeposition using a conventional additive-free sulfamate bath according to the deposition parameters and conditions described ­previously96.f p y All the microstructures were characterized using Electronic Back Scatter Diffraction (EBSD) and Transmis- sion Electronic Microscopy (TEM). EBSD analyses using an EDAX/ TSL OIM system coupled to an FEI Quanta 200 ESEM- FEG scanning electron microscope were used to characterize the crystallographic orientation (tex- ture, surface orientation) and grain boundaries character (misorientation, coincidence site lattice)28. Dislocation densities of each sample were investigated using a JEOL JEM 2011 electron microscope operating at 200 kV. Foils for transmission electron microscopy (TEM) were thinned in a double twin-jet electro-polisher using electrolyte under the conditions described ­previously97. An estimation of the dislocation density of the as-received material gave an average value of less than ­10−10 ­m2. Charging hydrogen and permeation conditions. The electrochemical permeation (EP) test and ther- mal desorption spectroscopy (TDS) were used to characterize macroscopically the diffusivity and concentration of hydrogen for different conditions. Conclusion d l d The diffusion coefficient ratio as a function of the hydrogen segregation energy following x, y and z directions for different GBs. Figure 8. The diffusion coefficient ratio as a function of the hydrogen segregation energy following x, y and z directions for different GBs. the trapping process occurs in grain boundaries of high segregation energy sites and high migration energies. We underline the importance of the elastic energy to global hydrogen incorporation energy in one specific site of grain boundaries. Due to the complex structure of the grain boundary, the distribution of the segregation energy depends not only on the hydrostatic strain energy but also on the deviatoric elastic energy and the long- range elastic distortion which can predominate far from GB. The present work can be extended to a large variety of grain boundaries to allow a more generic relationship between elastic, segregation and migration energies. the trapping process occurs in grain boundaries of high segregation energy sites and high migration energies. We underline the importance of the elastic energy to global hydrogen incorporation energy in one specific site of grain boundaries. Due to the complex structure of the grain boundary, the distribution of the segregation energy depends not only on the hydrostatic strain energy but also on the deviatoric elastic energy and the long- range elastic distortion which can predominate far from GB. The present work can be extended to a large variety of grain boundaries to allow a more generic relationship between elastic, segregation and migration energies. www.nature.com/scientificreports/ The atomistic investigation following the works of Hallil et al.46, 60 was carried out using LAMMPS code (Large-scale Atomic/Molecular Massively Parallel Simulator) for MD simulations at 0 K, or more precisely, molecular statics (MS). The atomistic simulations based on the relaxed configuration of atoms are found using the minimization of the total energy with an appropriate interatomic potential at 0 K and 0 Pa. For the H-Ni system, the interatomic potential was established by the embedded-atom method (EAM)53. The energy ­Ei of atom i (H or Ni) in the EAM method is given by the following equation: (8) Ei = Fα  i=j ρβ  rij  + 1 2  i=j ϕβ(rij) (8) where ­Fα is the embedding energy function, ρ is the partial electron density contribution, rij denotes the distance between atom i and j, φ is the pair potential, α and β are the element types of atom i and j. For a grain boundary in a bi-crystal system, the construction of grain boundary is built by finding the most optimal configuration. An example of a simulation cell for the grain boundary Σ5-36°87 〈100〉 {310} is illustrated in Fig. 9. θ is the misorientation angle between two identical nickel crystals around the symmetric tilt axis along the grain boundary plane. The direction along the GB plane is designated by the tilt z-axis [001] in the simulation cell and common for both grains. Each GB simulation cell contains two-grain lattices which are characterized by two distinct crystallographic orientations in x and y directions. The GB simulation cells are considered in 3D periodic boundary conditions, this representation provides existence of two GBs in each simulation cell: one in the middle of the simulation cell and on another counting for a mirror image in the bound parts of the simulation cell. The separation distance between each GB is chosen to be large enough so that there will be no energetic interaction between two GB interfaces. A rigid body translation parallel to the GB plane has been applied following x and z-axis, all translational vectors are in a primitive cell of the displacement shift complete (DSC) ­lattice59 and the lattice spacing in the planar directions of each grain is kept constant. The translation of one grain relative to the other yields to a re-arrangement of atoms at the GB plane. www.nature.com/scientificreports/ cathodic charging current density (5 to 100 mA/cm2) in 0.1 M NaOH. The detection side (exit side) of the EP was maintained with a constant anodic potential of ~ 630 mV/SSE in 0.1 M NaOH, and the hydrogen flux (current density) at the detection side was recorded to study the transport of hydrogen through the membrane. The tem- perature was maintained at 300 K and both solutions were continuously deaerated under argon flux at 1.4 bar. Before the permeation test, both surfaces of the sample were prepared by mechanical polishing up to grade 4000 SiC; the final thickness of the sample was about 200 ± 20 μm.h cathodic charging current density (5 to 100 mA/cm2) in 0.1 M NaOH. The detection side (exit side) of the EP was maintained with a constant anodic potential of ~ 630 mV/SSE in 0.1 M NaOH, and the hydrogen flux (current density) at the detection side was recorded to study the transport of hydrogen through the membrane. The tem- perature was maintained at 300 K and both solutions were continuously deaerated under argon flux at 1.4 bar. Before the permeation test, both surfaces of the sample were prepared by mechanical polishing up to grade 4000 SiC; the final thickness of the sample was about 200 ± 20 μm.h i To quantify the hydrogen concentration and the maximum solubility in nickel, we used Thermal Desorption Spectroscopy (TDS)15, 28. These analyses were performed with a Jobin Yvon Horiba EMGA-621W hydrogen analyzer composed of an impulsion furnace system coupled with a thermal conductivity detector. The procedure used consists involved measuring the hydrogen concentration in the pre-charged samples (0.1 M NaOH at 298 K under galvanostatic polarization − 10 mA/cm2 with different time) by fusion. For this purpose, after hydrogen charging, the specimen (dimension 8 × 4 × 0.3 ­mm3) were mechanically polished with 5 μm SiC grinding paper, and then cleaned in acetone before introduced into the furnace, where they were instantly heated to 2000 °C and maintained at this temperature for 75 s. The desorbed hydrogen was then detected and analyzed by gaseous Catharometry. The recorded curve corresponds to the amount of hydrogen detected as a function of time. The average concentration of hydrogen in the sample was estimated by measuring the area under the curve. Atomistic simulation. Methods M i l d The EP technique introduced by Devanathan and ­Stachurski98 is the main technique used to detect the mechanisms of diffusion and trapping of hydrogen in different ­microstructures15, 16, 28. This technique is composed of two cells separated by a membrane with an exposed surface in contact with an electrolytic solution. The charging side (entry side) of the EP was galvanostatically polarized at a constant https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ www.nature.com/scientificreports/ Received: 25 April 2021; Accepted: 30 June 2021 Received: 25 April 2021; Accepted: 30 June 2021 Received: 25 April 2021; Accepted: 30 June 2021 www.nature.com/scientificreports/ After testing hundreds of configurations, the one with the minimum energy at the grain boundary is obtained and the excess volume could be calculated using Voronoi tessellation method implemented in LAMMPS code.ih Since the GB structure is well defined, we can insert an atom of hydrogen into the GB. The initial position of the hydrogen atom is at the vicinity of a nickel atom. Then the energy minimization occurs and the hydrogen atom will find its stable position using atomic simulation analysis OVITO software. After determining several stable segregation sites in the core region, we focused on the diffusion paths among these different segregation sites using the nudged elastic band method (NEB)99. This method is used to detect saddle points and minimum energy paths (MEP) between the known initial and final states. Transition states of diffusion paths (referred to as images) identify the lowest possible energy while maintaining equal spacing to neighboring images. Once the images have converged sufficiently to the MEP, the image at the highest energy point is allowed to climb uphill along the MEP until it reaches the transition state enabling thus the transition geometry and energy to be accurately defined from the NEB ­method98. We have investigated the MEP and the energy barrier between the most stable segregation positions and the highest Voronoi volume positions (the volume occupied by the hydrogen atom) using LAMMPS code with NEB package. The GB energy is computed as the difference between the total energy of the relaxed GB atoms and the bulk energy in the whole system in GB plane. For a number of atoms Nat in the calculation, the grain boundary energy EGB is given as: (9) EGB = 1 2  ENi GB(Nat) −ENi Bulk(Nat)  /A0 (9) where ENi GB(Nat) is the total energy of the relaxed GB, ENi Bulk(Nat) is the total bulk energy, A0 is the area of the GB plane. VGB the excess volume of a GB can be accessed with (volume variation per unit of GB area): (10) VGB = 1 2  VTot GB (Nat) −VTot Bulk(Nat)  /A0 (10) where VTot GB (Nat) is the volume of GB, VTot Bulk(Nat) is the volume of bulk. where VTot GB (Nat) is the volume of GB, VTot Bulk(Nat) is the volume of bulk. where VTot GB (Nat) is the volume of GB, VTot Bulk(Nat) is the volume of bulk. Received: 25 April 2021; Accepted: 30 June 2021 www.nature.com/scientificreports/ https://doi.org/10.1038/s41598-021-94107-6 https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ Figure 9. (a) GB Σ5-36°87 〈001〉 {310} plane for nickel. (b) Schematic of a bi-crystal simulation cell. Figure 9. (a) GB Σ5-36°87 〈001〉 {310} plane for nickel. (b) Schematic of a bi-crystal simulation cell. When the stable configuration of GBs has been established, we started to insert the hydrogen atom in dif- ferent locations in the GBs. The insertion energy (in some publications called also the adsorption energy) of a hydrogen atom in the nickel lattice EIns H is given in [Eq. (11)]: (11) EIns H = ETot Ni+H −ETot Ni −1 2EH2 (11) ETot Ni+H is the total energy of nickel lattice with a hydrogen atom, ETot Ni is the total energy of nickel lattice with- out a hydrogen atom, EH2 is the chemical potential of the molecular hydrogen and the binding energy of one hydrogen atom can be obtained by the EAM potential in ­vacuum99 ( 1 2EH2 = − 2.36947 eV). This calculation has indicated that the octahedral site ( EIns Oct = 0.1775 eV) is more stable than the tetrahedral site ( EIns Tet = 0.586 eV). Similar results of EAM have been reported by Huang et al.54. Thus, the segregation energy relative to the octa- hedral site is written in [Eq. (12)]: (12) Eseg H =  ETot GB+H −ETot GB  −  Eoct Ni+H −Eoct Ni (12) ETot GB+H is the total energy of GB with a hydrogen atom, ETot GB is the total energy of the GB and EOct Ni+H is the otal energy of the nickel lattice with a hydrogen atom at the octahedral site. ETot GB+H is the total energy of GB with a hydrogen atom, ETot GB is the total energy of the GB and EOct Ni+H is the total energy of the nickel lattice with a hydrogen atom at the octahedral site. Impact statement. We clearly established the impact of the geometry and the deformation of hydrogen insertion site on the diffusion path and segregation process along grain-boundaries. The analyses offer the opportunity to clearly precise the trapping process on grain-boundary. A well-defined effect of deviatoric and hydrostatic elastic energies on segregation energy was identified at a short-range scale and additionally signifi- cant impact of a long-range elastic distortion observed for some GBs configurations. www.nature.com/scientificreports/ Cohesive zone simulation of grain size and misorientation effects on hydrogen embrittlement in nickel. Eng. Fail. Anal 81, 79–93 (2017). 41. Tehranchi, A. & Curtin, W. A. Atomistic study of hydrogen embrittlement of grain boundaries in nickel: II. Fract. Model. Simul. Mater. Sci. Eng. 25(7), 075013 (2017).ff 2. Zhou, X. W., Dingreville, R. & Karnesky, R. A. Molecular dynamics studies of irradiation effects on hydrogen isotope diffusion through nickel crystals and grain boundaries. Phys. Chem. Chem. Phys. 20(1), 520–534 (2017). 3. O’Brien, C. J. & Foiles, S. M. Hydrogen segregation to inclined twin grain boundaries in nickel. Philos. Mag. 96(26), 2808–2828 (2016). 44. Tehranchi, A. & Curtin, W. A. Atomistic study of hydrogen embrittlement of grain boundaries in nickel: I. Fracture. J. Mech. Phys. Solids 101, 150–165 (2017). 5. Hallil, A., Metsue, A., Oudriss, A., Bouhattate, J. & Feaugas, X. Segregation energy of the hydrogen at Ni Σ3 grain boundaries Some implications of the atomic volume and the interstitial self-stress. J. Mater. Sci. 53(7), 5356–5363 (2018). p 6. Moody, N. R. & Foiles, S. M. An atomistic study of the equilibrium segregation of hydrogen to tilt boundaries in nickel. Mater Res. Soc. Symp. Proc. 229, 179 (1991). y p 47. He, S., Ecker, W., Pippan, R. & Razumovskiy, V. I. Hydrogen-enhanced decohesion mechanism of the special Σ5(012)[100] grain boundary in Ni with Mo and C solutes. Comput. Mater. Sci. 167, 100–110 (2019). y 8. Yamaguchi, M., Shiga, M. & Kaburaki, H. Energetics of segregation and embrittling potency for non-transition elements in the Σ 5 (012) symmetrical tilt grain boundary: A first-principles study. J. Phys. Condens. Matter. 16(23), 3933 (2004). i 9. Yamaguchi, M., Shiga, M. & Kaburaki, H. First-principles study on segregation energy and embrittling potency of hydrogen in n Σ 5(012) tilt grain boundary. J. Phys. Soc. Jpn. 73(2), 441–449 (2004).f 50. Di Stefano, D., Mrovec, M. & Elsässer, C. First-principles investigation of hydrogen trapping and diffusion at grain boundaries in nickel. Acta Mater. 98(Supplement C), 306–312 (2015). 51. Young, G., et al. An atomistic modeling study of alloying element impurity element, and transmutation products on the cohesion of a nickel ni Σ 5 {001} twist grain boundary. Tech. Report LM-03K047, NASA. (2003). { } g y p ( ) 2. Angelo, J. E., Moody, N. R. & Baskes, M. I. Trapping of hydrogen to lattice defects in nickel. Model. Simul. Mater. Sc. 3(3), 289 (1995). www.nature.com/scientificreports/ www.nature.com/scientificreports/ 10. Seita, M., Hanson, J. P., Gradečak, S. & Demkowicz, M. J. The dual role of coherent twin boundaries in hydrogen embrittlement. Nat. Commun. 6, 6164 (2015). 1. Harris, Z. D. et al. Elucidating the contribution of mobile hydrogen-deformation interactions to hydrogen-induced intergranular cracking in polycrystalline nickel. Acta Mater. 158, 180–192 (2018). g p y y 12. Bertsch, K. M., Wang, S., Nagao, A. & Robertson, I. M. Hydrogen-induced compatibility constraints across grain boundaries drive intergranular failure of Ni. Mater. Sci. Eng. A 760, 58–67 (2019). g g y g intergranular failure of Ni. Mater. Sci. Eng. A 760, 58–67 (2019) g g 3. Lynch, S. A review of underlying reasons for intergranular cracking for a variety of failure modes and materials and examples of case histories. Eng. Fail. Anal. 100, 329–350 (2019).i case histories. Eng. Fail. Anal. 100, 329–350 (2019). 14. Koyama, M. et al. Recent progress in microstructural hydrogen mapping in steels: Quantification, kinetic analysis, and multi-scale h S T h l ( ) g 4. Koyama, M. et al. Recent progress in microstructural hydrogen mapping in steels: Quantification, kinetic analysis, and multi-scale characterization. Mater. Sci. Technol. 33, 1481–1496 (2017). 5. Oudriss, A., Martin, F. & Feaugas, X. Experimental techniques for dosage and detection of hydrogen. Mech. Microstruct. Corros Coupl. 11, 245–268 (2019).fhf p ( ) 6. Li, J., Oudriss, A., Metsue, A., Bouhattate, J. & Feaugas, X. Anisotropy of hydrogen diffusion in nickel single crystals: The effects of self-stress and hydrogen concentration on diffusion. Sci. Rep. 7, 45041 (2017).l p 16. Li, J., Oudriss, A., Metsue, A., Bouhattate, J. & Feaugas, X. Anisotropy of hydrogen diffusion of self-stress and hydrogen concentration on diffusion. Sci. Rep. 7, 45041 (2017).l y gf p 17. Hachet, G., Metsue, A., Oudriss, A. & Feaugas, X. Influence of hydrogen on the elastic properties of nickel single crystal: A numeri- cal and experimental investigation. Acta Mater. 148, 280–288 (2018).f l cal and experimental investigation. Acta Mater. 148, 280–288 (201 p g 18. Kirchheim, R., Somerday, B. & Sofronis, P. Chemomechanical effects on the separation of interfaces occurring during fracture with emphasis on the hydrogen-iron and hydrogen-nickel system. Acta Mater. 99, 87–98 (2015).hi yf p g g with emphasis on the hydrogen-iron and hydrogen-nickel system. Acta Mater. 99, 87–98 (2015).hi 19. Alvaro, A., Thue Jensen, I., Kheradman, N., Løvvik, O. M. & Olden, V. www.nature.com/scientificreports/ Observation of local hydrogen on nickel surface. Metall. Trans. A 14A, 2440–2442 (1983).h 31. Yao, J. & Cahoon, J. R. The use of silver decoration technique in the study of hydrogen transport in metallic materials. M Trans. 21A, 603 (1990).lf 32. Tanaka, T., Kawakami, K. & Hayashi, S. Visualization of deuterium flux and grain boundary diffusion in duplex stainless steel and Fe–30 %Ni alloy, using secondary ion mass spectrometry equipped with a Ga focused ion beam. J. Mater. Sci. 49, 3928 (2014). 33. Ma, Z. X. et al. Experimental study on the diffusion of hydrogen along individual grain boundaries in nickel. Electroch mun. 92, 24–28 (2018). 34. Tsuru, T. & Latanision, R. M. Grain boundary transport of hydrogen in nickel. Scr. Metall. Mater. 16, 575 (1982). 35. Fukushima, H. & Birnbaum, H. K. Surface and grain boundary 35. Fukushima, H. & Birnbaum, H. K. Surface and grain boundary segregation of deuterium in nickel. Acta Metall. 32, 851–859 (1984). 36. Ladna, B. & Birnbaum, H. K. SIMS study of hydrogen at the surface and grain boundaries of nickel bicrystals. Acta Metall. 35, 25352542 (1987).i 36. Ladna, B. & Birnbaum, H. K. SIMS study of hydrogen at the surface and grain boundaries of nickel bicrystals. Acta Metall. 35, 25352542 (1987). 7. Karnesky, R. A., Norman, J., Bartelt, C., Teslich, D. & Kumar, M. Imaging and quantification of hydrogen isotope trapping. Sandia Rep. 2012, 8539 (2012). 37. Karnesky, R. A., Norman, J., Bartelt, C., Teslich, D. & Kumar, M. Imaging and quantification of hydrogen isotope trapping. Sandia Rep. 2012, 8539 (2012). 38. Dingreville, R., Alsoy, D. & Spearot, D. E. A primer on selecting grain boundary sets for comparison of interfacial fracture proper- Rep. 2012, 8539 (2012). 38. Dingreville, R., Alsoy, D. & Spearot, D. E. A primer on selecting grain boundary sets for comparison of interfacial fracture proper- ti i l l d i i l ti S i R p 7 8332 (2017) 8. Dingreville, R., Alsoy, D. & Spearot, D. E. A primer on selecting grain boundary sets for comparison of interfacial fracture proper- ties in molecular dynamics simulations. Sci. Rep. 7, 8332 (2017). y p 39. Tehranchi, A. & Curtin, W. Atomistic study of hydrogen embrittlement of grain boundaries in nickel: I Fracture. J Mech. Phys. Solids 101(Supplement C), 150–165 (2017).f pp 0. Yu, H. et al. References e e e ces 1. Martin, F. et al. State of hydrogen in matter: Fundamental Ad/absorption, trapping and transport mechanisms. In Mechanics: Microstructure: Corrosion Coupling (eds Blanc, C. & Aubert, I.) 171–197 (Elsevier, 2019).f 1. Martin, F. et al. State of hydrogen in matter: Fundamental Ad/absorption, trapping and transport mechanisms. In Mechanics: Microstructure: Corrosion Coupling (eds Blanc, C. & Aubert, I.) 171–197 (Elsevier, 2019). p g 2. Ashby, M. F. & Hirth, J. P. Perspectives in Hydrogen in Metals: Collected Papers on the Effect of Hydrogen on the Properties of Metals and Alloys (Elsevier, 1986).f 3. Gangloff, B. P. S. (ed.) Gaseous Hydrogen Embrittlement of Materials in Energy Technologies: Mechanisms, Modelling and Future Development 247–285 (Woodhead Publishing LtD, 2012). p g 4. Robertson, I. M. Hydrogen embrittlement understood. Metall. Mater. Trans. A 46, 2323–2341 (2015). 5. Nagumo, M. Fundamentals of Hydrogen Embrittlement (Springer, 2016).f 6. Feaugas, X. & Delafosse, D. Hydrogen and crystal defects interactions: Effects on plasticity and fracture. Mech. Microstruct. Corros Coupl. 9, 199–222 (2019). p 7. Kimura, A. & Birnbaum, H. K. Hydrogen induced grain boundary fracture in high purity nickel and its alloys: Enhanced hydrogen diffusion along grain boundaries. Acta Metall. 36, 757–766 (1988). f g g , ( ) 8. Martin, M. L., Somerday, B. P., Ritchie, R. O., Sofronis, P. & Robertson, I. M. Hydrogen-induced intergranular failure in nickel revisited. Acta Mater. 60(6–7), 2739–2745 (2012). f g g , ( ) 8. Martin, M. L., Somerday, B. P., Ritchie, R. O., Sofronis, P. & Robertson, I. M. Hydrogen-induced intergranular failure in nicke revisited. Acta Mater. 60(6–7), 2739–2745 (2012). ( ), ( ) 9. Oudriss, A. et al. On the implication of hydrogen on inter-granular fracture. Procedia Mater. Sci. 3, 2030–2034 (2014). https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ Hydrogen embrittlement in nickel, visited by first prin modeling, cohesive zone simulation and nanomechanical testing. Int. J. Hydrogen 40, 16892–16900 (2015). h l f b d h d h d d d b l 0. Chen, J. & Dongare, A. M. Role of grain boundary character on oxygen and hydrogen segregation-induced embrittlement in polycrystalline Ni. J. Mater Sci. 52, 30 (2017). 1. Tehranchi, A. & Curtin, W. A. Atomistic study of hydrogen embrittlement of grain boundaries in nickel: I. Fracture. J. Mech. Phys Sol. 101, 150–165 (2017). 2. Hea, S., Eckera, W., Pippanc, R. & Razumovskiy, V. I. Hydrogen-enhanced decohesion mechanism of the special Σ5(012)[100] i b d i Ni ith M d C l t C M t S i 167 100 110 (2019) 22. Hea, S., Eckera, W., Pippanc, R. & Razumovskiy, V. I. Hydrogen-enhanced decohesion mechanism of the special Σ5(012)[ grain boundary in Ni with Mo and C solutes. Comp. Mater. Sci. 167, 100–110 (2019).f 2. Hea, S., Eckera, W., Pippanc, R. & Razumovskiy, V. I. Hydrogen enhanced decohesion mechanism of the special Σ5(012)[100] grain boundary in Ni with Mo and C solutes. Comp. Mater. Sci. 167, 100–110 (2019).f 23. Kuhr, B., Farkas, D. & Robertson, I. M. Atomistic studies of hydrogen effects on grain boundary structure and deformation response in FCC Ni. Comp. Mater. Sci. 122, 92–101 (2016).f p 24. Louthan, M. R., Donovan, J. A. & Caskey, G. R. Hydrogen diffusion and trapping in nickel. Acta Metall. 23, 745 (1975). hh k l k d b d ll k l 25. Kirchheim, R., Kownacka, I. & Filipek, S. M. Hydrogen segregation at grain boundaries in nanocrystalline nickel. Scr. Metall. 2 1229 (1993). 26 Brass A M & Chanfreau A Accelerated diffusion of hydrogen along grain boundaries in nickel Acta Mater 44 3823 (1996) 6. Brass, A. M. & Chanfreau, A. Accelerated diffusion of hydrogen along grain boundaries in nickel. Acta Mater. 44, 3823 (1996).hf f 7. Oudriss, A. et al. The diffusion and trapping of hydrogen along the grain boundaries in polycrystalline nickel. Scr. Mater. 66, 37–40 (2012).ff 28. Oudriss, A. et al. Grain size and grain-boundary effects on diffusion and trapping of hydrogen in pure nickel. Acta Mater. 60, 6814–6828 (2012). 29. Oudriss, A. et al. Meso-scale anisotropic hydrogen segregation near grain-boundaries in polycrystalline nickel characteriz EBSD/SIMS. Mater. Lett. 165, 217–222 (2016). 30. Schober, T. & Dieker, C. www.nature.com/scientificreports/ Stress and solubility for solutes with asymmetrical distortion fields. Acta Metall. 35, 2899–2903 (1985). 85. Feaugas, X. & Haddou, H. Grain-size effects on tensile behavior of nickel and AISI 316L stainless steel. Metall. Mater. Trans. A 34, 2329–2340 (2003).f hf 84. Kirchheim, R. & Hirth, J. P. Stress and solubility for solutes with asymmetrical distortion fields. Acta M i 85. Feaugas, X. & Haddou, H. Grain-size effects on tensile behavior of nickel and AISI 316L stainless steel. Metall. Mater. Trans. A 2329–2340 (2003).f 6. Vattré, A., Jourdan, T., Ding, H., Marinica, M. C. & Demkowicz, M. J. Non-random walk diffusion enhances the sink strength o semicoherent interfaces. Nat. Commun. 7, 10424 (2016). . McLean, D. Grain Boundaries in Metals (Clarendon Press, 1957 8. Huber, L., Hadian, R., Grabowski, B. & Neugebauer, J. A machine learning approach to model solute grain boundary segregation NPJ Comput. Mater. 4, 64 (2018).h 8. Huber, L., Hadian, R., Grabowski, B. & Neugebauer, J. A machine learning approach to model solute grain boundary segregation NPJ Comput. Mater. 4, 64 (2018). 9 Whit C & C hl W Th t f bi di i h t i b d ti M t ll M t T A 8 9. White, C. & Coghlan, W. The spectrum of binding energies approach to grain boundary segregation. Metall. Mater. Trans. A 8 1403–1412 (1977).h 90. von Pezold, J., Lymperakis, L. & Neugebeauer, J. Hydrogen-enhanced local plasticity at dilute bulk H concentrations: The role of H-H interactions and the formation of local hydrides. Acta Mater. 59(8), 2969–2980 (2011). 91. Song, J. & Curtin, W. A. A nanoscale mechanism of hydrogen embrittlement in metals. Acta Mater. 59, 1557–1569 (2011). 92. Lezzar, B., Khalfallah, O., Larere, A., Paidar, V. & Hardouin, D. O. Detailed analysis of the segregation driving forces for Ni(Ag) and Ag(Ni) in the Σ=11 113 and Σ=11 332 grain boundaries. Acta Mater. 52(9), 2809–2818 (2004). 93. Juhas, M. C., Priester, L. & Biscondi, M. Correlation between geometry and chemistry of a near Σ11 grain boundary. Mater Eng. A185, 71–76 (1994). g 4. Godon, A. et al. Characterization of electrodeposited nickel coatings from sulphamate electrolyte without additive. Mater. Charact 62(2), 164–173 (2011). 95. Oudriss, A. & Feaugas, X. Length scales and scaling laws for dislocation cells developed during monotonic deformation of (001) nickel single crystal. Int. J. Plast. 78, 187–202 (2016). 96. Foiles, S. M., Baskes, M. I., Melius, C. F. www.nature.com/scientificreports/ First-principles calculation of the elastic dipole tensor of a point defect: Application to hydrogen in α-zirconium. Phys. Rev. B 94, 241112 (2016). y 73. Clouet, E., Varvenne, C. & Jourdan, T. Elastic modeling of point-defects and their interaction. Comp. Mater. Sci. 147, 49–63 (2 h lb Th h f l d f l h y 73. Clouet, E., Varvenne, C. & Jourdan, T. Elastic modeling of point-defects and their interaction. Comp. Mater. Sci. 147, 49–63 (2018). 74. Eshelby, J. D. The continuum theory of lattice defects. Sol. State Phys. 3, 79 (1956).f g p p 74. Eshelby, J. D. The continuum theory of lattice defects. Sol. State Phys. 3, 79 (1956).f . Eshelby, J. D. The continuum theory of lattice defects. Sol. State The continuum theory of lattice defects. Sol. State Phys. 3, 79 (195 yh y y 75. Mura, T. Micromechanics of Defects in Solids (Martinus Nijhoff Publishers, 1987). f f jf 76. Sofronis, P. & Birnbaum, H. K. Mechanics of the hydrogen-dislocation-impurity interactions-I increasing shear modulus. J. Mech. Phys. Sol. 43(1), 49–90 (1995). 7. Chateau, J. P., Delafosse, D. & Magnin, T. Numerical simulations of hydrogen-dislocation interactions in FCC stainless steels: part I: Hydrogen-dislocation interactions in bulk crystals. Acta Mater. 50(6), 1507–1522 (2002).f 78. Delafosse, D. Hydrogen effects on the plasticity of face centred cubic (fcc) crystals. In Gaseous Hydrogen Embrittlement of Materials in Energy Technologies: Mechanisms, Modelling and Future Developments Vol. 2 (eds Gangloff, R. D. & Somerday, B. P.) 247–285 (Woodhead Publishing LtD, 2002).i g 9. Cai, W., Sills, R., Barnett, D. & Nix, W. Modeling a distribution of point defects as misfitting inclusions in stressed solids. J. Mech Phys. Solid. 66, 154–171 (2014).i y 0. Metsue, A., Oudriss, A. & Feaugas, X. Displacement field induced by a vacancy in nickel and some implications for the solubility of hydrogen. Philos. Mag. 94(34), 3978–3991 (2014).ffi 1. Wimmer, E. et al. Temperature-dependent diffusion coefficients from ab initio computations: Hydrogen in nickel. Phys. Rev. B 77, 134305 (2008). ( ) 82. Metsue, A., Oudriss, A. & Feaugas, X. Trapping/detrapping kinetic rates of hydrogen around a vacancy in nickel and some con- sequences on the hydrogen-vacancy clusters thermodynamic equilibrium. Comput. Mater. Sci. 151, 144–152 (2018).hf hf 4. Kirchheim, R. & Hirth, J. P. Stress and solubility for solutes with asymmetrical distortion fields. Acta Metall. 35, 2899–2903 (1985) hf 84. Kirchheim, R. & Hirth, J. P. www.nature.com/scientificreports/ www.nature.com/scientificreports/ 3. Huang, S., Chen, D., Song, J., McDowell, D. L. & Zhu, T. Hydrogen embrittlement of grain boundaries in nickel: An atomistic study. NPJ Comput. Mater. 3, 28 (2017). y p 4. Zhou, X., Marchand, D., McDowell, D. L., Zhu, T. & Song, J. Chemomechanical origin of hydrogen trapping at grain boundaries in fcc metals. Phys. Rev. Lett. 116(7), 075502 (2016). y ( ) ( ) 55. Zhou, X. & Song, J. Effects of alloying elements on vacancies and vacancy-hydrogen clusters at coherent twin boundaries in nickel alloys. Acta Mater. 148, 9–17 (2018).fi gf y g alloys. Acta Mater. 148, 9–17 (2018).fi A. P. & Balluffi, R. W. Interfaces in Crystalline Materials (Clarendon y 56. Sutton, A. P. & Balluffi, R. W. Interfaces in Crystalline Materiah fi f y 57. Priester, L. Grain Boundaries From Theory to Engineering (Springer, 2013). h 58. Hallil, A., Metsue, A., Bouhattate, J. & Feaugas, X. Correlation between vacancy formation and Σ3 grain boundary structu nickel from atomistic simulations. Philos. Mag. 96, 2088–2114 (2016). h Hallil, A., Metsue, A., Bouhattate, J. & Feaugas, X. Correlation between vacancy formation and Σ3 grain boundary structures in ickel from atomistic simulations Philos Mag 96 2088 2114 (2016) g 59. Dingreville, R. & Berbenni, S. On the interaction of solutes with grain boundaries. Acta Mater. 104, 237–249 (2016). 59. Dingreville, R. & Berbenni, S. On the interaction of solutes with grain boundaries. Acta Mater. 104, 237 249 (2016). 60. Du, Y. A. et al. First-principles study on the interaction of H interstitials with grain boundaries in α- and γ-Fe. Phys. Rev. B 84 144121 (2011). 60. Du, Y. A. et al. First-principles study on the interaction of H interstitials with grain boundaries in α- and γ-Fe. Phys. Rev. 144121 (2011).f ( ) 61. Zhou, X., Mousseau, N. & Song, J. Is hydrogen diffusion along grain boundaries fast or slow? Atomistic origin and mechanistic modeling. Phys. Rev. Lett. 122(21), 215501 (2019). g y 2. Olmsted, D. L., Foiles, S. M. & Holm, E. A. Survey of computed grain boundary properties in face-centered cubic metals: I. Grain boundary energy. Acta Mater. 57, 3694–3703 (2009). y gy ( ) 63. Lejcek, P. Grain Boundary Segregation in Metals (Springer, 2010).h y gy 63. Lejcek, P. Grain Boundary Segregation in Metals (Springer, 2010).h 64. Priester, L. Grain Boundaries: From Theory to Engineering (Sp h y g g p g 65. Kirchheim, R. www.nature.com/scientificreports/ & Hirth, J. P. Stress and solubility for solutes with asymmetrical distortion fields. Acta Metall. 35, 2899–2903 (1987). 66. Dyre, J. C. Dominance of shear elastic energy far from a point defect in a solid. Phys. Rev. B 75, 092102 (2007). h 65. Kirchheim, R. & Hirth, J. P. Stress and solubility for solutes with asymmetrical distortion fields. Acta Metall. 35, 2899–2903 (1 66 D J C D i f h l ti f f i t d f t i lid Ph R B 75 092102 (2007) h 65. Kirchheim, R. & Hirth, J. P. Stress and solubility for solutes with asymmetrical d J y yi ( ) 66. Dyre, J. C. Dominance of shear elastic energy far from a point defect in a solid. Phys. Rev. B 75, 092102 (2007). 66. Dyre, J. C. Dominance of shear elastic energy far from a po 66. Dyre, J. C. Dominance of shear elastic energy far from a point defect in a solid. Phys. Rev. B 75, 092102 (2007). 67. Traisnel, C., Metsue, A., Oudriss, A., Bouhattate, J. & Feaugas, X. About the crystallographic orientation impact on hydrogen solubility and diffusivity near surface of nickel single crystals: A connection between long range and short-range energies. Comput. Mater. Sci. 1, 110136 (2020).f ( ) 8. Taketomi, S., Matsumoto, R. & Miyazaki, N. Atomistic study of hydrogen distribution and diffusion around a {112}<111> edge dislocation in alpha iron. Acta Mater. 56, 3761–3769 (2008).li p 69. Matsumoto, R., Inoue, Y., Taketomi, S. & Miyazaki, N. Influence of shear strain on the hydrogen trapped in bcc-Fe: A first- principles-based study. Scr. Mater. 60, 555–558 (2009).i 69. Matsumoto, R., Inoue, Y., Taketomi, S. & Miyazaki, N principles-based study. Scr. Mater. 60, 555–558 (2009).i principles-based study. Scr. Mater. 60, 555–558 (2009).i p p y 70. Dudarev, S. L. & Ma, P. W. Elastic fields, dipole tensors and interaction between self-interstitial atom defects in transition metals. Phys. Rev. Mater. 2, 033602 (2018).f 1. Hachet, G. et al. A multi-scale analysis of the different interactions between defects T and hydrogen: A review on the contribution of the elastic fields. Eng. Fract. Mech. 218, 10662 (2019). 71. Hachet, G. et al. A multi-scale analysis of the different inte f of the elastic fields. Eng. Fract. Mech. 218, 10662 (2019). i 72. Nazarov, R. et al. www.nature.com/scientificreports/ https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ & Daw, M. S. Calculation of hydrogen dissociation pathways on nickel using the embedded atom method. J. Less Common Metals 130, 465–473 (1987).hf atom method. J. Less Common Metals 130, 465–473 (1987).h 7. Devanathan, M. A. V. & Stachurski, Z. The adsorption and diffusion of electrolytic hydrogen in palladium. Proc. R. Soc. A. 170 1340 (1962).i 98. Henkelman, G., Uberuaga, B. P. & Jónsson, H. A climbing image nudged elastic band method for finding saddle points and mini- mum energy paths. J. Chem. Phys. 113, 9901–9904 (2000). https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 | www.nature.com/scientificreports/ 99. Jónsson, H., Mills, G. & Jacobsen, K. W. Nudged elastic band method for finding minimum energy paths of transitions. Class. Quantum Dyn. Condens. Phase Simul. 1, 385–404 (1998). Competing interests h p g The authors declare no competing interests. Acknowledgementsh g The authors would like to acknowledge our colleagues E. Conforto, R. Milet and C. Berziou for their contribu- tions to the TEM and hydrogen charging facilities in the LaSIE Laboratory. This work is supported by the French National Research Agency (ANR) through Grant ANR-13-JS09-0015 CRISTALHYD. Author contributionsi J.B., A.O., A.M. and X.F. defined the selection of grain boundaries. J.L., A.O. and X.F. characterized the grain boundaries using TEM and SEM/EBSD and the hydrogen states based on TDS measurements. J.L. and M.H. conducted the atomistic calculation. X.F., A.M., A.O. and J.B. carried out the elastic calculation of stress field and elastic energy and manage this work. All the authors contribute to data interpretation and to manuscript preparation. © The Author(s) 2021 Additional information Correspondence and requests for materials should be addressed to X.F. Correspondence and requests for materials should be addressed to X.F. Reprints and permissions information is available at www.nature.com/reprints. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access  This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. © The Author(s) 2021 https://doi.org/10.1038/s41598-021-94107-6 Scientific Reports | (2021) 11:15533 |

https://openalex.org/W3193840685

https://iej.zaslavsky.com.ua/index.php/journal/article/download/1085/1777

Ukrainian

Pharmacokinetic characteristics and morphometric effects of sodium-glucose contransporter-2 inhibitors in men and women with type 2 diabetes mellitus (literature review and own results)

Mìžnarodnij endokrinologìčnij žurnal

cc-by

For citation: Mìžnarodnij endokrinologìčnij žurnal. 2021;17(4):293-303. doi: 10.22141/2224-0721.17.4.2021.237342 Резюме. Актуальність. Згідно з останніми міжнародними клінічними рекомендаціями, препарати групи глі­ флозинів — інгібіторів натрійзалежних котранспортерів глюкози 2-го типу (іНЗКТГ-2) показані як перораль­ні цукрознижувальні засоби 2–3-ї лінії терапії цукрового діабету (ЦД) 2-го типу. Завдяки незалежній від інсу­ ліну стимуляції глюкозурії гліфлозини демонструють такі позаглікемічні ефекти, як зменшення маси тіла, поліпшення регіонального розподілу жирової тканини, ліпідного спектра плазми крові, зниження рівня урике­ мії, що в цілому сприяє зменшенню ризику серцево-судинних ускладнень. Метою даної роботи була оцінка ефективності дапагліфлозину в терапії чоловіків і жінок, хворих на ЦД 2-го типу, з метаболічно нездоровим фенотипом. Матеріали та методи. До дослідження включено 17 хворих на ЦД 2-го типу (11 чоловіків і 6 жі­ нок) віком 58,0 ± 1,7 року (95% довірчий інтервал 53–62), яким проводилась оцінка показників композиції тіла методом біоелектричного імпедансу за допомогою аналізатора Tanita BC-545N (Японія). Пацієнти от­ римували терапію дапагліфлозином, антигіпертензивними й антигіперліпідемічними засобами (статинами). Результати. Тримісячне застосування дапагліфлозину в дозі 10 мг 1 раз на добу викликало в пацієнтів зни­ ження індексу маси тіла, окружності талії, поліпшення показників композиції тіла, зокрема зниження відсо­ тка загального жиру (вірогідність змін показників визначали з використанням парного t-тесту). Не виявлено вірогідних змін показників м’язової та кісткової маси, оцінки будови тіла, а також ліпідного комплексу та рівня урикемії. Особливістю групи жінок, на відміну від чоловіків, було зменшення рівня вісцерального жиру, що супроводжувалось покращенням водного забезпечення організму, зниженням розрахункового метаболічно­ го віку. Висновки. Терапія хворих на ЦД 2-го типу препаратами групи іНЗКТГ-2 вже впродовж трьох місяців забезпечила зменшення ступеня ожиріння та поліпшення деяких показників композиції тіла. Підтвердження цієї тенденції може бути отримано при подальшому спостереженні. Ключові слова:  цукровий діабет 2-го типу; дапагліфлозин; композиція тіла; вісцеральний жир; урикемія; Ключові слова:  цукровий діабет 2-го типу; дапагліфлозин; композиція тіла; вісцеральний жир; урикем ліпідний спектр published article, with the obligatory reference to the authors of original works and original publication in this journal. Для кореспонденції:  Прибила Ольга Володимирівна, молодший науковий співробітник відділу вікової ендокринології та клінічної фармакології, ДУ «Інститут ендокринології та обміну речовин ім. В.П. Комісаренка НАМН України», вул. Вишгородська, 69, м. Київ, 04114, Україна; e-mail: [email protected] For correspondence: Olha Prybyla, Junior Research Fellow at the Department of Age Endocrinology and Clinical Pharmacology, State Institution “V.P. Komisarenko Institute of Endocrinology and Meta­ bolism of the NAMS of Ukraine”, Vyshgorodska st., 69, Kyiv, 04114, Ukraine; e-mail: [email protected] Full list of author information is available at the end of the article. Огляд літератури ефектів, як збільшення маси тіла, гіпоглікемія, затрим­ ка рідини, шлунково-кишкові симптоми. Незважаючи на застосування сучасних протидіабе­ тичних засобів, що впливають на різні патофізіологічні механізми при цукровому діабеті (ЦД) 2-го типу, у де­ яких пацієнтів не вдається досягти стабільного гліке­ мічного контролю та запобігти розвитку діабетичних ускладнень [1, 2]. Терапевтична цінність багатьох кла­ сів препаратів знижується внаслідок таких побічних Інгібітори натрійзалежних котранспортерів глю­ кози 2-го типу (іНЗКТГ-2) є відносно новим класом пероральних антидіабетичних препаратів, що зни­ жують гіперглікемію шляхом посилення виведення глюкози з сечею незалежно від секреції або дії інсу­ ліну [3]. © 2021. The Authors. This is an open access article under the terms of the  Creative Commons Attribution 4.0 International License, CC BY,   which allows others to freely distribute the published article, with the obligatory reference to the authors of original works and original publication in this journal. DOI:  https://doi.org/10.22141/2224-0721.17.4.2021.237342 DOI:  https://doi.org/10.22141/2224-0721.17.4.2021.237342 УДК  616.379-008.64-082-056.76 Оригінальні дослідження /Original Researches/ © 2021. The Authors. This is an open access article under the terms of the  Creative Commons Attribution 4.0 International License, CC BY,   which allows others to freely distribute the published article, with the obligatory reference to the authors of original works and original publication in this journal. Інгібітори НЗКТГ-2 — препарати групи гліфлозинів Відомо, що за нормального фізіологічного ста­ ну нирки фільтрують і реабсорбують приблизно 180 г глюкози на день. Майже вся відфільтрована глюкоза реабсорбується у проксимальних згорнутих канальцях нефрона, головним чином через НЗКТГ-2, меншою мірою — через натрійзалежний котранспортер глюко­ зи 1-го типу (НЗКТГ-1). Встановлено, що при ЦД 2-го типу реабсорбція глюкози в ниркових канальцях є під­ вищеною [5]. Частково завершені клінічні дослідження й отрима­ но схвалення Управління з санітарного нагляду за якіс­ тю харчових продуктів та медикаментів (Food and Drug Administration — FDA) та Європейського агентства з лі­ карських засобів (European Medicines Agency — EMA): —  для канагліфлозину (торгова назва інвокана (Invokana), фірма Johnson & Johnson), зареєстровано FDA — 29.03.2013, EMA — 15.11.2013; Ідея про можливість лікування ЦД шляхом стиму­ ляції глюкозурії як варіант більш фізіологічного підходу до контролю глікемії висловлена давно. Першою знай­ деною субстанцією, що неселективно гальмує актив­ ність транспортерів глюкози в нирках, був природний О-глікозид флоризин, виділений 1835 року французь­ кими хіміками з кори яблуні. У панкреатомованих щу­ рів флоризин викликав глюкозурію, що призводила до гіпоглікемії, шляхом інгібування ниркових транспор­ терів глюкози, зокрема НЗКТГ-2 [6]. Протидіабетичні властивості препарату вивчалися в експериментальних дослідженнях in vitro і на різних видах тварин. У них було показано, що за рахунок пригнічення НЗКТГ-1 і НЗКТГ-2 флоризин підвищує секрецію глюкози з се­ чею, нормалізує її рівень у плазмі крові, не викликаючи гіпоглікемії. Недоліками були низька біодоступність при прийомі всередину і потенційна токсичність, у зв’язку з чим його розробка як антидіабетичного пре­ —  дапагліфлозину (комерційна назва форксига (Forxiga), фірма Bristol-Myers Squibb), зареєстровано FDA — 08.01.2014, EMA — 12.11.2012; —  емпагліфлозину (торгова назва джардинс (Jardiance), фірма Boehringer Ingelheim Pharmaceuticals, Inc, Boehringer Ingelheim Pharmaceuticals, Inc), зареє­ стровано FDA — 01.08.2014, EMA — 22.05.2014; —  ертугліфлозину (торгова назва стеглатро (Stegla­ tro), фірма Merck Sharp & Dohme Corp.), зареєстровано FDA — 19.12.2017, EMA — 03.2018. В Японії на даний час зареєстровано вже шість іНЗКТГ-2: іпрагліфлозин, дапагліфлозин, тофогліфло­ зин, канагліфлозин, емпагліфлозин, люзеогліфлозин. Із початком клінічного застосування дапагліфло­ зину  — високоселективного іНЗКТГ-2 виникли нові можливості покращити контроль глікемії у пацієнтів з ЦД 2-го типу як при застосуванні монотерапії [6, 7], Таблиця 1. Для кореспонденції:  Прибила Ольга Володимирівна, молодший науковий співробітник відділу вікової ендокринології та клінічної фармакології, ДУ «Інститут ендокринології та обміну речовин ім. В.П. Комісаренка НАМН України», вул. Вишгородська, 69, м. Київ, 04114, Україна; e-mail: [email protected] For correspondence: Olha Prybyla, Junior Research Fellow at the Department of Age Endocrinology and Clinical Pharmacology, State Institution “V.P. Komisarenko Institute of Endocrinology and Meta­ bolism of the NAMS of Ukraine”, Vyshgorodska st., 69, Kyiv, 04114, Ukraine; e-mail: [email protected] Full list of author information is available at the end of the article. Vol. 17, No. 4, 2021 http://iej.zaslavsky.com.ua 293 Оригінальні дослідження  /Original Researches/ парату була припинена. Оскільки зазначені недоліки флоризину пов’язували в основному з неселективною дією, на основі його структури були розроблені речо­ вини з переважним впливом на НЗКТГ-2 [6]. Хоча Американська діабетична асоціація та Євро­ пейська асоціація з вивчення діабету рекомендують препарати даної групи для лікування пацієнтів з ЦД 2-го типу та серцево-судинними захворюваннями [3, 4], особливо з хронічною серцевою недостатністю, гліфло­ зини мають достатньо недосліджених механізмів. парату була припинена. Оскільки зазначені недоліки флоризину пов’язували в основному з неселективною дією, на основі його структури були розроблені речо­ вини з переважним впливом на НЗКТГ-2 [6]. За останнє десятиліття синтезована низка сполук, які селективно гальмують НЗКТГ-2. Численні фарма­ цевтичні компанії досліджують застосування різних синтетичних похідних флоризину у вигляді монотера­ пії та в комбінаціях з іншими препаратами. На даний час більше десяти іНЗКТГ-2 знаходяться в стадії роз­ робки (табл. 1). Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Інгібітори НЗКТГ-2 — препарати групи гліфлозинів Гіпоурикемічна дія гліфлозинів здійснюється за ра­ хунок посилення екскреції сечової кислоти (СК) з се­ чею (урикозурії). Так, під час прийому дапагліфлозину урикозурія була підвищена (протягом 3–7 днів) і супро­ воджувалася стійким зменшенням концентрації СК у сироватці крові. Хоча точні механізми, за допомогою яких іНЗКТГ-2 викликають гальмування реабсорбції і зниження рівня СК у крові, поки повністю не розкриті, вважають, що вони можуть включати прямий вплив на нирковий транспорт уратів або непрямий ефект, вто­ ринний щодо зменшення реабсорбції натрію в прокси­ мальних канальцях. Оскільки сироваткова концентра­ ція СК змінюється паралельно з поглинанням натрію нирками, зниження реабсорбції і посилення екскреції СК може бути також результатом блокування реаб­ сорбції натрію і глюкози гліфлозинами [14, 15]. Клінічні дослідження показали, що гліфлозини в цілому при використанні в лікуванні пацієнтів з ЦД 1-го типу та ЦД 2-го типу з метаболічним синдромом (у вигляді монотерапії або на тлі лікування іншими протидіабетичними препаратами) демонструють зна­ чне поліпшення глікемічного контролю, тобто знижу­ ють рівень HbA1c, глюкози крові натще і постпранді­ ально. Важливою перевагою іНЗКТГ-2 порівняно з іншими пероральними цукрознижувальними препара­ тами є ефект зниження маси тіла (на 0,3–1,0 %, або на 2,0–3,4 кг), що обумовлено втратою калорій у зв’язку зі збільшенням ниркової елімінації глюкози; крім того, відзначають зниження систолічного та діастолічного артеріального тиску (АТ) (відповідно на 1,7–6,4 і 0,3– 2,0 мм рт.ст.) [11]. Інгібітори НЗКТГ-2 мають подібні фармакокіне­ тичні властивості (табл. 2), характеризуючись швид­ кою абсорбцією при пероральному прийомі, тривалим Таблиця 2. Інгібітори НЗКТГ-2 — препарати групи гліфлозинів Препарати групи іНЗКТГ-2, що знаходяться на різних стадіях клінічних досліджень Препарат Виробник Фаза клінічних досліджень Дапагліфлозин (Dapagliflozin) BMS-512148 Bristol-Myers Squibb/ AstraZeneca Фаза III — схвалено в Австралії (2012), Європі (2012), Мексиці (2013), Новій Зеландії (2013), Бразилії (2013), Аргентині (2013), США (2014), Японії (2014) Канагліфлозин (Canagliflozin) TA-7284, JAJ-28431754 Johnson and Johnson і Mitsubishi Tanabe Pharma Фаза III — схвалено в США (2013) та Європі (2013) Емпагліфлозин (Empagliflozin) BI-10773 Boehringer Ingelheim/Lilly Фаза III — схвалено в Європі (2014) Іпрагліфлозин (Ipragliflozin) ASP-1941 Astellas Pharma та Kotobuki Pharmaceutical Company Фаза III — схвалено в Японії (2014) Тофогліфлозин (Tofogliflozin) CSG452 Roche/Chugai Фаза III Люсеогліфлозин (Luseogliflozin) TS-071 Taisho Фаза III BI 44847 Boehringer Ingelheim Фаза III LX4211 Lexicon Pharmaceutical Фаза II PF-04971729 Pfizer Фаза II EGT0001442 Theracos Фаза II GW 869682 GlaxoSmithKline Фаза II Таблиця 1. Препарати групи іНЗКТГ-2, що знаходяться на різних стадіях клінічних дос Препарати групи іНЗКТГ-2, що знаходяться на різних стадіях клінічних досліджень Vol. 17, No. 4, 2021 294 Оригінальні дослідження  /Original Researches/ Інгібітори НЗКТГ-2 є ліпофільними і мають високу спорідненість до клітинної мембрани. Це означає, що препарати можуть діяти довше тривалості дії, визначеної виходячи з фармакологічного періоду напіввиведення, і, таким чином, можуть проявляти активність упродовж майже усього дня, після введення вранці чи ввечері, на­ віть у стані голодування або після прийому їжі. так і при додаванні до терапії метформіном, глімепіри­ дом або інсуліном [8, 9]. У цих же перших досліджен­ нях була продемонстрована помірна втрата маси тіла у пацієнтів, що пов’язують з діуретичним ефектом і постійною втратою калорій внаслідок стимуляції глю­ козурії (70–80 г/добу), на тлі зменшення рівня гліко­ ваного гемоглобіну (HbA1c) як основного ефекту. Крім того, з використанням методів магнітного резонансу показано, що дапагліфлозин впливає на регіональний розподіл жирової тканини і вміст ліпідів у печінці [10]. Гліфлозини проявляють також позитивні пле­ йотропні (позаглікемічні) метаболічні ефекти, не пов’язані з контролем рівня глюкози у крові: покра­ щують ліпідний спектр, знижують рівень урикемії на 5,9–17,8 %, знижують АТ і діабетичну клубочкову гі­ перфільтрацію, які можуть сприяти зменшенню ризи­ ку характерних для захворювання на ЦД ускладнень з боку нирок і серцево-судинної системи [12, 13]. Інгібування НЗКТГ-2 за допомогою гліфлозинів зменшує рівень канальцевої реабсорбції глюкози на 30–50 %, знижує нирковий поріг для глюкози і поси­ лює виведення глюкози із сечею, що призводить до зниження глюкози в плазмі крові у пацієнтів із гіпер­ глікемією, а також помірно стимулює осмотичний ді­ урез. Гіпоглікемічна дія не залежить від периферичних ефектів інсуліну і функції бета-клітин підшлункової залози [5]. Інгібітори НЗКТГ-2 — препарати групи гліфлозинів Повідомлялося, що приблизно дві тре­ тини втрати маси тіла при лікуванні канагліфлозином припадають на втрату жиру. Крім того, зниження маси тіла при лікуванні дапагліфлозином можна пояснити зменшенням загальної маси жиру, вісцеральної жиро­ вої тканини та об’єму підшкірної жирової клітковини. Початкова втрата маси тіла також може бути виклика­ на втратою рідини за рахунок легкого осмотичного ді­ урезу; однак останні дані доводять, що втрата маси тіла продовжується навіть після нормалізації діурезу [22]. Інгібітори НЗКТГ-2, як інгібітори осмотичного ді­ урезу, у першу чергу пригнічують котранспорт глюкози та натрію з просвітів проксимальних канальців до кро­ ві. За умов гальмування транспортерів НЗКТГ-2 у про­ ксимальних канальцях і петлі Генле відбуваються певні вторинні зміни. Проксимальні канальці мають кілька механізмів реабсорбції Na+, завдяки чому поглинання Na+ не припиняється повністю навіть після введення іНЗКТГ-2. З іншого боку, іНЗКТГ-2 інгібують реаб­ сорбцію глюкози, і, отже, глюкоза діє як речовина, яка не може бути резорбована. Проксимальні канальці є добре проникними для води, тому рідина в цих каналь­ цях залишається ізотонічною щодо крові [15]. Осмотичний діурез еквівалентний діурезу, виклика­ ному петльовими діуретиками. Коли концентрація Cl– знижується за рахунок осмотичного діурезу, канальцева реабсорбція в петлі Генле гальмується, внаслідок чого відбувається сольовий діурез (рис. 1). Цей механізм від­ різняється від дії петльового діуретика (фуросеміду), який зв’язується з Cl-місцем транспортера, але кінцеві результати аналогічні [15]. Хоча реабсорбція в прокси­ мальних канальцях також фактично гальмується через осмотичний діурез шляхом інгібування НЗКТГ-2, вплив осмотичного діурезу на канальцеву реабсорбцію кількіс­ но малий. Припускають, що сечогінна дія, індукована інгібуванням НЗКТГ-2, ґрунтується саме на вторинно­ му пригніченні реабсорбції в петлі Генле. Отже, інгібі­ тори НЗКТГ-2 діють як петльові діуретики, хоча прямих доказів цього немає. Що стосується активації ренін-ан­ гіотензин-адреналової системи (РААС), то інгібування НЗКТГ-2 викликає незначне підвищення плазмових рівнів таких гормонів, як альдостерон та ангіотензин II, але в межах діапазону, характерного для ЦД (парадокс РААС), і аналогічно збільшується сечовипускання [17]. Численні клінічні дослідження, що відповіда­ ють принципам доказової медицини, показали добру ефективність, переносимість і безпеку застосування іНЗКТГ-2 у хворих на ЦД 1-го і 2-го типу. Найбільш частими побічними ефектами були грибкові інфекції жіночих статевих органів [18–20]. У клінічному дослідженні 2010 р. глюкозурія збері­ галася протягом певного часу, тим часом як втрата маси тіла досягала плато і була набагато меншою від очікува­ ної втрати калорій, розрахованої за величиною щоден­ ної глюкозурії, що вказує на відсутність змін у спожи­ ванні енергії [23]. Інгібітори НЗКТГ-2 — препарати групи гліфлозинів Криві втрати маси тіла, що фактично спостерігались, почали розходитися з прогнозовани­ ми моделлю на 24-му тижні, коли маса тіла пацієнтів стабілізувалася. Таким чином, споживання калорій повинно було збільшитися, щоб пояснити цю невід­ повідність. Даний ефект відрізняє зниження маси тіла під впливом іНЗКТГ-2 від зниження при застосуванні агоністів рецепторів глюкагоноподібного пептиду-1 (ГПП-1). Хоча ефекти обох груп препаратів кількісно схожі, зниження маси тіла досягається при нормаль­ ному/підвищеному споживанні калорій під час терапії іНЗКТГ-2, тоді як антагоніст рецепторів ГПП-1 ви­ кликає відчуття ситості та зменшує споживання кало­ рій. Тим не менш втрата маси тіла, досягнута в перші тижні лікування іНЗКТГ-2, зберігається з часом [24]. Інгібітори НЗКТГ-2 — препарати групи гліфлозинів Порівняння фармакологічних характеристик інгібіторів НЗКТГ-2, схвалених у США та Європі Препарат Канагліфлозин (Invokana) Дапагліфлозин (Farxiga) Емпагліфлозин (Jardiance) Ертугліфлозин (Steglatro) Біодоступність, % 65 78 – 100 Tmax, год 1–2 < 2 1,5 1,0 Період напівжиття, год 10,6–13,1 10,6–13,1 12,4 16,6 Метаболізм У печінці (O-глюкуронізація через UGT1A9 та 2B4; 2 неактив­ них метаболіти) У печінці (O-глю­ куронізація через UGT1A9; 1 неактив­ ний метаболіт) У печінці (O-глю­ куронізація через UGT1A3, 2B7, 1А8, 1А9) У печінці (O-глюкуро­ нізація через UGT1A9 та 2B7; мінімальне (12 %) меди­йо­ване CYP окислення) Екскреція, % Фекалії — 60 % Сеча — 33 % Фекалії — 21 % Сеча — 75 % Фекалії — 41 % Сеча — 54 % Фекалії — 41 % Сеча — 50 % Вибірковість НЗКТГ-1 щодо НЗКТГ-2 1 : 414 1 : 1200 1 : 2500 1 : 2200 Пікові рівні погли­ нання після дозу­ вання 2,8–4,0 1,5 1,5 – Денні дозування 100–300 мг, перед першим денним прийомом їжі 5–10 мг, зранку незалежно від прийому їжі 10–25 мг, незалеж­ но від прийому їжі 5–15 мг, зранку неза­ лежно від прийому їжі Примітки: UGT — глюкуронозилтрансфераза; CYP — цитохром Р450. Таблиця 2. Порівняння фармакологічних характеристик інгібіторів НЗКТГ-2, схвалених у США та Європі Примітки: UGT — глюкуронозилтрансфераза; CYP — цитохром Р450. Vol. 17, No. 4, 2021 295 Оригінальні дослідження  /Original Researches/ періодом напіввиведення, що дозволяє приймати пре­ парат 1 раз на добу, ефективним метаболізмом у печінці (головним чином шляхом глюкуронізації з утворенням неактивних метаболітів), відсутністю клінічно значу­ щих лікарських взаємодій і низьким виведенням нир­ ками у формі вихідної речовини [16]. вік  — 57,4  ±  14,3 року (M  ±  SD), тривалість хворо­ би — 7,4 ± 3,1 року, з нормальним рівнем клубочкової фільтрації та різним ступенем альбумінурії. Пацієнти отримували іНЗКТГ-2 (дапагліфлозин або емпагліф­ лозин) та проходили оцінювання за допомогою методу біоімпедансу. Результати показали, що на початку до­ слідження більш ніж у 65  % пацієнтів спостерігалася надмірна гідратація без клінічних ознак. Після курсу лікування з використанням гліфлозинів у пацієнтів відзначено вірогідне зменшення маси тіла, яке демон­ струвало двофазну динаміку. Зменшення маси тіла за рахунок зменшення загального вмісту води в організмі спостерігалося при першому обстеженні (через 1 мі­ сяць) та було пов’язано з сечогінним ефектом. Через 6 місяців втрата маси тіла зберігалася або навіть збільшу­ валася за рахунок зменшення загальної жирової маси. Отже, якщо початкова втрата маси тіла викликана ді­ уретичною дією, то довготривале зниження маси тіла пояснюється зменшенням жирової маси через тривалу втрату калорій. Клінічні дослідження ефектів іНЗКТГ-2 на композицію тіла у хворих на ЦД 2-го типу Дослідження тенденцій щодо залежності титра­ ції дози та динаміки змін маси тіла для шести різних іНЗКТГ-2 показало, що при використанні дапагліф­ лозину існував статистично значущий взаємозв’язок між дозою препаратів та зниженням маси тіла пацієнтів (P < 0,05). Однак результати лікування канагліфлозином, емпагліфлозином, іпрагліфлозином, тофогліфлозином та люзеогліфлозином не дали таких результатів [25]. Позитивний вплив дапагліфлозину на композицію тіла продемонстровано в низці клінічних досліджень, дизайн яких відповідає сучасним вимогам щодо рівня доказовості. Дослідження з дапагліфлозином показа­ ло значне зменшення окружності талії, що відповідає зменшенню жирової маси за рахунок вісцерального або підшкірного жиру [10]. Аналіз ефективності іНЗКТГ-2 відповідно до роз­ поділу груп за етнічними характеристиками (азіатська та неазіатська етнічність) та індексу маси тіла (ІМТ) Наприклад, до обсерваційного дослідження, про­ веденого в Іспанії (2014–2016 рр.), включили 50 па­ цієнтів із ЦД 2-го типу, з них 28 чоловіків, середній Vol. 17, No. 4, 2021 296 Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Оригінальні дослідження  /Original Researches/ http://iej.zaslavsky.com.ua 297 Vol. 17, No. 4, 2021 Таблиця 3. Клінічні дослідження ефектів іНЗКТГ-2 на композицію тіла у хворих на ЦД 2-го типу 3) показав, що дозозалежні зміни маси тіла у пацієнтів з ІМТ < 30 кг/м2 були подібними до та­ ких у пацієнтів з ІМТ понад 30 кг/м2 [25]. патитом, підтвердженим шляхом біопсії, вивчали у відкритому нерандомізованому дослідженні, в якому після курсу лікування в пацієнтів виявлено вірогідне зменшення ІМТ, відсоткового вмісту жиру в організмі (P < 0,01), зменшення вісцерального ожиріння (оціню­ ваного за окружністю талії), без суттєвих змін безжиро­ вої маси тіла та вмісту води в організмі. На тлі поліп­ шення метаболічних показників (зниження інсулінемії та рівня глікемії натще), підвищення рівня адипонек­ тину зафіксовано суттєве покращення рівнів печінко­ вих ферментів, феритину та колагену ІV [27]. У 24-тижневому міжнародному багатоцентровому рандомізованому подвійному сліпому плацебо-конт­ рольованому дослідженні брали участь 182 пацієнти із ЦД 2-го типу (середній вік: жінки — 63,3 року, чо­ ловіки — 58,6 року; ІМТ — 31,9 кг/м2) із незадовіль­ ним контролем глікемії при лікуванні метформіном (HbA1c  — 7,17  %), яким призначали дапагліфлозин (10  мг/добу) або плацебо. Через 24 тижні лікування дапагліфлозином зафіксовано такі зміни показників (після корекції на відповідні рівні у групі плацебо): зниження маси тіла (–2,08 кг; 95% довірчий інтервал (ДІ) –2,84 — –1,31; P < 0,0001); зменшення окружнос­ ті талії (–1,52 см; 95% ДІ –2,74 — –0,31; Р < 0,0143). Методом двопроменевої радіоабсорбціометрії вияв­ лено зниження загальної жирової маси тіла (–1,48 кг; 95% ДІ –2,22 — –0,74; P < 0,0001), показників об’єму вісцеральної жирової тканини (–258,4 см3; 95% ДІ –448,1 — –68,6; P = 0,0084) та підшкірної жирової тка­ нини (–184,9 см3; 95% ДІ –359,7 — –10,1; P = 0,0385), тобто дапагліфлозин знижував масу тіла переважно за рахунок зменшення загального вмісту жиру в організ­ мі, кількості вісцерального та підшкірного жиру [10]. Щодо застосування іНЗКТГ-2 у літніх людей, то існує два негативних фактори, що зменшують прива­ бливість даної терапії, а саме: менша ефективність лі­ кування та більша частота деяких небажаних ефектів. Зокрема, причини більш частих переломів у пацієнтів, які лікувались канагліфлозином та дапагліфлозином, досі не з’ясовані. Не було зафіксовано помітних змін у сироватці або сечі рівнів кальцію, 1,25-дигідроксивіта­ міну D або паратиреоїдного гормону [10]. Однак у двох дослідженнях на тлі застосування канагліфлозину було зафіксовано незначне збільшення в сироватці крові од­ ного з маркерів резорбції кісток — колагену типу 1, а також β-карбокситермінального телопептиду [28, 29]. З огляду на ці результати, доки не з’являться додатко­ ві дані, слід з обережністю застосовувати іНЗКТГ-2 у пацієнтів з остеопорозом або переломами в анамнезі. Клінічні дослідження ефектів іНЗКТГ-2 на композицію тіла у хворих на ЦД 2-го типу , р р ру [ ] У поздовжньому обсерваційному дослідженні з використанням біоімпедансної спектроскопії аналі­ зували композицію тіла та рідинний стан (внутріш­ ньоклітинну гідратацію й позаклітинний вміст води) у 27 амбулаторних хворих на ЦД 2-го типу, які отри­ мували терапію іНЗКТ-2 (18 пацієнтів приймали ем­ пагліфлозин, 9  — дапагліфлозин). Групи порівняння становили 14 пацієнтів, які приймали гідрохлортіазид для лікування артеріальної гіпертензії, а також 16 здо­ рових осіб. Лікування іНЗКГТ упродовж шести міся­ ців сприяло зниженню HbA1c в середньому на 0,8 %, маси тіла та ІМТ — на 2,6 та 0,9 кг/м2 відповідно. Біо­ імпедансна спектроскопія виявила значне зниження маси жирової тканини та індексу жирової тканини, у той час як безжирова маса залишалася стабільною. По­ казники надмірної гідратації (овергідратації) та вмісту позаклітинної води вже на третій день зменшились на –0,5 л/1,73 м2 (–0,1; –0,9) і –0,4 л/1,73 м2 (–0,1; –0,8) відповідно, повертались до початкового значення че­ рез 3 та 6 місяців. Активність реніну в плазмі зросла в 2,1 раза (0,5; 3,6) за перший місяць і повернулася до по­ чаткового рівня через 3 та 6 місяців. Рідинний статус пацієнтів, які лікувались іНЗКТГ-2, після 6 місяців не відрізнявся від такого в групі пацієнтів, які приймали гідрохлортіазид, або в групі здорових людей. Застосу­ вання тестів Wilcoxon Signed-Rank не показало суттєвої різниці між підгрупами пацієнтів, які приймали емпа­ гліфлозин або дапагліфлозин. Результати показали, що зниження маси тіла під час лікування іНЗКТГ-2 обумов­лене зменшенням маси жирової тканини та тимчасовою втратою позаклітинної рідини, що супро­ воджується регуляцією РААС (рис. 2) [25, 26]. Відсутні рекомендації щодо корекції дози кана­ гліфлозину, дапагліфлозину, емпагліфлозину або ерту­ гліфлозину залежно від віку. Під час терапії іНЗКТГ-2 у пацієнтів віком 65 років і більше може бути підвище­ ний ризик виникнення внутрішньосудинних об’ємних побічних реакцій порівняно з молодшими пацієнта­ ми; для канагліфлозину це може відбуватись при дозі 300 мг, а більш помітне зростання захворюваності спо­ стерігалось у пацієнтів, яким було понад 75 років. По­ відомлялося про підвищений ризик розвитку інфекцій сечовивідних шляхів при лікуванні емпагліфлозином в осіб віком 75 років і старше. У молодших пацієнтів Люмінальна рідина Товста висхідна гілка петлі Генле Петльові діуретики АТР Na+ Na+ 2Cl– Cl– Mg++ Mg++ Са++ Са++ К+ К+ К+ Рисунок 1. Механізми реабсорбції в петлі Генле й осмотичний діурез [15] Примітки:  Са++ — катіон кальцію; К+ — катіон калію; Мg++ — катіон магнію; Na+ — катіон натрію; Cl– — ані­ он хлору; АТФ (ATP) — аденозинтрифосфат Рисунок 1. Клінічні дослідження ефектів іНЗКТГ-2 на композицію тіла у хворих на ЦД 2-го типу Зміна маси тіла від початкового коригованого плацебо рівня на тлі прийому різних препаратів іНЗКТГ-2 для різних етнічних груп і за різних показників початкового ІМТ (корекція за рівнями в групі плацебо) [25] ІНЗКТГ-2 (1 раз на добу) Зважена різниця середніх (ЗРС) від по­ чаткової маси тіла (кг) 95% ДІ Азіатська етногрупа Неазіатська етногрупа ІМТ < 30 кг/м2 ІМТ > 30 кг/м2 ЗРС (кг) (95% ДІ) ЗРС (кг) (95% ДІ) ЗРС (кг) (95% ДІ) ЗРС (кг) (95% ДІ) Дапагліфлозин (Dapagliflozin) 2,5 мг 21,30* 21,55–21,04 21,19 (21,26–21,12) 21,32 (21,67–20,96) 21,19 (21,26–21,12) 21,32 (21,67–20,96) 5 мг 21,51* 21,74–21,29 21,69 (22,32–21,06) 21,47 (21,73–21,21) 21,69 (22,32–21,06) 21,47 (21,73–21,21) 10 мг 21,79* 22,02–21,55 21,92 (22,04–21,80) 21,77 (22,05–21,49) 21,92 (22,04–21,80) 21,77 (22,05–21,49) 20 мг 22,24* 22,49–21,99 – 22,24 (22,49–21,99) – 22,24 (22,49–21,99) Канагліфлозин (Canagliflozin) 50 мг 21,20* 21,25–21,15 – 21,20 (21,25–21,15) – 21,20 (21,25–21,15) 100 мг 21,82* 22,10–21,53 21,74 (22,14–21,34) 21,85 (22,26–21,44) 21,74 (22,14–21,34) 21,85 (22,26–21,44) 200 мг 21,83* 22,37–21,30 21,95 (22,60–21,29) 21,60 (21,74–21,46) 21,95 (22,60–21,29) 21,60 (21,74–21,46) 300 мг 22,37* 22,82–21,93 22,01 (22,80–21,21) 22,47 (22,97–21,96) 22,01 (22,80–21,21) 22,47 (22,97–21,96) Емпагліфлозин (Empagliflozin) 10 мг 21,84* 21,98–21,69 22,79 (22,83–22,75) 21,84 (21,98–21,69) 21,89 (22,29–21,48) 22,03 (22,34–21,72) 25 мг 21,93* 22,08–21,77 22,15 (22,21–22,09) 21,90 (22,07–21,73) 21,93 (22,04–21,82) 21,94 (22,34–21,55) Іпрагліфлозин (Ipragliflozin) 12,5 мг 20,44* 20,63–20,26 – 20,44 (20,63–20,26) – 20,44 (20,63–20,26) 50 мг 21,40* 21,71–21,10 21,35 (21,66–21,04) 21,58 (21,77–21,38) 21,35 (21,66–21,04) 21,58 (21,77–21,38) 150 мг 21,49* 21,68–21,30 – 21,49 (21,68–21,30) – 21,49 (21,68–21,30) 300 мг 21,73* 21,92–21,54 – 21,73 (21,92–21,54) – 21,73 (21,92–21,54) Тофогліфлозин (Tofogliflozin) 10 мг 21,68* 22,06–21,31 21,68 (22,06–21,31) – 21,87 (22,03–21,72) 21,49 (21,66–21,30) 20 мг 22,15* 22,82–21,48 22,15 (22,82–21,48) – 22,50 (22,65–22,34) 21,81 (21,98–21,64) 40 мг 22,35* 22,87–21,83 22,35 (22,87–21,83) – 22,62 (22,77–22,46) 22,08 (22,25–21,92) Люзеогліфлозин (Luseogliflozin) 2,5 мг 21,54* 21,71–21,36 21,54 (21,71–21,36) – 21,54 (21,71–21,36) – 5 мг 21,92* 22,31–21,54 21,92 (22,31–21,54) – 21,92 (22,31–21,54) – Vol. 17, No. 4, 2021 http://iej.zaslavsky.com.ua 297 Оригінальні дослідження  /Original Researches/ патитом, підтвердженим шляхом біопсії, вивчали у відкритому нерандомізованому дослідженні, в якому після курсу лікування в пацієнтів виявлено вірогідне зменшення ІМТ, відсоткового вмісту жиру в організмі (P < 0,01), зменшення вісцерального ожиріння (оціню­ ваного за окружністю талії), без суттєвих змін безжиро­ вої маси тіла та вмісту води в організмі. На тлі поліп­ шення метаболічних показників (зниження інсулінемії та рівня глікемії натще), підвищення рівня адипонек­ тину зафіксовано суттєве покращення рівнів печінко­ вих ферментів, феритину та колагену ІV [27]. у групах (табл. Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Власне дослідження За мету даної роботи відділом вікової ендокриноло­ гії та клінічної фармакології ДУ «Інститут ендокрино­ логії та обміну речовин імені В.П. Комісаренка НАМН України» (м. Київ) визначено дослідження ефектив­ ності тримісячного застосування дапагліфлозину у те­ рапії чоловіків і жінок, хворих на ЦД 2-го типу, з мета­ болічно нездоровим фенотипом. Клінічні дослідження ефектів іНЗКТГ-2 на композицію тіла у хворих на ЦД 2-го типу Механізми реабсорбції в петлі Генле й осмотичний діурез [15] Ефективність застосування дапагліфлозину (5  мг/добу впродовж 24 тижнів) у терапії 16 хворих на ЦД 2-го типу із супутнім неалкогольним стеатоге­ Примітки:  Са++ — катіон кальцію; К+ — катіон калію; Мg++ — катіон магнію; Na+ — катіон натрію; Cl– — ані­ он хлору; АТФ (ATP) — аденозинтрифосфат Vol. 17, No. 4, 2021 298 Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Оригінальні дослідження  /Original Researches/ спостерігалося менше зниження рівня HbA1c відносно плацебо порівняно із пацієнтами старшого віку. Треба зазначити, що згадані контрольовані дослідження не включали достатню кількість пацієнтів, щоб визначити вірогідні зміни досліджуваних показників [30]. лікування та на проведення операції та знеболювання відповідно до Наказу Міністерства охорони здоров’я України від 14 лютого 2012 року № 110, розробленого на основі Гельсінської декларації 1975 року та її зміне­ ного й доповненого варіанта 2000 року. р р у Обстеження пацієнтів включало визначення антро­ пометричних параметрів (ріст, маса тіла, окружність талії та стегон  — ОТ, ОС). Обчислювали індекс цен­ трального ожиріння (ІЦО = ОТ/зріст). Композицію тіла оцінювали методом біоелектричного імпедансу. За допомогою аналізатора Tanita BC-545N (Японія) ви­ значали відсотковий вміст загального жиру в організмі (норма для чоловіків — 13–25 %, для жінок — 24–36 %) та води (норма для чоловіків — 50–65 %, для жінок — 45–60 %), вміст абдомінального жиру (в інтервалі від 1 до 59), м’язову та кісткову масу, а також сегментар­ ну структуру жирової та м’язової маси тулуба та кож­ ної кінцівки. Фізичний тип (будову тіла) оцінювали за шкалою від 1 до 9 балів, відповідно до критеріїв ана­ лізатора. Використовували функцію розрахунку ба­ зального метаболізму (кількість енергії, необхідної для підтримування основних життєвих функцій) та визна­ чення розрахункового метаболічного віку (в діапазоні від 12 до 90). Матеріали та методи До дослідження включено 17 хворих на ЦД 2-го типу (11 чоловіків і 6 жінок). Серед усіх пацієнтів 10 осіб мали загальне ожиріння (ІМТ ≥ 30 кг/м2, медіа­ на — 32,30 ± 1,63 кг/м2; 95% ДІ 24,2–29,4 кг/м2), решта 7 пацієнтів мали нормальну масу тіла (ІМТ < 30 кг/м2, медіана — 26,80 ± 0,85 кг/м2; 95% ДІ 29,9–35,5 кг/м2). Вік пацієнтів становив 58,0 ± 1,7 року; 95% ДІ 53–62 роки. Пацієнти отримували стабільну терапію гіпо­ глікемічними, атнигіпертензивними та антигіперлі­ підемічними засобами (статинами). Хворі проходили клінічне обстеження у відділенні вікової ендокрино­ логії та клінічної фармакології ДУ «Інститут ендокри­ нології та обміну речовин ім. В.П. Комісаренка НАМН України». Усі пацієнти підписували інформовану до­ бровільну згоду пацієнта на проведення діагностики, Для оцінки стану ліпідного обміну визначали концентрації в сироватці крові основних ліпідних фракцій: загального холестерину (ЗХС, нормальні величини  <  4,5 ммоль/л); тригліцеридів (ТГ, нор­ ма  <  1,7  ммоль/л); холестерину ліпопротеїнів низь­ кої щільності (ХС ЛПНЩ, середній нормальний рі­ вень < 3 ммоль/л; для хворих на ЦД 2-го типу ця норма Рисунок 2. Результати Пацієнтам призначали інгібітор НЗКТ2 дапагліф­ лозин у дозі 10 мг 1 раз на добу. Після трьох місяців терапії проводили повторне обстеження за всіма до­ сліджуваними показниками. Стан вуглеводного обміну характеризували за по­ казниками глікемії натще та HbA1c, рівнями інсуліну та С-пептиду в крові (результати оцінювали за норма­ ми, наданими сертифікованими лабораторіями м. Ки­ єва). Практично всі хворі мали некомпенсований ЦД 2-го типу (HbA1c > 7,5 %). З метою дослідження ефективності прийому да­ пагліфлозину у пацієнтів з некомпенсованим ЦД 2-го типу пацієнтам (із них 11 чоловіків і 6 жінок) призна­ чено лікування препаратом іНЗКТГ-2 тривалістю 3–4 місяці (табл. 4). Після лікування в усіх пацієнтів від­ значено зменшення ІМТ (Р < 0,05) — очікуваний ха­ рактерний ефект препаратів групи іНЗКТГ-2, але без вірогідного зменшення маси тіла. Дослідження схвалене комісією з біомедичної ети­ ки ДУ «Інститут ендокринології та обміну речовин іме­ ні В.П. Комісаренка НАМН України» (протокол № 1 від 05.02.2020). З огляду на невелику кількість пацієнтів для оцінки різниці показників у динаміці лікування застосовува­ ли парний t-тест, який дозволив виявити деякі суттєві зміни досліджуваних параметрів. Статистичний аналіз здійснювали методом варіа­ ційної статистики, використовуючи стандартні ста­ тистичні пакети MedStat v. 5.2. Результати наведені як медіана ± стандартна похибка (Ме ± m), з зазначенням нижнього і верхнього 95% довірчого інтервалу. Нор­ мальність розподілу отриманих результатів перевіряли З огляду на невелику кількість пацієнтів для оцінки різниці показників у динаміці лікування застосовува­ ли парний t-тест, який дозволив виявити деякі суттєві зміни досліджуваних параметрів. Не виявлено вірогідних змін показників м’язової та кісткової маси, оцінки будови тіла та рівня урикемії. Таблиця 4. Матеріали та методи Динаміка показників маси тіла (%), маси вісцерального жиру (%), зміни овергідратації (Δ, л/1,73 м2) та позаклітинної води (Δ, л/1,73 м2) у пацієнтів з ЦД 2-го типу на тлі прийому інгібіторів НЗКТГ-2 [25] Дні 100 95 90 85 0 n = 27 0 3 n = 26 –1,2 кг (–1,7; –0,6) 180 n = 14 –2,6 кг (–1,5; –9,3) 90 n = 16 –3,6 кг (–1,6; –8,4) 30 n = 16 –2,7 кг (–5,7; –1,0) Маса тіла (%) Зміни овергідратації (Δ, л/1,73 м2) Дні 0,5 0,0 –0,5 –1,0 0 n = 25 0 3 n = 24 –0,5 л/1,73 м2 (–0,9; –0,1) 180 n = 12 0,3 л/1,73 м2 (0,1; 0,5) 90 n = 13 –0,1 л/1,73 м2 (–0,4; 0,3) 30 n = 15 –0,2 л/1,73 м2 (–0,6; 0,2) Зміни позаклітинної води (Δ, л/1,73 м2) Дні 0,0 –0,5 –1,0 0 n = 25 0 3 n = 24 –0,4 л/1,73 м2 (–0,8; –0,1) 180 n = 12 0,01 л/1,73 м2 (–0,4; 0,1) 90 n = 13 –0,2 л/1,73 м2 (–0,8; 0,1) 30 n = 15 –0,4 л/1,73 м2 (–0,8; –0,0) Маса вісцерального жиру (%) Дні 100 90 80 70 0 n = 25 0 3 n = 27 –0,5 кг (–1,7; 0,5) 180 n = 14 –7,0 кг (–12,2; –2,4) 90 n = 16 –3,2 кг (–7,4; –0,3) 30 n = 17 –1,4 кг (–3,4; 0,7) Зміни позаклітинної води (Δ, л/1,73 м2) Дні 0,0 –0,5 –1,0 0 n = 25 0 3 n = 24 –0,4 л/1,73 м2 (–0,8; –0,1) 180 n = 12 0,01 л/1,73 м2 (–0,4; 0,1) 90 n = 13 –0,2 л/1,73 м2 (–0,8; 0,1) 30 n = 15 –0,4 л/1,73 м2 (–0,8; –0,0) Маса вісцерального жиру (%) Дні 100 90 80 70 0 n = 25 0 3 n = 27 –0,5 кг (–1,7; 0,5) 180 n = 14 –7,0 кг (–12,2; –2,4) 90 n = 16 –3,2 кг (–7,4; –0,3) 30 n = 17 –1,4 кг (–3,4; 0,7) Маса вісцерального жиру (%) Зміни позаклітинної води (Δ л/1 73 м2) Рисунок 2. Динаміка показників маси тіла (%), маси вісцерального жиру (%), зміни овергідратації (Δ, л/1,73 м2) та позаклітинної води (Δ, л/1,73 м2) у пацієнтів з ЦД 2-го типу на тлі прийому інгібіторів НЗКТГ-2 [25] Vol. 17, No. Матеріали та методи 4, 2021 http://iej.zaslavsky.com.ua 299 Оригінальні дослідження  /Original Researches/ становить < 2,5 ммоль/л); холестерину ліпопротеїнів високої щільності (ХС ЛПВЩ, норма > 1,2 ммоль/л); холестерину ліпопротеїнів дуже низької щільності (ХС ЛПДНЩ (за формулою Фридвальда), нормаль­ ний рівень 0,26–1,04 ммоль/л). Розраховували індекс атерогенності (ІА = [ЗХС – ХС ЛПВЩ]/ХС ЛПВЩ). Визначали концентрацію сечової кислоти в сироватці крові. за допомогою критерію Шапіро — Уїлка. Різницю між показниками визначали з використанням парного t-тесту Вілкоксона та вважали значущою при P < 0,05 за критерієм Стьюдента. Результати Показники композиції тіла й ліпідного спектра в чоловіків і жінок, хворих на ЦД 2-го типу, до та після лікування з використанням іНЗКТГ-2 (дапагліфлозин) (M ± m) Досліджуваний показник Чоловіки (n = 11) Жінки (n = 6) До лікування Після лікування До лікування Після лікування Композиція тіла й антро­ пометричні показники ІМТ, кг/м2 33,84 ± 1,89 30,92 ± 1,09* 36,42 ± 2,27 35,22 ± 2,21* Маса тіла, кг 106,83 ± 7,34 101,33 ± 6,61# 92,42 ± 5,64 91,85 ± 5,10 ОТ, см 113,00 ± 5,07 108,0 ± 5,1** 106,00 ± 5,45 103,00 ± 6,22* % загального жиру 29,80 ± 2,22 28,0 ± 1,5*** 43,27 ± 2,63 40,97 ± 2,68* % вісцерального жиру 16,28 ± 1,77 15,12 ± 1,45 12,47 ± 1,52 10,20 ± 1,34* % води 50,52 ± 1,66 52,34 ± 1,04* 42,52 ± 1,77 46,08 ± 2,06* М’язова маса 70,79 ± 2,94 68,39 ± 3,35 48,67 ± 1,44 50,90 ± 1,31 Оцінка тілобудови 3,33 ± 0,44 3,36 ± 0,34 2,83 ± 0,17 3,50 ± 0,50 Кісткова маса 3,64 ± 0,14 3,55 ± 0,16 2,60 ± 0,08 2,75 ± 0,05* Метаболічний вік 59,89 ± 3,90 60,82 ± 3,86 66,33 ± 4,72 56,25 ± 5,54* Показники вуглеводного обміну Глікемія натще, ммоль/л 9,88 ± 1,23 8,96 ± 0,81 12,58 ± 2,80 10,30 ± 0,68 HbA1c, % 9,88 ± 0,65 7,80 ± 0,20* 8,68 ± 0,83 8,57 ± 0,54 С-пептид, нмоль/л 1,99 ± 0,15 2,25 ± 0,16 2,52 ± 0,33 2,57 ± 0,31 Інсулін, мкОд/мл 9,48 ± 1,28 11,21 ± 1,13 10,65 ± 4,25 10,90 ± 2,89 Ліпідний спектр і рівні сечової кислоти в сироватці крові Холестерин, ммоль/л 5,00 ± 0,34 4,41 ± 0,33 5,67 ± 0,53 6,03 ± 0,43 Тригліцериди, ммоль/л 3,14 ± 0,59 2,62 ± 0,39 3,19 ± 0,88 3,15 ± 0,77 ХС ЛПВЩ, ммоль/л 1,06 ± 0,05 1,1 ± 0,9# 1,22 ± 0,03 1,29 ± 0,06# ХС ЛПНЩ, ммоль/л 3,02 ± 0,25 2,65 ± 0,30 3,27 ± 0,17 3,85 ± 0,40 ХС ЛПДНЩ, ммоль/л 0,92 ± 0,15 1,08 ± 0,15# 0,94 ± 0,22 1,13 ± 0,20# Індекс атерогенності 4,03 ± 0,44 3,17 ± 0,32* 3,76 ± 2,67 2,46 ± 1,96 Концентрація сечової кислоти 364,22 ± 27,40 364,22 ± 27,4 296,00 ± 25 302,00 ± 26 Примітки: вірогідність різниці між показниками до та після лікування: * — Р < 0,05; ** — Р < 0,01; *** — Р < 0,001; статистична тенденція: # — Р < 0,1. Таблиця 4. Висновки Препарати групи інгібіторів натрійзалежного ко­ транспортера глюкози на сьогодні зарекомендували себе як ефективні гіпоглікемічні засоби, що мають вира­ жені кардіовазопротекторні властивості, знижують жи­ рову масу. У даному дослідженні визначали особливості впливу препарату іНЗКТГ-2 тривалої дії дапагліфлозин в дозі 10 мг на добу. У 17 хворих на ЦД 2-го типу літнього віку на тлі 3–4 місяців лікування препаратом іНЗКТГ-2 дапагліфлозином отримано попередні результати щодо зниження ІМТ, ОТ, поліпшення показників композиції тіла, зокрема зниження відсотка загального жиру. У гру­ пі жінок, на відміну від чоловіків, відзначено зменшен­ ня рівня вісцерального жиру, зниження розрахункового метаболічного віку та збільшення вмісту води в орга­ нізмі. Отримані попередні дані вказують на позитивні тенденції застосування дапагліфлозину вже після три­ місячного прийому, що дає підстави очікувати кращих результатів при подальшому спостереженні. —  лише в чоловіків виявлено зменшення індексу атерогенності; —  в обох групах практично не змінилися показни­ ки ліпідного спектра, хоча прослідковується статис­ тична тенденція (P < 0,10) до підвищення рівня ЛПВЩ та зменшення ЛПДНЩ, і у разі збільшення кількості обстежуваних може мати вірогідність. Зазначимо, що в загальній групі обстежених жі­ нок, в яких було визначено показники композиції тіла (n = 42), відсотковий вміст води в організмі проявляв високу, вірогідну негативну кореляцію з відсотком жиру та рівнем вісцерального жиру (r = –0,98 та r = –0,79 від­ повідно, P  <  0,05). Серед чоловіків (n  =  75) виявлено аналогічний взаємозв’язок (r = –0,88 та r = –0,77 відпо­ відно, P < 0,05). Це може вказувати на те, що більшому вмісту жирової тканини (як загального жиру, так і абдо­ мінального) відповідає знижений вміст води в організмі, і навпаки, менший відносний вміст жиру пов’язаний з поліпшенням гідратації. Імовірно, відсутність вірогідної різниці до і після лікування дапагліфлозином показни­ ків ліпідного спектра, композиційних ознак в обох до­ сліджуваних групах пов’язана з недостатньою кількістю обстежених пацієнтів. Спостереження надалі буде про­ довжено із залученням більшої кількості обстежуваних. Конфлікт інтересів. Автор заявляє про відсутність конфлікту інтересів та власної фінансової зацікавле­ ності при підготовці даної статті. Інформація про фінансування. Стаття підготовлена в рамках бюджетного фінансування Національної акаде­ мії медичних наук України. Під впливом прийому 10 мг/добу дапагліфлозину на тлі зменшення в обох групах пацієнтів показників ІМТ, ОТ, відсотка загального жиру, перерозподілу води в організмі відзначалися такі відмінності ефектів між групами чоловіків і жінок: жуючи масу тіла, зокрема, це відбувається за рахунок зменшення кількості вісцерального жиру — важливого чинника ризику розвитку дисліпідемії та серцево-су­ динних ускладнень. Втрата позаклітинної рідини має тимчасовий характер на початку лікування іНЗКТ2 та повертається до початкового рівня при довгостроково­ му лікуванні. —  у жінок зафіксовано зменшення вмісту вісце­ рального жиру та зниження розрахункового мета­ болічного віку в середньому до 10 років. Також після лікування зафіксовано вірогідне збільшення  відсотка вмісту води в організмі (поліпшення водного забезпе­ чення), що могло частково нівелювати можливе змен­ шення маси тіла; —  у жінок зафіксовано зменшення вмісту вісце­ рального жиру та зниження розрахункового мета­ болічного віку в середньому до 10 років. Також після лікування зафіксовано вірогідне збільшення  відсотка вмісту води в організмі (поліпшення водного забезпе­ чення), що могло частково нівелювати можливе змен­ шення маси тіла; Результати Показники композиції тіла й ліпідного спектра в чоловіків і жінок, хворих на Ц до та після лікування з використанням іНЗКТГ-2 (дапагліфлозин) (M ± m) азники композиції тіла й ліпідного спектра в чоловіків і жінок, хворих на ЦД 2-го типу, о та після лікування з використанням іНЗКТГ-2 (дапагліфлозин) (M ± m) Vol. 17, No. 4, 2021 Vol. 17, No. 4 Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) 300 Оригінальні дослідження  /Original Researches/ 5.  DeFronzo RA, Eldor R, Abdul-Ghani M. Pathophys­ iologic approach to therapy in patients with newly diagnosed Обговорення A safety update on sodium glucose co-transporter 2 inhibitors. Diabetes Obes Metab. 2019 Apr;21 Suppl 2:34-42. doi:10.1111/dom.13611. 6. Meng W, Ellsworth BA, Nirschl AA, et al. Discovery of dapagliflozin: a potent, selective renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes. J Med Chem. 2008 Mar 13;51(5):1145-9. doi:10.1021/jm701272q. 19. McGill JB, Subramanian S. Safety of Sodium-Glu­ cose Co-Transporter 2 Inhibitors. Am J Cardiol. 2019 Dec 15;124 Suppl 1:S45-S52. doi:10.1016/j.amjcard.2019.10.029. 20. Scheen AJ. Pharmacodynamics, efficacy and safe­ ty of sodium-glucose co-transporter type 2 (SGLT2) inhibi­ tors for the treatment of type 2 diabetes mellitus. Drugs. 2015 Jan;75(1):33-59. doi:10.1007/s40265-014-0337-y. 7. Ferrannini E, Ramos SJ, Salsali A, Tang W, List JF. Dapagliflozin monotherapy in type 2 diabetic patients with in­ adequate glycemic control by diet and exercise: a randomized, double-blind, placebo-controlled, phase 3 trial. Diabetes Care. 2010 Oct;33(10):2217-24. doi:10.2337/dc10-0612. 21. Jabbour S, Seufert J, Scheen A, Bailey CJ, Karup C, Langkilde AM. Dapagliflozin in patients with type 2 diabetes mellitus: A pooled analysis of safety data from phase IIb/III clinical trials. Diabetes Obes Metab. 2018 Mar;20(3):620- 628. doi:10.1111/dom.13124. 8. Bailey CJ, Gross JL, Pieters A, Bastien A, List JF. Ef­ fect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with metformin: a randomised, double-blind, placebo-controlled trial. Lancet. 2010 Jun 26;375(9733):2223-33. doi:10.1016/S0140-6736(10)60407- 2. 22. Centers for Disease Control and Prevention. Diag­ nosed Diabetes. Available from: http://www.cdc.gov/diabetes/ statistics/prevalence_national.htm. 9. Strojek K, Yoon KH, Hruba V, Elze M, Langkilde AM, Parikh S. Effect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with glime­ piride: a randomized, 24-week, double-blind, placebo-con­ trolled trial. Diabetes Obes Metab. 2011 Oct;13(10):928-38. doi:10.1111/j.1463-1326.2011.01434.x. 23. Ferrannini E, Ramos SJ, Salsali A, Tang W, List JF. Dapagliflozin monotherapy in type 2 diabetic patients with in­ adequate glycemic control by diet and exercise: a randomized, double-blind, placebo-controlled, phase 3 trial. Diabetes Care. 2010 Oct;33(10):2217-24. doi:10.2337/dc10-0612. 10. Bolinder J, Ljunggren Ö, Kullberg J, et al. Effects of dapagliflozin on body weight, total fat mass, and region­ al adipose tissue distribution in patients with type 2 diabetes mellitus with inadequate glycemic control on metformin. J Clin Endocrinol Metab. 2012 Mar;97(3):1020-31. doi:10.1210/ jc.2011-2260. 24. Polidori D, Sanghvi A, Seeley RJ, Hall KD. How Strongly Does Appetite Counter Weight Loss? Quantification of the Feedback Control of Human Energy Intake. Obesity (Silver Spring). 2016 Nov;24(11):2289-2295. doi:10.1002/ oby.21653. 25. Обговорення 1. Alvarez Guisasola F, Mavros P, Nocea G, Alemao E, Alexander CM, Yin D. Glycaemic control among patients with type 2 diabetes mellitus in seven European countries: findings from the Real-Life Effectiveness and Care Patterns of Diabetes Management (RECAP-DM) study. Diabetes Obes Metab. 2008 Jun;10 Suppl 1:8-15. doi:10.1111/j.1463- 1326.2008.00881.x. 1. Alvarez Guisasola F, Mavros P, Nocea G, Alemao E, Alexander CM, Yin D. Glycaemic control among patients with type 2 diabetes mellitus in seven European countries: findings from the Real-Life Effectiveness and Care Patterns of Diabetes Management (RECAP-DM) study. Diabetes Obes Metab. 2008 Jun;10 Suppl 1:8-15. doi:10.1111/j.1463- 1326.2008.00881.x. До позаглікемічних ефектів гліфлозинів відносять їх гіпоурикемічну дію за рахунок посилення екскреції сечової кислоти з сечею (урикозурії). Під час прийому дапагліфлозіну урикозурія підвищувалася (протягом 3–7 днів) і супроводжувалася стійким зменшенням концентрації сечової кислоти в сироватці крові. Хоча точні механізми, за допомогою яких іНЗКТГ-2 викли­ кають гальмування реабсорбції і зниження рівня ури­ кемії, ще залишаються дискутабельними. Вважають, що вони можуть включати прямий вплив на нирковий транспорт сечової кислоти або непрямий ефект, вто­ ринний стосовно до зменшення реабсорбції натрію в проксимальних канальцях. Оскільки сироваткова кон­ центрація сечової кислоти змінюється паралельно з поглинанням натрію нирками, зниження реабсорбції і посилення екскреції СК може також бути результатом блокування реабсорбції натрію і глюкози гліфлозина­ ми [31–34]. 2. Hoerger TJ, Segel JE, Gregg EW, Saaddine JB. Is gly­ cemic control improving in U.S. adults? Diabetes Care. 2008 Jan;31(1):81-6. doi:10.2337/dc07-1572. 3. American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabe­ tes-2020. Diabetes Care. 2020 Jan;43(Suppl 1):S14-S31. doi:10.2337/dc20-S002. 4. Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/ NLA/PCNA Guideline on the Management of Blood Choles­ terol: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2019 Jun 25;73(24):3168-3209. doi:10.1016/j.jacc.2018.11.002. Таким чином, дані літератури та результати прове­ деного нами дослідження вказують на те, що препара­ ти групи іНЗКТ2 позитивно впливають на показники композиції тіла хворих на ЦД 2-го типу, вірогідно зни­ 5. DeFronzo RA, Eldor R, Abdul-Ghani M. Pathophys­ iologic approach to therapy in patients with newly diagnosed 5. DeFronzo RA, Eldor R, Abdul-Ghani M. Pathophys­ iologic approach to therapy in patients with newly diagnosed Vol. 17, No. 4, 2021 301 http://iej.zaslavsky.com.ua Оригінальні дослідження  /Original Researches/ type 2 diabetes. Diabetes Care. 2013 Aug;36 Suppl 2(Suppl 2):S127-38. doi:10.2337/dcS13-2011. Nov 26;373(22):2117-28. doi:10.1056/NEJMoa1504720. 18. Fitchett D. Обговорення Cai X, Yang W, Gao X, et al. The Association Be­ tween the Dosage of SGLT2 Inhibitor and Weight Reduction in Type 2 Diabetes Patients: A Meta-Analysis. Obesity (Silver Spring). 2018 Jan;26(1):70-80. doi:10.1002/oby.22066. 11. Schork A, Saynisch J, Vosseler A, et al. Effect of SGLT2 inhibitors on body composition, fluid status and re­ nin-angiotensin-aldosterone system in type 2 diabetes: a pro­ spective study using bioimpedance spectroscopy. Cardiovasc Diabetol. 2019 Apr 5;18(1):46. doi:10.1186/s12933-019- 0852-y. 26. Schork A, Saynisch J, Vosseler A, et al. Effect of SGLT2 inhibitors on body composition, fluid status and re­ nin-angiotensin-aldosterone system in type 2 diabetes: a pro­ spective study using bioimpedance spectroscopy. Cardiovasc Diabetol. 2019 Apr 5;18(1):46. doi:10.1186/s12933-019- 0852-y. 12. Giugliano D, Longo M, Scappaticcio L, Caru­ so P, Esposito K. Sodium-glucose transporter-2 inhibitors for prevention and treatment of cardiorenal complications of type 2 diabetes. Cardiovasc Diabetol. 2021 Jan 11;20(1):17. doi:10.1186/s12933-021-01213-w. 27. Tobita H, Sato S, Miyake T, Ishihara S, Kinoshi­ ta Y. Effects of Dapagliflozin on Body Composition and Liver Tests in Patients with Nonalcoholic Steatohepatitis Associated with Type 2 Diabetes Mellitus: A Prospective, Open-label, Un­ controlled Study. Curr Ther Res Clin Exp. 2017 Jul 8;87:13- 19. doi:10.1016/j.curtheres.2017.07.002. 13. Tsushima Y, Lansang MC, Makin V. The role of SGLT-2 inhibitors in managing type 2 diabetes. Cleve Clin J Med. 2020 Dec 31;88(1):47-58. doi:10.3949/ccjm.88a.20088. 14. Koval SM, Yushko KO, Snihurska IO, et al. Relations of angiotensin-(1-7) with hemodynamic and cardiac structural and functional parameters in patients with hypertension and type 2 diabetes. Arterial Hypertension 2019;23(3):183-189. doi:10.5603/AH.a2019.0012. 28. Kahn SE, Cooper ME, Del Prato S. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, pres­ ent, and future. Lancet. 2014 Mar 22;383(9922):1068-83. doi:10.1016/S0140-6736(13)62154-6. /   15. Chiang KM, Tsay YC, Vincent Ng TC, et al. Is Hy­ peruricemia, an Early-Onset Metabolic Disorder, Causally As­ sociated with Cardiovascular Disease Events in Han Chinese? J Clin Med. 2019 Aug 12;8(8):1202. doi:10.3390/jcm8081202. 16. Kalra S. Sodium Glucose Co-Transporter-2 (SGLT2) Inhibitors: A Review of Their Basic and Clinical Pharmacol­ ogy. Diabetes Ther. 2014 Dec;5(2):355-66. doi:10.1007/ s13300-014-0089-4. 29. Fulcher G, Matthews DR, Perkovic V, et al; CAN­ VAS trial collaborative group. Efficacy and safety of canagli­ flozin when used in conjunction with incretin-mimetic therapy in patients with type 2 diabetes. Diabetes Obes Metab. 2016 Jan;18(1):82-91. doi:10.1111/dom.12589. 15. Chiang KM, Tsay YC, Vincent Ng TC, et al. Information about author Information about author Olha Prybyla, Junior Research Fellow at the Department of Age Endocrinology and Clinical Pharmacology, State Institution “V.P. Komisarenko Institute of Endocrinology and Metabolism of the NAMS of Ukraine”, Kyiv, Ukraine; e-mail: [email protected]; https://orcid.org/0000-0003-2212-1172. Conflicts of interests. Author declares the absence of any conflicts of interests and their own financial interest that might be construed to influence the results or interpretation of their manuscript. Information about funding. The article was prepared within the budget funding of the National Academy of Medical Sciences of Ukraine. Обговорення Is Hy­ peruricemia, an Early-Onset Metabolic Disorder, Causally As­ sociated with Cardiovascular Disease Events in Han Chinese? J Clin Med. 2019 Aug 12;8(8):1202. doi:10.3390/jcm8081202. 16. Kalra S. Sodium Glucose Co-Transporter-2 (SGLT2) Inhibitors: A Review of Their Basic and Clinical Pharmacol­ ogy. Diabetes Ther. 2014 Dec;5(2):355-66. doi:10.1007/ s13300-014-0089-4. 30. Sinclair A, Bode B, Harris S, et al. Efficacy and safe­ ty of canagliflozin compared with placebo in older patients with type 2 diabetes mellitus: a pooled analysis of clinical studies. BMC Endocr Disord. 2014 Apr 18;14:37. doi:10.1186/1472- 6823-14-37. 17. Zinman B, Wanner C, Lachin JM, et al; EMPA-REG OUTCOME Investigators. Empagliflozin, Cardiovascular Out­ comes, and Mortality in Type 2 Diabetes. N Engl J Med. 2015 31. Griffin M, Rao VS, Ivey-Miranda J, et al. Empagli­ 31. Griffin M, Rao VS, Ivey-Miranda J, et al. Empagli­ Vol. 17, No. 4, 2021 302 Mìžnarodnij endokrinologìčnij žurnal, ISSN 2224-0721 (print), ISSN 2307-1427 (online) Оригінальні дослідження  /Original Researches/ cemia: An Evidence-Based Update. Medicina (Kaunas). 2021 Jan 10;57(1):58. doi:10.3390/medicina57010058. flozin in Heart Failure: Diuretic and Cardiorenal Effects. Cir­ culation. 2020 Sep 15;142(11):1028-1039. doi:10.1161/CIR­ CULATIONAHA.120.045691. Отримано/Received  01.04.2021 Рецензовано/Revised  22.04.2021 Прийнято до друку/Accepted  05.05.2021 32. Novikov A, Fu Y, Huang W, et al. SGLT2 inhi­ bition and renal urate excretion: role of luminal glucose, GLUT9, and URAT1. Am J Physiol Renal Physiol. 2019 Jan 1;316(1):F173-F185. doi:10.1152/ajprenal.00462.2018. 33. Yuan T, Liu S, Dong Y, Fu Y, Tang Y, Zhao W. Ef­ fects of dapagliflozin on serum and urinary uric acid levels in patients with type 2 diabetes: a prospective pilot trial. Diabetol Metab Syndr. 2020 Oct 27;12:92. doi:10.1186/s13098-020- 00600-9. 34. Cicero AFG, Fogacci F, Kuwabara M, Borghi C. Therapeutic Strategies for the Treatment of Chronic Hyperuri­ flozin in Heart Failure: Diuretic and Cardiorenal Effects. Cir­ culation. 2020 Sep 15;142(11):1028-1039. doi:10.1161/CIR­ CULATIONAHA.120.045691. http://iej.zaslavsky.com.ua Pharmacokinetic characteristics and morphometric effects of sodium-glucose contransporter-2 inhibitors in men and women with type 2 diabetes mellitus (literature review and own results) Abstract. Background. According to the latest international clinical guidelines, gliflozins — sodium-glucose cotransporter-2 inhibitors — are indicated as oral antidiabetic drugs of second- third-line therapy in type 2 diabetes mellitus. Due to insulin-in­ dependent stimulation of glucosuria, gliflozins have extraglycemic effects such as weight loss, improved adipose tissue distribution, better plasma lipid profile, and decreased uricemia that in gener­ ally reduce the risk of cardiovascular complications. The purpose of this study was to evaluate the effectiveness of dapagliflozin in the treatment of men and women with type 2 diabetes mellitus with a metabo­lically unhealthy phenotype. Materials and meth­ ods. The study included 17 individuals with diabetes mellitus type 2 (11 men and 6 women), aged 58.0 ± 1.7 years (95% confidence interval 53–62), whose body composition was evaluated by bio­ electric impedance using a Tanita analyzer BC-545N (Japan). Patients received therapy with dapagliflozin, antihypertensive and antihyperlipidemic drugs (statins). Results. A three-month use of dapagliflozin in a dose of 10 mg once daily caused a decrease in body mass index, waist circumference, improvement of body composition, in parti­cular a reduction in total body fat (the signif­ icance of changes was determined using a paired t-test). No sig­ nificant changes in muscle and bone mass, body composition, lip­ id profile, and uricemia level were observed. The group of women, in contrast to men, had a decreased level of visceral fat, which was accompanied by an improvement in the body’s water supply, and a reduction in the estimated metabolic age. Conclusions. Treatment of type 2 diabetes patients with sodium-glucose cotransporter-2 inhibitors for 3 months has reduced the degree of obesity and im­ proved some indices of body composition. Confirmation of this trend can be obtained in further observations. Keywords:  type 2 diabetes mellitus; dapagliflozin; body composi­ tion; visceral fat; uricemia; lipid profile Vol. 17, No. 4, 2021 http://iej.zaslavsky.com.ua 303

https://openalex.org/W2806555776

https://acp.copernicus.org/preprints/acp-2018-529/acp-2018-529.pdf

English

Fine particles from Independence Day fireworks events: chemical characterization and source apportionment

cc-by

Fine particles from Independence Day fireworks events: chemical characterization and source apportionment 25 Abstract: To study the impact of fireworks (FW) events on air quality, aerosol particles from FW displays were measured 10 using a High-Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-ToF-AMS) and collocated instruments during the Independence Day (July 4) holiday 2017 at Albany, NY, USA. Three FW events were identified through the potassium ion (K+) signals in the aerosol mass spectra. The largest FW event signal measured at two different sites was the Independence Day celebration in downtown Albany, with maximum hourly PM2.5 of about 55 μg m–3 at the downtown site (approximately 1 km Abstract: To study the impact of fireworks (FW) events on air quality, aerosol particles from FW displays were measured 10 using a High-Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-ToF-AMS) and collocated instruments during the Independence Day (July 4) holiday 2017 at Albany, NY, USA. Three FW events were identified through the potassium ion (K+) signals in the aerosol mass spectra. The largest FW event signal measured at two different sites was the Independence Day celebration in downtown Albany, with maximum hourly PM2.5 of about 55 μg m–3 at the downtown site (approximately 1 km Fine particles from Independence Day fireworks events: chemical characterization and source apportionment Jie Zhang1, Sara Lance1, Jeffrey M. Freedman1, Yele Sun2, Brian A. Crandall1, Xiuli Wei1,3, James J. Schwab1 1Atmospheric Sciences Research Center, University at Albany, State University of New York; 5 2State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China 3Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei, China Correspondence to:James J. Schwab ([email protected]) Abstract: To study the impact of fireworks (FW) events on air quality, aerosol particles from FW displays were measured 10 using a High-Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-ToF-AMS) and collocated instruments during the Independence Day (July 4) holiday 2017 at Albany, NY, USA. Three FW events were identified through the potassium ion (K+) signals in the aerosol mass spectra. The largest FW event signal measured at two different sites was the Independence Day celebration in downtown Albany, with maximum hourly PM2.5 of about 55 μg m–3 at the downtown site (approximately 1 km from the FW launch location), and 33.3 μg m–3 of non-refractory PM1 at the uptown site (approximately 8 km downwind). The 15 aerosol concentration peak measured at the uptown site occurred 2 hours later than at the downtown site. The Independence Day FW events resulted in significant increases in both organic and inorganic (K+, sulfate, chloride) chemical components. Positive Matrix Factorization (PMF) of organics mass spectra identified one FW related organic aerosol factor (FW-OOA) with a highly oxidized state. The intense emission of FW particles from the Independence Day celebration contributed about 79.0% (26.1 μg m–3) of total PM1 (33.0 μg m–3) measured at the uptown site during Independence Day FW event (07/04 23:00-07/05 02:00). 20 Aerosol measurements and wind LiDAR measurements showed two distinct pollution sources, one from the Independence Day FW event in Albany, and the other transported from the northeast, potentially associated with another city’s FW events. This study highlights the significant influence of FW burning on fine aerosol mass concentration and chemical characteristics, which is useful in quantifying the impacts of FW on air pollution, at a time when more than usual people are clustered together and breathing the outdoor air. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. 1 Introduction 1 Firework displays (FW) from national celebrations, such as Independence Day in United States, Spring and Lantern Festivals in China, Diwali Festival in India, Guy Fawkes Night in the UK, and worldwide New Year’s Eve celebrations are known to cause short-term very high-intensity air quality degradation, especially atmospheric particle matter (PM) pollution (Seidel et al., 2015; c⃝Author(s) 2018. CC BY 4.0 License. Dickerson et al., 2016; Wang et al., 2007; Jiang et al., 2015; Kong et al., 2015; Tian et al., 2014; Yang et al., 2014; Barman et al., 2008; Godri et al., 2010; Drewnick et al., 2006). Besides substantial visibility reductions lasting for hours and heavy emission of gaseous pollutants, such as nitrogen oxides and sulfur dioxide (Vecchi et al., 2008), FW events also produce large amounts of fine aerosol (PM2.5) that are rich in sulfate, organics, potassium, and heavy metals (e.g. Cu, Ba, Al) (Moreno et al., 2007; Jiang et al., 2015; Lin et al., 2016). The heavy metals come from the inorganic salts that are used in FW manufacturing to produce 5 different colors, while the compounds used for oxidizing firework mixtures produce abundant potassium salts, such as potassium nitrate, potassium chlorate, and potassium perchlorate. FW produced aerosols with metal compounds were found to have greater toxicity than traffic emitted aerosols (Godri et al., 2010), and to be harmful to public health, with detrimental impacts on the respiratory system (Joly et al., 2010; WHO, 2006). Dickerson et al., 2016; Wang et al., 2007; Jiang et al., 2015; Kong et al., 2015; Tian et al., 2014; Yang et al., 2014; Barman et al., 2008; Godri et al., 2010; Drewnick et al., 2006). Besides substantial visibility reductions lasting for hours and heavy emission of gaseous pollutants, such as nitrogen oxides and sulfur dioxide (Vecchi et al., 2008), FW events also produce large amounts of fine aerosol (PM2.5) that are rich in sulfate, organics, potassium, and heavy metals (e.g. Cu, Ba, Al) (Moreno et al., 2007; Jiang et al., 2015; Lin et al., 2016). The heavy metals come from the inorganic salts that are used in FW manufacturing to produce 5 different colors, while the compounds used for oxidizing firework mixtures produce abundant potassium salts, such as potassium nitrate, potassium chlorate, and potassium perchlorate. 1 Introduction FW produced aerosols with metal compounds were found to have greater toxicity than traffic emitted aerosols (Godri et al., 2010), and to be harmful to public health, with detrimental impacts on the respiratory system (Joly et al., 2010; WHO, 2006). The Independence Day holiday in the United States (July 4) has been celebrated for more than two hundred years and is 10 associated with grand FW displays around the whole country, from small towns to megacities. According to the American Pyrotechnics Association (APA) fireworks consumption figures, about 110.7 million kg of fireworks were detonated by U.S. consumers in 2016. Nationwide on Independence Day, there has been an average increase in 24-hr PM2.5 concentration by 42% across the U.S. on Independence Day comparing to the adjacent days (Seidel et al., 2015). This indicates the importance of studying the fine PM emission from the FW displays that are a feature of Independence Day celebrations. 15 Previous studies of FW fine PM were mainly based on filter methods to analyze the particle composition mass concentration (Kong et al., 2015; Yang et al., 2014), or bulk particle size distribution characteristics (Wehner et al., 2000). The long integration time for collection of filter samples (12 hours or 24 hours) loses the real-time information (composition and size distribution) of FW aerosol, and creates uncertainties in accurately quantifying these parameters (Jiang et al., 2015). The chemical composition and size distribution of particles, as measured by more advanced aerosol mass spectrometry technology (Jayne et al., 2000; 20 Decarlo et al., 2006, 2008; Sun et al., 2010; Sun et al., 2012), provides a new perspective on FW PM characteristics (Drewnick et al., 2006, Jiang et al., 2015). Firework particle emissions have been shown to have high potassium (K+) and organic mass ratios, and the observed mass spectra of organic compounds is similar to that of secondary organic aerosol (SOA) (Jiang et al., 2015, Jimenez et al., 2009; Ng et al., 2011). The Independence Day holiday in the United States (July 4) has been celebrated for more than two hundred years and is 10 associated with grand FW displays around the whole country, from small towns to megacities. According to the American Pyrotechnics Association (APA) fireworks consumption figures, about 110.7 million kg of fireworks were detonated by U.S. consumers in 2016. Nationwide on Independence Day, there has been an average increase in 24-hr PM2.5 concentration by 42% across the U.S. The high mass concentration and complex emission sources of background aerosol in megacities such as Beijing (Jiang et al., 25 2015) will to some extent mask information about the contribution of FW to ambient aerosol, and in particular the features of the organic aerosol species directly emitted from FW burning. Albany, the capital of New York State (a city with a population of 1 Introduction In this study, aerosol particles from FW displays were measured using a High-Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-ToF-AMS) and collocated instruments during the 2017 Independence Day holiday in Albany, NY. The aerosol mass concentration, chemical composition, and size distribution are characterized. The sources of different organic 10 aerosol, and their mass spectra character are studied. In addition, the role of transport is investigated based on LiDAR measurements and back trajectory analysis. approximately 100,000 about roughly 240 km north of New York City), features relatively clean air (PM mass concentration generally below 15 μg m–3, Zhang et al., 2018). Albany hosted a large FW display for the Independence Day holiday, which lasted for almost an hour from 22:00 to 23:00 local time (EDT: Eastern Daylight Time) on July 4, 2017 and was surrounded by smaller FW events in regional towns such as Bennington, Vermont and Saratoga Springs, New York. As will be shown, the weather conditions during Albany ’ s fireworks display provided a unique opportunity for a detailed investigation of the 5 contribution of FW displays to the ambient aerosol mass concentration and chemical content, the different sources of aerosols, and the health influence of these kinds of metal-rich aerosols on surrounding areas. In this study, aerosol particles from FW displays were measured using a High-Resolution Time-of-Flight Aerosol Mass Spectrometer (HR-ToF-AMS) and collocated instruments during the 2017 Independence Day holiday in Albany, NY. The aerosol mass concentration, chemical composition, and size distribution are characterized. The sources of different organic 10 aerosol, and their mass spectra character are studied. In addition, the role of transport is investigated based on LiDAR measurements and back trajectory analysis. 2.1 Measurement site and sampling instruments The main (“uptown”) measurement site is located at the Atmospheric Sciences Research Center (latitude: 42.7 °N, longitude: 15 73.8 °W, elevation: 81 m, hereafter ASRC site, as shown in Fig. S1), near the main campus of the University at Albany, which is located in the uptown (NW) section of Albany, about 8 km downwind of the Albany FW launch site (at Empire State Plaza) and near the intersection of interstates I90 and I87. The measurements of aerosol concentration were conducted from June 27 00:00 local time (EDT) to July 7 12:00, except the period from July 1 00:00 to July 2 20:00 for the maintenance of instruments. The measurement period covered the Independence Day holiday, during which an intense Albany FW event occurred. A second 20 (“downtown”) measurement site is the Albany County Health Department measurement site (latitude: 42.63 °N, longitude: 73.75 °W, elevation: 8 m, hereafter ACHD site), which is about 1 km southeast of Empire State Plaza and about 9 km from ASRC site. The key instrument is the HR-ToF-AMS, which was used to measure non-refractory (NR) submicron aerosol (NR-PM1) chemical composition (including organics, nitrate, sulfate, ammonium, and chloride), as well as the size distributions of aerosol chemical species (DeCarlo et al., 2006; Drewnick et al., 2005). This HR-ToF-AMS was operated under the sensitive “V-mode”, 25 and cycled through the mass spectrum (MS) mode and the particle-of-flight mode (PToF) every minute. The main (“uptown”) measurement site is located at the Atmospheric Sciences Research Center (latitude: 42.7 °N, longitude: 15 73.8 °W, elevation: 81 m, hereafter ASRC site, as shown in Fig. S1), near the main campus of the University at Albany, which is located in the uptown (NW) section of Albany, about 8 km downwind of the Albany FW launch site (at Empire State Plaza) and near the intersection of interstates I90 and I87. The measurements of aerosol concentration were conducted from June 27 00:00 local time (EDT) to July 7 12:00, except the period from July 1 00:00 to July 2 20:00 for the maintenance of instruments. The measurement period covered the Independence Day holiday, during which an intense Albany FW event occurred. A second 20 (“downtown”) measurement site is the Albany County Health Department measurement site (latitude: 42.63 °N, longitude: 73.75 °W, elevation: 8 m, hereafter ACHD site), which is about 1 km southeast of Empire State Plaza and about 9 km from ASRC site. 1 Introduction on Independence Day comparing to the adjacent days (Seidel et al., 2015). This indicates the importance of studying the fine PM emission from the FW displays that are a feature of Independence Day celebrations. 15 The Independence Day holiday in the United States (July 4) has been celebrated for more than two hundred years and is 10 associated with grand FW displays around the whole country, from small towns to megacities. According to the American Pyrotechnics Association (APA) fireworks consumption figures, about 110.7 million kg of fireworks were detonated by U.S. consumers in 2016. Nationwide on Independence Day, there has been an average increase in 24-hr PM2.5 concentration by 42% across the U.S. on Independence Day comparing to the adjacent days (Seidel et al., 2015). This indicates the importance of studying the fine PM emission from the FW displays that are a feature of Independence Day celebrations. 15 The high mass concentration and complex emission sources of background aerosol in megacities such as Beijing (Jiang et al., 25 2015) will to some extent mask information about the contribution of FW to ambient aerosol, and in particular the features of the organic aerosol species directly emitted from FW burning. Albany, the capital of New York State (a city with a population of 2 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. approximately 100,000 about roughly 240 km north of New York City), features relatively clean air (PM mass concentration generally below 15 μg m–3, Zhang et al., 2018). Albany hosted a large FW display for the Independence Day holiday, which lasted for almost an hour from 22:00 to 23:00 local time (EDT: Eastern Daylight Time) on July 4, 2017 and was surrounded by smaller FW events in regional towns such as Bennington, Vermont and Saratoga Springs, New York. As will be shown, the weather conditions during Albany ’ s fireworks display provided a unique opportunity for a detailed investigation of the 5 contribution of FW displays to the ambient aerosol mass concentration and chemical content, the different sources of aerosols, and the health influence of these kinds of metal-rich aerosols on surrounding areas. 2.1 Measurement site and sampling instruments The key instrument is the HR-ToF-AMS, which was used to measure non-refractory (NR) submicron aerosol (NR-PM1) chemical composition (including organics, nitrate, sulfate, ammonium, and chloride), as well as the size distributions of aerosol chemical species (DeCarlo et al., 2006; Drewnick et al., 2005). This HR-ToF-AMS was operated under the sensitive “V-mode”, 25 and cycled through the mass spectrum (MS) mode and the particle-of-flight mode (PToF) every minute. The main (“uptown”) measurement site is located at the Atmospheric Sciences Research Center (latitude: 42.7 °N, longitude: 15 73.8 °W, elevation: 81 m, hereafter ASRC site, as shown in Fig. S1), near the main campus of the University at Albany, which is located in the uptown (NW) section of Albany, about 8 km downwind of the Albany FW launch site (at Empire State Plaza) and near the intersection of interstates I90 and I87. The measurements of aerosol concentration were conducted from June 27 00:00 local time (EDT) to July 7 12:00, except the period from July 1 00:00 to July 2 20:00 for the maintenance of instruments. The 3 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. A TSI Scanning Mobility Particle Sizer (SMPS), which consists of an Electrostatic Classifier (EC, model 3080), a Differential Mobility Analyzer (DMA, model 3081), and a Condensation Particle Counter (CPC, model 3785), was used to measure particle number size distributions. Total PM2.5 mass concentration was also obtained from an optical scattering based aerosol mass measurement instrument (pDR-1500, Thermo Fisher Scientific Inc.), which measures the scattered light flux and uses this signal to estimate the aerosol mass concentrations. These aerosol measurement instruments sampled from a common silica gel dryer 5 (RH < 37%), which was downstream from a PM2.5 cyclone and a ~9 m long sample inlet with 10 cm inner diameter. A supplemental airflow was continually drawn through the sample inlet, and aerosols were drawn from the centerline of this inlet duct. Also at the ASRC site, a Leosphere Windcube 100S scanning LiDAR (hereafter LiDAR), operated from the roof of Center for Environmental Sciences and Technology Management building (about 15 m high). The LiDAR provided high resolution (25 m range gates) wind data and backscatter profile structure. 2.1 Measurement site and sampling instruments At the ACHD site, a Teledyne API Model T640 PM mass monitor 10 (hereafter ACHD T640) was used to provide the mass concentration of PM2.5, and black carbon was measured by a Teledyne API Aethalometer (model 633). 2.2 Meteorological parameters and back trajectory calculations Meteorological data was obtained from the Voorheesville New York State Mesonet station (NYSM, latitude: 42.65 °N, longitude: 73.92 °N, elevation: 100 m), located approximately 8 km southwest of the ASRC site. The Mesonet site provides meteorological 15 data in 5-minute intervals, and includes temperature (at 2 m and 9 m), relative humidity, redundant measurements of wind direction and wind speed, irradiance, and precipitation. On the night of the Independence Day FW event, the RH was above 90%, with maximum wind speeds generally below 2 m/s, indicating quiescent atmospheric conditions. Ten-hour air mass back trajectories were calculated using the NOAA ARL Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model, based on GDAS meteorological data (Draxler et al., 1998), and were used to study the effect of 20 transported FW burning aerosol on measured aerosol mass concentration at the uptown ASRC sampling site. The ending heights of 200 m, 500 m, and 1000 m were chosen to investigate the influence of different transport layers. 2.3 Data analysis 4 The HR-ToF-AMS data were analyzed using the standard HR-AMS data analysis software – SQUIRREL v1.59D and PIKA v1.19 (Allan et al., 2003; Canagaratna et al., 2007), to obtain the mass concentrations of different aerosol components (organics, 25 nitrates (NO3), sulfate (SO4), ammonium (NH4), chloride (Cl)), and their size distributions. The default relative ionization efficiency (RIE) values were used in the analysis (4 for ammonium, 1.1 for nitrate, 1.2 for sulfate, 1.3 for chloride, and 1.4 for 4 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. organics). The collection efficiencies (CE) used were those suggested in a previous study (Zhang et al., 2005), specifically a CE of 0.5 for inorganic compounds, and a CE of 0.7 for organic compounds based on the comparison with SMPS as shown in Fig. S2. The elemental analysis (oxygen-to-carbon (O:C) and hydrogen-to-carbon (H:C) ratio) is calculated following the improved method proposed by Canagaratna et al. (2015). Due to the slow evaporation and ionization of potassium (K) (Jiang et al., 2015; Slowik et al., 2010), quantification of K mass 5 concentrations (which are often relatively small) can be difficult and prone to large uncertainties, and potassium signals are often ignored when reporting ambient measurements. However, during FW events, K signals generally increase dramatically (Drewnick et al., 2006; Jiang et al., 2015), due to K-rich salts used in FW burning. To quantify the mass spectral signals and obtain the mass concentration of K, a RIE value needs to be specified. Previous work has used values of RIEk=10 (Slowik et al., Due to the slow evaporation and ionization of potassium (K) (Jiang et al., 2015; Slowik et al., 2010), quantification of K mass 5 concentrations (which are often relatively small) can be difficult and prone to large uncertainties, and potassium signals are often ignored when reporting ambient measurements. However, during FW events, K signals generally increase dramatically (Drewnick et al., 2006; Jiang et al., 2015), due to K-rich salts used in FW burning. To quantify the mass spectral signals and obtain the mass concentration of K, a RIE value needs to be specified. Previous work has used values of RIEk=10 (Slowik et al., 2010) or RIEk=2.9 (Drewnick et al., 2006; Jiang et al., 2015). 2.3 Data analysis In this work, an RIEK=2.9 would result in the highest 10 hourly-averaged value of K in the AMS data being as high as 36 μg m–3, and the total HR-ToF-AMS mass concentration would be about 58.7 μg m–3. That would be higher than the highest PM2.5 mass concentration measured in ACHD (55 μg m–3), a result that is not expected considering that HR-ToF-AMS only measures NR-PM1 species. Furthermore, a RIEK = 2.9 would result in a K/S (potassium to sulfur) ratio as high as 16, which deviates greatly from the value of 2.75 suggested by Drewnick et al (2006). A RIEk=10 results in lower K mass concentration by a factor of 3, and the resulting maximum for K hourly-averaged mass 15 concentration is 10.6 μg m–3, with the corresponding highest HR-ToF-AMS total aerosol mass concentration equal to 33.3 μg m–3, substantially lower than the ACHD PM2.5 mass concentration. For this RIE, the K/S ratio works out to be 5.3, nearer to the expected 2.75, and the K mass fraction during the maximum FW burning period is about 33.5%, which is consistent with the 30% result of Drewnick et al (2006). Therefore, an RIEK=10 for this study is reasonable. However, due to a lack of formal analysis about RIEk formulations, large uncertainties are still possible for the quantification of the mass concentration of K. For 20 the isotopic 41K+, a ratio of 0.0722 was used, and the total K mass concentration in the following section is inferred to be the combined mass concentration of K+ and 41K+ (Jiang et al., 2015). , p ( , 2010) or RIEk=2.9 (Drewnick et al., 2006; Jiang et al., 2015). In this work, an RIEK=2.9 would result in the highest 10 hourly-averaged value of K in the AMS data being as high as 36 μg m–3, and the total HR-ToF-AMS mass concentration would be about 58.7 μg m–3. That would be higher than the highest PM2.5 mass concentration measured in ACHD (55 μg m–3), a result that is not expected considering that HR-ToF-AMS only measures NR-PM1 species. Furthermore, a RIEK = 2.9 would result in a K/S (potassium to sulfur) ratio as high as 16, which deviates greatly from the value of 2.75 suggested by Drewnick et al (2006). 2010) or RIEk=2.9 (Drewnick et al., 2006; Jiang et al., 2015). 2.3 Data analysis In this work, an RIEK=2.9 would result in the highest 10 hourly-averaged value of K in the AMS data being as high as 36 μg m–3, and the total HR-ToF-AMS mass concentration would be about 58.7 μg m–3. That would be higher than the highest PM2.5 mass concentration measured in ACHD (55 μg m–3), a result that is not expected considering that HR-ToF-AMS only measures NR-PM1 species. Furthermore, a RIEK = 2.9 would result in a K/S (potassium to sulfur) ratio as high as 16, which deviates greatly from the value of 2.75 suggested by Drewnick et al (2006). A RIEk=10 results in lower K mass concentration by a factor of 3, and the resulting maximum for K hourly-averaged mass 15 concentration is 10.6 μg m–3, with the corresponding highest HR-ToF-AMS total aerosol mass concentration equal to 33.3 μg m–3, substantially lower than the ACHD PM2.5 mass concentration. For this RIE, the K/S ratio works out to be 5.3, nearer to the expected 2.75, and the K mass fraction during the maximum FW burning period is about 33.5%, which is consistent with the 30% result of Drewnick et al (2006). Therefore, an RIEK=10 for this study is reasonable. However, due to a lack of formal analysis about RIEk formulations, large uncertainties are still possible for the quantification of the mass concentration of K. For 20 the isotopic 41K+, a ratio of 0.0722 was used, and the total K mass concentration in the following section is inferred to be the combined mass concentration of K+ and 41K+ (Jiang et al., 2015). The HR-ToF-AMS organic mass spectra were analyzed using positive matrix factorization (PMF) (Paatero and Tapper, 1994) to resolve different organic aerosol factors, and the solutions were evaluated using the PMF Evaluation Tool (PET, v2.08D, Ulbrich et al., 2009). In this study, a five-factor solution with fpeak=0.2 (Q/Qexpected=3.9, Fig. S3) was chosen as the optimal result, based 25 on the evaluation of spectral profiles, diurnal variation, and correlations with external tracers (Fig. S4) (Zhang et al., 2011). 5 5 3 Result and discussion The time series of aerosol hourly-averaged mass concentration measured by all instruments exhibited the same general behavior during the measurement period (Fig. 1a). From June 29 12:00 until late in the day of June 30, the aerosol mass concentration displayed an increasing trend which also corresponded to high RH conditions (generally > 75%, Fig. 1b), likely due in part to hygroscopic aerosol growth and liquid-phase organic formation (Sun et al., 2011). This is supported by the steady increase in the 5 aerosol median diameter and the growth trend shown in the aerosol number size distribution (Fig. S5). A dramatic decrease in aerosol concentration occurred at June 30 14:00, due to scavenging by rain (Fig. 1b). From the overnight hours of July 2 onward, there were several aerosol concentration peaks at nighttime, and the biggest one was during the night of July 4, with maximum recorded concentration of 55 μg m–3 for ACHD T640 PM2.5 data and 33.3 μg m–3 for ASRC HR-ToF-AMS NR-PM1, which temporally extended into the early morning of July 5. These PM peaks are hypothesized to 10 be dominated by FW events, which are readily identifiable using the K signal. There are two reasons why the HR-ToF-AMS observed lower PM concentrations than the ACHD T640. One reason is that HR-ToF-AMS only measures non-refractory aerosol species below 1 μm (DeCarlo et al., 2006). The second is due to diffusion and dispersal of the aerosol plume during the transport from the source location of the FW burning to the uptown ASRC site. A time difference of about two hours between the peaks of the ACHD T640 data and ASRC instruments was observed, indicating the transport time of aerosol from the launch site 15 downtown to the ASRC measurements site, and providing for an advection velocity of about 1 m s–1. Excluding the high points during the night of July 4 (the FW event period), the high coefficient of determination (R2=0.72) between ACHD T640 and ASRC HR-ToF-AMS suggests generally similar conditions - typical of regional aerosol pollution processes during normal non-FW days. 3.1 Identification of firework events and aerosol composition 20 3.1 Identification of firework events and aerosol composition 20 Among the PM peaks, three were identified as significant FW events based on the prominent potassium signal peaks (Jiang et al., 2015; Drewnick et al., 2006). As shown in Fig. 2a, these three K peaks occurred during the nighttime hours of July 2, July 3, and July 4, with the highest peak occurring on the night of July 4. Observed concentrations of K were as high as 11.33 μg m–3 (July 5 00:00 EDT), which is about 63 times of the background value (0.18 μg m–3, averaged July 4 09:00 to 17:00 EDT). The high K signal was present until about 09:00 EDT the following morning. Based on the time variation of K, we define the Independence 25 Day FW event time period as July 4 23:00 to July 5 02:00 EDT. Apart from these three prominent peaks, there were still some spikes of K during the measurement period, likely related to small-scale, localized, or more distant firework burning. Among the PM peaks, three were identified as significant FW events based on the prominent potassium signal peaks (Jiang et al., 2015; Drewnick et al., 2006). As shown in Fig. 2a, these three K peaks occurred during the nighttime hours of July 2, July 3, and July 4, with the highest peak occurring on the night of July 4. Observed concentrations of K were as high as 11.33 μg m–3 (July 5 00:00 EDT), which is about 63 times of the background value (0.18 μg m–3, averaged July 4 09:00 to 17:00 EDT). The high K signal was present until about 09:00 EDT the following morning. Based on the time variation of K, we define the Independence 25 Day FW event time period as July 4 23:00 to July 5 02:00 EDT. Apart from these three prominent peaks, there were still some spikes of K during the measurement period, likely related to small-scale, localized, or more distant firework burning. Among the PM peaks, three were identified as significant FW events based on the prominent potassium signal peaks (Jiang et al., 2015; Drewnick et al., 2006). As shown in Fig. 2a, these three K peaks occurred during the nighttime hours of July 2, July 3, and July 4, with the highest peak occurring on the night of July 4. 3.1 Identification of firework events and aerosol composition 20 Observed concentrations of K were as high as 11.33 μg m–3 (July 5 00:00 EDT), which is about 63 times of the background value (0.18 μg m–3, averaged July 4 09:00 to 17:00 EDT). The high K 6 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. During the Independence Day FW event period (July 4 23:00 to July 5 02:00 EDT as defined above), the averaged mass concentrations of total aerosol, organics, K, SO4, NO3, and Cl were all clearly elevated compared to earlier that day (July 4 9:00 to 17:00 EDT), increasing from 6.0 μg m–3 to 27.2 μg m–3 (4.5 times), 5.0 μg m–3 to 12.1 μg m–3 (2.4 times), 0.2 μg m–3 to 9.1 μg m–3 (45.5 times), 0.5 μg m–3 to 4.1 μg m–3 (8.2 times), 0.09 μg m–3 to 0.8 μg m–3 (8.8 times), and 0.01 μg m–3 to 0.84 μg m–3 (84 times) respectively (Fig. 2a). These enhancements are similar to the results of the Drewnick et al. (2006) study. Absolute 5 concentrations of chloride from the FW burning were relatively low (compared to potassium) due to the semi-refractory character of metal chlorides (Drewnick et al., 2006). Still, the rise in chloride concentration was dramatic, with values nearly 84 times greater than the daily averaged value. The observed FW events had only minor effects on NH4, as in the Drewnick et al. (2006) study, due to a lack of NH4-containing material in the fireworks, and also indicates the increased SO4 and NO3 came from K-rich salt instead of (NH4)2SO4 and NH4NO3. The time series shows that organics and NO3 sustained broader peaks than K and 10 SO4, possibly due to contribution of the vehicular traffic emission to organics and NO3 and the nighttime NO3 formation (Sun et al., 2011; Xu et al., 2015). Before and after these three FW events (before July 1 00:00 and after July 5 12:00 EDT), the mass fraction of the HR-ToF-AMS chemical components was fairly stable, with organic compounds constituting the major fraction of NR-PM1, with 70.5% (before July 1) and 81.1% (after July 5) on average, followed by SO4 (16.4% and 8.5% respectively, Fig. 2b and Fig. S6). 3.2.1 PMF results 5 Based on the evaluation of spectral profiles, diurnal variation, and correlations with external tracers, five organic aerosol (OA) components, including hydrocarbon-like OA (HOA), a biomass burning OA (BBOA), a semi-volatile oxygenated OA (SV-OOA), and two low-volatility oxygenated OAs (LV-OOAs, one is named as FW-OOA and another one is named as LV-OOA) are identified, as shown in Fig. 4 and Fig. S3. HOA and BBOA are the two smallest components in this data set. The HOA is characterized by hydrocarbon-like ions (CxHy+ family), with evident morning and evening rush hour peaks. The BBOA 10 is characterized by prominent signals at m/z 60 (C2H4O2+) and m/z 73 (C3H5O2+), with increasing concentrations during evening hours. The sum of HOA and BBOA had similar variation in its time series as the BC data from ADHC (Fig. S4a) with a moderate R2 (0.38). These two components are mainly identified with vehicular traffic and wood-fire emissions, such as bonfires, barbeques, or other small-scale celebrations. The SV-OOA spectral profile is characterized by a O:C ratio of roughly 0.6, and a high f43/f44 ratio of ~1 (Ng et al., 2010). SV-OOA also exhibited daily maxima during mid-afternoon, and it was moderately 15 correlated with concurrent NO3 before July 1 00:00 EDT, with a R2 = 0.48, giving some indication of the local source characteristics of SV-OOA formation (Zhou et al., 2015). The mass spectra of the two LV-OOAs were similar in this study, as shown in Fig. S3c, but they demonstrated different time series behaviour (Fig. S3d), indicating the presence of different types of OOAs. In comparison to LV-OOA, FW-OOA contains a higher H:C ratio (1.33 vs. 1.20), and higher signals at m/z 29 (CHO+: 0.05 vs. 0.01), m/z 60 (C2H4O2+: 0.006 vs. 0.0004) and m/z 20 73 ( C3H5O2+: 0.004 vs. 0.0008), as shown in Fig. 5. The FW-OOA is believed to be related to the organic aerosol released by FW due to similar variation trends in the K signal (Fig. S4c), with a high coefficient of determination (R2=0.70). Compared to previous AMS fireworks papers, this is the first time that this form of organic aerosol has been separated from overall FW burning emission and contributed ~51.2% (6.2 μg m–3) of the organic aerosol (12.1 μg m–3) during the Independence Day FW event period. 3.1 Identification of firework events and aerosol composition 20 The stable 15 component ratio indicates relatively constant PM1 sources in and near Albany, suggesting a stable aerosol evolution process during these two time periods. A higher ratio of organics after FW events may be the result of the primary emissions of organics during the FW events, or the formation of new secondary organic aerosol (Li et al., 2013; Kong et al., 2015, Wang et al., 2007). During the FW events, and especially during the Independence Day FW event, most aerosol components clearly increased, and there was a large difference in the mass fractions of aerosol components compared to before and after periods, highlighting the 20 influence of FW burning. The indicator of FW events, K, displayed large increases during the FW events, contributing roughly 33% to the ambient aerosol during the four-hour FW event period, and up to 45 % of ambient aerosol at the peak hour of July 4 23:00 EDT. 7 Most of the Independence Day FW PM1 aerosol located at the size range of 200 to 500 nm (volume mobility diameter, Fig. 3a), and FW-averaged size distribution of NR-PM1 aerosol chemical components (Fig. 3b) showed externally mixed characteristics, 25 with sulfate and nitrate peaking at ~550 nm (Dva: vacuum aerodynamic diameter), organics peaking at ~400 nm (Dva). The size distribution of K is complicated due to poorly characterized surface ionization properties of the evaporated K-containing species and is not presented here. Different peaks for the different chemical components suggest a non-uniform mixing state for organics, 7 Most of the Independence Day FW PM1 aerosol located at the size range of 200 to 500 nm (volume mobility diameter, Fig. 3a), and FW-averaged size distribution of NR-PM1 aerosol chemical components (Fig. 3b) showed externally mixed characteristics, 25 with sulfate and nitrate peaking at ~550 nm (Dva: vacuum aerodynamic diameter), organics peaking at ~400 nm (Dva). The size distribution of K is complicated due to poorly characterized surface ionization properties of the evaporated K-containing species and is not presented here. Different peaks for the different chemical components suggest a non-uniform mixing state for organics, Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Atmos. Chem. Phys. 3.1 Identification of firework events and aerosol composition 20 Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. and sulfate/nitrate, possibly related to different kinds of fresh emitted aerosols from FW. As discussed above, the SO4 and NO3 were likely mainly from K-rich inorganic salt, such as K2SO4 or KNO3, and organics are from burning organic materials or reaction with FW oxidizer, which will be discussed in next section. 3.2.1 PMF results 5 FW-OOA is likely directly emitted from FW (Jiang et al., 2014, Drewnick et al, 2006), and produced by the 25 chemical reaction of binding agents, such as dextrin, with the oxidizers. 8 8 The LV-OOA shows a more traditional spectral character comparable to those seen in earlier studies. It has a high O:C ratio (0.84), low f43/f44 ratio (0.25), and a high coefficient of determination (R2=0.86) with concurrent SO4 before July 1 00:00 EDT. This is thought to be related to highly oxidized secondary organic aerosol (Ng et al., 2010). Here we only considered the time period before the FW events to calculate correlation coefficients, since the SO4 and NO3 from burning fireworks will change the basic relationship with background organics. In this study, SV-OOA showed a high coefficient of determination (R2=0.94) with 5 LV-OOA before FW events (Fig. S4d), indicating that the bulk of the LV-OOA at Albany was likely produced locally through oxidation of SV-OOA during the measurement period, instead of being carried in via long-distance transport. 3.2.2 Mass closure and source apportionment from FW events (July 4th) Assuming that the aerosol is charge balanced and that the ionic species can be identified or assumed from the mass spectra (Wang et al., 2016), the different inorganic salts can be estimated using ion-balance considerations. NH4 is apportioned first into 10 (NH4)2SO4 and then into NH4NO3, based on the equivalent ratio of SO4 to NO3. Cl is ignored in this context due to its very low concentration before FW events.. The residual SO4, (calculated by subtracting SO4-in-(NH4)2SO4 from total SO4), the residual NO3, and Cl are combined with measured K to form K2SO4, KNO3, and KCl, which are used to oxidize firework mixtures. These K compounds would be assumed to comprise inorganic K, while the residual K, calculated by subtracting inorganic K from total (Wang et al., 2016), the different inorganic salts can be estimated using ion-balance considerations. NH4 is apportioned first into 10 (NH4)2SO4 and then into NH4NO3, based on the equivalent ratio of SO4 to NO3. Cl is ignored in this context due to its very low concentration before FW events.. The residual SO4, (calculated by subtracting SO4-in-(NH4)2SO4 from total SO4), the residual NO3, and Cl are combined with measured K to form K2SO4, KNO3, and KCl, which are used to oxidize firework mixtures. These K compounds would be assumed to comprise inorganic K, while the residual K, calculated by subtracting inorganic K from total 9 K, would be treated as organopotassium. Organopotassium is strongly correlated with FW-OOA, as shown in Fig. S4c. The 15 density for each inorganic salt was averaged together, and assuming the density of organics factors (including those containing organopotassium) is 1.2 g m–3, the density of the ambient aerosol bulk was then estimated to be mostly in the range 1.3 – 1.5 g m–3, as shown in Fig. S2. Figure S2 also shows that HR-ToF-AMS and SMPS determined mass concentrations exhibit similar behavior, except that the mass concentration measured by the SMPS was as much as 12 μg m–3 larger than that from the AMS during the Independence Day FW event. This may be caused by the uncertainty of the RIE for K, the semi-refractory character of 20 other inorganic metal salts, and/or the exiting of BC producing from the black powder explosives (Drewnick et al., 2015; Wang et al., 2016). Figure 6 shows the time series of each inorganic salt from the ion balance calculation, as well as the different OA components. 3.2.2 Mass closure and source apportionment from FW events (July 4th) Here, OrgK is the sum of Organopotassium and FW-OOA, OOA is the sum of LV-OOA and SV-OOA, and “others” indicates the difference between SMPS and HR-ToF-AMS (the aerosols not measured by HR-ToF-AMS). OOA, (NH4)2SO4, and NH4NO3 25 showed no significant trends in concentration from July 4 12:00 to the end of the period of study, indicating that these species comprised the local background concentration, and contributing about 13.8% of total PM during the FW event period (07/04 23:00-07/05 02:00). BBOA+HOA exhibited consistently elevated concentrations from the night of July 4 to early morning of K, would be treated as organopotassium. Organopotassium is strongly correlated with FW-OOA, as shown in Fig. S4c. The 15 density for each inorganic salt was averaged together, and assuming the density of organics factors (including those containing organopotassium) is 1.2 g m–3, the density of the ambient aerosol bulk was then estimated to be mostly in the range 1.3 – 1.5 g m–3, as shown in Fig. S2. Figure S2 also shows that HR-ToF-AMS and SMPS determined mass concentrations exhibit similar behavior, except that the mass concentration measured by the SMPS was as much as 12 μg m–3 larger than that from the AMS during the Independence Day FW event. This may be caused by the uncertainty of the RIE for K, the semi-refractory character of 20 other inorganic metal salts, and/or the exiting of BC producing from the black powder explosives (Drewnick et al., 2015; Wang et al., 2016). K, would be treated as organopotassium. Organopotassium is strongly correlated with FW-OOA, as shown in Fig. S4c. The 15 density for each inorganic salt was averaged together, and assuming the density of organics factors (including those containing organopotassium) is 1.2 g m–3, the density of the ambient aerosol bulk was then estimated to be mostly in the range 1.3 – 1.5 g m–3, as shown in Fig. S2. Figure S2 also shows that HR-ToF-AMS and SMPS determined mass concentrations exhibit similar behavior, except that the mass concentration measured by the SMPS was as much as 12 μg m–3 larger than that from the AMS during the Independence Day FW event. This may be caused by the uncertainty of the RIE for K, the semi-refractory character of 20 other inorganic metal salts, and/or the exiting of BC producing from the black powder explosives (Drewnick et al., 2015; Wang et al., 2016). 3.2.2 Mass closure and source apportionment from FW events (July 4th) 9 Figure 6 shows the time series of each inorganic salt from the ion balance calculation, as well as the different OA components. Here, OrgK is the sum of Organopotassium and FW-OOA, OOA is the sum of LV-OOA and SV-OOA, and “others” indicates the difference between SMPS and HR-ToF-AMS (the aerosols not measured by HR-ToF-AMS). OOA, (NH4)2SO4, and NH4NO3 25 showed no significant trends in concentration from July 4 12:00 to the end of the period of study, indicating that these species comprised the local background concentration, and contributing about 13.8% of total PM during the FW event period (07/04 23:00-07/05 02:00). BBOA+HOA exhibited consistently elevated concentrations from the night of July 4 to early morning of Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. July 5, likely be related to traffic emissions, wood, and charcoal smoke, a result that is supported by the large numbers of smaller particles emitted before and after FW event. BBOA+HOA contributed a relatively small amount, about 7.2% of the total PM. The biggest contribution came from FW burning, as inorganic and organic K-rich aerosol (including “others” compounds) comprised 79% of the total (26.1 μg m–3 of 33.0 μg m–3). Assuming this ratio did not change during transport from the launch site, the metal-rich aerosol increases caused by the FW burning in downtown Albany (the ACHD site) would have reached a averaged 5 value as high as 43.0 μg m–3 (07/04 21:00-07/04 23:00), higher than the NAAQS PM2.5 24-hr threshold value of 35 μg m–3, and considered harmful to public health. After the FW display, the high LV-OOA mass concentration and low SO4/NO3 mass concentration persisted to the end of the period of study, with an increased f44 ratio and decreasing f43 ratio (indicators of aged aerosol, as shown in Fig. S7). This implies that FW-OOA may be converted into LV-OOA in the ambient atmosphere, and demonstrates the persistent effects of FW burning (Kong et al., 2014). 10 3.3 Pollution processing and meteorological conditions The wind LiDAR on the roof of the ASRC uptown site reports data as the carrier-to-noise ratio (CNR) and wind direction, and this data is used to describe the aerosol transport and mixing process (Aitken et al., 2012). The LiDAR data for the Independence Day FW event is shown in Fig. 7. During the late evening on Independence Day (July 4, 20:00- 23:00 EDT, July 5 00:00-02:00 UTC), the winds above the ASRC sampling site were weak easterlies, and from the late night onward (July 4 23:00 local time, 15 Jul 5 03:00 UTC) there was a temporal maximum in low-level CNR as shown in Fig. 7, concurrent with a high-concentration aerosol cluster over the site at about 400 meters above observation level, as indicated by relatively cool colors representing higher CNR values. This burst of high CNR values is thought to result from the high-altitude FW aerosol transport from the celebrations in downtown Albany, east of the site. In the following half hour, a lower altitude aerosol cluster occurred above the ASRC sampling site, coincident with the increases in aerosol mass concentrations measured by the ASRC instruments. This 20 lower-altitude aerosol cluster may have been caused by subsidence or diffusion, which is the mixing the high-altitude aerosol with the lower atmosphere, or it may have been caused by the aerosol emission of low-altitude FW transported westward from Empire State Plaza. The aerosol cluster lingered over the ASRC site for almost ten hours, matching up with the persistent and slowly decaying aerosol mass concentrations measured by the other instruments. At about 07:00 EDT (11 UTC) on July 5, there was another high-altitude (1000 m) and near-ground aerosol cluster that passed over the ASRC site, producing a second aerosol 25 peak measured on-site. It is quite plausible that this is related to transport of FW aerosol from other areas. A 10-hour back trajectory shows the 500 m wind passing over the town of Bennington, Vermont (Fig. S8), which had its own FW celebration on the evening of July 4. From July 4 23:00 – Jul 5 08:00 EDT, the lower surface layer wind speed was low, suggesting that the The wind LiDAR on the roof of the ASRC uptown site reports data as the carrier-to-noise ratio (CNR) and wind direction, and this data is used to describe the aerosol transport and mixing process (Aitken et al., 2012). 3.3 Pollution processing and meteorological conditions The LiDAR data for the Independence Day FW event is shown in Fig. 7. During the late evening on Independence Day (July 4, 20:00- 23:00 EDT, July 5 00:00-02:00 UTC), the winds above the ASRC sampling site were weak easterlies, and from the late night onward (July 4 23:00 local time, 15 Jul 5 03:00 UTC) there was a temporal maximum in low-level CNR as shown in Fig. 7, concurrent with a high-concentration aerosol cluster over the site at about 400 meters above observation level, as indicated by relatively cool colors representing higher CNR values. This burst of high CNR values is thought to result from the high-altitude FW aerosol transport from the celebrations in downtown Albany, east of the site. In the following half hour, a lower altitude aerosol cluster occurred above the ASRC sampling site, coincident with the increases in aerosol mass concentrations measured by the ASRC instruments. This 20 lower-altitude aerosol cluster may have been caused by subsidence or diffusion, which is the mixing the high-altitude aerosol with the lower atmosphere, or it may have been caused by the aerosol emission of low-altitude FW transported westward from Empire State Plaza. The aerosol cluster lingered over the ASRC site for almost ten hours, matching up with the persistent and slowly decaying aerosol mass concentrations measured by the other instruments. At about 07:00 EDT (11 UTC) on July 5, there was another high-altitude (1000 m) and near-ground aerosol cluster that passed over the ASRC site, producing a second aerosol 25 peak measured on-site. It is quite plausible that this is related to transport of FW aerosol from other areas. A 10-hour back trajectory shows the 500 m wind passing over the town of Bennington, Vermont (Fig. S8), which had its own FW celebration on the evening of July 4. From July 4 23:00 – Jul 5 08:00 EDT, the lower surface layer wind speed was low, suggesting that the 10 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. aerosol cluster was near static, and about 12 hours was required (Jul 4 23:00 – July 5 11:00 EDT) to return to normal aerosol concentration levels (~6 μg m–3). 4 Conclusion Chemical characterization and source apportionment of the submicron aerosols from FW displays during the 2017 Independence Day holiday are quantified here for the city of Albany, New York. The hourly maximum mass concentration of NR-PM1 measured using the ASRC HR-ToF-AMS was 33.3 μg m–3, almost 5 times the daytime background mass concentration, while 10 PM2.5 at the downtown ACHD site was ~67% higher (55 ug m–3). There were significantly elevated aerosol chemical component mass concentrations (organics, K, sulfate, chloride) as a result of the Independence Day FW display, especially K, which spiked from 0.2 μg m–3 to 9.1 μg m–3, signifying the influence of burning fireworks. The size distribution of aerosol compounds showed externally mixed characteristics, with sulfate and nitrate peaking at ~550 nm (Dva), organics peaking at ~400 nm (Dva). PMF analysis revealed unique oxidized organic aerosol compounds (FW-OOA), which showed a high 15 correlation (R2=0.8) with the variations in K. FW-OOA shows a higher ratio at m/z 60 and m/z 73, and a lower m/z 43 and m/z 44 than the general high oxidized secondary organic aerosol does, and contributed about 51.2% (6.2 μg m–3 of 12.1 μg m–3) of total organics during FW events. Our study shows that the total contribution from fireworks to the total aerosol concentration could be as high as 79% (26.1 μg m–3 of 33.3 μg m–3) based on source apportionment calculations, while congested vehicular traffic and wood-burning for smaller events like such as bonfires and barbeques contribute another 7.2% during the 20 Independence Day FW event (07/04 23:00-07/05 02:00 EDT). The morning following the FW display, another aerosol mass peak occurred, which is likely the transport of FW aerosol from other towns, most likely from Bennington, Vermont. Under this kind of weak advective mixing, during the first several hours after FW displays, enhanced aerosols with rich metal materials could be very harmful to the nearby area residents, especially to the people nearest the display area and living in the downwind direction, and this condition could be more severe for areas of high population density. 25 Chemical characterization and source apportionment of the submicron aerosols from FW displays during the 2017 Independence Day holiday are quantified here for the city of Albany, New York. 3.3 Pollution processing and meteorological conditions After July 5 09:00 EDT (13 UTC), the aerosol layer was again elevated, impacted by the increasing planetary boundary layer height, and ASRC instruments recorded rapid reductions in aerosol mass concentration. The firework emissions, which occurred in an environment with a static planetary boundary layer, had combined with the transported FW aerosol from other celebrations in the region and resulted in high aerosol mass concentrations over the Albany area for 5 several hours, lasting through the night. aerosol cluster was near static, and about 12 hours was required (Jul 4 23:00 – July 5 11:00 EDT) to return to normal aerosol concentration levels (~6 μg m–3). After July 5 09:00 EDT (13 UTC), the aerosol layer was again elevated, impacted by the increasing planetary boundary layer height, and ASRC instruments recorded rapid reductions in aerosol mass concentration. The firework emissions, which occurred in an environment with a static planetary boundary layer, had combined with the transported FW aerosol from other celebrations in the region and resulted in high aerosol mass concentrations over the Albany area for 5 several hours, lasting through the night. References: Allan, J. D., Jimenez, J. L., Williams, P. I., Alfarra, M. R., Bower, K. N., Jayne, J. T., Coe, H., and Worsnop, D. R.: Quantitative sampling using an Aerodyne aerosol mass spectrometer 1. Techniques of data interpretation and error analysis, Journal of Geophysical Research: Atmospheres, 108, D34090, 1-10, doi:10.1029/2002JD002358, 2003. Allan, J. D., Jimenez, J. L., Williams, P. I., Alfarra, M. R., Bower, K. N., Jayne, J. T., Coe, H., and Worsnop, D. R.: Quantitative sampling using an Aerodyne aerosol mass spectrometer 1. Techniques of data interpretation and error analysis, Journal of Geophysical Research: Atmospheres, 108, D34090, 1-10, doi:10.1029/2002JD002358, 2003. Barman, S. C., Singh, R., Negi, M. P., and Bhargava, S. K.: Ambient air quality of Lucknow City (India) during use of fireworks 5 on Diwali Festival, Environmental monitoring and assessment, 137, 495-504, doi:10.1007/s10661-007-9784-1, 2008. Canagaratna M R, Jayne J T, Jimenez J L, Allan, J. D., Alfarra, M. R., Zhang, Q., Onasch, T. B., Drewnick, F., Coe, H., Middlebrook, A., Delia, A., Williams, L. R., Trimborn, A. M., Northway, M. J., DeCarlo, P. F., Kolb, C. E., Davidovits, P., and Worsnop, D. R.: Chemical and microphysical characterization of ambient aerosols with the aerodyne aerosol mass spectrometer, Mass Spectrometry Reviews, 26, 185-222, doi:10.1002/mas.20115, 2007. 10 Canagaratna, M. R., Jimenez, J. L., Kroll, J. H., Chen, Q., Kessler, S. H., Massoli, P., Hildebrandt Ruiz, L., Fortner, E., Williams, L. R., Wilson, K. R., Surratt, J. D., Donahue, N. M., Jayne, J. T., and Worsnop, D. R.: Elemental ratio measurements of organic compounds using aerosol mass spectrometry: characterization, improved calibration, and implications, Atmospheric Chemistry and Physics, 15, 253-272, doi:10.5194/acp-15-253-2015, 2015. Dall'Osto, M, and Harrison, R. M.: Urban organic aerosols measured by single particle mass spectrometry in the megacity of 15 London, Atmospheric Chemistry and Physics, 12, 4127-4142, doi:10.5194/acp-12-4127-2012, 2012. DeCarlo, P. F., Kimmel, J. R., Trimborn, A., Northway, M. J., Jayne, J. T., Aiken, A. C., Gonin, M., Fuhrer, K., Horvath Thomas., Docherty, K. S., Worsnop, D. R., and Jimenez, J. L.: Field-deployable, high-resolution, time-of-flight aerosol mass spectrometer, Analytical chemistry, 78, 8281-8289, doi: 10.1021/ac061249n, 2006. DeCarlo, P. F., Dunlea, E. J., Kimmel, J. R., Aiken, A. C., Sueper, D., Crounse, J., Wennberg, P. O., Emmons, L., Shinozuka, Y., 20 Clarke, A., Zhou, J., Tomlinson, J., Collins, D. R., Knapp, D., Weinheimer, A. J., Montzka, D. D., Campos, T., and Jimenez, J. 4 Conclusion The hourly maximum mass concentration of NR-PM1 measured using the ASRC HR-ToF-AMS was 33.3 μg m–3, almost 5 times the daytime background mass concentration, while 10 PM2.5 at the downtown ACHD site was ~67% higher (55 ug m–3). There were significantly elevated aerosol chemical component mass concentrations (organics, K, sulfate, chloride) as a result of the Independence Day FW display, especially K, which spiked from 0.2 μg m–3 to 9.1 μg m–3, signifying the influence of burning fireworks. The size distribution of aerosol compounds showed externally mixed characteristics, with sulfate and nitrate peaking at ~550 nm (Dva), organics peaking at compounds showed externally mixed characteristics, with sulfate and nitrate peaking at ~550 nm (Dva), organics peaking at ~400 nm (Dva). PMF analysis revealed unique oxidized organic aerosol compounds (FW-OOA), which showed a high 15 correlation (R2=0.8) with the variations in K. FW-OOA shows a higher ratio at m/z 60 and m/z 73, and a lower m/z 43 and m/z 44 than the general high oxidized secondary organic aerosol does, and contributed about 51.2% (6.2 μg m–3 of 12.1 μg m–3) of total organics during FW events. Our study shows that the total contribution from fireworks to the total aerosol concentration could be as high as 79% (26.1 μg m–3 of 33.3 μg m–3) based on source apportionment calculations, while congested vehicular traffic and wood-burning for smaller events like such as bonfires and barbeques contribute another 7.2% during the 20 11 11 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Acknowledgements. This work has been supported by the New York State Energy Research and Development Authority (NYSERDA) contract number 48971. Special thanks go to New York State Department of Health for providing data from the ACHD site. Competing interests. The authors declare that they have no conflict of interest. 5 12 12 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. c⃝Author(s) 2018. CC BY 4.0 License. References: W., Dall’Osto, M., Beddows, D. C., Kelly, F. J., Harrison, R. M., and Particulate oxidative burden associated with firework activity, Environmental science & technology, 44, 8295-8301, doi: 10 10.1021/es1016284, 2010. Jayne, J. T., Leard, D. C., Zhang, X. F., Davidovits, P., Smith, K. A. , Kolb, C. E., and Worsnop, D. R.: Development of an aerosol mass spectrometer for size and composition analysis of submicron particles, Aerosol Science & Technology, 33, 49-70, doi: 10.1080/027868200410840, 2000. Jayne, J. T., Leard, D. C., Zhang, X. F., Davidovits, P., Smith, K. A. , Kolb, C. E., and Worsnop, D. R.: Development of an aerosol mass spectrometer for size and composition analysis of submicron particles, Aerosol Science & Technology, 33, 49-70, doi: 10.1080/027868200410840, 2000. Jiang, Q., Sun, Y. L., Wang, Z., and Yin, Y.: Aerosol composition and sources during the Chinese Spring Festival: fireworks, 15 secondary aerosol, and holiday effects, Atmospheric Chemistry and Physics, 15, 6023-6034, doi:10.5194/acp-15-6023-2015, 2015. Jimenez, J. L., Canagaratna, M. R., Donahue, N. M., Prevot, A. S. H., Zhang, Q., Kroll, J. H., DeCarlo, P. F., Allan, J. D., Coe, H., Ng, N. L., Aiken, A. C., Docherty, K. S., Ulbrich, I. M., Grieshop, A. P., Robinson, A. L., Duplissy, J., Smith, J. D., Wilson, Jiang, Q., Sun, Y. L., Wang, Z., and Yin, Y.: Aerosol composition and sources during the Chinese Spring Festival: fireworks, 15 secondary aerosol, and holiday effects, Atmospheric Chemistry and Physics, 15, 6023-6034, doi:10.5194/acp-15-6023-2015, 2015. Jimenez, J. L., Canagaratna, M. R., Donahue, N. M., Prevot, A. S. H., Zhang, Q., Kroll, J. H., DeCarlo, P. F., A H., Ng, N. L., Aiken, A. C., Docherty, K. S., Ulbrich, I. M., Grieshop, A. P., Robinson, A. L., Duplissy, J., K. R., Lanz, V. A., Hueglin, C., Sun, Y. L., Tian, J., Laaksonen, A., Raatikainen, T., Rautiainen, J., Vaattovaara, P., Ehn, M., 20 Kulmala, M., Tomlinson, J. M., Collins, D. R., Cubison, M. J., Dunlea, E. J., Huffman, J. A., Onasch, T. B., Alfarra, M. R., Williams, P. I., Bower, K., Kondo, Y., Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demerjian, K., Salcedo, D., Cottrell, L., Griffin, R., Takami, A., Miyoshi, T., Hatakeyama, S., Shimono, A., Sun, J. Y., Zhang, Y. M., Dzepina, K., Kimmel, J. R., Sueper, D., Jayne, J. T., Herndon, S. C., Trimborn, A. M., Williams, L. R., Wood, E. C., Middlebrook, A. M., Kolb, C. E., Baltensperger, U., and Worsnop, D. References: L.: Fast airborne aerosol size and chemistry measurements above Mexico City and Central Mexico during the MILAGRO campaign, Atmospheric Chemistry and Physics, 8, 4027-4048, doi:10.5194/acp-8-4027-2008, 2008. Dickerson, A. S., Benson, A. F., Buckley, B., and Chan, E. A. W.: Concentrations of individual fine particulate matter ickerson, A. S., Benson, A. F., Buckley, B., and Chan, E. A. W.: Concentrations of individual fine parti components in the USA around July 4th, Air Quality, Atmosphere & Health, 10, 349-358, doi:10.1007/s11869-016-0433-0, 2017. 25 Draxler, R. R., and Hess, G. D.: An overview of the HYSPLIT_4 modelling system for trajectories. Australian meteorological magazine, 47, 295-3, 1998. 13 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Drewnick, F., Hings, S. S., DeCarlo, P., Jayne, J. T., Gonin, M., Fuhrer, K., Weimer S., Jimenez, J. L., Demerjian, K. L., Borrmann, S., and Worsnop, D. R.: A new time-of-flight aerosol mass spectrometer (ToF-AMS) - Instrument description and first field deployment, Aerosol Science and Technology, 39, 637-658, doi: 10.1080/02786820500182040, 2005. Drewnick, F., Hings, S. S., Curtius J, Eerdekens, G., and Williams, J.: Measurement of fine particulate and gas-phase species during the New Year's fireworks 2005 in Mainz, Germany, Atmospheric Environment, 40, 4316-4327, 5 10.1016/j.atmosenv.2006.03.040, 2006. Holstius, D. M., Pillarisetti, A., Smith, K. R., and Seto, E.: Field Calibrations of a Low-Cost Aerosol Sensor at a Regulatory Monitoring Site in California, Atmos. Meas. Tech. 7, 1121-1131, doi:10.5194/amt-7-1121-2014, 2014. Godri K. J., Green, D. C., Fuller, G. W., Dall’Osto, M., Beddows, D. C., Kelly, F. J., Harrison, R. M., and Mudway, I. S.: Drewnick, F., Hings, S. S., DeCarlo, P., Jayne, J. T., Gonin, M., Fuhrer, K., Weimer S., Jimenez, J. L., Demerjian, K. L., Borrmann, S., and Worsnop, D. R.: A new time-of-flight aerosol mass spectrometer (ToF-AMS) - Instrument description and first field deployment, Aerosol Science and Technology, 39, 637-658, doi: 10.1080/02786820500182040, 2005. Holstius, D. M., Pillarisetti, A., Smith, K. R., and Seto, E.: Field Calibrations of a Low-Cost Aerosol Sensor Holstius, D. M., Pillarisetti, A., Smith, K. R., and Seto, E.: Field Calibrations of a Low-Cost Aerosol Sensor at a Regulatory Monitoring Site in California, Atmos. Meas. Tech. 7, 1121-1131, doi:10.5194/amt-7-1121-2014, 2014. Monitoring Site in California, Atmos. Meas. Tech. 7, 1121-1131, doi:10.5194/amt-7-1121-2014, 2014. Monitoring Site in California, Atmos. Meas. Tech. 7, 1121-1131, doi:10.5194/amt-7-1121-2014, 2014. Godri K. J., Green, D. C., Fuller, G. References: R.: Evolution of organic aerosols in the atmosphere, Science, 326, 1525 – 1529, 25 doi:10.1126/science.1180353, 2009. Joly, A., Smargiassi, A., Kosatsky, T., Fournier, M., Dabek-Zlotorzynska, E., Celo, V., Mathieu, D., Servranckx, R., D'amours, R., Malo, A., and Brook, J.. Characterisation of particulate exposure during fireworks displays, Atmospheric Environment, 44, 4325-4 9, doi:10.1016/j.atmosenv.2009.12.010, 2010. 14 c⃝Author(s) 2018. CC BY 4.0 License. Kong, S. F., Li, L., Li, X. X., Yin, Y., Chen, K., Liu, D. T., Yuan, L., Zhang, Y. J., Shan, Y. P., and Ji, Y. Q.: The impacts of firework burning at the Chinese Spring Festival on air quality: insights of tracers, source evolution and aging processes, Atmospheric Chemistry and Physics, 15, 2167-2184, doi:10.5194/acp-15-2167-2015, 2015. Kong, S. F., Li, L., Li, X. X., Yin, Y., Chen, K., Liu, D. T., Yuan, L., Zhang, Y. J., Shan, Y. P., and Ji, Y. Q.: The impacts of firework burning at the Chinese Spring Festival on air quality: insights of tracers, source evolution and aging processes, Atmospheric Chemistry and Physics, 15, 2167-2184, doi:10.5194/acp-15-2167-2015, 2015. Lin, C. C.: A review of the impact of fireworks on particulate matter in ambient air, Journal of the Air & Waste Management Association 66 1171 1182 doi: 10 1080/10962247 2016 1219280 2016 5 Lin, C. C.: A review of the impact of fireworks on particulate matter in ambient air, Journal of the Air & Waste Management Association, 66, 1171-1182, doi: 10.1080/10962247.2016.1219280, 2016. 5 Moreno T, Querol X, Alastuey A, Minguillón, M. C., Pey, J., Rodriguez, S., Miró, J. V., Felis, C., and Gibbons, W.: Recreational atmospheric pollution episodes: inhalable metalliferous particles from firework displays, Atmospheric Environment, 41, 913-922, doi:10.1016/j.atmosenv.2006.09.019, 2007. Ng N L, Canagaratna M R, Zhang Q, Jimenez, J. L., Tian, J., Ulbrich, I. M., Kroll, J. H., Docherty, K. S., Chhabra, P. S., Ng N L, Canagaratna M R, Zhang Q, Jimenez, J. L., Tian, J., Ulbrich, I. M., Kroll, J. H., Docherty, K. S., Chhabra, P. S., Ng N L, Canagaratna M R, Zhang Q, Jimenez, J. L., Tian, J., Ulbrich, I. M., Kroll, J. H., Docherty, K. S. Bahreini, R., Murphy, S. M., Seinfeld, J. H., Hildebrandt, L., Donahue, N. M., DeCarlo, P. F., Lanz, V. A., Prévôt, A. S. H., 10 Dinar, E., Rudich, Y., and Worsnop, D. R.: Organic aerosol components observed in Northern Hemispheric datasets from Aerosol Mass Spectrometry, Atmospheric Chemistry and Physics, 10(10): 4625-4641, doi:10.5194/acp-10-4625-2010, 2010. Paatero, P. References: and Tapper, U.: Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values, Environmetrics, 5, 111–126, doi:10.1002/env.3170050203, 1994. Paatero, P. and Tapper, U.: Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values, Environmetrics, 5, 111–126, doi:10.1002/env.3170050203, 1994. Seidel, D. J., and Birnbaum, A. N.: Effects of Independence Day fireworks on atmospheric concentrations of fine particulate 15 matter in the United States. Atmospheric Environment, 115, 192-198, doi:10.1016/j.atmosenv.2015.05.065, 2015. Seidel, D. J., and Birnbaum, A. N.: Effects of Independence Day fireworks on atmospheric concentrations of fine particulate 15 matter in the United States. Atmospheric Environment, 115, 192-198, doi:10.1016/j.atmosenv.2015.05.065, 2015. Sun, J., Zhang, Q., Canagaratna, M. R., Zhang, Y., Ng, N. L., Sun Y. L. Jayne, J. T., Zhang, X., Zhang, X., Worsnop, D. R. . Highly time-and size-resolved characterization of submicron aerosol particles in Beijing using an Aerodyne Aerosol Mass Spectrometer, Atmospheric Environment, 44, 131-140, doi:10.1016/j.atmosenv.2009.03.020, 2010. Sun, Y. L., Zhang, Q., Schwab, J. J., Chen, W. N., Bae, M. S., Lin, Y. C., Hung, H. M., and Demerjian, K. L.: A case study of 20 aerosol processing and evolution in summer in New York City, Atmospheric Chemistry and Physics, 11, 12737-12750, doi:10.5194/acp-11-12737-2011, 2011. Sun, Y. L., Zhang, Q., Schwab, J. J., Chen, W. N., Bae, M. S., Hung, H. M., Lin, Y. C, Ng, N. L., Jayne, J., Massoli, P., Williams, L. R., and Demerjian, K. L.: Characterization of near-highway submicron aerosols in New York City with a Sun, Y. L., Zhang, Q., Schwab, J. J., Chen, W. N., Bae, M. S., Lin, Y. C., Hung, H. M., and Demerjian, K. L.: A case study of 20 aerosol processing and evolution in summer in New York City, Atmospheric Chemistry and Physics, 11, 12737-12750, doi:10.5194/acp-11-12737-2011, 2011. doi:10.5194/acp-11-12737-2011, 2011. Sun, Y. L., Zhang, Q., Schwab, J. J., Chen, W. N., Bae, M. S., Hung, H. M., Lin, Y. C, Ng, N. L., Jayne, J., Massoli, P., Williams, L. R., and Demerjian, K. L.: Characterization of near-highway submicron aerosols in New York City with a high-resolution aerosol mass spectrometer, Atmospheric Chemistry and Physics, 12, 2215-2227, doi:10.5194/acp-12-2215-2012, 25 2012. Slowik, J. G., Stroud, C., Bottenheim, J. W., Brickell, P. C., Chang, R. Y.-W., Liggio, J., Makar, P. A., Martin, R. V., Moran, M. D., Shantz, N. C., Sjostedt, S. J., Donkelaar A. van, Vlasenko A., Wiebe, H. A., Xia, A. References: G., Zhang, J., Leaitch, W. R., and 15 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Abbatt. J. P. D.: Characterization of a large biogenic secondary organic aerosol event from eastern Canadian forests, Atmospheric Chemistry and Physics, 10, 2825-2845, doi:10.5194/acp-10-2825-2010, 2010. Tian, Y. Z., Wang, J., Peng, X., Shi, G. L., and Feng, Y. C.: Estimation of the direct and indirect impacts of fireworks on the physicochemical characteristics of atmospheric PM 10 and PM 2.5, Atmospheric Chemistry and Physics, 14, 9469-9479, doi:10.5194/acp-14-9469-2014, 2014. 5 Ulbrich, I. M., Canagaratna, M. R., Zhang, Q.,Worsnop, D. R., and Jimenez, J. L.: Interpretation of organic components from Positive Matrix Factorization of aerosol mass spectrometric data, Atmos. Chem. Phys., 9, 2891-2918, doi:10.5194/acp-9-2891-2009, 2009. Vecchi, R., Bernardoni, V., Cricchio, D., D’Alessandro, A., Fermo, P., Lucarelli, F., Nava, S., Piazzalunga, A., and Valli, G.: The impact of fireworks on airborne particles, Atmospheric Environment, 42, 1121-1132, doi:10.1016/j.atmosenv.2007.10.047, 10 2008. Wang, J. F., Ge X, Chen Y, Shen, Y. F., Zhang, Q., Sun, Y. L., Xu, J. Z., Ge, S., Yu, H, and Chen, M. D.: Highly time-resolved urban aerosol characteristics during springtime in Yangtze River Delta, China: insights from soot particle aerosol mass spectrometry, Atmospheric Chemistry and Physics, 16, 9109-9127, doi: 10.5194/acp-16-9109-2016, 2016. Wang, J. F., Ge X, Chen Y, Shen, Y. F., Zhang, Q., Sun, Y. L., Xu, J. Z., Ge, S., Yu, H, and Chen, M. D.: Highly time-resolved urban aerosol characteristics during springtime in Yangtze River Delta, China: insights from soot particle aerosol mass spectrometry, Atmospheric Chemistry and Physics, 16, 9109-9127, doi: 10.5194/acp-16-9109-2016, 2016. Wang, Y., Zhuang, G., Xu, C., and An, Z. S.: The air pollution caused by the burning of fireworks during the lantern festival in 15 Beijing, Atmospheric Environment, 41, 417-431, doi:10.1016/j.atmosenv.2006.07.043, 2007. Wehner, B., Wiedensohler, A., and Heintzenberg, J.: Submicrometer aerosol size distributions and mass concentration of the millennium fireworks 2000 in Leipzig, Germany, Journal of Aerosol Science, 31, 1489-1493, doi:10.1016/S0021-8502(00)00039-2, 2000. WHO: Air Quality Guidelines: Global Update 2005. Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide, World 20 Health Organization, 2006. Xu, L., Suresh, S., Guo, H., Weber, R. J., and Ng, N. References: L.: Aerosol characterization over the southeastern United States using high-resolution aerosol mass spectrometry: spatial and seasonal variation of aerosol composition and sources with a focus on organic nitrates, Atmospheric Chemistry and Physics, 15, 7307-7336, doi:10.5194/acp-15-7307-2015, 2015. WHO: Air Quality Guidelines: Global Update 2005. Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide, World 20 Health Organization, 2006. Yang, L., Gao, X., Wang, X., Nie, W., Wang, J., Gao, R., Xu, P., Shou, Y., Zhang, Q., and Wang, W.: Impacts of firecracker 25 burning on aerosol chemical characteristics and human health risk levels during the Chinese New Year Celebration in Jinan, China. Science of the Total Environment, 476, 57-64, doi: 10.1016/j.scitotenv.2013.12.110, 2014. Zhang, J., Marto, P. J., and Schwab, J. J.: Exploring the applicability and limitations of selected optical scattering instruments for PM mass measurement. Atmos. Meas. Tech., 11, 2995-3005, doi: 10.5194/amt-11-2995-2018, 2018. Zhang, J., Marto, P. J., and Schwab, J. J.: Exploring the applicability and limitations of selected optical scattering instruments for PM mass measurement. Atmos. Meas. Tech., 11, 2995-3005, doi: 10.5194/amt-11-2995-2018, 2018. Zhang, J., Marto, P. J., and Schwab, J. J.: Exploring the applicability and limitations of selected optical scatt PM mass measurement. Atmos. Meas. Tech., 11, 2995-3005, doi: 10.5194/amt-11-2995-2018, 2018. 16 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Zhang, Q., Canagaratna, M. R., Jayne, J. T., Worsnop, D. R., and Jimenez, J. L.: Time‐and size‐resolved chemical composition of submicron particles in Pittsburgh: Implications for aerosol sources and processes. Journal of Geophysical Research: Atmospheres, 110, D07S09, 1-19, doi:10.1029/2004JD004649, 2005. Zhang, Q., Canagaratna, M. R., Jayne, J. T., Worsnop, D. R., and Jimenez, J. L.: Time‐and size‐resolved chemical composition of submicron particles in Pittsburgh: Implications for aerosol sources and processes. Journal of Geophysical Research: Atmospheres, 110, D07S09, 1-19, doi:10.1029/2004JD004649, 2005. Zhang, Q., Jimenez, J., Canagaratna, M., Ulbrich, I., Ng, N., Worsnop, D., and Sun, Y.: Understanding atmospheric organic aerosols via factor analysis of aerosol mass spectrometry: a review, Anal. Bioanal. Chem., 401, 3045-3067, 5 doi:10.1007/s00216-011-5355-y, 2011. Zhou, S., Collier, S., Xu, J., Mei, F., Wang, J., Lee, Y. N., Sedlacek, A. J., Springston, S. R., Sun, Y. References: L., and Zhang, Q.: Influences of upwind emission sources and atmospheric processing on aerosol chemistry and properties at a rural location in the Northeastern US, Journal of Geophysical Research: Atmospheres, 121, 6049-6065, doi:10.1002/2015JD024568, 2016. Zhang, Q., Jimenez, J., Canagaratna, M., Ulbrich, I., Ng, N., Worsnop, D., and Sun, Y.: Understanding atmospheric organic aerosols via factor analysis of aerosol mass spectrometry: a review, Anal. Bioanal. Chem., 401, 3045-3067, 5 doi:10.1007/s00216-011-5355-y, 2011. Zhou, S., Collier, S., Xu, J., Mei, F., Wang, J., Lee, Y. N., Sedlacek, A. J., Springston, S. R., Sun, Y. L., and Zhang, Q.: Influences of upwind emission sources and atmospheric processing on aerosol chemistry and properties at a rural location in the Northeastern US, Journal of Geophysical Research: Atmospheres, 121, 6049-6065, doi:10.1002/2015JD024568, 2016. 17 17 17 Figure 1. Time series of (a) aerosol hourly-averaged mass concentration measured by ASRC AMS (PM1), DRX (PM2.5), PDR (PM2.5), and ACHD T640 (PM2.5) (μg m–3); (b) NYSM meteorological parameters, with relative humidity (%), wind direction, wind speed (m/s), and precipitation (mm) os. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 uscript under review for journal Atmos. Chem. Phys. ussion started: 6 June 2018 uthor(s) 2018. CC BY 4.0 License. Figure 1. Time series of (a) aerosol hourly-averaged mass concentration measured by ASRC AMS (PM1), DRX (PM2.5), PDR (PM2.5), and ACHD T640 (PM2.5) (μg m–3); (b) NYSM meteorological parameters, with relative humidity (%), wind direction, wind speed (m/s), and precipitation (mm) Figure 2. The time series of (a) the mass concentrations of aerosol component (Organic, SO4, NO3, NH4, Chl, and K) from AMS; and (b) the mass fraction of each component. Figure 2. The time series of (a) the mass concentrations of aerosol component (Organic, SO4, NO3, NH4, Chl, and K) from AMS; and (b) the mass fraction of each component. 18 Figure 3. (a): Mass size distribution (mobility diameter, density is set to be 1 g cm-3 here) measured by SMPS from July 4 12:00 to July 6 12:00; (b) AMS measured size distribution (vacuum aerodynamic diameter) for SO4, organics and K compounds from July 4 12:00 to July 6 12:00 (a) s. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 script under review for journal Atmos. Chem. Phys. ssion started: 6 June 2018 uthor(s) 2018. CC BY 4.0 License. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. References: ): Mass size distribution (mobility diameter, density is set to be 1 g cm-3 here) measured by SMPS from Jul Figure 3. (a): Mass size distribution (mobility diameter, density is set to be 1 g cm-3 here) measured by SMPS from July 4 12:00 to July 6 12:00; (b) AMS measured size distribution (vacuum aerodynamic diameter) for SO4, organics and K Figure 3. (a): Mass size distribution (mobility diameter, density is set to be 1 g cm-3 here) measured by SMPS from July 4 12:00 to July 6 12:00; (b) AMS measured size distribution (vacuum aerodynamic diameter) for SO4, organics and K compounds from July 4 12:00 to July 6 12:00 compounds from July 4 12:00 to July 6 12:00 compounds from July 4 12:00 to July 6 12:00 Figure 4. Time series of the five organic aerosol factors determined from PMF analysis. Figure 4. Time series of the five organic aerosol factors determined from PMF analysis. 19 Figure 5. The difference of the mass spectra of FW-OOA and LV-OOA. Here, both FW-OOA and LV-OOA mass spectra signals were normalized to be 1. Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Figure 5. The difference of the mass spectra of FW-OOA and LV-OOA. Here, both FW-OOA and LV-OOA mass spectra signals were normalized to be 1. Figure 5. The difference of the mass spectra of FW-OOA and LV-OOA. Here, both FW-OOA and LV-OOA mass spectra signals were normalized to be 1. Figure 6. Time series of different kinds of inorganic salt and organic compounds estimated based on the ion-balance calculation. Figure 6. Time series of different kinds of inorganic salt and organic compounds estimated based on the ion-balance calculation. 20 Figure 7. Time series of LiDAR data on Independence Day (to match the wind back trajectory time in Fig.8, UTC time was used here) Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-529 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 6 June 2018 c⃝Author(s) 2018. CC BY 4.0 License. Figure 7. Time series of LiDAR data on Independence Day (to match the wind back trajectory time in Fig.8, UTC time was used here) 21 21

https://openalex.org/W3120414804

https://jhoonline.biomedcentral.com/track/pdf/10.1186/s13045-020-01014-w

English

NK cell-based cancer immunotherapy: from basic biology to clinical development

Journal of hematology & oncology

cc-by

© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat​iveco​ mmons​.org/publi​cdoma​in/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Abstract Natural killer (NK) cell is a specialized immune effector cell type that plays a critical role in immune activation against abnormal cells. Different from events required for T cell activation, NK cell activation is governed by the interaction of NK receptors with target cells, independent of antigen processing and presentation. Due to relatively unsophisticated cues for activation, NK cell has gained significant attention in the field of cancer immunotherapy. Many efforts are emerging for developing and engineering NK cell-based cancer immunotherapy. In this review, we provide our cur- rent understandings of NK cell biology, ongoing pre-clinical and clinical development of NK cell-based therapies and discuss the progress, challenges, and future perspectives. Keywords:  NK cell, Cancer immunotherapy, Clinical trials, iPSC of NK cell infusion, even in the allogeneic setting [4–7]. The feasibility of utilizing allogeneic NK cells, the estab- lished safety profiles, and the fast-acting nature of NK cells largely have led to the emerging effort to develop “off-the-shelf” NK cell-based cancer immunother- apy. However, there are many challenges to overcome, such as difficulty to meet clinical-grade ex  vivo expan- sion, limited in  vivo persistence, limited infiltration to solid tumors, and tumor editing to evade NK cell activ- ity. Various strategies are being employed to overcome these challenges to improve the efficacy of NK cell-based therapy, such as ex vivo pre-conditioning with cytokines and/or small molecular drugs, engineering an “off-the- shelf” or iPSC-differentiated chimeric antigen receptor (CAR)-NK. There has been an explosion of NK-based immunotherapies in pre-clinical development and clini- cal development. Herein, we will provide an updated overview of the emerging endeavors for developing NK cell-based cancer immunotherapy from pre-clinical *Correspondence: [email protected] †Sizhe Liu, Vasiliy Galat, Yekaterina Galat, and Yoo Kyung Annie Lee have contributed equally 1 Department of Urology, Feinberg School of Medicine, Northwestern University, 303 E. Superior St., Lurie Research Building 6‑117, Chicago, IL 60611, USA Full list of author information is available at the end of the article Liu et al. J Hematol Oncol (2021) 14:7 https://doi.org/10.1186/s13045-020-01014-w Liu et al. J Hematol Oncol (2021) 14:7 https://doi.org/10.1186/s13045-020-01014-w Main text NK cell biology NK cells were identified over four decades ago as lym- phocytes with innate ability to lyse tumor cells without the need for prior sensitization [8–10]. NK cells can trigger target cell death by releasing cytotoxic granules containing granzymes and perforin and through death receptor-mediated pathways (e.g., FasL/Fas) [11]. NK cells also play immunomodulatory functions by secreting chemokines and cytokines, such as RANTES and IFN-γ [12, 13].i p [ ] NK cells originate from ­CD34+ hematopoietic stem cells. Bone marrow is considered to be the primary site of NK development. More recent evidence had indicated that NK cells can also develop and mature at secondary lymphoid organs, including tonsils, spleen, and lymph nodes [19]. NK cell progenitors progress through distinct developmental stages and gradually acquire the expres- sion of surface receptors that define NK cell identity such as NK1.1 and CD56 and/or regulate their effector func- tions such as CD16 and NKp46 [19]. Distinct from T cells, there is no master transcriptional factor that controls NK cell development. Instead, a combination of transcription factors, including T-bet, Eomes, E4BP4, Id2, and BLIMP, was identified to instruct NK development and matura- tion [23]. Common gamma chain cytokines such as IL2, IL-7, and IL15 and their receptor components including CD122 and CD127 have been shown to play essential roles in NK development and homeostasis [19]. Despite decades of work, the ontogeny of NK cells in humans is still not fully elucidated. The conventional linear model suggests that mature NK cells arise from common lym- phoid progenitors (CLPs) by progressing through a linear continuum [24]. The linear model proposes that CD56 marks a transition from immature into a more mature status and that immature ­CD56bright NK cells further dif- ferentiate into mature ­CD56dim populations in human [19, 24]. Recent evidence has challenged this model and suggests a possibility of more branched development in the form of both CLPs and common myeloid progenitors (CMPs) giving rise to NK cell progenitors. The branched model also proposes that distinct precursor populations independently develop into different mature NK subsets [24]. In humans, NK cells are traditionally identified by the absence of CD3 and the presence of CD56 on their sur- face as characterized by flow cytometry. In mouse, the lack of CD3 and the presence of NK1.1 are canonical criterion for distinguishing NK cells. In mouse strains lacking NK1.1 expression (e.g., BALB/c), CD49b is used for NK cell identification. Main text NK cell biology The natural cytotoxic receptor NKp46 is also often used to identify mouse and human NK cells in combination with the absence of CD3 expres- sion. Notably, certain tissues such as the mucosal barri- ers possess subsets of recently identified innate lymphoid cells (ILCs) that also share canonical markers of NK cells. For example, a subset of IL-22 secreting human ILC3s is ­CD56+NKp46+CD3− [14]. Additional markers such as lack of c-kit can be used to distinguish human NK (c-kit-) from ILC3s (c-kit +) [14]. In mouse, NK and ILC1 are NK1.1+CD3− but can be further characterized by CD49a and Eomes expression. NK cells are ­CD49a−Eomes+, whereas ILC1s are ­CD49a+Eomes− [15, 16]. NK cells are found both in blood at levels of 5%-15% of circulating lymphocytes and in various lymphoid and non-lymphoid organs such as the spleen, lung, and liver [17, 18]. Based on characterization of NK cells in the peripheral blood, human NK cells are conventionally sub-divided into two major subsets: ­CD56brightCD16dim/− and ­CD56dimCD16+, with the former classically believed to be less mature and a potent cytokine producer and the latter more mature and the most cytotoxic [17, 19]. Most NK cells in the blood are ­CD56dim, whereas the ­CD56bright subset only represents less than 15% of total circulating NK cells [17]. The relative proportion of ­CD56bright and ­CD56dim NK cells in tissues can be very different from that observed in the peripheral blood [17]. Notably, many tissue-resident NK subsets are now shown to be phe- notypically and functionally distinct from conventional peripheral blood NK cells [20]. For example, uterine NK cells, which constitute the majority of lymphocyte in the uterus during the first trimester, are ­CD56super bright and Background Natural killer (NK) cells are an essential part of tumor immunosurveillance, evidenced by higher cancer sus- ceptibility and metastasis in association with diminished NK activity in mouse models and clinical studies [1–3]. Using an array of germline-encoded surface receptors, NK cells are able to recognize and rapidly act against malignant cells without prior sensitization. Upon acti- vation, NK cells release cytotoxic granules containing perforin and granzymes to directly lyse tumor cells, in a similar fashion to activated cytotoxic T cells. NK cells are also potent producers of chemokines and cytokines such as interferon gamma (IFN-γ) and tumor necrosis factor alpha (TNF-α) and thereby are essential in modulating adaptive immune responses. Due to their innate ability to eliminate tumor cells, NK cell-based immunothera- pies against cancer have been investigated for decades. Early clinical trials have demonstrated the overall safety Liu et al. J Hematol Oncol (2021) 14:7 Page 2 of 17 Liu et al. J Hematol Oncol (2021) 14:7 Page 2 of 17 conceptual development, clinical grade expansion, and ongoing clinical development. play important roles in pregnancies by regulating placen- tal vascular remodeling [17, 20]. Conventional NK cells are short-lived innate lympho- cytes that lack antigen specificity. Recent studies revealed that subset of mature NK cells can elicit long-lived “adaptive”-like nature in the specific context of CMV infection [21]. It is now understood that the “adaptive”- like NK cells express the activating receptor NKG2C and that the “adaptive” nature of ­NKG2C+ NK cells attributes to the non-classical MHC I molecule HLA-E presenting CMV-specific viral peptide to NKG2C [22]. The ­NKG2C+NK cells can be found in the circulation of HCMV-seropositive individuals [22]. NK cell receptors and NK activation NK cells are mounted with a repertoire of inhibitory and activating surface receptors (Table 1) [25–29]. Distinctly different from T cell receptors, NK cell receptors are germline-encoded and “hard-wired” receptors without a requirement for “V(D)J” recombination. Upon ligation, Liu et al. J Hematol Oncol (2021) 14:7 Page 3 of 17 Table 1  NK cell receptors and their ligands in human Ligands NK activating receptor NKG2D MHC class I chain-related protein A (MICA) and B (MICB), UL16-binding proteins (ULBP1-6) DNAM1 PVR(CD155), nectin-2 (CD112) NKp30 (NCR3) pp65, B7-H6, galectin-3, BAG6, viral hemagglutinin (HA), heparan sulfate (HS) glycosaminoglycans (GAGs), (DBL)-1a domain of Plasmodium falciparum erythrocyte membrane protein-1 NKp44 (NCR2) PDGF-DD, 21spe-MLL5, PCNA, Syndecan-4, Nidogen-1, viral HA, HS GAGs NKp46 (NCR1) Complement factor P, viral HA, HS GAGs, (DBL)-1a domain of Plasmodium falciparum erythrocyte membrane protein-1, vimentin CD16 (FcγRIII) Fc portion of IgG antibodies NK inhibitory receptors CD94/NKG2A HLA-E KIR2DL1 HLA-C, group 2 KIR2DL2/3 HLA-C, group 1 KIR3DL1 HLA-Bw4 KIR3DL2 HLA-A3, A11 Table 1  NK cell receptors and their ligands in human Ligands Table 1  NK cell receptors and their ligands in human these receptors transmit either inhibitory or activating signals to control NK activation. The integration and balance of the activating and inhibitory signals from the ligand/receptor interactions dictates the status of NK cell activation. For instance, healthy cells express no or mini- mal level of ligands for NK cell activating receptors, but express high levels of the major histocompatibility com- plex class I molecules (MHC I), also known as human leukocyte antigen (HLA), that ligates to the killer immu- noglobulin-like (KIR) family inhibitory receptors on NK cells to protect them from NK attack [29]. Conversely, tumorigenic cells or virally infected cells have downregu- lated MHC I expression but upregulated levels of ligands for NK cell activating receptors and thus trigger NK cell activation due to the lack of inhibitory signals and/or the presence of activating signals [11]. In allogeneic trans- fer settings, the concept of KIR and HLA mismatches between donor and recipient was the original strategy aimed at enhancing the activation of donor NK cells to eradicate patient’s tumor cells. NK cells also play a signif- icant role in antibody-mediated cancer therapies by uti- lizing the Fcγ receptor to carry out antibody-dependent cellular cytotoxicity (ADCC) [30]. protected bone marrow-transplanted AML patients from AML relapse while sparing graft versus host diseases (GVHD) [31, 32]. Miller et al. NK cell receptors and NK activation further pioneered the use of NK cells in non-transplant settings. They showed that infusion of NK cells from HLA-haploidentical donors combined with subcutaneous IL-2 administration after a pre-conditioning regimen of high-dose cyclophos- phamide and fludarabine resulted in successful in  vivo expansion of donor NK cells and the induction of com- plete remission in 5 out of 19 patients with poor-prog- nosis acute myeloid leukemia (AML) [5]. Miller et  al. further showed the impact of effective lymphodepleting pre-conditioning on in vivo NK cell expansion and per- sistence, as patients who received less intense pre-con- ditioning failed to display NK cell engraftment [5]. It is currently believed that the success of adoptive transfer requires the creation of a lymphopenic environment to provide a niche for donor cells to survive and proliferate. Initial successes of adoptive NK cell transfer in treat- ing hematological cancers prompted clinical endeavors in using the strategy against solid cancers. NK cells are cyto- toxic against a wide range of tumor cells of solid cancer types in vitro. Anti-tumor activities of adoptively trans- ferred NK cells in vivo have been demonstrated as well in pre-clinical xenograft mouse models of ovarian cancer, glioblastoma, and metastatic colorectal cancer [33–36]. The safety of NK cell-based therapy has been demon- strated in both autologous and allogeneic haploidentical settings [4–7]. Clinical efficacy of this strategy has proven to be thus far be limited. The following section summa- rizes the current pre-clinical efforts to enhance the effi- cacy of NK cell-based therapy. CAR‑NK cell as an alternative to CAR‑T therapy “native” NK cell signaling molecule, such as NKG2D-2B4, exhibited superior in vitro and in vivo anti-tumor activi- ties in comparison with which contains CD28-CD137 [42]. The NKG2D-2B4 containing CAR-NK had elevated Syk and Erk1/2 phosphorylation [42]. y T cells equipped with CARs have been shown to pro- vide clinical benefit for patients with select liquid cancers. Two CD19-targeting CAR-T products were approved by the FDA for treatment of B cell lympho- mas and acute lymphoblastic leukemia (ALL). However, CAR-T therapy has two major challenges: (i) require- ment of a substantiate length of time to generate a ther- apeutic dose of autologous CAR-T cells which limits its feasibility to treat patients with rapidly progressing dis- eases; (ii) difficulty to obtain sufficient number of autol- ogous T cells for CAR-T cell generation from heavily pre-treated and often lymphopenic cancer patients [37]. As an alternative to CAR-T cell therapy, CAR-NK cell therapy not only circumvents these challenges but also presents additional major advantages: (i) the ability to use unlimited allogeneic NK source without concern of GVHD [6, 31, 32]; (ii) the potential to generate “off- shelf” product with NK cell line or iPSC-NK [37–41]; (iii) relatively shortened production time; iv) recog- nition and killing tumor cells through NK cell native receptors independent of the CAR engineering [42], less likely allowing disease escape through downregu- lation of the CAR antigens as shown with CAR-T cell therapy [37]. Beyond directing CAR-NK cytotoxicity against tumor cells by targeting tumor antigens, it has been proposed that CAR-NK cells can be used to eliminate immuno- suppressive immune cells in the tumor microenviron- ment that include myeloid-derived suppressor cells (MDSCs) and M2 tumor-associated macrophages (TAMs) [77, 78]. Parihar and colleagues designed CAR- NK cells expressing the activating receptor NKG2D as the antigen recognition to target human MDSCs. They demonstrated in vitro and in a mouse xenograft model grafted with human neuroblastoma and MDSCs that NKG2D CAR-NK cells can reduce human MDSCs effi- ciently [77]. Moreover, the CAR-NK cells secrete pro- inflammatory cytokines and chemokines which may improve the infiltration and functions of subsequently infused CAR-T cells in the mouse model [77]. It has been proposed that CAR-NK cells may be able to be combined with T cell-based therapies for solid tumors [77]. Similar to CAR-T cells, CAR-NK cells are genetically modified to express CARs that recognize a specific anti- gen uniquely expressed or overexpressed by target cells. CAR‑NK cell as an alternative to CAR‑T therapy In most pre-clinical studies, lentiviral or retroviral-based transduction was used to achieve stable and sustained CAR expression in NK cells. Non-viral vector-based delivery methods such as transposon systems and elec- troporation of mRNA have been used as well [41, 43–45]. A wide range of tumor antigens have been targeted by CAR-NK cells in pre-clinical studies for hematologi- cal malignancies and solid tumors [38, 39, 41–76]. They are also summarized in Table 2. The antigen recognition domain usually consists of a single-chain fragment (scFV) molecule derived from a monoclonal antibody, and nano- body-based constructs have been used in limited studies to date [72]. For hematological cancers, CD19 remains a major target. Antigens such EGFRvIII, mesothelin, and Her2 have been targeted by CAR-NK cells for the treat- ment of solid cancers including colorectal cancer, ovar- ian cancer, breast cancer, and glioblastoma [41, 48]. The signaling domains of CAR-NK cells are very similar to those in CAR-T cells, typically composed of the fusion of CD3ζ with one or two TCR co-stimulatory molecule (s), such as CD28, 4-1BB, 2B4, DNAM1, and NKG2D. Among these TCR cell co-stimulatory molecules, 4-1BB, DNAM1, 2B4, and NKG2D were also expressed by NK cells as native activating receptors. It was shown that the hMesothelin-CAR-NK cells containing the shared Preclinical development of NK cell‑based cancer immunotherapy Current preclinical development of NK cell-based ther- apy was largely inspired by early clinical studies. With the understanding of how NK cells are activated, the ini- tial NK cell-based therapy was pioneered in the clinical setting of hematopoietic stem cell transplants (HSCTs) whereby NK cells were shown to have the capacity to exert a graft versus leukemia effect. The Ruggeri group showed that KIR-mismatched alloreactive donor NK cells Liu et al. J Hematol Oncol (2021) 14:7 Page 4 of 17 Liu et al. J Hematol Oncol (2021) 14:7 Killer cell engagers unleash NK cytotoxicity against tumor cells Although CAR-NK cells hold great promise as a future “off-the-shelf” drug, there are NK cell-specific chal- lenges, such as potential loss of viability and/or activ- ity with freeze–thaw process and lack of in  vivo persistence. To circumvent these challenges, Bi- and tri-specific killer engagers, BiKEs and TriKEs, are in development as a complemental approach. The BiKEs or TrikEs are composed of two or three linked single- chain antibody variable fragments of different antigen specificities [79]. The current in-development BiKEs or TrikEs simultaneously engage with CD16 and tumor antigens, thereby inducing the formation of immune synapses and NK-mediated ADCC [79]. In pre-clinical studies, various BiKEs and TriKEs have been designed to target a number of tumor antigens: CD19, CD20, and CD33 for hematological cancers, HER2, EGFR, and EpCAM for solid tumors [79]. Cytokines, such as IL15, have been incorporated into killer engagers to further enhance NK cell functions [80]. Vallera et  al. reported that a CD16/IL-15/CD33 TriKE not only enhances NK-mediated immunity against ­CD33+ tar- gets, but also promotes the in vivo persistence, activa- tion, and survival of NK cells by delivering IL-15 [80]. This TriKE design (GTB-3550) is currently in clinical trial for treating high-risk hematological malignancies Liu et al. Killer cell engagers unleash NK cytotoxicity against tumor cells J Hematol Oncol (2021) 14:7 Page 5 of 17 Table 2  CAR-NK cells that have been evaluated preclinically Target CAR construct(s) Source of NK Method(s) Cancer type(s) References Hematological cancers CD19 CD19-scFv-CD3ζ, CD19-scFv-CD28-CD3ζ, CD19-scFv-41BB-CD3ζ, CD19-scFv-DAP10- CD3ζ, (iC9*).CD19-scFv.CD28- CD3ζ-(IL15*) NK-92, NKL, Cord Blood, peripheral blood Retrovirus, lentivirus B cell malignancies [38, 39, 49, 52, 63] FLT3 FLT3-scFV-CD28-CD3ζ NK-92 Lentivirus B cell acute lympho- blastic leukemia (B-ALL) [62] CS1 CS1-scFv-CD28-CD3ζ NK-92 Lentivirus Multiple Myeloma [53] CD38 Nb(CD38)a-CD28-41BB- CD3ζ NK-92 (with CD38 knocked out) Retrovirus Multiple Myeloma [72] CD4 CD4-scFv-CD28-41BB- CD3ζ NK-92 Lentivirus Peripheral T cell lym- phoma [73] CD5 CD5-scFv-2B4-CD3ζ NK-92 Lentivirus T cell malignancies [46] CD7 Nb(CD7)-CD28-41BB- CD3ζ NK-92MI PiggyBac Transposon System T cell acute lympho- blastic leukemia (T-ALL) [67] Solid cancers Wild Type EGFR and/or EGFRvIII EGFR-scFv-CD28-CD3ζ, EGFR-scFv-CD28-41BB- CD3ζ, EGFRvIII-scFv-CD28- CD3ζ, EGFRVIII-scFv-DAP12 NK-92 NKL, YTS Lentivirus Glioblastoma, renal cell carcinoma, breast cancer [47, 51, 58, 59, 71, 76] ErbB2/HER2 ErbB2-scFv-CD3ζ, ErbB2-scFv-CD28-CD3ζ, ErbB2-scFv-41BB- CD3ζ NK-92 Lentivirus, Retrovirus Breast carcinoma, glioblastoma [48, 60, 70] GD2 GD2-scFv-CD3ζ NK-92 Retrovirus Neuroblastoma, melanoma, breast carcinoma [65] Glypican-3 (GPC3) GPC3-scFv-CD28-41BB- CD3ζ, GPC3-scFv-CD3ζ, GPC3-scFv-CD28-CD3ζ, GPC3-scFv-DNAM1- CD3ζ, GPC3-scFv-DNAM1- 2B4-CD3ζ iPSC, NK-92 Lentivirus Ovarian cancer, hepato- cellular cancer [57, 66] EpCAM EpCAM-scFV-41BB- CD3ζ, EpCAM-scFv-CD28- CD3ζ NK-92 Lentivirus Colorectal Cancer [47, 50, 61] Mesothelin Mesothelin-scFv- (NKG2D*)b-2B4-CD3z, Mesothelin-scFv-CD28- 41BB-CD3ζ NK-92, iPSC PiggyBac Transposon System; Lentivirus Ovarian cancer [42, 75] Prostate Stem Cell Ag (PSCA) PSCA-scFv-DAP12 YST cell line, primary NK Lentivirus Prostate Cancer [55] Carcinoembryonic antigen (CEA) CEA-scFv-CD3ζ NK-92MI Retrovirus Colorectal Cancer [69] CD133 CD133-CD28-41BB- CD3ζ NK-92 Lentivirus CD133 + cancer stem cells, ovarian cancer [64] c-MET c-MeT-scFv-41BB– DAP12 Peripheral blood Lentivirus Liver cancer [56] Table 2  CAR-NK cells that have been evaluated preclinically Liu et al. J Hematol Oncol (2021) 14:7 Page 6 of 17 Table 2  (continued) Target CAR construct(s) Source of NK Method(s) Cancer type(s) References NKG2D ligands NKG2D-CD3ζ, NKG2D-CD28- CD3ζ, NKG2D-CD28-41BB- CD3ζ, NKG2D-DAP10-CD3ζ Peripheral blood, NK-92 RNA electroporation, PiggyBac Transposon System NKG2DL + cancer cells, Ovarian cancer [43, 44, 54] PD-L1 PD1-(NKG2D*)c-41BB NK-92 Lentivirus PD-L1 + tumor cells [68] Tissue Factor (TF) Factor VII light chain (FvIIL)-CD28-41BB- CD3ζ NK-92MI (transduced to express CD16) Lentivirus Triple-negative breast cancer [74] For CAR constructs, only the antigen recognition domain and signaling domain(s) are listed unless otherwise specified. Other functional domains will be in parentheses and marked by asterisks (*). Killer cell engagers unleash NK cytotoxicity against tumor cells Abbreviations: scFv = single-chain variable fragment; FLT3 = FMS-like tyrosine kinase 3; a  Nanobody for CD38 b  transmembrane domain of NKG2D c  hinge region and transmembrane domain of NKG2D For CAR constructs, only the antigen recognition domain and signaling domain(s) are listed unless otherwise specified. Other functional domains will be in parentheses and marked by asterisks (*). Abbreviations: scFv = single-chain variable fragment; FLT3 = FMS-like tyrosine kinase 3; a  Nanobody for CD38 b  transmembrane domain of NKG2D c  hinge region and transmembrane domain of NKG2D In addition to cytokine-only strategies, Cichocki and colleagues demonstrated that pharmacologic inhibition of GSK3 kinase with CHIR99021 drives late-stage mat- uration of ex  vivo-expanded human peripheral blood NK cells and enhanced their anti-tumor effector func- tions both in  vitro and in  vivo [87]. Human NK cells expanded with IL-15 in the presence GSK3 inhibition had increased expression of the NK cell maturation marker CD57 and transcription factors associated with late-stage NK cell maturation including T-bet, ZEB2, and BLIMP-1 as compared to those expanded with IL-15 alone [87]. The expanded human NK cells showed potent ADCC activities in  vitro and superior tumor control in a mouse xenograft model of ovarian cancer when combined with Herceptin, an anti-HER2 anti- body [87]. This strategy of GSK3 inhibition was used to generate FATE-NK100, which is being evaluated in an ongoing phase 1 clinical trial either as a monother- apy or in combination with monoclonal antibodies in patients with advanced solid tumors (NCT03319459). (NCT03214666). Tri-functional NK cell engagers (NKCEs), which crosslink both NKp46 and CD16, have demonstrated superior in vitro and in vivo anti-tumor activities compared to conventional monoclonal anti- bodies targeting the same antigens [81]. (NCT03214666). Tri-functional NK cell engagers (NKCEs), which crosslink both NKp46 and CD16, have demonstrated superior in vitro and in vivo anti-tumor activities compared to conventional monoclonal anti- bodies targeting the same antigens [81]. Methods to enhance the infiltration and functions of infused NK cells Source and clinical-scale expansion of NK cells with preserved cytotoxic activity are the major challenges for developing clinical-scale NK cell-based therapy. Cur- rently, strategies vary depending on the clinical setting and source of cells. Freshly isolated, activated, or in vitro expanded NK cell populations display phenotypic and functional differences. The differences also arise from activation approaches, such as the choice of interleukins, their combinations, type of feeder cells, and some other factors [95–97]. The current source and characteristics of NK cells for immunotherapies, expansion, and activation approaches are given in Tables 3 and 4. Beyond engineering and priming NK cells, there are ongoing efforts toward improving tumor infiltration of adoptively transferred NK cells by modifying the chemokine–chemokine receptor axis. Lee et al. devel- oped an NK cell recruiting protein-conjugated antibody (NRPbody) containing a mesothelin-specific tumor tar- geting domain Meso-scFv and the chemokine CXCL16 linked by a furin cleavage sites [90]. Mesothelin is a tumor differentiation antigen that is highly overex- pressed in several human cancers including malignant mesothelioma, pancreatic, ovarian, and lung adenocar- cinoma [91]. It was hypothesized that once bound to mesothelin-overexpressing tumor cells, furin-mediated cleavage would release CXCL16 from the NRPbody and thereby recruit NK cells to the tumor sites [90]. The cleavable CXCL16 containing NRPbody was shown to promote NK cell migration in vitro and infiltration into the tumor sites in  vivo in xenografted mouse models of orthotopic and metastatic pancreatic cancer [90]. In the xenograft models, NK cell infusion combined with intraperitoneal injection of the NRPbody signifi- cantly reduced tumor burden as compared to NK infu- sion combined with the non-cleavable control [90]. It remains to be determined how NRPbody will interact with a more complex and realistic immune contexture in the TME which consists of more than NK cells, as CXCL16 was shown in other tumor models to correlate with the infiltration of monocytes and M2-macrophages as well [92, 93]. Some investigations have sought to enhance NK migration toward the tumor by genetically modifying NK cells to overexpress chemokine receptors [54, 59, 94]. In one study, Ng et al. showed that CAR- NK cells genetically modified to express the chemokine receptor CXCR1 had enhanced migration in vitro and in vivo as compared to control CAR-NK cells [54]. NK cells “priming” strategies While the CAR technology seeks to enhance NK cell functions by genetically directing their target specificity, there are active investigations on other strategies to effec- tively “prime” NK cells ex vivo or in vivo for optimal anti- tumor functions after their infusion. It has been shown that freshly isolated, resting NK cells are generally less lytic as compared to NK cells primed via various strate- gies [82]. g Cytokine-mediated activation is frequently employed and currently under extensive investigation. In many studies, IL-2 or IL-15 is supplemented during ex  vivo expansion of NK cells and sometimes after NK cell infu- sion as well [5]. ALT-803, an IL-15/IL-15Ra fusion com- plex, was shown to enhance NK cell functions both in vitro and in vivo [83]. Short-term (18–20 h) pre-acti- vation with ALT-803 augmented cytotoxicity and ADCC of NK cells in vitro. In a xenograft mouse model of lym- phoma, co-administration of ALT-803 with NK cells significantly enhanced anti-CD20-triggered, NK cell- mediated ADCC effects [83]. Recently, several groups demonstrated that mouse and human NK cells pre- activated with a cocktail of IL-12/15/18 had enhanced and sustained anti-tumor effector functions in vitro and in vivo after infusion [84–86]. It was proposed that the cytokine pre-activated NK cells were “memory-like” with an enhanced response to cytokine or activating receptor re-stimulation weeks or months after the initial pre-acti- vation [84] While the majority of investigation of ex  vivo NK priming strategies rely on one or more cytokines, some studies indicate that cytokine supplement is not as crit- ical for NK cell priming. One study showed that over- night co-incubation with a leukemia cell line CTV-1 or its lysate without cytokines was sufficient to prime potent NK cells in vitro cytotoxicity against otherwise NK-insensitive tumor cells but not normal hemopoi- etic cells [88]. The mechanism of CTV-1-mediated NK priming is unclear. It was suggested that induced expression of CD69 on NK cells is important for the cytotoxicity of the CTV-1 tumor cell-activated NK [88]. Based on the pre-clinical findings, CTV-1 lysate- primed human NK cells (CNDO-109-NK cells) derived from HLA-haploidentical donors were evaluated for safety in a phase I clinical trial (NCT01520558) in Liu et al. J Hematol Oncol (2021) 14:7 Page 7 of 17 CD73 [43]. Adenosine impairs the anti-tumor functions of both T and NK cells [43]. Wang et al. NK cells “priming” strategies demonstrated that antibody-mediated blockade of CD73 significantly enhanced the anti-tumor activities of NKG2D-engin- neered CAR-NK92 in vitro and in vivo, with improved tumor infiltration by CAR-NK cells in vivo [43]. high-risk AML patients with AML after first complete remission [89]. CNDO-109-NK cells were found well tolerated without occurrence of dose-limiting toxici- ties [89]. Three out of 12 patients had durable complete remissions [89], providing evidence for further clinical evaluation of this approach. Methods to enhance the infiltration and functions of infused NK cells With the subcutaneous hypopharyngeal tumor xenograft model, it was shown that CXCR1-expressing CAR-NK cells had enhanced tumor infiltration and tumor con- trol as compared to control CAR-NK cells [54]. Tar- geting immunosuppressive components in the TME to re-invigorate NK functions is also under investigation. One major immunosuppressive factor in the TME is the metabolite adenosine, whose production is catalyzed in a sequential manner by the ectoenzymes CD39 and Donor‑derived NK sourcesh The main source of donor NK cells is peripheral blood collected by apheresis. Using allogeneic NKs for adop- tive transfer without expansion after a short overnight incubation with cytokines was shown to be sufficient to activate NK cells and enhance their cell cytotoxicity against tumor targets [86]. However, the dose of NK cells being used for therapy is limited. As the activation and expansion methods are improving, it is becoming pos- sible to prepare increasingly higher dosages of NK cells for adoptive transfer from a single-donor phlebotomy. In order to avoid critical side effects, such as GVHD caused by alloreactive T cells [98] or a passenger lymphocyte syndrome caused by donor-derived B cells [99], purifica- tion is recommended for allogeneic NK cells before the expansion to restrict contaminating total T cells to less than 1–5 × 105/Kg. The purification of NK cells is typi- cally achieved by magnetic depletion of CD3-expressing cells and subsequent enrichment for CD56-expressing cells [100] or by fluorescence-activated cell sorting [101]. The umbilical cord blood (UCB) [102–105] or pla- centa [106] represents other notable sources of NK cells that have been pursued for clinical applications. Both UCB and placenta-derived cells contain some propor- tion of NK cell progenitors that have the capacity to dif- ferentiate into NK cells during maturation and expansion stages. Typically, a dose of UCB or placenta donor can be expended to an amount sufficient for one adoptive transfer procedure. For instance, 21-day NK culture of placenta-isolated NKs yields an average of 1.2 × 109 NK cells with around 80% viability [106] and 1.59 × 1010 NK cells with an average purity of 92.37% from UCB [107]. Page 8 of 17 Liu et al. Limitations compared to other methods producing only up to 4% of ­CD34+. The emergence of this technology provided a completely new framework for clinical-scale NK cell production by allowing for the genetic modifications and unlimited expansion to be performed at the pluripotent cell state. CAR-engineered iPSC-NK cells already have demonstrated effectiveness in targeting human tumors in preclinical studies [41]. The other important feature of iPSC-NK cell technology is the ability to manipulate the differentiation strategy, thus shaping the phenotype and functionality of the resulting product. For instance, the enhancement of the Wnt signaling pathway with GSK3b inhibitor induces definitive hematopoiesis [112, 113] NK cells developed in such conditions had more pronounced inflammatory cytokine production phenotype, whereas Wnt independent NK subsets, similar to primary fetal NK cells, formed a bias for increased cytotoxicity [110]. Such ability to alter the course of differentiation opens a possibility to pursue resident or organ-specific pheno- types of NK cells.h iPSC‑NK cell sourceh The donor-derived NK cells have certain limitations mainly because of their variability in functional com- petence and expansion potential. In the clinical setting, each batch requires validation, which results in addi- tional lag time before a patient receives infusion. More- over, restricted expansion capacity poses a difficulty for improving the efficacy of NK cells through genetic engi- neering. CAR-NK is one of the approaches in develop- ment to overcome this limitation. The approach of using the modified NK cell line CAR-NK92, which can be easily expanded to large dose, has been in clinical trial. How- ever, NK92 is a transformed cell line that has limitations associated with its tumorigenic nature, and cytogenetic abnormalities, thus requiring irradiation for clinical use, which limits its life-time activity. A novel source of NK cells has emerged to circumvent many of the challenges associated with NK cell therapy. It is iPSC-differentiated NK cells (iPSC-NK). Conceptu- ally, iPSC-NK can provide a homogenously differentiated NK cell population that can be expanded to clinical scale as an “off-the-shelf” supply, overcoming the limitation of the NK-92 cell line. Several groups have demonstrated in  vitro derivation of functional NK cells from human embryonic stem cells (hESCs) and iPSCs [108–110]. Typically, the differentiation was induced in embry- oid bodies or by OP9 mouse stroma co-culture [111]. Recently, the differentiation method was also established in adherent monolayer cultures [108]. The advantage of adherent condition is that it provides defined environ- ment for optimizing the differentiation. This method has achieved over 15% of ­CD34+ hematopoietic progenitors The high proliferation capacity of pluripotent stem cells allows for the introduction of various genetic modifica- tions and for the development libraries of off-the-shelf haplotype-specific cells for treating a range of diseases. There are a number of ongoing clinical trials for cancer immunotherapy using engineered iPSC-NK cells, which are summarized in the following sections. Donor‑derived NK sourcesh J Hematol Oncol (2021) 14:7 Table 3  Comparison of commonly used allogeneic NK cell sources NK source Advantages Limitations PB-NK cells Relatively easy to collect Good in vivo expansion Good clinical track record Heterogeneous cell population Challenging to genetically modify Can only give one dose UB-NK cells NK Progenitors and CD34 + present Higher percentage of NK cells The ability to cryopreserve UCB Heterogeneous cell population NK92 cells Defined, homogeneous cell population Easy to expand Easy to genetically modify Can give multiple doses Tumor cells Irradiated Lack certain receptors, e.g., CD16 limited in vivo expansion iPSC-NK cells Defined, homogeneous cell population Circumvent issues with donor sourced cells (donor selection, contami- nating T, B) Potential for in depth preclinical testing Defined genetic makeup Easy to genetically modify at iPSC stage Can give multiple doses Can engineer multiple enhancements Don’t need to irradiate- good in vivo survival Suitable for “off-the-shelf” multicancer NK cell therapy More complicated to produce Table 3  Comparison of commonly used allogeneic NK cell sources Cytokine‑induced human NK expansion and activation Cytokines are the critical components of NK mainte- nance system and activation as discussed earlier. They induce short-term activation of NK cells but do not Page 9 of 17 Liu et al. J Hematol Oncol (2021) 14:7 Table 4  Summary of NK expansion and activation strategies support effective expansion without feeder cells. Inter- leukin (IL)-2 (IL-2) is one of the first and most impor- tant cytokines used for NK maintenance and is integral the treatment of several malignant diseases [114]. Cytokine‑induced human NK expansion and activation IL-2 was used to induce lymphokine-activated killer (LAK) cells a heterogeneous population of cells consisting pri Table 4  Summary of NK expansion and activation strategies Stimulation substance Expansion criteria Used clinically Reference Considerations Cytokines alone: applied separately or in combinations of two IL-2, IL-15, IL-2/IL-15, IL-2/IL-21  ~ 5 (two weeks) Yes [7, 100, 120] Generate highly activated NK cells Possibility of dependence on cytokine Expansion is facilitated in the presence of autologous PBMC IL-2/IL-15/IL-21  ~ 8 (two weeks) No [119, 121] Lower rate of NK cell expansion com- pared to feeder cell Only IL-2 cytokine is GMP-grade IL-15/IL-18/IL-27  ~ 17 (two weeks) No [122] IL-2, IL-18  ~ 500 (two weeks) No [123] Autologous feeder cells OK432, RN-T cells  ~ 600 (three weeks) Yes [4] RN-T cells were established by activation PBMC with OKT-3 and RetroNectin FN-CH296 Autologous feeder + Activating Abs Anti-CD335 (NKp46) and anti-CD2  ~ 3800 (three weeks) No Patent, 2013, EP2824112B [153] CD2 and CD335 coated nanomatrices with commercially available cell stimu- lation beads (Miltenyi Biotec Kit) OKT-3 (Anti CD3), anti-CD 52 1,537 (18 days) Yes [146] PBMCs are typically irradiated 25 Gy or more GMP-grade antibody Anti CD3 is avail- able OKT-3 (Anti CD3)  ~ 1000 (two weeks) Yes [142, 143] Anti CD16  > 500 (two weeks) No [147] Allogeneic feeder cells PBMC  + PHA, Ionomycin  ~ 100 No [139] Without selection final product may contain up to 40% T cell PBMCs are typically irradiated 25 Gy or more  + ConA  ~ 100 Yes [138]  + anti CD3  ~ 300 Yes [144] Allogeneic feeder cells (tumor) Wilms tumor cell line (HFWT),  ~ 113 (two weeks) Yes [134] Feeder can be genetically modified to enhance activation Feeder cells require irradiation and GMP- grade production Final product needs to be feeder free assured Jurkat  ~ 100 (two weeks) No [135] Risk of bacterial and viral contamination derived from feeder cells Transformed feeder cells Epstein-Barr lymphoblastoid cell line (EBV-LCL),  ~ 3000 (two weeks) Yes [136] Feeder cells require irradiation Safety considerations associated with feeder Engineered feeder K562 4-1BB + IL15  ~ 1200 (two weeks) Yes [38, 125–127] Increased apoptosis of NK cells noted after extensive expansion Engineered feeder K562 4-1BB + IL21  ~ 30,000 (three weeks) Yes [128–130] Greatest rate of expansion reported so far Lower dose of supportive IL-2 required Feeder particles K562 4-1BB + IL21  ~ 250 (two weeks) [141] Avoids the safety considerations associ- ated with feeder cells Laborious to produce Group A-Streptococcus and zole- dronate  ~ 1,560 (three weeks) No [107]  > 90% of NK cells. Cytokine‑induced human NK expansion and activation For example, using the membrane particles of K562 cells with membrane-bound IL-21 and 4-1BB ligand as the feeder has achieved the activation and 250- fold expansion of NK cells after approximately two weeks of ex vivo culture [141]. response of LAK cells was shown attributed mainly to NK cells [116]. Noteworthy, IL-2 primarily activates NK cytotoxicity, while supporting proliferation of both NK and T cells. The expansion of NK cells using IL-2 alone is relatively modest and typically results in only several cell divisions in medium containing 1000 U/mL of IL-2 [100]. IL-21 has various effects on human NK cells. On the one hand, it was shown to enhance interferon gamma (IFN-γ) production, cytotoxic functions, and antibody- dependent cellular cytotoxicity (ADCC) responses [117]. On the other hand, it has limited viability support and can trigger proliferative arrest and apoptosis of NK cells at higher dosage (50  ng/mL) [118]. IL-21 has cumula- tive activation effect in combination with IL-2 [119] or IL-15. It was shown that of CD3-depleted peripheral blood mononuclear cells (PBMC) with IL-21 and IL-15 for 13–20 days resulted in 3.7-fold expansion of NK cells with clinical activity in delaying leukemia progression [120]. The combination of IL-2/IL-15/IL-21 can sup- port up to eightfold expansion of NK cells [121]. Efforts are still ongoing to identify the most optimal cytokine combination for NK cell expansion. Most recent stud- ies showed that ex  vivo  stimulation of human NK cells with the combination of IL-15/IL-18/IL-27 can achieve 17-fold expansion [122], and that the combination of IL-2 with IL-18 can achieve approximately 500-fold expansion over two-week period [123]. The presence of autologous feeder cells (typically CD3-depleted PBMCs) in culture additionally facilitates NK expansion [124]. It is believed that membrane-bound interleukins are able to stimulate the expansion of NK cells more effec- tively than the soluble form. A study by Campana and coworkers has shown that stimulation of NK cells with gene-modified K562 expressing the NK-stimulatory molecules 4-1BB ligand and IL-15 induced a median 21.6-fold expansion during a 7-day culture period. It yielded a greater than 1000-fold expansion of NK cells after 3 weeks of culture [38, 125, 126]. An even greater expansion of NK cells, of over 30,000-fold in a period of 3  weeks, was achieved with K562 membrane-bound IL-21 and 4-1BB ligand [127–130]. Cytokine‑induced human NK expansion and activation May not require mag- netic cell sorting Components IL2, streptococcus and zoledronate are FDA approved support effective expansion without feeder cells. Inter- leukin (IL)-2 (IL-2) is one of the first and most impor- tant cytokines used for NK maintenance and is integral for NK cell survival. It is one of the two cytokines, IL-2 and interferon alpha (IFN-α), approved by the FDA for the treatment of several malignant diseases [114]. IL-2 was used to induce lymphokine-activated killer (LAK) cells, a heterogeneous population of cells consisting pri- marily of NK, NKT, and T cell for autologous killer cell- based cancer therapy decades ago [115]. The anti-tumor Liu et al. J Hematol Oncol (2021) 14:7 Liu et al. J Hematol Oncol (2021) 14:7 Page 10 of 17 derivatives, have been explored for enhancing ex  vivo NK cell expansion. Exposure of NK to unmodified NK cell-sensitive leukemia cells (K562) stimulates expan- sion and short-term proliferation [133]. Over 100-fold expansion was achieved with Wilms tumor HFWT [134] and immortalized T lymphocyte Jurkat cell lines [135]. Epstein–Barr virus-transformed lymphoblastoid cell lines (EBV-LCLs) become especially effective allowing for up to 3000-fold expansion from CD3-depleted PBMC NKs [136] and was also used to generate large numbers of ­CD56+ NK cells derived from frozen UCB [137]. Chemi- cal stimuli such as Concanavalin A  (ConA) [138], Phy- tohemagglutinin (PHA), and ionomycin [139] were also used in combination with irradiated allogeneic PBMCs to facilitate the activation. Although lethal irradiation of feeder cells before use is required, for safety concern, the residual contamination of feeder cells should be assessed. The release criteria should be developed with unique sig- nature to distinguish the feeder cells from expanded NK cells to ensure no feeder cell contamination. For instance, in the use of CD19-modified K562 to propagate NK cells, the contamination was assessed by flow cytometry detec- tion of surface expression of the NK cell endogenous molecule CD32 and the K562 transgene CD19 to distin- guish NK cells from the feeder cells [140]. Other methods may include transgene of suicide gene or expression of a fluorescent marker in the feeder cell. Thus, feeder-free approaches are an alternative or a more desirable method to avoid safety concerns associated with the clini- cal application of cancer cell-derived feeder cells. One approach to address the safety concern is the use of lysed cell product. Cytokine‑induced human NK expansion and activation This protocol created a possibility to generate a substantially higher number of NK cells from a single dose of peripheral blood [131] and is currently in phase I/II clinical trial (NCT01787474) with expanded haploid-identical NK cells for treating relapsed or refractory AML. Moreover, highly cytotoxic NK cells derived using such method are capable of pro- ducing endogenous cytokines that improve their survival, proliferation, and function [132]. Stimulating antibodies is typically used along with irra- diated allogeneic PBMC to further promote NK activa- tion and expansion. OKT3, an anti-CD3 mAb, has been commonly added to the irradiated autologous PBMC feeder, which can promote the expansion of NK cells up to over 1000-fold [4, 142–144]. The anti-CD3 mAb pre- sumably activates T cells in feeder to secret cytokines which subsequently create a milieu favorable for NK cell expansion [145]. Masuyama et al. reported an approximately 1500-fold expansion of NK cells after PBMC stimulation with a combination of anti-CD3 and anti-CD56 mAbs [146]. Using irradiated autologous PBMCs and anti-CD16 mAb, Lee et al. demonstrated a more than 500-fold NK expan- sion with over 98% purity within 2 weeks and a greater than 5000-fold NK expansion over a 3-week period [147]. Other methods to induce human NK cell expansion Other methods to induce human NK cell expansion Beyond cytokines, other stimulants, including tumor cells, allogeneic PBMCs, antibodies, and microbiol Liu et al. J Hematol Oncol (2021) 14:7 Page 11 of 17 Page 11 of 17 A more simplified NK cell expansion method has been used by combining group A  streptococcus and zoledronate with IL-2 to stimulate UCB-derived mono- nuclear cells. This method resulted in a 1,560-fold expan- sion of NK cells with a purity of 92.37% after 21 days of ex vivo culture [107]. This method was advantageous in that it did not require magnetic cell sorting, feeder cells, or multiple cytokines, potentially lowering the cost of production. Furthermore, IL-2, streptococcus A group, and zoledronate have all been approved for human use. Clinical evaluation of safety and efficacy of NK cells under this expansion is warranted. Other methods to induce human NK cell expansion Table 5  Completed and ongoing clinical trial of NK cell-based therapy for hematological malignancies *  Culture medium of every expansion protocol contains IL-2 or more cytokines Cancer type Therapy Phase NCT Status Acute myeloid leukemia Allogeneic, IL-2 UCB-NK cells I/II NCT04347616 Ongoing Allogeneic NK cell I NCT04220684 Ongoing Alloreactive NK cell N/A NCT03955848 Ongoing Poor prognosis non-AML hematologic malignancies Allogeneic haploidentical NK cell infu- sions I NCT00697671 Completed No outcome reported Refractory non-B lineage hematologic malignancies Allogeneic haploidentical NK Cell Infu- sions I NCT00640796 Completed No outcome reported CD33-positive acute myeloid leukemia Anti-CD33 CAR-NK cells I/II NCT02944162 PMID: 28054442 Found to be effective and prevented both tumor relapses and graft versus host disease Acute myeloid leukemia & advanced hematological malignancies Alloreactive, IL-2 activated NK cells I/II NCT01220544 Unknown No outcome reported Hematological malignancy patients who received fate therapeutics Genetically modified NK cell N/A NCT04093622 Ongoing High-risk tumor and lymphoma Allogeneic haploidentical NK cell infu- sion combined with autologous stem cell transplantation I NCT02130869 Completed No outcome reported Pediatric acute leukemia Activated and expanded NK cells (NKAEs) II NCT02074657 PMID: 29477379 Found to be safe and feasible Chronic Lymphocytic Leukemia (CLL) NK cell with rituximab and Rhu-GMCSF I NCT00383994 Completed No outcome reported NK cell I NCT02280525 Ongoing CD19 + Leukemia Allogeneic anti-CD19 CAR-NK cells I/II NCT02892695 PMID: 28054442 Found major improvements in treating leukemia Multiple myeloma KIR-ligand mismatched NK cells from a haploidentical donor I NCT00089453 Completed No outcome reported Acute myeloid leukemia & acute lymphoblastic leukemia Expanded haploidentical NK cells I NCT04327037 Ongoing Ph + acute lymphoblastic leukemia Autologous NK cell I NCT02185781 Unknown No outcome reported Lymphoma, myeloma, and leukemia HLA-I haplotype mismatched NK cell I NCT00660166 Completed No outcome reported Chronic myeloid leukemia NK cell I/II NCT03348033 Enrolling B cell non-Hodgkin’s lymphoma Cord blood-derived expanded allogeneic NK cells combined with rituximab, high-dose chemotherapy, and stem cell transplant II NCT03019640 Ongoing CAR.CD19-CD28-zeta-2A-iCasp9-IL15- transduced cord blood NK cells com- bined with high-dose chemotherapy and stem cell transplant I/II NCT03579927 Withdrawn B cell lymphoma NK cells with rituximab I/II NCT02843061 Completed No outcome reported 5  Completed and ongoing clinical trial of NK cell-based therapy for hematological malignancies ng clinical trial of NK cell-based therapy for hematological malignancies Liu et al. Other methods to induce human NK cell expansion J Hematol Oncol (2021) 14:7 Table 6  (continued) Cancer type Therapy Phase NCT Status Recurrent malignant solid tumor NK cell combined with Bevacizumab I/II NCT02857920 Completed No outcome reported NK cells I NCT03619954 Unknown No outcome reported Pediatric solid tumor Autologous NK cell I NCT01875601 Completed No outcome reported Clinical development of NK cell‑based cancer therapy The “off-shelf” NK cell therapeutic product oNKord, the allogeneic partial HLA-matched NK cells derived from UCB-CD34+ progenitors, has received an orphan drug designation from EMA and FDA for treating AML patients who were not eligible for allogeneic stem cell transplantation. This approval was based on clinical study demonstrating that oNKord improves survival in year 1 of 80% vs. 35% in the control arm. Recently, the FDA has approved the investigational new drug (IND) for the use of placenta-expanded NK cells (CYNK-001) against glioblastoma (GBM). The success has encouraged many ongoing clinical investigations of NK cell-based can- cer therapy alone or in combination with other regimes. Clinicaltrials.gov currently lists over 100 clinical trials of NK cell-based cancer immunotherapy. Herein we high- light current evaluations for hematological malignancies and solid tumors. NK cell‑based clinical trial for solid tumors There are great numbers of clinical trials on NK cell immunotherapy to treat solid tumors. For example, there is an ongoing phase 2 clinical trial to evaluate the safety and efficacy of human HLA-haploidentical hematopoi- etic cell transplantation (HCT) followed by an early, post- transplant infusion of donor NK  cells (NCT02100891). The subjects of the study were patients with high-risk solid tumors, including Ewing Sarcoma, Neuroblastoma, Rhabdomyosarcoma, Osteosarcoma, and CNS tumors. There are also many ongoing clinical trials to evaluate the safety and efficacy of tumor-targeting CAR-NK cells, including using HER2-specific CAR-NK cell to treat advanced or metastatic HER2-expressing solid tumors (NCT04319757), ROBO1-specific CAR-NK cell to treat a broad spectrum of solid tumors (NCT03940820), and MUC1-specific CAR-pNK to treat patients with MUC1-positive relapsed or refractory solid tumor (NCT02839954). MUC1-specific CAR-pNK therapy presented good safety profile and preliminary efficacy in preventing both tumor relapses and graft versus host disease [151]. There are numerous ongoing clinical tri- als exploring the safety and efficacy of NK-based therapy in combination with other modalities. Phase I/II clinical trials are ongoing to evaluate the safety and efficacy of NK cell combined with nimotuzumab to treat late-stage malignancies (NCT03554889) and of NK cell infusion on patients with advanced malignant tumors following multi-line therapies (NCT03619954). NK cell‑based clinical trial for hematological malignancieshfi NK cell‑based clinical trial for hematological malignancies The safety and efficacy of allogeneic or autologous donor- derived NK cell-based therapy for treating hematological malignancies, such as AML, have been well established [5, 29, 148]. With the ongoing effort to improve the treat- ment response and new methods of generating more feasible clinical scales of NK cells, emerging clinical tri- als are being designed to evaluate these new modalities and to expand their indications. A first-in-human clini- cal trial of CAR NK-92 cells in 3 patients with relapsed or refractory AML showed that CAR NK-92 can be infused at doses up to 5 billion cells per patient without caus- ing significant adverse effects [149]. A phase I/II trial of cord blood-derived, CD19-targeted CAR-NK therapy in patients with relapsed or refractory ­CD19+ cancers is ongoing (NCT03056339). The interim results showed that 8 out of 11 patients had an objective response to treatment without development of major toxic effects [150]. Table 5 summarizes the clinical trials of NK cell- based therapy for hematological malignancies to date. Other methods to induce human NK cell expansion J Hematol Oncol (2021) 14:7 Page 12 of 17 Table 6  Completed and ongoing clinical trial of NK cell-based therapy for solid tumors Cancer type Therapy Phase NCT Status Gastric cancer Allogeneic UCB-NK cells N/A NCT04385641 Ongoing Autologous NK cells combined with Trastuzumab I/II NCT02030561 Unknown No outcome reported Pancreatic cancer ROBO1 CAR-NK cells I/II NCT03941457 Ongoing ROBO1 specific BiCAR-NK/T cells I/II NCT03931720 Ongoing Tongue cancer Cryosurgery combined with NK cells I/II NCT02849379 Completed No outcome reported Esophageal cancer Cryosurgery combined with NK cells I/II NCT02843581 Completed No outcome reported Laryngeal cancer Cryosurgery combined with NK cells I/II NCT02849314 Completed No outcome reported Pharyngeal cancer Cryosurgery combined with NK cells I/II NCT02849327 Completed No outcome reported Cancer lack of MHC-I expression Autologous-induced T cell like NK cells I/II NCT03882840 Ongoing Small cell lung cancer Autologous NK cells II NCT03410368 Unknown No outcome reported Non-small cell lung cancer CCCR-modified NK92 cell I NCT03656705 Enrolling Hsp70-peptide TKD/IL-2 activated, autologous NK cells II NCT02118415 Suspended Autologous NK cells I NCT03662477 Ongoing Cryosurgery combined with allogeneic NK cells I/II NCT02843815 PMID: 28508945 Showed preliminary efficacy Cetuximab combined with NK I/II NCT02845856 Completed No outcome reported Renal cancer Cryosurgery combined with NK I/II NCT02843607 Completed No outcome reported Breast cancer Cryosurgery combined with NK I/II NCT02844335 Completed No outcome reported Trastuzumab combined with NK I/II NCT02843126 Completed No outcome reported Ovarian cancer Cryosurgery combined with NK I/II NCT02849353 Completed No outcome reported Cervical cancer Cryosurgery combined with NK I/II NCT02849340 Completed No outcome reported Neuroblastoma Autologous NK combined with antibody ch14.18 and lenalidomide I NCT02573896 Ongoing Allogeneic NK combined with the anti-GD2 antibody I/II NCT03242603 Unknown No outcome reported Liver cancer Cryosurgery combined with NK I/II NCT02843802 Completed No outcome reported Allogeneic NK I NCT01147380 Completed No outcome reported Irreversible electroporation and autologous NK I/II NCT03008343 Completed No outcome reported Cryosurgery combined with NK I/II NCT02849015 Completed No outcome reported High-risk solid tumor Allogeneic NK cells with human leukocyte antigen (HLA)-haploidentical hematopoietic cell transplanta- tion (HCT) II NCT02100891 Ongoing Metastatic HER2-expressing solid tumor Allogeneic ACE1702 (anti-HER2 oNK cells) I NCT04319757 Ongoing Solid tumor expressing ROBO1 ROBO1 CAR-NK cells I/II NCT03940820 Ongoing MUC1-positive solid tumor Allogeneic anti-MUC1 CAR-pNK cells I/II NCT02839954 PMID: 28054442 Show preliminary efficacy Late-stage malignancies Autologous NK cell combined with nimotuzumab I NCT03554889 Unknown No outcome reported Table 6  Completed and ongoing clinical trial of NK cell-based therapy for solid tumors Page 13 of 17 Liu et al. Other methods to induce human NK cell expansion Table  6 summa- rizes updated clinical trials of NK cell-based therapy for all solid tumors. Acknowledgements Not applicable. 9. Lanier LL, et al. Natural killer cells: definition of a cell type rather than a function. J Immunol. 1986;137(9):2735–9. 10. Herberman RB, Nunn ME, Lavrin DH. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic acid allogeneic tumors. I. Dis- tribution of reactivity and specificity. Int J Cancer. 1975; 16(2): 216–29. 10. Herberman RB, Nunn ME, Lavrin DH. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic acid allogeneic tumors. I. Dis- tribution of reactivity and specificity. Int J Cancer. 1975; 16(2): 216–29. Author details NKG2D and DNAM-1 activating receptors and their ligands in NK-T cell interactions: role in the NK cell-mediated nega- tive regulation of T cell responses. Front Immunol. 2012;3:408. 27. Parodi M, et al. NKp44-NKp44 ligand interactions in the regulation of natural killer cells and other innate lymphoid cells in humans. Front Immunol. 2019;10:719. 27. Parodi M, et al. NKp44-NKp44 ligand interactions in the regulation of natural killer cells and other innate lymphoid cells in humans. Front Immunol. 2019;10:719. Competing interest 18. Bjorkstrom NK, Ljunggren HG, Michaelsson J. Emerging insights into natural killer cells in human peripheral tissues. Nat Rev Immunol. 2016;16(5):310–20. 18. Bjorkstrom NK, Ljunggren HG, Michaelsson J. Emerging insights into natural killer cells in human peripheral tissues. Nat Rev Immunol. 2016;16(5):310–20. 19. Abel AM, et al. Natural killer cells: development, maturation, and clinical utilization. Front Immunol. 2018;9:1869. 19. Abel AM, et al. Natural killer cells: development, maturation, and clinical utilization. Front Immunol. 2018;9:1869. Author details 1 Department of Urology, Feinberg School of Medicine, Northwestern Uni- versity, 303 E. Superior St., Lurie Research Building 6‑117, Chicago, IL 60611, USA. 2 Department of Pathology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA. 3 Robert H. Lurie Comprehensive Cancer Center, Northwestern University Feinberg School of Medicine, Chicago, IL, USA. 4 Department of Pediatrics, Stanley Manne Children’s Research Institute, Ann & Robert H. Lurie Children’s Hospital, Northwestern University Feinberg School of Medicine, Chicago, IL, USA. 5 Biology Program, Emory University, Atlanta, USA. 6 Departments of Neurological Surgery, Medicine‑Hematology and Oncology, Microbiology‑Immunology, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA. 7 Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences, Pushchino, Russia. 20. Hashemi E, Malarkannan S. Tissue-resident NK cells: development, maturation, and clinical relevance. Cancers (Basel), 2020; 12(6). 20. Hashemi E, Malarkannan S. Tissue-resident NK cells: development, maturation, and clinical relevance. Cancers (Basel), 2020; 12(6). 21. Cerwenka A, Lanier LL. Natural killer cell memory in infection, inflam- mation and cancer. Nat Rev Immunol. 2016;16(2):112–23. 22. Hammer Q, et al. Peptide-specific recognition of human cytomeg- alovirus strains controls adaptive natural killer cells. Nat Immunol. 2018;19(5):453–63. 22. Hammer Q, et al. Peptide-specific recognition of human cytomeg- alovirus strains controls adaptive natural killer cells. Nat Immunol. 2018;19(5):453–63. 23. Luevano M, Madrigal A, Saudemont A. Transcription factors involved in the regulation of natural killer cell development and function: an update. Front Immunol. 2012;3:319. 23. Luevano M, Madrigal A, Saudemont A. Transcription factors involved in the regulation of natural killer cell development and function: an update. Front Immunol. 2012;3:319. 24. Cichocki F, Grzywacz B, Miller JS. Human NK cell development: one road or many? Front Immunol. 2019;10:2078. 24. Cichocki F, Grzywacz B, Miller JS. Human NK cell development: one road or many? Front Immunol. 2019;10:2078. Received: 14 October 2020 Accepted: 30 November 2020 25. Dhar P, Wu JD. NKG2D and its ligands in cancer. Curr Opin Immunol. 2018;51:55–61. 25. Dhar P, Wu JD. NKG2D and its ligands in cancer. Curr Opin Immunol. 2018;51:55–61. Received: 14 October 2020 Accepted: 30 November 2020 26. Zingoni A, et al. NKG2D and DNAM-1 activating receptors and their ligands in NK-T cell interactions: role in the NK cell-mediated nega- tive regulation of T cell responses. Front Immunol. 2012;3:408. 26. Zingoni A, et al. Conclusions NK cell represents a specialized immune effector cell population equipped with fast-acting and potent anti- tumor capacity. The concept of adoptive NK cell cancer immunotherapy was proven a decade ago from pioneer- ing clinical studies against hematological malignan- cies. The difficulty to obtain large quantity of NK cells, to expand to clinical scale ex vivo, and to sustain in vivo survival and activity of infused NK cells has encumbered the progress. With the new tools of iPSC-NK and genetic Liu et al. J Hematol Oncol (2021) 14:7 Liu et al. J Hematol Oncol (2021) 14:7 Page 14 of 17 4. Sakamoto N, et al. Phase I clinical trial of autologous NK cell therapy using novel expansion method in patients with advanced digestive cancer. J Transl Med. 2015;13:277. 4. Sakamoto N, et al. Phase I clinical trial of autologous NK cell therapy using novel expansion method in patients with advanced digestive cancer. J Transl Med. 2015;13:277. engineering approach as well as new understandings of NK cell biology, it is the time to re-explore the thera- peutic potentials of NK cells. Although challenges exist, perspectives are enticed by currently approved NK cell- based therapies and emerging pre-clinical and clinical studies. 5. Miller JS, et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood. 2005;105(8):3051–7. 5. Miller JS, et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood. 2005;105(8):3051–7. 6. Rubnitz JE, et al. NKAML: a pilot study to determine the safety and fea- sibility of haploidentical natural killer cell transplantation in childhood acute myeloid leukemia. J Clin Oncol. 2010;28(6):955–9. 7. Iliopoulou EG, et al. A phase I trial of adoptive transfer of allogeneic natural killer cells in patients with advanced non-small cell lung cancer. Cancer Immunol Immunother. 2010;59(12):1781–9. Authors’ contributions Jennifer Wu helped in conception and design. Sizhe Liu, Vasiliy Galat, Yekat- erina Galat, Yoo Kyung Annie Lee, Derek Wainwright, and Jennifer Wu wrote and reviewed the manuscript. Jennifer Wu supervised the study. All authors read and approved the final manuscript. yi y 11. Smyth MJ, et al. Activation of NK cell cytotoxicity. Mol Immunol. 2005;42(4):501–10. 11. Smyth MJ, et al. Activation of NK cell cytotoxicity. Mol Immunol. 2005;42(4):501–10. 12. Roda JM, et al. Natural killer cells produce T cell-recruiting chemokines in response to antibody-coated tumor cells. Cancer Res. 2006;66(1):517–26. 12. Roda JM, et al. Natural killer cells produce T cell-recruiting chemokines in response to antibody-coated tumor cells. Cancer Res. 2006;66(1):517–26. Availability of data and materials Not applicable. 14. Spits H, et al. Innate lymphoid cells: a proposal for uniform nomen- clature. Nat Rev Immunol. 2013;13(2):145–9. Availability of data and materials Not applicable. 15. Gao Y, et al. Tumor immunoevasion by the conversion of effec- tor NK cells into type 1 innate lymphoid cells. Nat Immunol. 2017;18(9):1004–15. 15. Gao Y, et al. Tumor immunoevasion by the conversion of effec- tor NK cells into type 1 innate lymphoid cells. Nat Immunol. 2017;18(9):1004–15. Ethics approval and consent to participate Not applicable. 3. Lopez-Soto A, et al. Control of metastasis by NK Cells. Cancer Cell. 2017;32(2):135–54. Ethics approval and consent to participate 16. Colonna M. Innate lymphoid cells: diversity, plasticity, and unique functions in immunity. Immunity. 2018;48(6):1104–17. 16. Colonna M. Innate lymphoid cells: diversity, plasticity, and unique functions in immunity. Immunity. 2018;48(6):1104–17. Consent for publication All authors gave consent for publication of the manuscript. Consent for publication All authors gave consent for publication of the manuscript. 2. Imai K, et al. Natural cytotoxic activity of peripheral-blood lymphocytes and cancer incidence: an 11-year follow-up study of a general popula- tion. Lancet. 2000;356(9244):1795–9. Consent for publication y y 17. Freud AG, et al. The broad spectrum of human natural killer cell diver- sity. Immunity. 2017;47(5):820–33. 17. Freud AG, et al. The broad spectrum of human natural killer cell diver- sity. Immunity. 2017;47(5):820–33. 1. Guerra N, et al. NKG2D-deficient mice are defective in tumor surveillance in models of spontaneous malignancy. Immunity. 2008;28(4):571–80. Funding 13. Bottcher JP, et al. NK cells stimulate recruitment of cDC1 into the tumor microenvironment promoting cancer immune control. Cell. 2018; 172(5): 1022–1037. 13. Bottcher JP, et al. NK cells stimulate recruitment of cDC1 into the tumor microenvironment promoting cancer immune control. Cell. 2018; 172(5): 1022–1037. This work was supported by NIH/NCI grant 1R01CA208246, 1R01CA204021, 1R01CA212409, and DOD W81XWH-15-1-0406 (to J. D. Wu). 14. Spits H, et al. Innate lymphoid cells: a proposal for uniform nomen- clature. Nat Rev Immunol. 2013;13(2):145–9. 4. Sakamoto N, et al. Phase I clinical trial of autologous NK cell therapy using novel expansion method in patients with advanced digestive cancer. J Transl Med. 2015;13:277. Abbreviations NK N l k ll NK: Natural killer; MHC I: Major histocompatibility complex I; CAR​: Chimeric antigen receptor; IL: Interleukin; ADCC: Antibody-dependent cell cytotoxic- ity; AML: Acute myeloid leukemia; iPSC: Induced pluripotent stem cell; LAK: Lymphokine-activated killer. 8. Kiessling R, Klein E, Wigzell H. Natural" killer cells in the mouse. I. Cyto- toxic cells with specificity for mouse Moloney leukemia cells. Specificity and distribution according to genotype. Eur J Immunol. 1975;5(2): 112–7. 9. Lanier LL, et al. Natural killer cells: definition of a cell type rather than a function. J Immunol. 1986;137(9):2735–9. Liu et al. J Hematol Oncol (2021) 14:7 Liu et al. J Hematol Oncol (2021) 14:7 Page 15 of 17 Page 15 of 17 Liu et al. J Hematol Oncol (2021) 14:7 31. Ruggeri L, et al. Role of natural killer cell alloreactivity in HLA- mismatched hematopoietic stem cell transplantation. Blood. 1999;94(1):333–9. 55. Topfer K, et al. DAP12-based activating chimeric antigen receptor for NK cell tumor immunotherapy. J Immunol. 2015;194(7):3201–12. 56. Liu B, et al. Development of cMET-specific chimeric antigen receptor- engineered natural killer cells with cytotoxic effects on human liver cancer HepG2 cells. Mol Med Rep. 2019;20(3):2823–31. 32. Ruggeri L, et al. Effectiveness of donor natural killer cell allo- reactivity in mismatched hematopoietic transplants. Science. 2002;295(5562):2097–100. 57. Huang Y, et al. DNAM1 and 2B4 costimulatory domains enhance the cytotoxicity of anti-GPC3 chimeric antigen receptor-modified natural killer cells against hepatocellular cancer cells in vitro. Cancer Manag Res. 2020;12:3247–55. 33. Gras Navarro A, et al. NK cells with KIR2DS2 immunogenotype have a functional activation advantage to efficiently kill glioblastoma and prolong animal survival. J Immunol. 2014;193(12):6192–206. 34. Veluchamy JP, et al. In vivo efficacy of umbilical cord blood stem cell-derived NK cells in the treatment of metastatic colorectal cancer. Front Immunol. 2017;8:87. 58. Genssler S, et al. Dual targeting of glioblastoma with chimeric antigen receptor-engineered natural killer cells overcomes heterogeneity of target antigen expression and enhances antitumor activity and survival. Oncoimmunology. 2016;5(4):e1119354. 35. Geller MA, et al. Intraperitoneal delivery of human natural killer cells for treatment of ovarian cancer in a mouse xenograft model. Cyto- therapy. 2013;15(10):1297–306. 59. Muller N, et al. Engineering NK cells modified with an EGFRvIII- specific chimeric antigen receptor to overexpress CXCR4 improves immunotherapy of CXCL12/SDF-1alpha-secreting glioblastoma. J Immunother. 2015;38(5):197–210. y 36. Lee SJ, et al. Natural killer (NK) cells inhibit systemic metastasis of glioblastoma cells and have therapeutic effects against glioblastomas in the brain. BMC Cancer. 2015;15:1011. 60. Zhang C., et al. ErbB2/HER2-specific NK cells for targeted therapy of glioblastoma. J Natl Cancer Inst. 2016; 108(5). 37. Rezvani K, et al. Engineering natural killer cells for cancer immuno- therapy. Mol Ther. 2017;25(8):1769–81. 61. Sahm C, Schonfeld K, Wels WS. Expression of IL-15 in NK cells results in rapid enrichment and selective cytotoxicity of gene-modified effectors that carry a tumor-specific antigen receptor. Cancer Immu- nol Immunother. 2012;61(9):1451–61. 38. Imai C, Iwamoto S, Campana D. Genetic modification of primary natural killer cells overcomes inhibitory signals and induces specific killing of leukemic cells. Blood. 2005;106(1):376–83. 39. Liu et al. J Hematol Oncol (2021) 14:7 Xu Y, et al. 2B4 costimulatory domain enhancing cytotoxic ability of anti-CD5 chimeric antigen receptor engineered natural killer cells against T cell malignancies. J Hematol Oncol. 2019;12(1):49. 69. Shiozawa M, et al. Pharmacologically upregulated carcinoembryonic antigen-expression enhances the cytolytic activity of genetically- modified chimeric antigen receptor NK-92MI against colorectal cancer cells. BMC Immunol. 2018;19(1):27. 47. Schnalzger TE. et al. 3D model for CAR-mediated cytotoxicity using patient-derived colorectal cancer organoids. EMBO J, 2019; 38(12). 70. Uherek C, et al. Retargeting of natural killer-cell cytolytic activity to ErbB2-expressing cancer cells results in efficient and selective tumor cell destruction. Blood. 2002;100(4):1265–73. 48. Schonfeld K, et al. Selective inhibition of tumor growth by clonal NK cells expressing an ErbB2/HER2-specific chimeric antigen receptor. Mol Ther. 2015;23(2):330–8. 71. Zhang Q, et al. Synergistic effects of cabozantinib and EGFR- specific CAR-NK-92 cells in renal cell carcinoma. J Immunol Res. 2017;2017:6915912. 49. Romanski A, et al. CD19-CAR engineered NK-92 cells are sufficient to overcome NK cell resistance in B-cell malignancies. J Cell Mol Med. 2016;20(7):1287–94. 50. Zhang Q, et al. Combination therapy with EpCAM-CAR-NK-92 cells and regorafenib against human colorectal cancer models. J Immunol Res. 2018;2018:4263520. 72. Hambach, J., et al., Targeting CD38-expressing multiple myeloma and Burkitt lymphoma cells in vitro with nanobody-based chimeric antigen receptors (Nb-CARs). Cells, 2020; 9(2). g p ( ) , ; ( ) 73. Pinz KG, et al. Targeting T-cell malignancies using anti-CD4 CAR NK-92 cells. Oncotarget. 2017;8(68):112783–96. 74. Hu Z. Tissue factor as a new target for CAR-NK cell immunotherapy of triple-negative breast cancer. Sci Rep. 2020;10(1):2815. 51. Chen X, et al. A combinational therapy of EGFR-CAR NK cells and oncolytic herpes simplex virus 1 for breast cancer brain metastases. Oncotarget. 2016;7(19):27764–77. 73. Pinz KG, et al. Targeting T-cell malignancies using anti-CD4 CAR NK-92 cells. Oncotarget. 2017;8(68):112783–96. 73. Pinz KG, et al. Targeting T-cell malignancies using anti-CD4 CAR NK-92 cells. Oncotarget. 2017;8(68):112783–96. 74. Hu Z. Tissue factor as a new target for CAR-NK cell immunotherapy of triple-negative breast cancer. Sci Rep. 2020;10(1):2815. g 52. Oelsner S, et al. Continuously expanding CAR NK-92 cells display selec- tive cytotoxicity against B-cell leukemia and lymphoma. Cytotherapy. 2017;19(2):235–49. 75. Cao B, et al. Use of chimeric antigen receptor NK-92 cells to target mesothelin in ovarian cancer. Biochem Biophys Res Commun. 2020;524(1):96–102. 53. Chu J, et al. CS1-specific chimeric antigen receptor (CAR)-engineered natural killer cells enhance in vitro and in vivo antitumor activity against human multiple myeloma. Liu et al. J Hematol Oncol (2021) 14:7 Liu E, et al. Cord blood NK cells engineered to express IL-15 and a CD19-targeted CAR show long-term persistence and potent antitumor activity. Leukemia. 2018;32(2):520–31. 62. Oelsner S, et al. Genetically engineered CAR NK cells display selective cytotoxicity against FLT3-positive B-ALL and inhibit in vivo leukemia growth. Int J Cancer. 2019;145(7):1935–45. 40. Ni Z, et al. Expression of chimeric receptor CD4zeta by natural killer cells derived from human pluripotent stem cells improves in vitro activity but does not enhance suppression of HIV infection in vivo. Stem Cells. 2014;32(4):1021–31. 63. Muller S, et al. High cytotoxic efficiency of lentivirally and alpharet- rovirally engineered CD19-specific chimeric antigen receptor natural killer cells against acute lymphoblastic leukemia. Front Immunol. 2019;10:3123. 41. Li Y, et al. Human iPSC-derived natural killer cells engineered with chimeric antigen receptors enhance anti-tumor activity. Cell Stem Cell. 2018;23(2):181-192.e5. 64. Klapdor R, et al. Improved killing of ovarian cancer stem cells by combining a novel chimeric antigen receptor-based immunotherapy and chemotherapy. Hum Gene Ther. 2017;28(10):886–96.i 65. Esser R, et al. NK cells engineered to express a GD2 -specific antigen receptor display built-in ADCC-like activity against tumour cells of neuroectodermal origin. J Cell Mol Med. 2012;16(3):569–81. 42. Li Y, et al. Human iPSC-derived natural killer cells engineered with chimeric antigen receptors enhance anti-tumor activity. Cell Stem Cell. 2018; 23(2): 181–192. 66. Ueda T, et al. Non-clinical efficacy, safety and stable clinical cell processing of induced pluripotent stem cell-derived anti-glypican-3 chimeric antigen receptor-expressing natural killer/innate lymphoid cells. Cancer Sci. 2020;111(5):1478–90. 43. Wang J, et al. Purinergic targeting enhances immunotherapy of CD73(+) solid tumors with piggyBac-engineered chimeric antigen receptor natural killer cells. J Immunother Cancer. 2018;6(1):136.i 44. Chang YH, et al. A chimeric receptor with NKG2D specificity enhances natural killer cell activation and killing of tumor cells. Cancer Res. 2013;73(6):1777–86. 67. You F, et al. A novel CD7 chimeric antigen receptor-modified NK-92MI cell line targeting T-cell acute lymphoblastic leukemia. Am J Cancer Res. 2019;9(1):64–78. 45. Shimasaki N, et al. A clinically adaptable method to enhance the cyto- toxicity of natural killer cells against B-cell malignancies. Cytotherapy. 2012;14(7):830–40. 68. Lu C, et al. A novel chimeric PD1-NKG2D-41BB receptor enhances antitumor activity of NK92 cells against human lung cancer H1299 cells by triggering pyroptosis. Mol Immunol. 2020;122:200–6. 68. Lu C, et al. A novel chimeric PD1-NKG2D-41BB receptor enhances antitumor activity of NK92 cells against human lung cancer H1299 cells by triggering pyroptosis. Mol Immunol. 2020;122:200–6. 46. References 28. Barrow AD, Martin CJ, Colonna M. The natural cytotoxicity receptors in health and disease. Front Immunol. 2019;10:909. 29. Handgretinger R, Lang P, Andre MC. Exploitation of natural killer cells for the treatment of acute leukemia. Blood. 2016;127(26):3341–9. 30. Wang W, et al. NK Cell-mediated antibody-dependent cellular cyto- toxicity in cancer immunotherapy. Front Immunol. 2015;6:368. 30. Wang W, et al. NK Cell-mediated antibody-dependent cellular cyto- toxicity in cancer immunotherapy. Front Immunol. 2015;6:368. 3. Lopez-Soto A, et al. Control of metastasis by NK Cells. Cancer Cell. 2017;32(2):135–54. Liu et al. J Hematol Oncol (2021) 14:7 Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion. Blood. 2006;107(1):159–66. 108. Galat Y, et al. Cytokine-free directed differentiation of human pluri- potent stem cells efficiently produces hemogenic endothelium with lymphoid potential. Stem Cell Res Ther. 2017;8(1):67. 83. Rosario M, et al. The IL-15-based ALT-803 complex enhances FcgammaRIIIa-triggered NK cell responses and in vivo clearance of B cell lymphomas. Clin Cancer Res. 2016;22(3):596–608. 109. Knorr DA, et al. Clinical-scale derivation of natural killer cells from human pluripotent stem cells for cancer therapy. Stem Cells Transl Med. 2013;2(4):274–83. 84. Ni J, et al. Sustained effector function of IL-12/15/18-preactivated NK cells against established tumors. J Exp Med. 2012;209(13):2351–65. 110. Dege C, et al. Potently cytotoxic natural killer cells initially emerge from erythro-myeloid progenitors during mammalian development. Dev Cell. 2020;53(2):229–39. 85. Romee, R., et al. Cytokine-induced memory-like natural killer cells exhibit enhanced responses against myeloid leukemia. Sci Transl Med. 2016; 8(357): p. 357ra123. 111. Zeng J, et al. Generation of “off-the-shelf” natural killer cells from peripheral blood cell-derived induced pluripotent stem cells. Stem Cell Reports. 2017;9(6):1796–812. 86. Leong JW, et al. Preactivation with IL-12, IL-15, and IL-18 induces CD25 and a functional high-affinity IL-2 receptor on human cytokine- induced memory-like natural killer cells. Biol Blood Marrow Transpl. 2014;20(4):463–73. p 112. Galat Y, et al. Application of small molecule CHIR99021 leads to the loss of hemangioblast progenitor and increased hematopoiesis of human pluripotent stem cells. Exp Hematol. 2018;65:38-48.e1. 87. Cichocki F, et al. GSK3 inhibition drives maturation of NK cells and enhances their antitumor activity. Cancer Res. 2017;77(20):5664–75. 113. Ditadi A, Sturgeon CM, Keller G. A view of human haematopoi- etic development from the Petri dish. Nat Rev Mol Cell Biol. 2017;18(1):56–67. 113. Ditadi A, Sturgeon CM, Keller G. A view of human haematopoi- etic development from the Petri dish. Nat Rev Mol Cell Biol. 2017;18(1):56–67. 88. North J, et al. Tumor-primed human natural killer cells lyse NK-resistant tumor targets: evidence of a two-stage process in resting NK cell activa- tion. J Immunol. 2007;178(1):85–94. 114. Berraondo P, et al. Cytokines in clinical cancer immunotherapy. Br J Cancer. 2019;120(1):6–15. 89. Fehniger TA, et al. A phase 1 trial of CNDO-109-activated natural killer cells in patients with high-risk acute myeloid leukemia. Biol Blood Mar- row Transpl. 2018;24(8):1581–9. 115. Lotze MT, et al. Lysis of fresh and cultured autologous tumor by human lymphocytes cultured in T-cell growth factor. Cancer Res. Liu et al. J Hematol Oncol (2021) 14:7 1981;41(11 Pt 1):4420–5. 90. Lee J, et al. An antibody designed to improve adoptive NK-cell therapy inhibits pancreatic cancer progression in a murine model. Cancer Immunol Res. 2019;7(2):219–29. 116. Grimm, E.A., et al., Lymphokine-activated killer cell phenomenon. Lysis of natural killer-resistant fresh solid tumor cells by interleukin 2-acti- vated autologous human peripheral blood lymphocytes. J Exp Med, 1982; 155(6): 1823–41. 91. Hassan R, Ho M. Mesothelin targeted cancer immunotherapy. Eur J Cancer. 2008;44(1):46–53. 117. McMichael EL, et al. IL-21 enhances natural killer cell response to cetuximab-coated pancreatic tumor cells. Clin Cancer Res. 2017;23(2):489–502. 92. Allaoui R, et al. Cancer-associated fibroblast-secreted CXCL16 attracts monocytes to promote stroma activation in triple-negative breast cancers. Nat Commun. 2016;7:13050. 93. Kim MJ, et al. CXCL16 positively correlated with M2-macrophage infil- tration, enhanced angiogenesis, and poor prognosis in thyroid cancer. Sci Rep. 2019;9(1):13288. 118. Li Q, et al. Multiple effects of IL-21 on human NK cells in ex vivo expan- sion. Immunobiology. 2015;220(7):876–88. 119. Heinze, A., et al. The synergistic use of IL-15 and IL-21 for the genera- tion of NK cells from CD3/CD19-depleted grafts improves their ex vivo expansion and cytotoxic potential against neuroblastoma: perspective for optimized immunotherapy post haploidentical stem cell transplan- tation. Front Immunol. 2019; 10(2816). 94. Carlsten M, et al. Efficient mRNA-based genetic engineering of human NK cells with high-affinity CD16 and CCR7 augments rituximab- induced ADCC against lymphoma and targets NK cell migration toward the lymph node-associated chemokine CCL19. Front Immunol. 2016;7:105. 120. Choi I, et al. Donor-derived natural killer cells infused after human leukocyte antigen-haploidentical hematopoietic cell transplantation: a dose-escalation study. Biol Blood Marrow Transpl. 2014;20(5):696–704. 95. Childs RW, Berg M. Bringing natural killer cells to the clinic: ex vivo manipulation. Hematology Am Soc Hematol Educ Program. 2013;2013(1):234–46. 121. Oberschmidt O, et al. Development of automated separation, expan- sion, and quality control protocols for clinical-scale manufacturing of primary human NK cells and alpharetroviral chimeric antigen receptor engineering. Hum Gene Ther Methods. 2019;30(3):102–20. 96. Granzin M, et al. Shaping of natural killer cell antitumor activity by ex vivo cultivation. Front Immunol. 2017;8:458. 96. Granzin M, et al. Shaping of natural killer cell antitumor activity by ex vivo cultivation. Front Immunol. 2017;8:458. 97. Chabannon C, et al. Manufacturing natural killer cells as medicinal products. Front Immunol. 2016;7:504. 97. Chabannon C, et al. Manufacturing natural killer cells as medicinal products. Front Immunol. 2016;7:504. 122. Choi YH, et al. Liu et al. J Hematol Oncol (2021) 14:7 Leukemia. 2014;28(4):917–27. 76. Han J, et al. CAR-engineered NK cells targeting wild-type EGFR and EGFRvIII enhance killing of glioblastoma and patient-derived glio- blastoma stem cells. Sci Rep. 2015;5:11483. 76. Han J, et al. CAR-engineered NK cells targeting wild-type EGFR and EGFRvIII enhance killing of glioblastoma and patient-derived glio- blastoma stem cells. Sci Rep. 2015;5:11483. 54. Ng YY, Tay JCK, Wang S. cxcr1 expression to improve anti-cancer efficacy of intravenously injected CAR-NK cells in mice with peritoneal xenografts. Mol Ther Oncolytics. 2020;16:75–85. Liu et al. J Hematol Oncol (2021) 14:7 Liu et al. J Hematol Oncol (2021) 14:7 Page 16 of 17 Page 16 of 17 77. Parihar R, et al. NK cells expressing a chimeric activating receptor eliminate MDSCs and rescue impaired CAR-T cell activity against solid tumors. Cancer Immunol Res. 2019;7(3):363–75. natural killer cells for multiple infusions to leukemia patients. Cyto- therapy. 2010;12(6):750–63. y 102. Mehta RS, Shpall EJ, Rezvani K. Cord blood as a source of natural killer cells. Front Med (Lausanne). 2015;2:93. 78. Zhang P, et al. Effects of CSF1R-targeted chimeric antigen receptor- modified NK92MI & T cells on tumor-associated macrophages. Immunotherapy. 2018;10(11):935–49. 103. Nham T, et al. Ex vivo-expanded NK cells from blood and ascites of ovarian cancer patients are cytotoxic against autologous primary ovar- ian cancer cells. Cancer Immunol Immunother. 2018;67(4):575–87. 79. Felices M, et al. Generation of BiKEs and TriKEs to improve NK cell-mediated targeting of tumor cells. Methods Mol Biol. 2016;1441:333–46. 104. Sarvaria A, et al. Umbilical cord blood natural killer cells, their character- istics, and potential clinical applications. Front Immunol. 2017;8:329. 105. Spanholtz J, et al. Clinical-grade generation of active NK cells from cord blood hematopoietic progenitor cells for immunotherapy using a closed-system culture process. PLoS ONE. 2011;6(6):e20740. 80. Vallera DA, et al. IL15 Trispecific Killer Engagers (TriKE) Make natural killer cells specific to CD33+ targets while also inducing persis- tence, in vivo expansion, and enhanced function. Clin Cancer Res. 2016;22(14):3440–50. 106. Kang L, et al. Characterization and ex vivo expansion of human placenta-derived natural killer cells for cancer immunotherapy. Front Immunol. 2013;4:101. 81. Gauthier, L., et al. Multifunctional natural killer cell engagers target- ing NKp46 trigger protective tumor immunity. Cell, 2019; 177(7): 1701–1713. 107. Mu YX, et al. A simple method for in vitro preparation of natural killer cells from cord blood. BMC Biotechnol. 2019;19(1):80. 82. Bryceson YT, et al. 125. Fujisaki H, et al. Expansion of highly cytotoxic human natural killer cells for cancer cell therapy. Cancer Res. 2009;69(9):4010–7. Liu et al. J Hematol Oncol (2021) 14:7 Correction to: IL-27 enhances IL-15/IL-18-mediated acti- vation of human natural killer cells. J Immunother Cancer. 2019;7(1):211 p 98. Ferrara JL, et al. Graft-versus-host disease. Lancet. 2009;373(9674):1550–61. 98. Ferrara JL, et al. Graft-versus-host disease. Lancet. 2009;373(9674):1550–61. 123. Tanaka Y, et al. Ex vivo-expanded highly purified natural killer cells in combination with temozolomide induce antitumor effects in human glioblastoma cells in vitro. PLoS ONE. 2019;14(3):e0212455. 99. Skeate R, et al. Hemolytic anemia due to passenger lymphocyte syndrome in solid malignancy patients treated with allogeneic natural killer cell products. Transfusion. 2013;53(2):419–23. 99. Skeate R, et al. Hemolytic anemia due to passenger lymphocyte syndrome in solid malignancy patients treated with allogeneic natural killer cell products. Transfusion. 2013;53(2):419–23. 124. Torelli GF, et al. A good manufacturing practice method to ex vivo expand natural killer cells for clinical use. Blood Transfus. 2015;13(3):464–71. 100. Koehl U, et al. Clinical grade purification and expansion of NK cell prod- ucts for an optimized manufacturing protocol. Front Oncol. 2013;3:118. 100. Koehl U, et al. Clinical grade purification and expansion of NK cell prod- ucts for an optimized manufacturing protocol. Front Oncol. 2013;3:118. 101. Siegler U, et al. Good manufacturing practice-compliant cell sorting and large-scale expansion of single KIR-positive alloreactive human 125. Fujisaki H, et al. Expansion of highly cytotoxic human natural killer cells for cancer cell therapy. Cancer Res. 2009;69(9):4010–7. 101. Siegler U, et al. Good manufacturing practice-compliant cell sorting and large-scale expansion of single KIR-positive alloreactive human Liu et al. J Hematol Oncol (2021) 14:7 Liu et al. J Hematol Oncol (2021) 14:7 Page 17 of 17 126. Szmania S, et al. Ex vivo-expanded natural killer cells demonstrate robust proliferation in vivo in high-risk relapsed multiple myeloma patients. J Immunother. 2015;38(1):24–36. 140. Lee DA. Regulatory considerations for NK cells used in human immuno- therapy applications. Methods Mol Biol. 2016;1441:347–61. 141. Oyer JL, et al. Natural killer cells stimulated with PM21 particles expand and biodistribute in vivo: clinical implications for cancer treatment. Cytotherapy. 2016;18(5):653–63. 127. Denman CJ, et al. Membrane-bound IL-21 promotes sustained ex vivo proliferation of human natural killer cells. PLoS ONE. 2012;7(1):e30264 142. Parkhurst MR, et al. Adoptive transfer of autologous natural killer cells leads to high levels of circulating natural killer cells but does not medi- ate tumor regression. Clin Cancer Res. 2011;17(19):6287–97. 128. Ojo EO, et al. Membrane bound IL-21 based NK cell feeder cells drive robust expansion and metabolic activation of NK cells. Sci Rep. Liu et al. J Hematol Oncol (2021) 14:7 First-in-man clinical trial of CAR NK-92 cells: safety test of CD33-CAR NK-92 cells in patients with relapsed and refractory acute myeloid leukemia. Am J Cancer Res. 2018;8(6):1083–9. 136. Granzin M, et al. Highly efficient IL-21 and feeder cell-driven ex vivo expansion of human NK cells with therapeutic activity in a xenograft mouse model of melanoma. Oncoimmunology. 2016;5(9):e1219007. 150. Liu E, et al. Use of CAR-transduced natural killer cells in CD19-positive lymphoid tumors. N Engl J Med. 2020;382(6):545–53. 150. Liu E, et al. Use of CAR-transduced natural killer cells in CD19-positive lymphoid tumors. N Engl J Med. 2020;382(6):545–53. 151. Del Zotto G, et al. Markers and function of human NK cells in normal and pathological conditions. Cytometry B Clin Cytom. 2017;92(2):100–14. 137. Vasu S, et al. A novel method to expand large numbers of CD56(+) natural killer cells from a minute fraction of selectively accessed cryopreserved cord blood for immunotherapy after transplantation. Cytotherapy. 2015;17(11):1582–93. Liu et al. J Hematol Oncol (2021) 14:7 2019;9(1):14916. 129. Shah N, et al. Phase I study of cord blood-derived natural killer cells combined with autologous stem cell transplantation in multiple myeloma. Br J Haematol. 2017;177(3):457–66. 143. Yang Y, et al. Phase I study of random healthy donor-derived allogeneic natural killer cell therapy in patients with malignant lymphoma or advanced solid tumors. Cancer Immunol Res. 2016;4(3):215–24. 130. Ciurea SO, et al. Phase 1 clinical trial using mbIL21 ex vivo-expanded donor-derived NK cells after haploidentical transplantation. Blood. 2017;130(16):1857–68. 144. Kim EK, et al. Ex vivo activation and expansion of natural killer cells from patients with advanced cancer with feeder cells from healthy volun- teers. Cytotherapy. 2013;15(2):231-241.e1. 131. Lee DA, Verneris MR, Campana D. Acquisition, preparation, and func- tional assessment of human NK cells for adoptive immunotherapy. Methods Mol Biol. 2010;651:61–77. 145. Huijskens MJ, et al. Ascorbic acid promotes proliferation of natural killer cell populations in culture systems applicable for natural killer cell therapy. Cytotherapy. 2015;17(5):613–20. 132. Koehl U, et al. Advances in clinical NK cell studies: donor selection, man- ufacturing and quality control. Oncoimmunology. 2016;5(4):e1115178.f 146. Masuyama J, et al. Ex vivo expansion of natural killer cells from human peripheral blood mononuclear cells co-stimulated with anti-CD3 and anti-CD52 monoclonal antibodies. Cytotherapy. 2016;18(1):80–90. 133. Phillips JH, Lanier LL. A model for the differentiation of human natural killer cells. Studies on the in vitro activation of Leu-11+ granular lymphocytes with a natural killer-sensitive tumor cell, K562. J Exp Med. 1985; 161(6): 1464–82. y y 147. Lee HR, et al. Expansion of cytotoxic natural killer cells using irradi- ated autologous peripheral blood mononuclear cells and anti-CD16 antibody. Sci Rep. 2017;7(1):11075. 147. Lee HR, et al. Expansion of cytotoxic natural killer cells using irradi- ated autologous peripheral blood mononuclear cells and anti-CD16 antibody. Sci Rep. 2017;7(1):11075. y p 148. Myers JA, Miller JS. Exploring the NK cell platform for cancer immuno- therapy. Nat Rev Clin Oncol. 2020. y 148. Myers JA, Miller JS. Exploring the NK cell platform for cancer immuno- therapy. Nat Rev Clin Oncol. 2020. 134. Ishikawa E, et al. Autologous natural killer cell therapy for human recur- rent malignant glioma. Anticancer Res. 2004;24(3b):1861–71. 135. Lim SA, et al. Ex vivo expansion of highly cytotoxic human NK cells by cocultivation with irradiated tumor cells for adoptive immunotherapy. Cancer Res. 2013;73(8):2598–607. 149. Tang X, et al. Publisher’s Note S i N i y y 138. Lister J, et al. Autologous peripheral blood stem cell transplantation and adoptive immunotherapy with activated natural killer cells in the immediate posttransplant period. Clin Cancer Res. 1995;1(6):607–14. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. 139. Luhm J, et al. Large-scale generation of natural killer lymphocytes for clinical application. J Hematother Stem Cell Res. 2002;11(4):651–7. 139. Luhm J, et al. Large-scale generation of natural killer lymphocytes for clinical application. J Hematother Stem Cell Res. 2002;11(4):651–7. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from:

https://openalex.org/W2489225126

https://www.nature.com/articles/ncomms12175.pdf

English

Deletion of Wiskott–Aldrich syndrome protein triggers Rac2 activity and increased cross-presentation by dendritic cells

Nature communications

cc-by

1 Department of Microbiology Tumor and Cell biology, Karolinska Institutet, Stockholm 171 77, Sweden. 2 Institute for Virology and Immunobiology, University of Wu¨rzburg, 97078 Wu¨rzburg, Germany. 3 Experimental Physiopathology, Department of Sciences/Experimental Physiopatholgy, Medical School, University of Sa˜o Paulo, Sa˜o Paulo, Brazil. 4 Department of Medicine Solna, Translational Immunology Unit, Karolinska Institutet and Karolinska University Hospital, Stockholm 171 76, Sweden. 5 University College London Institute of Child Health, London WC1N 1EH, UK. 6 Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm 171 76, Sweden. 7 Department of Paediatrics, Yong Loo Lin School of Medicine, National University of Singapore, Singapore 119228, Singapore. 8 Khoo Teck Puat-National University Children’s Medical Institute, The National University Health System, Singapore 119228, Singapore. 9 Gastroenterology Division, Children’s Hospital, Harvard Medical School, Boston MA 02115, USA. 10Department of Medicine Solna, Dermatology and Venereology Unit, Karolinska Institutet, Stockholm 171 76, Sweden. 11Department of Immunology, Royal Free London NHS Foundation Trust, London NW3 2QG, UK. 12University College London Institute of Immunity and Transplantation, London WC1E 6BT, UK. 13Department of Clinical Microbiology, Division of Immunology, Umeå University, Umeå 901 87, Sweden. Correspondence and requests for materials should be addressed to L.S.W. (email: [email protected]). ARTICLE Received 22 Oct 2015 | Accepted 8 Jun 2016 | Published 18 Jul 2016 Received 22 Oct 2015 | Accepted 8 Jun 2016 | Published 18 Jul 2016 Received 22 Oct 2015 | Accepted 8 Jun 2016 | Published 18 Jul 2016 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE We detected similar number of Ki67 þ proliferating dermal DCs in wild-type and WASp KO mice (Supplementary Fig. 1a). Wild-type mice had similar numbers of CD4 þ and CD8 þ T cells in the skin before and after Der p 2 challenge (Fig. 1d). WASp KO mice had decreased number of CD4 þ T cells in unchallenged skin. Upon Der p 2 challenge, the number of CD4 þ T cells was increased in WASp KO mice and reached similar number to Der p 2-challenged wild-type mice (Fig. 1d). In contrast, unchallenged WASp KO mice had increased number of CD8 þ T cell in the skin compared with wild-type mice and the CD8 þ T-cell population was further increased in WASp KO mice after Der p 2 challenge (Fig. 1d). To corroborate the results from histological analysis, we performed flow cytometry analysis of the 1 cm2 back skin challenged with Der p 2 (containing both macroscopically inflamed and non-inflamed skin). Upon Der p 2 challenge, wild-type and WASp KO mice had increased number of CD45 þ hematopoietic cells in the skin when compared with unchallenged mice (Supplementary Fig. 1c). Der p 2 challenge induced increased number of CD11b þCD11c þ and EpCAM þ DCs in the WASp KO skin when compared with Der p 2-challenged wild-type skin (Fig. 1e). However, due to the fact that we assessed both lesional and non-lesional skin by flow cytometry, Der p 2 challenge did not induce a significant increase of CD4 þ and CD8 þ T cells in wild-type or WASp KO mice (Fig. 1e). WASp KO mice had fewer IFNg-producing CD4 þ T cells when compared with wild-type mice both before and after Der p 2 challenge (Supplementary Fig. 1c). Moreover, WASp KO mice had a tendency to increased number of IFNg- producing CD8 þ T cells and CD11c þEpCAM CD103 þ DCs W 1 3 W iskott–Aldrich syndrome (WAS) is a severe X-linked primary immunodeficiency caused by loss-of-function mutations in the gene encoding the WAS protein (WASp)1–3. More than 80% of WAS patients develop skin rash characterized as atopic eczema during infancy and childhood1–4. One possible reason for development of skin rash is the reduced function of WASp-deficient regulatory T cells that have poor suppressive activity in vitro and in vivo5–8. Surprisingly in the context of eczema, WASp-deficient mast cells have decreased capacity to degranulate upon triggering of the FceR1 (ref. 9). ARTICLE WASp-deficient Langerhans cells have decreased emigration from the epidermis upon activation with hapten10, and it has been suggested that skin pathology in WAS may be caused by local accumulation of dendritic cells (DCs) in the skin1,10–13. WASp belongs to the WASp family of proteins also including neuronal (N)-WASp and WASp-family verprolin-homologous protein (WAVE)/suppressor of the cyclic AMP receptor (SCAR) 1-3, and they together coordinate receptor signalling to changes in the actin cytoskeleton1–3. At rest, WASp and N-WASp resides in an auto-inhibited conformation. Upon binding of the small Rho GTPase Cdc42, the auto-inhibited conformation is released and exposes the carboxy-terminal verprolin-cofilin-acidic (VCA) domain that allows for recruitment of the Arp2/3 complex and actin polymerization14,15. The small Rho GTPases Rac1 and Rac2 regulate activation of the WAVE/Scar proteins to stimulate actin polymerization by the VCA domain16,17. p y y Langerhans cells and CD103 þ DCs in the skin, and CD8 þ DCs and to a lesser extent CD8  DCs in the draining lymph nodes (dLNs) and spleen possess a unique capacity for presenting exogenous antigen on major histocompatibility (MHC) class I in a process termed cross-presentation18–20. The superior capacity of specific DCs to cross-present antigens results from that DCs in contrast to other phagocytic cells can maintain a near neutral pH in phagocytic and endocytic vesicles20. Rac2 is a key component for cross-presentation of soluble antigens and localize to intracellular membranes in which Rac2 regulates the NADPH complex, thereby maintaining a near neutral pH in phagosomes and endosomes21. Moreover, CD8 DCs can take up antigen in the form of immune complex by Fc receptors and efficiently shuttle exogenous antigens efficiently into the cross-presentation pathway22–24. The role of WASp in cross-presentation has been investigated by direct targeting of antigen to the DEC205 receptor expressed on CD8þ DCs13. WASp KO CD8þ DCs induced shorter contact duration with wild-type CD8 þ T cells in vivo leading to decreased early activation of CD8þ T cells13. In the specific anti-viral response, WASp KO mice have decreased capacity to mount an antigen- specific CD8þ T cell response to lymphocytic choriomeningitis virus (LCMV) infection25 and influenza26,27. Here, we examined the response of WASp KO mice to skin challenge. Our findings show that WASp KO mice can respond to allergens and parasite infiltration in the skin. However, the immune response is skewed to DC-mediated activation of CD8 þ T cells that produce IFNg. ARTICLE Upon Der p 2 challenge, wild-type and WASp KO mice had increased number of CD45 þ hematopoietic cells in the skin when compared with unchallenged mice (Supplementary Fig. 1c). Der p 2 challenge induced increased number of CD11b þCD11c þ and EpCAM þ DCs in the WASp KO skin when compared with Der p 2-challenged wild-type skin (Fig. 1e). However, due to the fact that we assessed both lesional and non-lesional skin by flow cytometry, Der p 2 challenge did not induce a significant increase of CD4 þ and CD8 þ T cells in wild-type or WASp KO mice (Fig. 1e). WASp KO mice had fewer IFNg-producing CD4 þ T cells when compared with wild-type mice both before and after Der p 2 challenge (Supplementary Fig. 1c). Moreover, WASp KO mice had a tendency to increased number of IFNg- producing CD8 þ T cells and CD11c þEpCAM CD103 þ DCs (Supplementary Fig. 1c). h h d we compared shaved mice patched with Der p 2 to that of unshaved unchallenged mice. To examine skin pathology at day 50, a 4 mm2 punch biopsy was taken from macroscopically inflamed back skin. Epidermal thickening is a hallmark of atopic dermatitis and wild-type mice showed, after three patches with Der p 2, epidermal hyperplasia (Fig. 1a)29,30 with increased number of Ki67 þ proliferating epidermal keratinocytes (Supplementary Fig. 1a). Der p 2-challenged WASp KO mice showed less epidermal hyperplasia (Fig. 1a; Supplementary Fig. 1a). We prepared epidermal sheets and found that Langerhans cells in both wild-type mice and WASp KO mice were decreased in epidermis after Der p 2 challenge (Fig. 1b). To identify dermal DCs and Langerhans cells, we quantified the number of DC subsets in dermis including CD11c þEpCAM þ (epithelial cell adhesion molecule), Langerin þ and CD11c þEpCAM Langerin  DCs18,31,32. Wild-type mice had similar number of dermal DCs before and after Der p 2 challenge (Fig. 1c; Supplementary Fig. 1b), suggesting that the marked reduction of epidermal Langerhans cells after Der p 2 challenge in wild-type mice is caused by egress of Langerhans cells from the skin to the dLNs. Unchallenged WASp KO mice had decreased number of dermal Langerin þ DCs (Fig. 1c), while after Der p 2 challenge WASp KO mice had increased number of CD11cþEpCAMþ, Langerinþ and CD11cþEpCAMLangerin DCs in the dermis (Fig. 1c; Supplementary Fig. 1b). ARTICLE We provide evidence for that WASp KO CD8  DCs upregulate the molecular machinery to cross- present antigens and activate CD8 þ T cells. Our data suggests that downregulation of cross-presentation by WASp may be an active process that is essential to prevent over-activation of CD8 þ T cells. On treatment with the TLR7 agonist imiquimod, wild-type mice exhibited increased numbers of migratory MHCIIhighDEC205 þCD8  DCs and CCR7 þDEC205 þCD8  DCs in the dLNs, whereas WASp KO DCs failed to migrate to the dLNs (Fig. 1f). This implied that WASp KO skin DCs had decreased capacity to egress and therefore accumulated in the dermis. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 we compared shaved mice patched with Der p 2 to that of unshaved unchallenged mice. To examine skin pathology at day 50, a 4 mm2 punch biopsy was taken from macroscopically inflamed back skin. Epidermal thickening is a hallmark of atopic dermatitis and wild-type mice showed, after three patches with Der p 2, epidermal hyperplasia (Fig. 1a)29,30 with increased number of Ki67 þ proliferating epidermal keratinocytes (Supplementary Fig. 1a). Der p 2-challenged WASp KO mice showed less epidermal hyperplasia (Fig. 1a; Supplementary Fig. 1a). We prepared epidermal sheets and found that Langerhans cells in both wild-type mice and WASp KO mice were decreased in epidermis after Der p 2 challenge (Fig. 1b). To identify dermal DCs and Langerhans cells, we quantified the number of DC subsets in dermis including CD11c þEpCAM þ (epithelial cell adhesion molecule), Langerin þ and CD11c þEpCAM Langerin  DCs18,31,32. Wild-type mice had similar number of dermal DCs before and after Der p 2 challenge (Fig. 1c; Supplementary Fig. 1b), suggesting that the marked reduction of epidermal Langerhans cells after Der p 2 challenge in wild-type mice is caused by egress of Langerhans cells from the skin to the dLNs. Unchallenged WASp KO mice had decreased number of dermal Langerin þ DCs (Fig. 1c), while after Der p 2 challenge WASp KO mice had increased number of CD11cþEpCAMþ, Langerinþ and CD11cþEpCAMLangerin DCs in the dermis (Fig. 1c; Supplementary Fig. 1b). We detected similar number of Ki67 þ proliferating dermal DCs in wild-type and WASp KO mice (Supplementary Fig. 1a). Wild-type mice had similar numbers of CD4 þ and CD8 þ T cells in the skin before and after Der p 2 challenge (Fig. 1d). WASp KO mice had decreased number of CD4 þ T cells in unchallenged skin. Upon Der p 2 challenge, the number of CD4 þ T cells was increased in WASp KO mice and reached similar number to Der p 2-challenged wild-type mice (Fig. 1d). In contrast, unchallenged WASp KO mice had increased number of CD8 þ T cell in the skin compared with wild-type mice and the CD8 þ T-cell population was further increased in WASp KO mice after Der p 2 challenge (Fig. 1d). To corroborate the results from histological analysis, we performed flow cytometry analysis of the 1 cm2 back skin challenged with Der p 2 (containing both macroscopically inflamed and non-inflamed skin). Deletion of Wiskott–Aldrich syndrome protein triggers Rac2 activity and increased cross-presentation by dendritic cells Marisa A.P. Baptista1,2, Marton Keszei1, Mariana Oliveira1, Karen K.S. Sunahara1,3, John Andersson4, Carin I.M. Dahlberg1, Austen J. Worth5, Agne Liede´n6, I-Chun Kuo7,8, Robert P.A. Wallin1, Scott B. Snapper9, Liv Eidsmo10, Annika Scheynius4, Mikael C.I. Karlsson1, Gerben Bouma5, Siobhan O. Burns5,11,12, Mattias N.E. Forsell1,13, Adrian J. Thrasher5, Susanne Nyle´n1 & Lisa S. Westerberg1 Wiskott–Aldrich syndrome (WAS) is caused by loss-of-function mutations in the WASp gene. Decreased cellular responses in WASp-deficient cells have been interpreted to mean that WASp directly regulates these responses in WASp-sufficient cells. Here, we identify an exception to this concept and show that WASp-deficient dendritic cells have increased activation of Rac2 that support cross-presentation to CD8 þ T cells. Using two different skin pathology models, WASp-deficient mice show an accumulation of dendritic cells in the skin and increased expansion of IFNg-producing CD8 þ T cells in the draining lymph node and spleen. Specific deletion of WASp in dendritic cells leads to marked expansion of CD8 þ T cells at the expense of CD4 þ T cells. WASp-deficient dendritic cells induce increased cross-presentation to CD8 þ T cells by activating Rac2 that maintains a near neutral pH of phagosomes. Our data reveals an intricate balance between activation of WASp and Rac2 signalling pathways in dendritic cells. 1 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications URE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test LC Langerhans cells; WT wild type; WKO WASp KO a WT Derm Epi WKO Unchallenged a 60 Epidermis thickness 40 20 0 Thickness (μm) WT WT WT L i Derm Epi WKO WKO * WT WKO WKO Unchallenged Unchallenged Der p 2 Der p 2 1,500 Langerhans cells 1,000 500 0 LC mm–2 WT WKO *** ** WT WKO Unchallenged Der p 2 a b WT WKO Unchallenged 60 Epidermis thickness 40 20 0 Thickness (μm) WT WKO * WT WKO Unchallenged Der p 2 1,500 Langerhans cells 1,000 500 0 LC mm–2 WT WKO *** ** WT WKO Unchallenged Der p 2 Langerin b b 300 200 100 Cells mm–2 Cells mm–2 0 Langerin+ cells CD11c+EpCAM+ cells 1,000 800 600 400 200 0 * ** ** ** WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 Langerin CD11c EpCAM Derm Epi c Cells mm–2 Cells mm–2 CD8+ T cells CD4+ T cells 400 300 200 100 0 400 300 200 100 0 * ** ** ** ** WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 CD3 CD4 CD8 Derm Epi d d 200 150 100 50 0 Cell number (×103) 15 10 5 0 Cell number (×103) CD8+ T cells CD4+ T cells WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 30 150 100 50 0 20 10 0 Cell number Cell number CCR7+ DEC205+ CD8- DCs MHCllhi DEC205+ CD-DCs * * ** WT WKO WT WKO Non-treated Imiquimod WT WKO WT WKO Non-treated Imiquimod f 600 400 200 150 100 50 0 200 Cell number (×103) Cell number (×103) 0 EpCAM+ DCs CD11b+ CD11c+DCs ** * WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 e e Cell number (×103) WT WKO WT WKO Non-treated Imiquimod Non-treated Imiquimod Der p 2 induces skin pathology in WASp KO mice. (a) Whole-skin sections (10 mm) from 4 mm2 punch biopsies of back pathology in WASp KO mice. (a) Whole-skin sections (10 mm) from 4 mm2 punch biopsies of back skin from day 50 w stained with hematoxylin and eosin. Epidermal thickening is indicated in mm. (b) Langerhans cells in epidermis by histology. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (Fig. 2b), but Der p 2 challenge induced higher number of memory/effector CD8 þ T cells in the spleen of WASp KO mice when compared with wild-type mice (Fig. 2b). CD4 þ and CD8 þ T cells were similar in wild-type and WASp KO mice (Fig. 2a). Compared to LNs and spleen, WASp KO mice had a reversed CD4/CD8 ratio in blood, suggesting that WASp KO CD8 þ T cells preferentially accumulated in tissues (Supplementary Fig. 2a). Wild-type and WASp KO mice had similar number of memory/effector CD4 þ T cells before and after Der p 2 challenge as determined by CD44high/CD62L  cells Together, these results suggests that Der p 2-challenged WASp KO mice had an accumulation of DCs in the dermis and an altered systemic T-cell balance with increased number of effector/memory CD8 þ T cells. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 p g y y 60 Epidermis thickness 40 20 0 Thickness (μm) WT WT WT Langerin Langerin CD11c EpCAM CD3 CD4 CD8 Derm Epi Derm Epi Derm Epi 300 200 100 Cells mm–2 Cells mm–2 Cells mm–2 Cells mm–2 0 600 400 200 150 100 50 0 200 Cell number (×103) Cell number (×103) 200 150 100 50 0 Cell number (×103) 15 30 150 100 50 0 20 10 0 10 5 0 Cell number (×103) Cell number Cell number 0 WKO WKO * WT WKO WKO Unchallenged Unchallenged Der p 2 Der p 2 1,500 Langerhans cells Langerin+ cells CD11c+EpCAM+ cells CD8+ T cells CCR7+ DEC205+ CD8- DCs MHCllhi DEC205+ CD-DCs CD8+ T cells CD4+ T cells EpCAM+ DCs CD11b+ CD11c+DCs CD4+ T cells 1,000 1,000 800 600 400 200 400 300 200 100 0 400 300 200 100 0 0 500 0 LC mm–2 WT WKO * * * * * ** * ** ** ** ** ** ** ** ** ** ** WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Unchallenged Der p 2 WT WKO WT WKO Non-treated Imiquimod WT WKO WT WKO Non-treated Imiquimod a b c d e f Figure 1 | Der p 2 induces skin pathology in WASp KO mice. (a) Whole-skin sections (10 mm) from 4 mm2 punch biopsies of back skin from day 50 wer stained with hematoxylin and eosin. Epidermal thickening is indicated in mm. (b) Langerhans cells in epidermis by histology. Epidermal sheets labelled wit Langerin (CD207, green) for identification of Langerhans cells. The mean number of Langerhans cells per mm2 of epidermis at day 50 is indicated. (c,d) DCs and T cells in dermis by histology. (c) The mean number of total Langerin þ (including Langerhans and dermal DCs, green), CD11c þEpCAM þ (mature Langerhans cells, CD11c in red and EpCAM in blue) DCs and (d) CD4 þCD3 þ (red and green, respectively) and CD8b þ (blue) Tcells per mm2 indicated. Examples of counted cells are magnified in the white boxes. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (a–d) Bar represents mean value and each dot represents one mouse (a,b) or on picture (c,d). Results are a pool of two separate experiments. (a–d) WTunchallenged n ¼ 3–4; WKO unchallenged n ¼ 3–6; WT Der p 2 n ¼ 3–9; WKO De p 2 n ¼ 4–10. Scale bar, 50 mm. (e) DCs and T cells in skin by flow cytometry analysis. Absolute numbers of cells in the back skin at day 50 from unchallenged and Der p 2-challenged WTand WASp KO mice as measured by flow cytometry. WT unchallenged n ¼ 4; WKO unchallenged n ¼ 5; WT De p 2 n ¼ 6; WASP KO Der p 2 n ¼ 8. (f) DC egress from the skin. Imiquimod was applied on the ear and 48 h later, MHCIIhighDEC205 þCD8  DCs an CCR7 þDEC205 þCD8  DCs were analysed by flow cytometry in dLNs. WT unchallenged n ¼ 4; WKO unchallenged n ¼ 4; WT imiquimod n ¼ 4; WAS KO Imiquimod n ¼ 4. (a–f) Results are representative of two separate experiments. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications Results Der p 2 induces skin pathology in WASp KO mice. To induce an eczema-like phenotype, mice were shaved and treated by epicutaneous patching on the back skin with Der p 2, a major allergen from the house dust mite Dermatophagoides pteronyssinus28. Since the shaving in itself mimics mechanical injury inflicted by scratching of dry itchy skin in human eczema, When we analyzed the Der p 2 response in dLNs and spleen, individual WASp KO mice showed a consistent increase in CD8 þ T cells over CD4 þ T cells leading to a skewed CD4/CD8 T-cell ratio, although the mean total number of LN and spleen NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications 2 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (a,b) LN and spleen Tcells by flow cytometry. Absolute numbers of (a) total and (b) effector/memory (CD44hiCD62L ), CD4 þ and CD8 þ T cells from day 50 spleens and dLNs from unchallenged and Der p 2-challenged WT and WASp KO mice on Balb/c background as measured by flow cytometry. (c) In vitro stimulation of spleen cells. Total splenocytes from unchallenged or Der p 2-challenged mice at day 50 were either unstimulated or stimulated with PMA plus ionomycin for 4 h or Der p 2 for 48 h (c). Absolute numbers of total CD4 þIFNg þ and CD8 þIFNg þ Tcells after Der p 2 and PMA plus ionomycin stimulation as measured by flow cytometry. (a–c) Bar represents mean value and each dot represents one mouse. (a,b) Results are a pool of two separate experiments and (c) representative of two separate experiments. (a,b) WT unchallenged n ¼ 4–9; WKO unchallenged n ¼ 6–10; WT Der p 2 n ¼ 10–11; WKO Der p 2 n ¼ 8. (c) WTunchallenged n ¼ 3–7; WKO unchallenged n ¼ 5–8; WT Der p 2 n ¼ 5–6; WKO Der p 2 n ¼ 6. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, wild type; WKO, WASp KO. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Epidermal sheets labelled with Langerin (CD207, green) for identification of Langerhans cells. The mean number of Langerhans cells per mm2 of epidermis at day 50 is indicated. (c,d) DCs and T cells in dermis by histology. (c) The mean number of total Langerin þ (including Langerhans and dermal DCs, green), CD11c þEpCAM þ (mature Langerhans cells, CD11c in red and EpCAM in blue) DCs and (d) CD4 þCD3 þ (red and green, respectively) and CD8b þ (blue) Tcells per mm2 is indicated. Examples of counted cells are magnified in the white boxes. (a–d) Bar represents mean value and each dot represents one mouse (a,b) or one picture (c,d). Results are a pool of two separate experiments. (a–d) WTunchallenged n ¼ 3–4; WKO unchallenged n ¼ 3–6; WT Der p 2 n ¼ 3–9; WKO Der p 2 n ¼ 4–10. Scale bar, 50 mm. (e) DCs and T cells in skin by flow cytometry analysis. Absolute numbers of cells in the back skin at day 50 from unchallenged and Der p 2-challenged WTand WASp KO mice as measured by flow cytometry. WT unchallenged n ¼ 4; WKO unchallenged n ¼ 5; WT Der p 2 n ¼ 6; WASP KO Der p 2 n ¼ 8. (f) DC egress from the skin. Imiquimod was applied on the ear and 48 h later, MHCIIhighDEC205 þCD8  DCs and CCR7 þDEC205 þCD8  DCs were analysed by flow cytometry in dLNs. WT unchallenged n ¼ 4; WKO unchallenged n ¼ 4; WT imiquimod n ¼ 4; WASP KO Imiquimod n ¼ 4. (a–f) Results are representative of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. LC, Langerhans cells; WT, wild type; WKO, WASp KO. 3 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Der p 2 induces expansion of WASp KO CD8 þIFNc þ T cells. To examine the cytokine profile in WASp KO mice, we measured cytokines in serum and skin. Similar quantities of TNFa, IL-4, there was no detectable IFNg production in freshly isolated cells from either wild-type or WASP KO mice (Fig. 2c,d). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 However, after 4 h stimulation with phorbol 12-myristate CD8 CD4 WT Der p 2 Effector/memory CD8+ T cells CD44 CD62L Effector/memory CD4+ T cells 11,6% 10,2% Effector/memory CD8+ T cells WT WKO WT WKO 0.0 0.2 0.4 0.6 0.8 1.0 ** Der p 2 Unchallenged Cell number (×106) Effector/memory CD4+ T cells WT WKO WT WKO 0 2 4 6 8 Der p 2 Unchallenged Cell number (×106) CD4+ T cells WT WKO WT WKO 0 5 10 15 20 Der p 2 Unchallenged Cell number (×106) CD8+ T cells WT WKO WT WKO 0 2 4 6 8 Der p 2 Unchallenged Cell number (×106) CD4+/CD8+ T cell ratio WT WKO WT WKO 0 2 4 6 Der p 2 Unchallenged ** * CD4/CD8 62.6% 29.6% 53.4% 38.3% WKO Der p 2 CD4+ T cells WT WKO WT WKO 0 1 2 3 4 5 Cell number (×106) Der p 2 Unchallenged CD8+ T cells WT WKO WT WKO 0.0 0.5 1.0 1.5 2.0 Cell number (×106) Der p 2 Unchallenged CD4+/CD8+ T cell ratio WT WKO WT WKO 0 2 4 6 Der p 2 unchallenged ** CD4/CD8 LN Spleen Effector/memory CD4+ T cells WT WKO WT WKO 0 5 10 15 Cell number (×104) Der p 2 Unchallenged Effector/memory CD8+ T cells WT WKO WT WKO 0 2 4 6 8 Cell number (×104) Der p 2 Unchallenged LN Spleen IFNγ+ CD4+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD4+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD8+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD4+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD8+ T cells * WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged ** * IFNγ+ CD8+ T cells Cell number (×103) No treatment 4 h PMA+ionomycin 48 h Der p 2 IFNγ+ CD4+ T cells 4 h PMA+ionomycin WT WKO WT WKO 0 20 40 60 80 Der p 2 Unchallenged % of cells WT WKO WT WKO 0 20 40 60 80 Der p 2 Unchallenged % of cells * IFNγ+ CD8+ T cells 48 h Der p 2 No treatment IFNγ+ CD4+ T cells IFNγ+ CD8+ T cells WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged ** IFNγ+ CD4+ T cells WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged * * IFNγ+ CD8+ T cells WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged a b c d Figure 2 | Der p 2 induces expansion of WASp KO CD8 þIFNc þ Tcells. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (a,b) LN and spleen Tcells by flow cytometry. Absolute numbers of (a) total and hi þ þ Figure 2 | Der p 2 induces expansion of WASp KO CD8 þIFNc þ Tcells. (a,b) LN and spleen Tcells by flow cytometry. Absolute numbers of (a) total and (b) effector/memory (CD44hiCD62L ), CD4 þ and CD8 þ T cells from day 50 spleens and dLNs from unchallenged and Der p 2-challenged WT and WASp KO mice on Balb/c background as measured by flow cytometry. (c) In vitro stimulation of spleen cells. Total splenocytes from unchallenged or Der p Figure 2 | Der p 2 induces expansion of WASp KO CD8 þIFNc þ Tcells. (a,b) LN and spleen Tcells by flow cytometry on of WASp KO CD8 þIFNc þ Tcells. (a,b) LN and spleen Tcells by flow cytometry. Absolute numbers of (a) total and g | p p p c ( , ) p y y y ( ) (b) effector/memory (CD44hiCD62L ), CD4 þ and CD8 þ T cells from day 50 spleens and dLNs from unchallenged and Der p 2-challenged WT and WASp KO mice on Balb/c background as measured by flow cytometry. (c) In vitro stimulation of spleen cells. Total splenocytes from unchallenged or Der p 2-challenged mice at day 50 were either unstimulated or stimulated with PMA plus ionomycin for 4 h or Der p 2 for 48 h (c). Absolute numbers of total CD4 þIFNg þ and CD8 þIFNg þ Tcells after Der p 2 and PMA plus ionomycin stimulation as measured by flow cytometry. (a–c) Bar represents mean value and each dot represents one mouse. (a,b) Results are a pool of two separate experiments and (c) representative of two separate experiments. (a,b) WT unchallenged n ¼ 4–9; WKO unchallenged n ¼ 6–10; WT Der p 2 n ¼ 10–11; WKO Der p 2 n ¼ 8. (c) WTunchallenged n ¼ 3–7; WKO unchallenged n ¼ 5–8; WT Der p 2 n ¼ 5–6; WKO Der p 2 n ¼ 6. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, wild type; WKO, WASp KO. ( ) / y ( ), y p g WASp KO mice on Balb/c background as measured by flow cytometry. (c) In vitro stimulation of spleen cells. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 CD8 CD4 WT Der p 2 CD4+ T cells WT WKO WT WKO 0 5 10 15 20 Der p 2 Unchallenged Cell number (×106) CD8+ T cells WT WKO WT WKO 0 2 4 6 8 Der p 2 Unchallenged Cell number (×106) CD4+/CD8+ T cell ratio WT WKO WT WKO 0 2 4 6 Der p 2 Unchallenged ** * CD4/CD8 62.6% 29.6% 53.4% 38.3% WKO Der p 2 CD4+ T cells WT WKO WT WKO 0 1 2 3 4 5 Cell number (×106) Der p 2 Unchallenged CD8+ T cells WT WKO WT WKO 0.0 0.5 1.0 1.5 2.0 Cell number (×106) Der p 2 Unchallenged CD4+/CD8+ T cell ratio WT WKO WT WKO 0 2 4 6 Der p 2 unchallenged ** CD4/CD8 LN Spleen a CD8 CD4 WT Der p 2 62.6% 29.6% 53.4% 38.3% WKO Der p 2 a a CD4+ T cells WT WKO WT WKO 0 5 10 15 20 Der p 2 Unchallenged Cell number (×106) CD8+ T cells WT WKO WT WKO 0 2 4 6 8 Der p 2 Unchallenged Cell number (×106) CD4+/CD8+ T cell ratio WT WKO WT WKO 0 2 4 6 Der p 2 Unchallenged ** * CD4/CD8 3% CD4+ T cells WT WKO WT WKO 0 1 2 3 4 5 Cell number (×106) Der p 2 Unchallenged CD8+ T cells WT WKO WT WKO 0.0 0.5 1.0 1.5 2.0 Cell number (×106) Der p 2 Unchallenged CD4+/CD8+ T cell ratio WT WKO WT WKO 0 2 4 6 Der p 2 unchallenged ** CD4/CD8 LN Spleen Effector/memory CD8+ T cells CD44 CD62L Effector/memory CD4+ T cells 11,6% 10,2% b b Effector/memory CD4+ T cells WT WKO WT WKO 0 2 4 6 8 Der p 2 Unchallenged Cell number (×106) Effector/memory CD4+ T cells WT WKO WT WKO 0 5 10 15 Cell number (×104) Der p 2 Unchallenged LN Spleen Effector/memory CD8+ T cells WT WKO WT WKO 0.0 0.2 0.4 0.6 0.8 1.0 ** Der p 2 Unchallenged Cell number (×106) Effector/memory CD8+ T cells WT WKO WT WKO 0 2 4 6 8 Cell number (×104) Der p 2 Unchallenged IFNγ+ CD4+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD4+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD8+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD4+ T cells WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged Cell number (×103) IFNγ+ CD8+ T cells * WT WKO WT WKO 0 50 100 150 200 Der p 2 Unchallenged ** * IFNγ+ CD8+ T cells Cell number (×103) No treatment 4 h PMA+ionomycin 48 h Der p 2 c IFNγ+ CD4+ T cells 4 h PMA+ionomycin WT WKO WT WKO 0 20 40 60 80 Der p 2 Unchallenged % of cells WT WKO WT WKO 0 20 40 60 80 Der p 2 Unchallenged % of cells * IFNγ+ CD8+ T cells 48 h No treatment IFNγ+ CD4+ T cells IFNγ+ CD8+ T cells WT WK 0 20 40 60 80 % of cells Unchallenge IFNγ+ C WT WK 0 20 40 60 80 % of cells Unchalleng * IFNγ+ C WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged d d c 48 h Der p 2 WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged ** IFNγ+ CD4+ T cells WT WKO WT WKO 0 20 40 60 80 % of cells Der p 2 Unchallenged * * IFNγ+ CD8+ T cells WT WKO WT WKO 0 Der p 2 Unchallenged WT WKO WT WKO Der p 2 Unchallenged Figure 2 | Der p 2 induces expansion of WASp KO CD8 þIFNc þ Tcells. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 major) infect dermal macrophages and induce a massive Th1 response char- acterized by CD4 þ T cells producing IFNg33,34. When compared with wild-type mice, WASp KO mice had a delayed response to L. major infection at 2 weeks post infection as evidenced by smaller lesion size (Fig. 3a; Supplementary Fig. 3a) and decreased Increased WASp KO CD8 þIFNg þ T cells upon L. major infection. We next investigated how WASp KO mice would respond to dermal infection. Leishmania major (L. major) infect dermal macrophages and induce a massive Th1 response char- acterized by CD4 þ T cells producing IFNg33,34. When compared with wild-type mice, WASp KO mice had a delayed response to L. major infection at 2 weeks post infection as evidenced by smaller lesion size (Fig. 3a; Supplementary Fig. 3a) and decreased p y g In the T-cell compartment of dLNs, WASp KO mice had significantly lower number of CD4 þ T cells both at 2 and 6 weeks post infection when compared with wild-type mice WT WKO WT WKO 1,500 50 10 8 6 4 2 200 150 100 50 0 * ** 0 40 30 20 10 0 1,000 500 0 WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO Control Leishmania 2 weeks dLN MHCIIhi DCs Ears MHCIIhi DCs dLN CD4+ T cells dLN CD8+ T cells dLN IFNγ+CD4+T cells dLN IFNγ+CD8+T cells ** ** ** dLN CD4 /CD8 T cell ratio * ** ** ** ** 5 4 2.0 4 3 2 1 150 50 40 30 20 10 0 15 10 5 0 100 50 0 0 1.5 1.0 0.5 0.0 3 2 1 0 ** Ears CD4+ T cells Ears CD8+ T cells dLN CD103+ DCs Negative control (FMO) 0.0% MHCII CD11c 0.6% CD103 CD11c 1.6% 0.8% 0.63% 1.8% 800 600 400 200 0 WT L. major WKO L. major Negative control (FMO) WT L. major WKO L. major WT L. major WKO L. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 major WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number (×103) dLN CD4+ T cells dLN CD8+ T cells d * ** 5 4 2.0 4 3 2 1 0 1.5 1.0 0.5 0.0 3 2 1 0 2 weeks 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT Co Cell number (×106) Cell number (×106) CD4/CD8 ratio d d dLN CD4 /CD8 T cell ratio ** ** ** 4 3 2 1 0 WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks CD4/CD8 ratio dLN IFNγ+CD4+T cells ** 150 15 10 5 0 100 50 0 WT L. major WKO L. major Isotype control 0.9% CD8 IFNγ 8.8% 14.4% WT WKO WT WKO WT WT WKO Control Leishmania 2 weeks Con Leishmania 6 weeks Cell number (×103) Cell number (×103) e e dLN IFNγ+CD4+T cells ** 150 100 50 0 WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number (×103) dLN IFNγ+CD8+T cells ** ** 50 40 30 20 10 0 15 10 5 0 WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number (×103) Figure 3 | L. major induces increased number of WASp KO CD8 þIFNc þ T cells. (a) Ear infiltration of cells. (a) Ears from WT and WASp KO control or L. major infected mice on Balb/c background after 6 weeks. (b) Absolute numbers in ear of total MHCIIhiCD11c þ DCs; total CD4 þCD3 þ and CD8 þCD3 þ T cells, measured by flow cytometry. (c–e) dLN infiltration of cells. Absolute numbers in dLN of total MHCIIhi DCs; total CD4 þCD3 þ and CD8 þCD3 þ T cells; CD4 þ/CD8 þ T-cell ratio; IFNg þCD4 þCD3 þ and IFNg þCD8 þCD3 þ cells, measured by flow cytometry. (a–e) Bar represents mean value and each dot represents one ear or dLNs. Results from week 2 and week 6 are representative of two separate experiments. WT control n ¼ 3–4; WKO control n ¼ 4; WT L. major 2 weeks n ¼ 6; WASp KO L. major 2 weeks n ¼ 6; WT L. major 6 weeks n ¼ 10; WASp KO L. major 6 weeks n ¼ 7. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Total sple 2-challenged mice at day 50 were either unstimulated or stimulated with PMA plus ionomycin for 4 h or Der p 2 for CD4 þIFNg þ and CD8 þIFNg þ Tcells after Der p 2 and PMA plus ionomycin stimulation as measured by flow cytomet and each dot represents one mouse. (a,b) Results are a pool of two separate experiments and (c) representative of tw unchallenged n ¼ 4–9; WKO unchallenged n ¼ 6–10; WT Der p 2 n ¼ 10–11; WKO Der p 2 n ¼ 8. (c) WTunchallenged n WT Der p 2 n ¼ 5–6; WKO Der p 2 n ¼ 6. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, w Der p 2 induces expansion of WASp KO CD8 þIFNc þ T cells. To examine the cytokine profile in WASp KO mice, we measured cytokines in serum and skin. Similar quantities of TNFa, IL-4, IL-5, IL-6, IL-10, IL-13, IFNg and TGFb in wild-type and WASp KO mice were detected both before and after Der p 2 challenge (Supplementary Fig. 2b). When we examined the CD8 þ T cells, there was no detectable IFNg production in freshly isolated cells from either wild-type or WASP KO mice (Fig. 2c,d). However, after 4 h stimulation with phorbol 12-myristate 13-acetate (PMA) and ionomycin, Der p 2-challenged WASp KO mice exhibited a high proportion of IFNg-producing CD8 þ T cells (Fig. 2c,d). In splenocytes stimulated for 48 h with Der p 2 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications 4 4 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 CD4 þ T-cell infiltration (Fig. 3b). At 6 weeks post L. major infection, both wild-type and WASp KO mice had large lesions (Fig. 3a; Supplementary Fig. 3a) with considerable infiltration of MHC class IIhi DCs, CD4 þ and CD8 þ T cells and macrophages (Fig. 3b; Supplementary Fig. 3b,c). At 6 weeks, dLNs in wild-type mice had increased number of MHC class IIhigh DCs, which had likely emigrated from the infected skin (Fig. 3c). Moreover, wild-type mice had increased numbers of CD103 þ, CD8a þ and CD8a  DCs capable of cross-presenting exogenous antigen and activate CD8 þ T cells (Fig. 3c; Supplementary Fig. 3d). In contrast, WASp KO mice showed no increased numbers of MHC class IIhigh DCs or CD103 þ, CD8a þ and CD8a  DCs in the dLNs upon infection (Fig. 3c; Supplementary Fig. 3d). Together with increased accumulation of DCs in the dermis of WASp KO mice after Der p 2 challenge, this suggests that WASp KO DCs have decreased capacity to egress from dermis. to examine the allergen-specific response, both unchallenged and Der p 2-challenged WASp KO mice had increased number of IFNg-producing CD8 þ T cells (Fig. 2c,d). WASp KO T cells from unchallenged mice, not previously stimulated with Der p 2, also showed expansion of CD8 þIFNg þ T cells upon Der p 2 treatment in vitro. Since few naive T cells will contain the Der p 2 specificity, this suggests that naive WASp KO CD8 þ T cells, but not CD4 þ T cells, were prone to produce IFNg irrespective of antigen specificity. to examine the allergen-specific response, both unchallenged and Der p 2-challenged WASp KO mice had increased number of IFNg-producing CD8 þ T cells (Fig. 2c,d). WASp KO T cells from unchallenged mice, not previously stimulated with Der p 2, also showed expansion of CD8 þIFNg þ T cells upon Der p 2 treatment in vitro. Since few naive T cells will contain the Der p 2 specificity, this suggests that naive WASp KO CD8 þ T cells, but not CD4 þ T cells, were prone to produce IFNg irrespective of antigen specificity. Increased WASp KO CD8 þIFNg þ T cells upon L. major infection. We next investigated how WASp KO mice would respond to dermal infection. Leishmania major (L. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 major * * ** Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Isotype control 0.9% CD8 IFNγ 8.8% 14.4% Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO Control Leishmania 2 weeks Control Leishmania 2 weeks Leishmania 6 weeks Leishmania 6 weeks Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number Cell number (×103) Cell number (×103) Cell number (×106) Cell number (×106) Cell number (×103) Cell number (×103) CD4/CD8 ratio Cell number (×103) Cell number (×103) Control L. major a b c d e Figure 3 | L. major induces increased number of WASp KO CD8 þIFNc þ T cells. (a) Ear infiltration of cells. (a) Ears from WT and WASp KO control or L. major infected mice on Balb/c background after 6 weeks. (b) Absolute numbers in ear of total MHCIIhiCD11c þ DCs; total CD4 þCD3 þ and CD8 þCD3 þ T cells, measured by flow cytometry. (c–e) dLN infiltration of cells. Absolute numbers in dLN of total MHCIIhi DCs; total CD4 þCD3 þ and CD8 þCD3 þ T cells; CD4 þ/CD8 þ T-cell ratio; IFNg þCD4 þCD3 þ and IFNg þCD8 þCD3 þ cells, measured by flow cytometry. (a–e) Bar represents mean value and each dot represents one ear or dLNs. Results from week 2 and week 6 are representative of two separate experiments. WT control n ¼ 3–4; WKO control n ¼ 4; WT L. major 2 weeks n ¼ 6; WASp KO L. major 2 weeks n ¼ 6; WT L. major 6 weeks n ¼ 10; WASp KO L. major 6 weeks n ¼ 7. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. FMO, fluorescence minus one (negative control for MHC class II and CD103); WT, wild type; WKO, WASp KO. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 major Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell numbe Cell number (× Cell number (×103) Cell number (× Cell number (×103) c 200 150 100 50 0 * ** W W W W W W Control Leishmania 2 weeks dLN MHCIIhi DCs ** dLN CD103+ DCs Negative control (FMO) 0.0% MHCII CD11c 0.6% CD103 CD11c 1.6% 0.8% 0.63% 1.8% 800 600 400 200 0 WT L. major WKO L. major Negative control (FMO) WT L. major WKO L. major Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Control Leishmania 2 weeks Leishmania 6 weeks W W W Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number (×103) Cell number (×103) c 2 weeks dLN MHCIIhi DCs ** Negative control (FMO) 0.0% MHCII CD11c 0.6% CD103 CD11c 1.6% 0.63% 1.8% 800 600 400 200 0 WT L. major WKO L. major Negative control (FMO) WT 6 weeks WT WKO WT WKO WT WKO Control Leishmania Leishmania Cell number (×103) c c 200 150 100 50 0 * dLN MHCIIhi DCs ** dLN CD103+ DCs 0.6% CD103 CD11c 1.6% 0.8% % 1.8% 800 600 400 200 0 or WKO L. major Negative control (FMO) WT L. major WKO L. major WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT Control Leishmania 2 weeks Leis 6 Cell number (×103) Cell number (×103) 200 150 100 50 0 * ** dLN CD103+ DCs 0.6% CD103 CD11c 1.6% 0.8% Negative control (FMO) WT L. major WKO L. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications 5 1,500 50 10 8 6 4 2 0 40 30 20 10 0 1,000 500 0 WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO Control Leishmania 2 weeks Ears MHCIIhi DCs Ears CD4+ T cells Ears CD8+ T cells * * ** Leishmania 6 weeks Control Leishmania 2 weeks Leishmania 6 weeks WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number Cell number (×103) Cell number (×103) b WT WKO WT WKO Control L. major a 1,500 50 40 30 20 10 0 1,000 500 0 WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO Control Leishmania 2 weeks Ears MHCIIhi DCs Ears CD4+ T cells * * Leishmania 6 weeks Control Leishmania 2 weeks Leishmania 6 weeks Cell number Cell number (×103) b 1,500 1,000 500 0 WT WKO WT WKO WT WKO Control Leishmania 2 weeks Ears MHCIIhi DCs Leishmania 6 weeks Cell number b b WT WKO Control a a 10 8 6 4 2 0 Ears CD8+ T cells ** WT WKO WT WKO WT WKO Control Leishmania 2 weeks Leishmania 6 weeks Cell number (×103) 50 40 30 20 10 0 WT WKO WT WKO WT WKO Ears CD4+ T cells * * Control Leishmania 2 weeks Leishmania 6 weeks Cell number (×103) 6 4 2 200 150 100 50 0 * ** 0 30 20 10 0 1,000 500 0 WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO Control Leishmania 2 weeks dLN MHCIIhi DCs ** dLN CD103+ DCs Negative control (FMO) 0.0% MHCII CD11c 0.6% CD103 CD11c 1.6% 0.8% 0.63% 1.8% 800 600 400 200 0 WT L. major WKO L. major Negative control (FMO) WT L. major WKO L. ARTICLE (Fig. 3d). While the total number of CD8 þ T cells was similar in wild-type and WASp KO dLNs upon L. major infection, WASp KO mice showed a consistent failure to accumulate CD4 þ T cells in dLNs leading to a skewed CD4/CD8 T-cell ratio irrespective of L. major infection (Fig. 3d). We detected similar number of IFNg-producing CD4 þ and CD8 þ T cells in the dLNs of wild-type mice before and after L. major infection (Fig. 3e). In contrast, WASp KO mice had increased number of IFNg-producing CD4 þ and CD8 þ T cells in the dLNs (Fig. 3e). Together, this data suggests that WASp KO mice, despite having less antigen-presenting DCs in dLNs, can activate IFNg-producing CD4 þ and CD8 þ T cells upon L. major infection. when compared with wild-type CD8  DCs as determined by dilution of CFSE (Fig. 5c) and 3H-thymidine incorporation (Supplementary Fig. 5). CD8  DCs, and to a lesser extent CD8 þ DCs, can efficiently take up antigen in the form of immune complex by Fc receptors and shuttle the antigen into the cross-presentation pathway23,24. When compared with wild-type CD8  DCs, WASp KO CD8  DCs had reduced capacity to cross-present antigen from immune complexes leading to decreased proliferation of OT-I CD8 þ T cells (Supplementary Fig. 7a–c). We next examined antigen processing and presentation on MHC class II using soluble ovalbumin. Wild-type and WASp KO CD8 þ had similar capacity to stimulate proliferation of ovalbumin-specific OT-II CD4 þ T cells. In contrast, WASp KO CD8  DCs induced less proliferation when compared with wild-type CD8  DCs (Supplementary Fig. 8). This suggests that WASp KO CD8  DCs favoured presentation of soluble ovalbumin on MHC class I molecules. DC-specific WASp deletion induces increased CD8 þ T cells. To determine if WASp KO DCs induce expansion of wild-type CD8 þ T cells, we took advantage of mice harbouring a conditionally targeted loxP-flanked WASp allele35 bred to mice expressing Cre recombinase under the DC-expressed CD11c promoter (CD11cCre/wt mice), referred to as DC/cWKO mice (Fig. 4a). In DC/cWKO mice, WASp expression was efficiently deleted in CD11c þ DCs and a large fraction of CD8 þ and CD8  DCs had decreased to absent expression of WASp (Fig. 4b,c; images in Fig. 4b have been cropped for presentation. Full size images are presented in Supplementary Fig. 4). ARTICLE To evaluate presentation of ovalbumin peptides on MHC class I molecules, DCs were incubated with soluble ovalbumin and labelled with an antibody specific for the ovalbumin peptide SIINFEKL presented by MHC class I molecule H-2Kb. Wild-type and WASp KO CD8 þ DCs had similar expression of SIINFEKL–H-2Kb on the surface (Fig. 5g). When compared with wild-type CD8  DCs, WASp KO CD8  DCs presented more SIINFEKL in the context of H-2Kb (Fig. 5g). Increased cross-presentation by WASp KO DCs. The increased expansion of CD8 þ T cells in WASp KO and DC/cWKO mice led us to investigate if WASp KO DCs would have increased capacity to cross-present exogenous antigen and activate CD8 þ T cells. To address this possibility, wild-type CD8 þ T cells from OT-I TCR transgenic mice were used in which all CD8 þ T cells recognize the SIINFEKL peptide from ovalbumin presented by H-2Kb MHC class I molecules. Wild-type and WASp KO CD8 þ DCs and CD8  DCs, loaded with soluble ovalbumin, had similar capacity to form immune synapses with OT-I CD8 þ T cells (Fig. 5a). When comparing direct presentation of the SIINFEKL peptide, we detected similar proliferation of OT-I CD8 þ T cells when stimulated with wild-type or WASp KO CD8 þ or CD8  DCs (Fig. 5b). These findings suggest that WASp KO DCs could form immune synapses with CD8 þ T cells to facilitate their activation. To examine the cross-presenting capacity, we com- pared CD8 þ DCs with high capacity to cross-present soluble antigens and CD8  DCs with less capacity to cross-present soluble antigens21. Wild-type and WASp KO CD8 þ DCs had similar capacity to induce OT-I CD8 þ T-cell proliferation in response to soluble ovalbumin (Fig. 5c). In contrast, WASp KO CD8  DCs induced increased OT-I CD8 þ T-cell proliferation To precisely determine the cross-presentation and acidification capacity of DCs, we used latex beads that were coated with ovalbumin (Fig. 6a). To circumvent differences in uptake of beads, DCs that had taken up one ovalbumin-coated bead (Fig. 6a) were sorted by flow cytometry and examined for their capacity to induce cross-presentation. Wild-type and WASp KO CD8 þ DCs induced high proliferation of OT-I CD8 þ T cells when compared with wild-type CD8  DCs (Fig. 6b). ARTICLE CD11c-driven deletion of WASp led to the absence of WASp in 10% of CD4 þ T cells and 30% of CD8 þ T cells (Fig. 4b,c). When comparing maturation and activation phenotype to wild-type CD8 þ DCs, DC/cWKO CD8 þ DCs had lower expression of MHC class II and CD86, while expression of MHC class I and II, and CD86 was similar in wild-type and DC/cWKO CD8  DCs (Supplementary Fig. 5). When compared with wild-type mice, DC/cWKO mice had increased numbers of CD8 þ T cells and decreased number of CD4 þ T cells both in the LNs and spleen, creating a skewed CD4/CD8 T-cell ratio (Fig. 4d,e). Moreover, DC/cWKO mice had increased number of effector/memory CD8 þ T cells both in the LNs and spleen when compared with wild-type mice while effector/memory CD4 þ T cells were similar in wild-type and DC/cWKO mice (Fig. 4f). This data indicates that reduced expression of WASp in CD11c þ DCs was associated with expansion and activation of wild-type CD8 þ T cells. We next tried to identify the cause for the increased capacity of WASp KO CD8  DCs to cross-present soluble ovalbumin and stimulate proliferation of CD8 þ T cells. Wild-type and WASp KO DCs had similar maturation phenotype as assessed by upregulation of CD86 and MHC class I molecules (Supplementary Fig. 5); normal uptake of soluble ovalbumin (Fig. 5d); and normal degradation of ovalbumin as determined by usage of DQ-OVA that start to emit fluorescence when ovalbumin is processed in the cell (Fig. 5e). For cross- presentation to occur, DCs need to maintain a near neutral pH of the phagosome that allows for escape of proteins to the cytosol and loading on MHC class I molecules20. To examine acidification, we used pH rodo-ovalbumin that emits fluorescence at low pH (pH 3–5), indicating when ovalbumin reaches lysosomes. Wild-type CD8 þ DCs had a lower percentage of pHrodohigh cells when compared with wild-type CD8  DCs (Fig. 5f), showing the low capacity of CD8 þ DCs to acidify endosomes. WASp KO CD8 þ DCs showed similar low percentage of pHrodohigh cells as wild-type CD8 þ DCs (Fig. 5f). In contrast to wild-type CD8  DCs with increased percentage of pHrodohigh cells, WASp KO CD8  DCs had decreased percentage of pHrodohigh cells (Fig. 5f). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. FMO, fluorescence minus one (negative control for MHC class II and CD103); WT, wild type; WKO, WASp KO. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications 5 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Mice containing the WAS allele flanked by loxP sites were bred with CD11c-Cre mice to generate DC/cWKO mice on C57Bl/6 background. (b,c) WASp expression as determined by (b) western blotting and (c) flow cytometry of CD11c þCD8 þ, CD11c þCD8 , CD4 þCD3 þ and CD8 þCD3 þ cells from spleen. (d,e) Flow cytometry analysis of total CD4 þCD3 þ and CD8 þCD3 þ T cells in the (d) LNs and (e) spleen and (f) total effector/memory (CD44hiCD62L ) CD4 þCD3 þ and CD8 þCD3 þ T cells in LNs and spleen. (c) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 1; DC/cWKO n ¼ 3. (d,f) WT n ¼ 6–7; DC/cWKO n ¼ 6. The data is representative of (b) one, (c) two, (d,e) four and (f) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. DC/cWKO, WASpfl/flCD11cCre/wt; fl, floxed (LoxP flanked); WT, wild type; WKO, WASp KO. Figure 4 | DC-specific WASp deletion induces increased CD8 þ T cells. (a) Schematic representation of the targeting strategy. Mice containing the WAS allele flanked by loxP sites were bred with CD11c-Cre mice to generate DC/cWKO mice on C57Bl/6 background. (b,c) WASp expression as determined by (b) western blotting and (c) flow cytometry of CD11c þCD8 þ, CD11c þCD8 , CD4 þCD3 þ and CD8 þCD3 þ cells from spleen. (d,e) Flow cytometry analysis of total CD4 þCD3 þ and CD8 þCD3 þ T cells in the (d) LNs and (e) spleen and (f) total effector/memory (CD44hiCD62L ) CD4 þCD3 þ and CD8 þCD3 þ T cells in LNs and spleen. (c) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 1; DC/cWKO n ¼ 3. (d,f) WT n ¼ 6–7; DC/cWKO n ¼ 6. The data is representative of (b) one, (c) two, (d,e) four and (f) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. DC/cWKO, WASpfl/flCD11cCre/wt; fl, floxed (LoxP flanked); WT, wild type; WKO, WASp KO. CD8 þ DCs (Fig. 6c). For CD8  DCs, wild-type and WASp KO DCs showed similar phagosomal pH at 30 min and 2 h (Fig. 6c). To examine the later acidification events, we used beads coated with pH rodo-ovalbumin that measures phagolysosomal pH 3–5. When compared with wild-type CD8 þ DCs, WASp KO CD8 þ DCs showed decreased percentage of pHrodohigh cells after 6 h (Fig. 6d). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 CD4 CD8 WT DCc/WKO 26.3% 63.8% 22.4% 57.8% WT DC cWKO 0 2 4 6 8 ** WT DC cWKO 0 2 4 6 8 10 ** WT DC cWKO 0 1 2 3 ** WT DC cWKO 0.0 0.5 1.0 1.5 2.0 * WT DC cWKO 0.0 0.5 1.0 1.5 2.0 ** WT DC cWKO 0.0 0.5 1.0 1.5 2.0 ** CD4 CD8 41.6% 83.7% 11.2% 54.9% WASP GAPDH WT WKO WASPfl/fl WASP GAPDH Cre 1 2 3456 7 8 9 10 11 12 5′ 3′ X WT DC cWKO 0 20 40 60 80 100 WT DC cWKO 0 20 40 ** WT DCc/WKO WT DC cWKO 0 2 4 6 8 10 WT DC cWKO 0 2 4 6 8 * Spleen LN WASp WT CD8+ T cells CD4+ T cells WASp WT WKO DC cWKO 0 2000 4000 6000 8000 ** WT WKO DC cWKO 0 2000 4000 6000 ** 0 50 100 150 % of WASp positive cells WT WKO DC cWKO WT WKO DC cWKO 0 50 100 150 % of WASp positive cells Was locus LoxP LoxP CD11c promotor CD8+ cells CD11c+ cells WASPfl/fl xCD11Cre+ CD8+DC CD8+DC CD8–DC CD8–DC DCcWKO WKO WT DCcWKO WKO WASp MFI WASp MFI CD8+ T cells CD4+ T cells Cell number (×106) Cell number (×106) Cell number (×106) Cell number (×106) LN CD4+ T cells LN CD8+ T cells CD4/CD8 LN CD4+/CD8+ T cell ratio Spleen CD4+T cells Spleen CD8+T cells CD4/CD8 Spleen CD4+/CD8+ T cell ratio Effector/memory CD4+ T cells Cell number (×104) Cell number (×104) Cell number (×104) Cell number (×104) Effector/memory CD8+ T cells Effector/memory CD4+ T cells Effector/memory CD8+ T cells a b c d e f Figure 4 | DC-specific WASp deletion induces increased CD8 þ T cells. (a) Schematic representation of the targeting strategy. Mice containing the WAS allele flanked by loxP sites were bred with CD11c-Cre mice to generate DC/cWKO mice on C57Bl/6 background. (b,c) WASp expression as determined by (b) western blotting and (c) flow cytometry of CD11c þCD8 þ, CD11c þCD8 , CD4 þCD3 þ and CD8 þCD3 þ cells from spleen. (d,e) Flow cytometry analysis of total CD4 þCD3 þ and CD8 þCD3 þ T cells in the (d) LNs and (e) spleen and (f) total effector/memory (CD44hiCD62L ) CD4 þCD3 þ and CD8 þCD3 þ T cells in LNs and spleen. (c) Bar represents mean±s.d. ARTICLE Again, WASp KO CD8  DCs induced higher proliferation of OT-I CD8 þ T cells when compared with wild-type CD8  DCs (Fig. 6b). To assess if increased cross-presentation by WASp KO CD8  DCs was associated with changes in acidification upon phagocytosis, we first examined the early acidification events using beads bearing a mixture of pH-sensitive (FITC) and pH-insensitive (Alexa-fluor647) dyes that allows detection of phagosomal pH 5–8 (ref. 36). Using this approach, WASp KO CD8 þ DCs maintained their phagosomes at a higher pH during the first 2 h after phagocytosis when compared with wild-type NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications 6 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 of WT n ¼ 3; WKO n ¼ 1; DC/cWKO n ¼ 3. (d,f) WT n ¼ 6–7; DC/cWKO n ¼ 6. The data is representative of (b) one, (c) two, (d,e) four and (f) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. DC/cWKO, WASpfl/flCD11cCre/wt; fl, floxed (LoxP flanked); WT, wild type; WKO, WASp KO. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 ARTICLE a WASp WT CD8+ T cells CD4+ T cells WASp WT WKO DC cWKO 0 2000 4000 6000 8000 ** WT WKO DC cWKO 0 2000 4000 6000 ** 0 50 100 150 % of WASp positive cells WT WKO DC cWKO WT WKO DC cWKO 0 50 100 150 % of WASp positive cells CD8+DC CD8+DC CD8–DC CD8–DC DCcWKO WKO WT DCcWKO WKO WASp MFI WASp MFI CD8+ T cells CD4+ T cells c b c CD4 CD8 41.6% 83.7% 11.2% 54.9% WT DCc/WKO d WT DC cWKO 0.0 0.5 1.0 1.5 2.0 ** CD4 CD8 41.6% 83.7% 11.2% 54.9% WT DCc/WKO Cell number (×106) LN CD4+ T cells d d WT DC cWKO 0.0 0.5 1.0 1.5 2.0 * Cell number (×106) LN CD8+ T cells WT DC cWKO 0.0 0.5 1.0 1.5 2.0 ** Cell number (×106) LN CD4+ T cells WT DC cWKO 0.0 0.5 1.0 1.5 2.0 ** CD4/CD8 LN CD4+/CD8+ T cell ratio e CD4 CD8 WT DCc/WKO 26.3% 63.8% 22.4% 57.8% WT DC cWKO 0 2 4 6 8 10 ** Cell number (×106) Spleen CD4+T cells e f WT DC cWKO 0 1 2 3 ** CD4/CD8 Spleen CD4+/CD8+ T cell ratio WT DC cWKO 0 2 4 6 8 ** Cell number (×106) Spleen CD8+T cells WT DC cWKO 0 2 4 6 8 10 ** Cell number (×106) Spleen CD4+T cells f WT DC cWKO 0 2 4 6 8 * Cell number (×104) Effector/memory CD8+ T cells WT DC cWKO 0 2 4 6 8 10 LN Effector/memory CD4+ T cells Cell number (×104) f WT DC cWKO 0 20 40 60 80 100 Spleen Cell number (×104) Effector/memory CD4+ T cells WT DC cWKO 0 20 40 ** Cell number (×104) Effector/memory CD8+ T cells DC cWKO Figure 4 | DC-specific WASp deletion induces increased CD8 þ T cells. (a) Schematic representation of the targeting strategy. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Enriched CD8 þ and CD8  DCs from Flt3L tumour cell-injected mice on C57Bl/6 ackground were incubated overnight with 2 mg ml  1 SIINFEKL peptide and co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells for 72 otal number of OT-I CD8 þ T cells is indicated. (c) CD8 þ T-cell proliferation with ovalbumin. Equal numbers of FACS-sorted splenic CD8 þ DCs a D8  DCs from wild-type and WASp KO mice on C57Bl/6 background were incubated overnight with ovalbuminm, co-cultured with CFSE-labelled O Vb5.1/5.2 þ) CD8 þ T cells, and proliferation determined at 72 h. Total number of OT-I CD8 þ T cells is indicated. (d) Ovalbumin uptake. DCs were ncubated with soluble ovalbumin-Alexa594 to assess uptake of ovalbumin. (e) Ovalbumin degradation. DCs were incubated with soluble DQ-ovalbumin ssess the capacity to process antigen. Note that increased DQ-ovalbumin mean fluorescence intensity indicates increased degradation. (f) Ovalbum cidification. DCs were incubated with soluble pH rodo-ovalbumin. Note that increased pH rodo-ovalbumin mean fluorescence intensity indicates ecreased pH value. (g) CD8 þ DCs and CD8  DCs from wild-type and WASp KO mice were incubated overnight with ovalbumin. The presentation IINFEKL peptide on MHC class I H-2Kb molecules was assessed by flow cytometry and fold increase in expression was determined using the MFI va cquired for 0 mg ml  1 ovalbumin set to 1 (dotted line) (a) A total of 39–113 conjugates per mouse was analyzed. (a–g) Bar represents mean±s.d. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Also for CD8  DCs, WASp KO CD8  DCs showed lower percentage of pHrodohigh cells when compared with wild-type CD8  DCs (Fig. 6d). To confirm that the pH rodo fluorescence represented acidification, cells were pre-treated with NH4Cl that completely abolished acidification of pH rodo- ovalbumin-coated beads (Fig. 6e). Together, this data suggests that WASp KO CD8  DCs had decreased capacity to acidify endocytic and phagocytic vesicles and this was associated with cross-presentation on MHC class I and increased proliferation of CD8 þ T cells. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 7 7 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 WT WKO 80 60 40 40 30 20 10 0 WT WKO WT WKO 50 20 0 150 100 50 30 25 20 15 10 5 0 30,000 20,000 10,000 0 WT WKO WT WKO CD8–DC CD8–DC CD8–DC CD8–DC *** *** ** CD8+DC % of synapses per total conjugates 16μg/ml 50μg/ml 150μg/ml 16 μg ml–1 50 μg ml–1 150 μg ml–1 16 μg ml–1 50 μg ml–1 DQ OVA CD8–DC DQ OVA CD8+DC 150 μg ml–1 16 μg ml–1 50μg ml–1 150 μg ml–1 16 μg ml–1 50 μg ml–1 150 μg ml–1 ** ** ** OT1 proliferation with CD8+DC OVA-Alexa594+ CD8+DC OVA-Alexa594+ CD8–DC OT1 proliferation with CD8–DC CD8+ DC CD8+ DC CD8+ DC WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO WT WKO 10 8 6 4 2 0 10 8 6 4 2 0 40min 2h 6h 40min 2h ** ** * 6h % of cells % of cells OVA-pH rodohi CD8+DC OVA-pH rodohi CD8–DC pHrodohi pHrodohi CD4+T CD8+T MHCII MHCI pH7 pH5–3 TCR TCR DC pH rodo Percent of max pH rodo Percent of max WT 3 2 1 0 MFI (H-2Kb-SIINFEKL) 3 2 1 0 MFI (H-2Kb-SIINFEKL) WKO WT WKO WT WKO 16 μg ml–1 50 μg ml–1 150 μg ml–1 16 μg ml–1 50 μg ml–1 150 μg ml–1 WT WKO WT WKO CFSE Cell number CFSE Cell number CD8–DC CD8+DC OT1 proliferation with SIINFEKL Cell number (×103) Cell number (×103) MFI 30,000 20,000 10,000 10,000 15,000 5,000 0 0 MFI MFI 10,000 15,000 5,000 0 MFI 150 100 50 30 25 20 15 10 5 0 Cell number (×103) CD8+DC CD8–DC γTubulin actin 16 μg ml–1 50μg ml–1 150 μg ml–1 WT WKO WT WKO WT WKO a b c d e f g igure 5 | Increased cross-presentation by WASp KO DCs. (a) Immune synapse. Enriched CD8 þ and CD8  DCs from Flt3L tumour cell-injected m n C57Bl/6 background were pulsed with ovalbumin and incubated with OT-I CD8 þ Tcells. Percentage of synapses was measured by counting the num f conjugates with polarized actin (red) and microtubule organizing center g-tubulin (green) towards the synapse and divided by the total number of onjugates. (b) CD8 þ T-cell proliferation with SIINFEKL peptide. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 of W 3; WKO n 3 per group The data is representative of (a b g) two experiments (c) four experiments and (d f) three separate experiments *Po0 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12 WT WKO 80 60 40 40 30 20 10 0 WT WKO WT WKO 50 20 0 WT WKO WT WKO CD8–DC CD8+DC % of synapses per total conjugates CD8–DC CD8+DC OT1 proliferation with SIINFEKL Cell number (×103) CD8+DC CD8–DC γTubulin actin a b 80 60 40 40 30 20 10 0 WT WKO WT WKO 50 20 0 WT WKO WT WKO CD8–DC CD8+DC % of synapses per total conjugates CD8–DC CD8+DC OT1 proliferation with SIINFEKL Cell number (×103) b b a CD8+ DC WT WKO CFSE Cell number c c 16 μg ml–1 50 μg ml–1 150 μg ml–1 ** ** ** OT1 proliferation with CD8–DC O WT WKO 150 100 50 30 25 20 15 10 5 0 Cell number (×103) 150 100 50 30 25 20 15 10 5 0 16μg/ml 50μg/ml 150μg/ml OT1 proliferation with CD8+DC WT WKO Cell number (×103) CD8–DC WT WKO CFSE Cell number 30,000 20,000 10,000 0 16 μg ml–1 50 μg ml–1 150 μg ml–1 OVA-Alexa594+ CD8+DC WT WKO MFI d d 16 μg ml–1 50μg ml–1 150 μg ml–1 μg μg μg CD8+ DC WT 3 2 1 0 MFI (H-2Kb-SIINFEKL) WKO WT WKO WT WKO g 16 μg ml–1 50 μg ml–1 150 μg ml–1 OVA-Alexa594+ CD8–DC WT WKO 30,000 20,000 10,000 0 MFI g μg ml–1 μg ml–1 μg ml–1 DQ OVA CD8+DC WT WKO 16 μg ml–1 50 μg ml–1 150 μg ml–1 10,000 15,000 5,000 0 MFI e f e e CD8–DC *** *** ** 3 2 1 0 MFI (H-2Kb-SIINFEKL) 16 μg ml–1 50μg ml–1 150 μg ml–1 WT WKO WT WKO WT WKO DQ OVA CD8–DC WT WKO 16 μg ml–1 50 μg ml–1 150 μg ml–1 10,000 15,000 5,000 0 MFI WT WKO 10 8 6 4 2 0 40min 2h ** ** * 6h % of cells OVA-pH rodohi CD8–DC CD8+ DC WT WKO pHrodohi pH rodo Percent of max CD8–DC WT WKO WT WKO 10 8 6 4 2 0 40min 2h 6h % of cells OVA-pH rodohi CD8+DC pHrodohi pH rodo Percent of max CD4+T CD8+T MHCII MHCI pH7 pH5–3 TCR TCR DC f Percent of max CD8+T Figure 5 | Increased cross-presentation by WASp KO DCs. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (a) Immune synapse. Enriched CD8 þ and CD8  DCs from Flt3L tumour cell-injected mice on C57Bl/6 background were pulsed with ovalbumin and incubated with OT-I CD8 þ Tcells. Percentage of synapses was measured by counting the number of conjugates with polarized actin (red) and microtubule organizing center g-tubulin (green) towards the synapse and divided by the total number of cell conjugates. (b) CD8 þ T-cell proliferation with SIINFEKL peptide. Enriched CD8 þ and CD8  DCs from Flt3L tumour cell-injected mice on C57Bl/6 background were incubated overnight with 2 mg ml  1 SIINFEKL peptide and co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells for 72 h. Total number of OT-I CD8 þ T cells is indicated. (c) CD8 þ T-cell proliferation with ovalbumin. Equal numbers of FACS-sorted splenic CD8 þ DCs and CD8  DCs from wild-type and WASp KO mice on C57Bl/6 background were incubated overnight with ovalbuminm, co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation determined at 72 h. Total number of OT-I CD8 þ T cells is indicated. (d) Ovalbumin uptake. DCs were incubated with soluble ovalbumin-Alexa594 to assess uptake of ovalbumin. (e) Ovalbumin degradation. DCs were incubated with soluble DQ-ovalbumin to assess the capacity to process antigen. Note that increased DQ-ovalbumin mean fluorescence intensity indicates increased degradation. (f) Ovalbumin acidification. DCs were incubated with soluble pH rodo-ovalbumin. Note that increased pH rodo-ovalbumin mean fluorescence intensity indicates decreased pH value. (g) CD8 þ DCs and CD8  DCs from wild-type and WASp KO mice were incubated overnight with ovalbumin. The presentation of SIINFEKL peptide on MHC class I H-2Kb molecules was assessed by flow cytometry and fold increase in expression was determined using the MFI value acquired for 0 mg ml  1 ovalbumin set to 1 (dotted line) (a) A total of 39–113 conjugates per mouse was analyzed. (a–g) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3 per group. The data is representative of (a,b,g) two experiments, (c) four experiments and (d–f) three separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. MFI, mean fluorescence intensity; WT, wild type; WKO, WASp KO. WASp KO CD8  DCs activate Rac2 and ROS production. CD8 þ DCs assemble a phagosomal complex consisting of Rac2 and the NADPH complex to maintain a neutral pH of the phagosome21,36. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (a) Ovalbumin-bead uptake by DCs. (b) CD8 þ T-cell proliferation with ovalbumin-coated beads. FACS-sorted DCs that had taken up one ovalbumin-coated bead were co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation determined as CFSE dilution at 72 h. (c) DCs were incubated with latex beads coupled with pH-sensitive (FITC) and pH-insensitive (Alexa647) dyes. FITC and Alexa647 intensities were measured at the specified time points and the pH was determined as described in the materials and methods. (d,e) Ovalbumin-bead acidification. (d) DCs were incubated with pH rodo-ovalbumin-coated beads to assess acidification of antigen in phagocytic vesicles. Note that increased pH rodo mean fluorescence intensity translates into decreased pH value. (e) DCs were pre-treated with NH4Cl to abolish acidification before addition of pH rodo-ovalbumin beads. (b–e) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. The data are representative of (b,d) three and (c,e) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, wild type; WKO, WASp KO. KO DCs (Fig. 7a). Wild-type CD8 þ DCs had high expression of Rac2 that co-localized with the phagocytosed bead together with the NADPH component gp91phox (Fig. 7a,b, upper left image; Supplementary Fig. 9a–c). Wild-type CD8  DCs had less Rac2 expression and did not co-localize Rac2 around the bead (Fig. 7a,b, lower left image; Supplementary Fig. 9a–c)21. WASp KO CD8 þ DCs showed similar expression and co-localization of Rac2 when compared with wild-type CD8 þ DCs (Fig. 7a,b, upper right image; Supplementary Fig. 9a–c). The phenotype of WASp KO CD8  DCs was strikingly similar to wild-type CD8 þ DCs with increased expression of Rac2 and increased co-localization of Rac2 and gp91phox to the phagocytosed bead (Fig. 7a,b, lower right image; Supplementary Fig. 9a–c). To quantify Rac1/2 activity, we measured active GTP-bound Rac1/2 in protein lysates of DCs incubated with ovalbumin- coated beads. Wild-type CD8 þ DCs had lower GTP-Rac1/2 when compared with WASp KO CD8 þ DCs (Fig. 7c). When compared with wild-type CD8  DCs, WASp KO CD8  DCs KO DCs (Fig. 7a). Wild-type CD8 þ DCs had high expression of Rac2 that co-localized with the phagocytosed bead together with the NADPH component gp91phox (Fig. 7a,b, upper left image; Supplementary Fig. 9a–c). Wild-type CD8  DCs had less Rac2 expression and did not co-localize Rac2 around the bead (Fig. 7a,b, lower left image; Supplementary Fig. 9a–c)21. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 To study Rac2 expression and co-localization to the phagosome, we incubated wild-type and WASp KO CD8 þ and CD8  DCs with ovalbumin-coated beads and stained for Rac1 and Rac2. Rac1 expression was higher in CD8 þ DCs than in CD8  DCs and this was detected both in wild-type and WASp NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications 8 ARTICLE pH 5-3 pHrodo-OVA FSC SSC WT CD8-DCs WKO CD8-DCs WT CFSE Cell number WT WKO WT WKO 0 2,000 4,000 6,000 8,000 OT1 proliferation * CD8+DC CD8–DC WT WKO WT WKO 5 6 7 8 9 10 pH 2h 30min ** WT WKO WT WKO 5 6 7 8 9 10 pH 2h 30 min WT WKO WT WKO WT WKO WT WKO 0 10 20 30 40 2h 6h ** WT WKO WT WKO 2 h 6 h ** * pHrodo Percent of max pHrodo Percent of max ** WT WKO WT WKO 0 10 20 30 WT WKO WT WKO % of cells 0 10 20 30 % of cells CD8+ DC CD8+DC CD8– DC CD8–DC CD8+DC CD8–DC Cell number pHrodohi pHrodohi % of cells 0 10 20 30 40 % of cells pH rodohi CD8+DCs pH rodohi CD8–DC 6h pH-rodohi CD8+DCs 6h pH-rodohi CD8–DCs a b c d e WKO WT WKO RE COMMUNICATIONS | DOI: 10.1038/ncomms12175 AR NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 pH 5-3 pHrodo-OVA FSC SSC WT CD8-DCs WKO CD8-DCs WT CFSE Cell number WT WKO WT WKO 0 2,000 4,000 6,000 8,000 OT1 proliferation * CD8+DC CD8–DC WT WKO WT WKO 5 6 7 8 9 10 pH 2h 30min ** WT WKO WT WKO 5 6 7 8 9 10 pH 2h 30 min WT WKO WT WKO WT WKO WT WKO 0 10 20 30 40 2h 6h ** WT WKO WT WKO 2 h 6 h ** * pHrodo Percent of max pHrodo Percent of max ** WT WKO WT WKO 0 10 20 30 +NH4Cl –NH4Cl WT WKO WT WKO +NH4Cl –NH4Cl % of cells 0 10 20 30 % of cells CD8+ DC CD8+DC CD8– DC CD8–DC CD8+DC CD8–DC Cell number pHrodohi pHrodohi % of cells 0 10 20 30 40 % of cells pH rodohi CD8+DCs pH rodohi CD8–DC 6h pH-rodohi CD8+DCs 6h pH-rodohi CD8–DCs a b c d e WKO WT WKO Figure 6 | The phagosomal compartment in WASp KO DCs. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Note that increased pH rodo mean fluorescence intensity translates into decreased pH value. (e) DCs were pre-treated with NH4Cl to abolish acidification before addition of pH rodo-ovalbumin beads. (b–e) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. The data are representative of (b,d) three and (c,e) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, wild type; WKO, WASp KO. 2h 30min 2h 30 min WT WKO WT WKO WT WKO WT WKO 0 10 20 30 40 2h 6h ** WT WKO WT WKO 2 h 6 h ** * pHrodo Percent of max pHrodo Percent of max CD8+DC CD8–DC pHrodohi pHrodohi % of cells 0 10 20 30 40 % of cells pH rodohi CD8+DCs pH rodohi CD8–DC d d e e ** WT WKO WT WKO 0 10 20 30 +NH4Cl –NH4Cl WT WKO WT WKO +NH4Cl –NH4Cl % of cells 0 10 20 30 % of cells 6h pH-rodohi CD8+DCs 6h pH-rodohi CD8–DCs e Figure 6 | The phagosomal compartment in WASp KO DCs. Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were incubated with ovalbumin-coated latex beads overnight. (a) Ovalbumin-bead uptake by DCs. (b) CD8 þ T-cell proliferation with ovalbumin-coated beads. FACS-sorted DCs that had taken up one ovalbumin-coated bead were co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation determined as CFSE dilution at 72 h. (c) DCs were incubated with latex beads coupled with pH-sensitive (FITC) and pH-insensitive (Alexa647) dyes. FITC and Alexa647 intensities were measured at the specified time points and the pH was determined as described in the materials and methods. (d,e) Ovalbumin-bead acidification. (d) DCs were incubated with pH rodo-ovalbumin-coated beads to assess acidification of antigen in phagocytic vesicles. Note that increased pH rodo mean fluorescence intensity translates into decreased pH value. (e) DCs were pre-treated with NH4Cl to abolish acidification before addition of pH rodo-ovalbumin beads. (b–e) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. The data are representative of (b,d) three and (c,e) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, wild type; WKO, WASp KO. Figure 6 | The phagosomal compartment in WASp KO DCs. Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were incubated with ovalbumin-coated latex beads overnight. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were incubated with ovalbumin-coated latex beads overnight. (a) Ovalbumin-bead uptake by DCs. (b) CD8 þ T-cell proliferation with ovalbumin-coated beads. FACS-sorted DCs that had taken up one ovalbumin-coated bead were co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation determined as CFSE dilution at 72 h. (c) DCs were incubated with latex beads coupled with pH-sensitive (FITC) and pH-insensitive (Alexa647) dyes. FITC and Alexa647 intensities were measured at the specified time points and the pH was determined as described in the materials and methods. (d,e) Ovalbumin-bead acidification. (d) DCs were incubated with pH rodo-ovalbumin-coated beads to assess acidification of antigen in phagocytic vesicles. Note that increased pH rodo mean fluorescence intensity translates into decreased pH value. (e) DCs were pre-treated with NH4Cl to abolish acidification before addition of pH rodo-ovalbumin beads. (b–e) Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. The data are representative of (b,d) three and (c,e) two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. WT, wild type; WKO, WASp KO. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 pH 5-3 pHrodo-OVA FSC SSC WT CD8-DCs WKO CD8-DCs a pH 5-3 pHrodo-OVA FSC SSC WT CD8-DCs a pH 5-3 pHrodo-OVA FSC SSC WT CD8-DCs WKO CD8-DCs a pH 5-3 FSC SSC WT CFSE Cell number WT WKO WT WKO 0 2,000 4,000 6,000 8,000 OT1 proliferation * CD8+DC CD8–DC CD8+ DC CD8– DC Cell number b WKO WT WKO FSC S WT CFSE Cell number WT WKO WT WKO 0 2,000 4,000 6,000 8,000 OT1 proliferation * CD8+DC CD8–DC CD8+ DC CD8– DC Cell number b WKO WT WKO WT CFSE Cell number CD8+ DC b WKO b WT WKO WT WKO 0 2,000 4,000 6,000 8,000 OT1 proliferation * CD8+DC CD8–DC CD8– DC Cell number WT WKO WT WKO WT WKO 5 6 7 8 9 10 pH 2h 30min ** WT WKO WT WKO 5 6 7 8 9 10 pH 2h 30 min CD8+DC CD8–DC c WT WKO WT WKO 5 6 2h 30min WT WKO WT WKO 5 6 2h 30 min WT WKO WT WKO WT WKO WT WKO 0 10 20 30 40 2h 6h ** WT WKO WT WKO 2 h 6 h ** * pHrodo Percent of max pHrodo Percent of max ** WT WKO WT WKO 0 10 20 30 +NH4Cl –NH4Cl WT WKO WT WKO +NH4Cl –NH4Cl % of cells 0 10 20 30 % of cells CD8+DC CD8–DC pHrodohi pHrodohi % of cells 0 10 20 30 40 % of cells pH rodohi CD8+DCs pH rodohi CD8–DC 6h pH-rodohi CD8+DCs 6h pH-rodohi CD8–DCs d e Figure 6 | The phagosomal compartment in WASp KO DCs. Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were incubated with ovalbumin-coated latex beads overnight. (a) Ovalbumin-bead uptake by DCs. (b) CD8 þ T-cell proliferation with ovalbumin-coated beads. FACS-sorted DCs that had taken up one ovalbumin-coated bead were co-cultured with CFSE-labelled OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation determined as CFSE dilution at 72 h. (c) DCs were incubated with latex beads coupled with pH-sensitive (FITC) and pH-insensitive (Alexa647) dyes. FITC and Alexa647 intensities were measured at the specified time points and the pH was determined as described in the materials and methods. (d,e) Ovalbumin-bead acidification. (d) DCs were incubated with pH rodo-ovalbumin-coated beads to assess acidification of antigen in phagocytic vesicles. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Wild-type and WASp KO CD8 þ and CD8  DCs from mice ed intracellularly with anti-Rac1 and anti-Rac2 antibodies and analysed by flow cytometry. Bar represents mean±s.d. Figure 7 | WASp KO CD8  DCs activate Rac2 and ROS production. (a) Rac1/2 expression. Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were stained intracellularly with anti-Rac1 and anti-Rac2 antibodies and analysed by flow cytometry. Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. (b) Rac1/2 localization. DCs were incubated with ovalbumin-coated beads for 2 h to allow phagocytosis, transferred to fibronectin-coated glass and stained for Rac1-FITC (green) and Rac2-Alexa555 (red) antibodies and analysed by confocal microscopy. Both panels show Rac1 and Rac2 staining to the left and bright field to the right, with the phagocytosed bead marked with an asterisk. (Left panel) Rac2 co-localization with the phagosome was calculated as: [(beads with Rac2)/(cells with beads)]  100. Bars represent mean±s.d. of 3–4 mice; 7–16 pictures with total 19–119 cells per mouse. (right panel) The MFI from the middle of the cell towards the bead (a) or in the opposite direction (b) was measured using the ImageJ software. The (MFI a/MFI b) is shown as Rac2 intensity around the bead. Bars represent mean±s.d. of 3–4 mice; 3–4 pictures with total 11–21 cells per mouse. (c) Rac1/2 activity. Quantification of active GTP-bound Rac1/2 and GTP-bound Rac2. Bars represent mean±s.d. of 3–6 mice. (d) NADPH induced ROS production. DCs from wild type, WASp KO, Ncf1*, and WASp KO  Ncf1* mice on C57Bl/6 background were enriched and incubated with DHR-coated beads for 1–2 h and analysed by flow cytometry for ROS production. The dotted line indicates background DHR intensity upon DPI treatment. The data in (a) are representative of three and (b–d) of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. Scale bar, 5 mm. MFI, mean fluorescence intensity; WT, wild type; WKO, WASp KO. species (ROS), we measured ROS in CD8 þ and CD8  DCs. Enriched DCs from the spleen were incubated with beads coated with dihydrorhodamine 123 (DHR) that emits fluorescence in the presence of ROS. CD8 þ and CD8  DCs from WASp KO mice showed increased ROS production at 1 and 2 h when compared with corresponding DCs in wild-type mice (Fig. 7d). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 WASp KO CD8 þ DCs showed similar expression and co-localization of Rac2 when compared with wild-type CD8 þ DCs (Fig. 7a,b, upper right image; Supplementary Fig. 9a–c). The phenotype of 9 a b Rac2 expression (CD8-DC) Rac2 CD8+DC CD8–DC WT WKO Rac2 expression WT WKO WT WKO 0 50 100 150 * CD8+DC CD8–DC MFI * * * * * * * * Rac1 Rac1 expression (CD8-DC) Rac1 expression WT WKO WT WKO 0 200 400 600 MFI CD8+DC CD8–DC WT WKO WT WKO WT WKO WT WKO 0 20 40 60 80 CD8+DC CD8-DC * Rac2+ phagosomes Rac2+ beads per DC with beads (%) Rac1/2-GTP WT WKO WT WKO 0.0 0.5 1.0 1.5 OD 490nm CD8+DC CD8–DC * * c Rac1 Rac2 WT WKO WT WKO 0.0 0.1 0.2 0.3 0.4 Rac2-GTP OD 450nm * CD8–DC CD8+DC a b (MFIa)/(MFIb) WT WKO WT WKO 0 1 2 3 4 MFI(A/B) CD8+DC CD8–DC * Rac2 intensity d DHR Percent of max CD8+ DC WT WKO CD8– DC WT WKO DHR Percent of max ROS production CD8+DC MFI DHR 2 h 1 h ** ** WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* 0 500 1,000 1,500 2,000 2,500 0 500 1,000 1,500 2,000 2,500 ROS production CD8–DC MFI DHR ** 2 h 1 h ** Figure 7 | WASp KO CD8  DCs activate Rac2 and ROS production. (a) Rac1/2 expression. Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were stained intracellularly with anti-Rac1 and anti-Rac2 antibodies and analysed by flow cytometry. Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. (b) Rac1/2 localization. DCs were incubated with ovalbumin-coated beads for 2 h to allow phagocytosis, transferred to fibronectin-coated glass and stained for Rac1-FITC (green) and Rac2-Alexa555 (red) antibodies and analysed by confocal microscopy. Both panels show Rac1 and Rac2 staining to the left and bright field to the right, with the phagocytosed bead marked with an asterisk. (Left panel) Rac2 co-localization with the phagosome was calculated as: [(beads with Rac2)/(cells with beads)]  100. Bars represent mean±s.d. of 3–4 mice; 7–16 pictures with total 19–119 cells per mouse. (right panel) The MFI from the middle of the cell towards the bead (a) or in the opposite direction (b) was measured using the ImageJ software. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 The (MFI a/MFI b) is shown as Rac2 intensity around the bead. Bars represent mean±s.d. of 3–4 mice; 3–4 pictures with total 11–21 cells per mouse. (c) Rac1/2 activity. Quantification of active GTP-bound Rac1/2 and GTP-bound Rac2. Bars represent mean±s.d. of 3–6 mice. (d) NADPH induced ROS production. DCs from wild type, WASp KO, Ncf1*, and WASp KO  Ncf1* mice on C57Bl/6 background were enriched and incubated with DHR-coated beads for 1–2 h and analysed by flow cytometry for ROS production. The dotted line indicates background DHR intensity upon DPI treatment. The data in (a) are representative of three and (b–d) of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. Scale bar, 5 mm. MFI, mean fluorescence intensity; WT, wild type; WKO, WASp KO. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (a) Rac1/2 expression. Wild-type and WASp KO CD8 þ and CD8  DCs from mice on C57Bl/6 background were stained intracellularly with anti-Rac1 and anti-Rac2 antibodies and analysed by flow cytometry. Bar represents mean±s.d. of WT n ¼ 3; WKO n ¼ 3. (b) Rac1/2 localization. DCs were incubated with ovalbumin-coated beads for 2 h to allow phagocytosis, transferred to fibronectin-coated glass and stained for Rac1-FITC (green) and Rac2-Alexa555 (red) antibodies and analysed by confocal microscopy. Both panels show Rac1 and Rac2 staining to the left and bright field to the right, with the phagocytosed bead marked with an asterisk. (Left panel) Rac2 co-localization with the phagosome was calculated as: [(beads with Rac2)/(cells with beads)]  100. Bars represent mean±s.d. of 3–4 mice; 7–16 pictures with total 19–119 cells per mouse. (right panel) The MFI from the middle of the cell towards the bead (a) or in the opposite direction (b) was measured using the ImageJ software. The (MFI a/MFI b) is shown as Rac2 intensity around the bead. Bars represent mean±s.d. of 3–4 mice; 3–4 pictures with total 11–21 cells per mouse. (c) Rac1/2 activity. Quantification of active GTP-bound Rac1/2 and GTP-bound Rac2. Bars represent mean±s.d. of 3–6 mice. (d) NADPH induced ROS production. DCs from wild type, WASp KO, Ncf1*, and WASp KO  Ncf1* mice on C57Bl/6 background were enriched and incubated with DHR-coated beads for 1–2 h and analysed by flow cytometry for ROS production. The dotted line indicates background DHR intensity upon DPI treatment. The data in (a) are representative of three and (b–d) of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. Scale bar, 5 mm. MFI, mean fluorescence intensity; WT, wild type; WKO, WASp KO. d DHR Percent of max CD8+ DC WT WKO CD8– DC WT WKO DHR Percent of max ROS production CD8+DC MFI DHR 2 h 1 h ** ** WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* 0 500 1,000 1,500 2,000 2,500 0 500 1,000 1,500 2,000 2,500 ROS production CD8–DC MFI DHR ** 2 h 1 h ** d WT WKO Ncf1* WKOxNcf1* WT WKO 0 500 1,000 1,500 2,000 2,500 ROS production CD8–DC MFI DHR ** 2 1 h ** 2 h ctivate Rac2 and ROS production. (a) Rac1/2 expression. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 a Rac2 expression (CD8-DC) Rac2 Rac2 expression WT WKO WT WKO 0 50 100 150 * CD8+DC CD8–DC MFI Rac1 Rac1 expression (CD8-DC) Rac1 expression WT WKO WT WKO 0 200 400 600 MFI CD8+DC CD8–DC WT WKO WT WKO a Rac1 Rac1 expression (CD8-DC) WT WKO a Rac2 expression (CD8-DC) Rac2 Rac2 expression WT WKO WT WKO 0 50 100 150 * CD8+DC CD8–DC MFI Rac1 expression WT WKO WT WKO 0 200 400 600 MFI CD8+DC CD8–DC WT WKO b Rac2 expression (CD8 DC) Rac2 CD8+DC CD8–DC WT WKO Rac2 expression WT WKO WT WKO 0 50 100 150 * CD8+DC CD8–DC MFI * * * * * * * * Rac1 Rac1 expression (CD8-DC) Rac1 expression WT WKO WT WKO 0 200 400 600 MFI CD8+DC CD8–DC WT WKO WT WKO WT WKO WT WKO 0 20 40 60 80 CD8+DC CD8-DC * Rac2+ phagosomes Rac2+ beads per DC with beads (%) Rac1/2-GTP WT WKO WT WKO 0.0 0.5 1.0 1.5 OD 490nm CD8+DC CD8–DC * * c Rac1 Rac2 WT WKO WT WKO 0.0 0.1 0.2 0.3 0.4 Rac2-GTP OD 450nm * CD8–DC CD8+DC a b (MFIa)/(MFIb) WT WKO WT WKO 0 1 2 3 4 MFI(A/B) CD8+DC CD8–DC * Rac2 intensity b CD8+DC CD8–DC WT WKO * * * * * * * * WT WKO WT WKO 0 20 40 60 80 CD8+DC CD8-DC * Rac2+ phagosomes Rac2+ beads per DC with beads (%) Rac1 Rac2 a b (MFIa)/(MFIb) WT WKO WT WKO 0 1 2 3 4 MFI(A/B) CD8+DC CD8–DC * Rac2 intensity b MFI(A/B) CD8–DC * * * * WT WKO WT WKO 0 20 40 CD8+DC CD8-DC Rac2+ bead with bea Rac1/2-GTP WT WKO WT WKO 0.0 0.5 1.0 1.5 OD 490nm CD8+DC CD8–DC * * c Rac1 Rac2 WT WKO WT WKO 0.0 0.1 0.2 0.3 0.4 Rac2-GTP OD 450nm * CD8–DC CD8+DC b (MFIa)/(MFIb) WT WKO WT WKO 0 1 2 MFI( CD8+DC CD8–DC Rac1/2-GTP WT WKO WT WKO 0.0 0.5 1.0 1.5 OD 490nm CD8+DC CD8–DC * * c WT WKO WT WKO 0.0 0.1 0.2 0.3 0.4 Rac2-GTP OD 450nm * CD8–DC CD8+DC c c d DHR Percent of max CD8+ DC WT WKO CD8– DC WT WKO DHR Percent of max ROS production CD8+DC MFI DHR 2 h 1 h ** ** WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* WT WKO Ncf1* WKOxNcf1* 0 500 1,000 1,500 2,000 2,500 0 500 1,000 1,500 2,000 2,500 ROS production CD8–DC MFI DHR ** 2 h 1 h ** Figure 7 | WASp KO CD8  DCs activate Rac2 and ROS production. NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 ARTICLE CD8 SSC SSC SSC Percent of max % of cells % of cells Cell number Cell number % of cells Cell number CD11c GFP GFP GFP pHrodo WASpWT GFP+ WASp KO GFP– WASpWT GFP+ WASp KO GFP– WASpΔVCA GFP+ WASp KO GFP– WASpΔVCA GFP+ WASp KO GFP– 0 20 40 60 40x104 OT1 proliferation OT1 proliferation WASpWT OT1 proliferation WASpΔVCA 30x104 * 20x104 10x104 0 1×104 2×104 3×104 4×104 50 40 30 ** ** ** * 20 10 0 50 40 30 20 10 0 5×104 0 1×104 2×104 3×104 4×104 5×104 0 * pH rodohi pHrodohi WASpWT Negative control WASpWT WASpΔVCA pHrodohi WASpΔVCA CD11c pHrodo-OVA CD11c pHrodo-OVA 86% 84% WT WT 10μm WKO WKO WKO WKO WT WT WKO WT a b c d e f Figure 8 | WASp and the WASp-VCA domain in cross-presentation. (a–c) Wild-type and WASp KO BM DCs from mice on C57Bl/6 background were analysed for capacity to acidify pH rodo-ovalbumin and induce proliferation of OT-I (Vb5.1/5.2 þ) CD8 þ T cells. (d–f) Wild-type BM DCs at day 6 of culture were treated with CK666 to inhibit activity of the Arp2/3 complex or treated with a-amanitin to inhibit polymerase II gene transcription. (g,h) WASp KO BM DCs at day 6 of culture were Amaxa transfected with wild-type WASp (GFP-WASpWT) or WASp lacking the VCA domain (GFP-WASpDVCA) and sorted into GFP þ and GFP  cells 6 h after transfection. (a) Expression of CD11c and CD8 on BM DCs at day 6 of culture and analysis of acidification using uptake of soluble pH rodo-ovalbumin and confocal microscopy. (b,e,h) DCs were incubated with pH rodo-ovalbumin-coated beads and gated for DCs that had taken up one bead and pH rodo intensity was analysed by flow cytometry. (c,f,i) DCs were incubated with 0.5 mg ml  1 soluble ovalbumin and co-cultured with OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation was assessed as total number of OT-I CD8 þ T cells by flow cytometry. (d) Wild-type BM DCs were treated for 6 h with CK666 to inhibit Arp2/3 activity and a-amanitin to inhibit RNA polymerase II transcription. Note the drop in polymerized actin measured using phalloidin upon CK666 treatment and the reduced expression of cyclin A2 upon a-amanitin treatment. (g) Gating strategy for sorting of GFP þ and GFP  WASp KO BM DCs. Negative control indicates non-transfected cells. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (g) Gating strategy for sorting of GFP þ and GFP  WASp KO BM DCs. Negative control indicates non-transfected cells. The data in (a–c) are representative of three and (d–h) of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. Scale bar, 10 mm. WASpDVCA, WASp lacking the verprolin-cofilin-acidic domain; WT, wild type; WKO, WASp KO. Figure 8 | WASp and the WASp-VCA domain in cross-presentation. (a–c) Wild-type and WASp KO BM DCs from Figure 8 | WASp and the WASp-VCA domain in cross-presentation. (a–c) Wild- WASp and the WASp-VCA domain in cross-presentation. To address if WASp directly reduces cross-presentation in DCs, we used the fact that bone marrow (BM)-derived DCs almost exclusively consists of CD11c þCD8  DCs. BM-derived DCs from wild-type and WASp KO mice showed similar proportion of CD11c þCD8  DCs (Fig. 8a). Similar to WASp KO spleen CD8  DCs, WASp KO BM DCs had reduced capacity to acidify antigen and induced increased proliferation of OT-I CD8 þ T cells when compared with wild-type BM DCs (Fig. 8b,c). To test if re-expression of WASp would reduce cross-presentation, WASp KO BM DCs were transfected with GFP-WASpWT and GFP þ and GFP  cells were sorted after 6 h (Fig. 8d). WASp KO BM DCs expressing GFP-WASpWT showed increased acidification and lower induction OT-I CD8 þ T-cell proliferation when compared with GFP  WASp KO BM DCs (Fig. 8e,f). Finally, we tested if the WASp-VCA domain was required to dampen cross-presentation. WASp KO BM DCs expressing GFP-WASpDVCA showed increased acidification and lower induction OT-I CD8 þ T-cell proliferation when compared with were pre-treated with diphenyleneiodonium (DPI) that inhibits ROS production followed by incubation with DHR-coated beads. In the presence of DPI, the DHR signal was quenched in all DCs (indicated as dotted line in Fig. 7d). To understand if the observed increase in ROS production was directly caused by the NADPH oxidase activity, WASp KO mice were bred to Ncf1m1J/m1J mice that have reduced NADPH oxidase activity37, herein referred to as Ncf1* mice. WASp KO  Ncf1* mice showed abolished ROS production in CD8 þ and CD8  DCs and reached the DHR intensity of wild-type CD8 þ and CD8  DCs (Fig. 7d). This suggests that increased ROS production in WASp KO DCs is dependent on a functional NADPH oxidase. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 To confirm that the increased DHR signal was due to increased ROS, DCs had higher quantity of active GTP-bound Rac1/2 (Fig. 7c). We next quantified Rac2 activity specifically and found that WASp KO CD8  DC had the highest quantity of active GTP-bound Rac2 when compared with wild-type CD8  DCs, and wild-type and WASp KO CD8 þ DCs (Fig. 7c). Because Rac2 activity is linked to the assembly of the NADPH complex that directly regulates the production of reactive oxygen 10 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 (c,f,i) DCs were incubated with 0.5 mg ml  1 soluble ovalbumin and co-cultured with OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation was assessed as total number of OT-I CD8 þ T cells by flow cytometry. (d) Wild-type BM DCs were treated for 6 h with CK666 to inhibit Arp2/3 activity and a-amanitin to inhibit RNA polymerase II transcription. Note the drop in polymerized actin measured using phalloidin upon CK666 treatment and the reduced expression of cyclin A2 upon a-amanitin treatment. (g) Gating strategy for sorting of GFP þ and GFP  WASp KO BM DCs. Negative control indicates non-transfected cells. The data in (a–c) are representative of three and (d–h) of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. Scale bar, 10 mm. WASpDVCA, WASp lacking the verprolin-cofilin-acidic domain; WT, wild type; WKO, WASp KO. Figure 8 | WASp and the WASp-VCA domain in cross-presentation. (a–c) Wild-type and WASp KO BM DCs from mice on C57Bl/6 background were analysed for capacity to acidify pH rodo-ovalbumin and induce proliferation of OT-I (Vb5.1/5.2 þ) CD8 þ T cells. (d–f) Wild-type BM DCs at day 6 of culture were treated with CK666 to inhibit activity of the Arp2/3 complex or treated with a-amanitin to inhibit polymerase II gene transcription. (g,h) WASp KO BM DCs at day 6 of culture were Amaxa transfected with wild-type WASp (GFP-WASpWT) or WASp lacking the VCA domain (GFP-WASpDVCA) and sorted into GFP þ and GFP  cells 6 h after transfection. (a) Expression of CD11c and CD8 on BM DCs at day 6 of culture and analysis of acidification using uptake of soluble pH rodo-ovalbumin and confocal microscopy. (b,e,h) DCs were incubated with pH rodo-ovalbumin-coated beads and gated for DCs that had taken up one bead and pH rodo intensity was analysed by flow cytometry. (c,f,i) DCs were incubated with 0.5 mg ml  1 soluble ovalbumin and co-cultured with OT-I (Vb5.1/5.2 þ) CD8 þ T cells, and proliferation was assessed as total number of OT-I CD8 þ T cells by flow cytometry. (d) Wild-type BM DCs were treated for 6 h with CK666 to inhibit Arp2/3 activity and a-amanitin to inhibit RNA polymerase II transcription. Note the drop in polymerized actin measured using phalloidin upon CK666 treatment and the reduced expression of cyclin A2 upon a-amanitin treatment. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 The data in (a–c) are representative of three and (d–h) of two separate experiments. *Po0.05; **Po0.01 as calculated by the unpaired Student’s t-test. Scale bar, 10 mm. WASpDVCA, WASp lacking the verprolin-cofilin-acidic domain; WT, wild type; WKO, WASp KO. CD8 Percent of max % of cells Cell number CD11c pHrodo 0 20 40 60 40x104 OT1 proliferation 30x104 * 20x104 10x104 0 * pH rodohi CD11c pHrodo-OVA CD11c pHrodo-OVA 86% 84% WT WT 10μm WKO WKO WKO WKO WT WT WKO WT a b c Percent of max % of cells Cell number pHrodo 0 20 40 60 40x104 OT1 proliferation 30x104 * 20x104 10x104 0 * pH rodohi WKO WKO WT WT WKO WT b c b CD8 CD11c 86% 84% WT WKO a c a Cell number WASpWT GFP+ WASp KO GFP– OT1 proliferation WASpWT 1×104 2×104 3×104 4×104 ** 5×104 0 f % of cells WASpWT GFP+ WASp KO GFP– 50 40 30 ** 20 10 0 pHrodohi WASpWT e SSC SSC SSC GFP GFP GFP Negative control WASpWT WASpΔVCA d O f f d e % of cells Cell number WASpΔVCA GFP+ WASp KO GFP– WASpΔVCA GFP+ WASp KO GFP– OT1 proliferation WASpΔVCA ** * 50 40 30 20 10 0 1×104 2×104 3×104 4×104 5×104 0 pHrodohi WASpΔVCA Figure 8 | WASp and the WASp-VCA domain in cross-presentation. (a–c) Wild-type and WASp KO BM DCs from mice on C57Bl/6 background were analysed for capacity to acidify pH rodo-ovalbumin and induce proliferation of OT-I (Vb5.1/5.2 þ) CD8 þ T cells. (d–f) Wild-type BM DCs at day 6 of culture were treated with CK666 to inhibit activity of the Arp2/3 complex or treated with a-amanitin to inhibit polymerase II gene transcription. (g,h) WASp KO BM DCs at day 6 of culture were Amaxa transfected with wild-type WASp (GFP-WASpWT) or WASp lacking the VCA domain (GFP-WASpDVCA) and sorted into GFP þ and GFP  cells 6 h after transfection. (a) Expression of CD11c and CD8 on BM DCs at day 6 of culture and analysis of acidification using uptake of soluble pH rodo-ovalbumin and confocal microscopy. (b,e,h) DCs were incubated with pH rodo-ovalbumin-coated beads and gated for DCs that had taken up one bead and pH rodo intensity was analysed by flow cytometry. Discussion d f Studies of WAS patient cells and WASp-deficient mice have provided critical insights into the role of cell trafficking and cell-to-cell communication during an immune response1–3. Decreased migratory or cell-to-cell interaction responses in WASp-deficient cells have been interpreted to mean that WASp directly regulates these responses in WASp-sufficient cells. The results presented in the current study provide an alternative explanation. We provide evidence for that WASp deficiency skews intracellular signalling to Rac2 activation that locally maintains a near neutral pH of endosomes and phagosomes, necessary for cross-presentation. We propose that deletion of key proteins, such as WASp, in patients and mice may induce alternative signalling pathways for cell survival and function that leads to altered biological responses. WASp serves an important role in the immunological synapse between DCs and CD4 þ T cells. WASp KO CD4 þ T cells display normal immunological synapse formation but are unable to reform the immunological synapse after each migration phase38. At the DC side of the synapse, WASp KO bone marrow- derived DCs form less stable interactions with ovalbumin-specific wild-type CD4 þ T cells in vitro resulting in decreased T-cell activation39. The role of WASp in CD8 þ DCs has been addressed by direct targeting of ovalbumin to the DEC205 receptor, uniquely expressed on CD8 þ DCs. Using this approach, wild- type CD8 þ DCs induced long-lasting contacts with ovalbumin- specific wild-type CD8 þ T cells in vivo, while WASp KO CD8 þ DCs formed much shorter contacts, suggesting decreased activation of CD8 þ T cells13. However, this study did not distinguish between changes in uptake of ovalbumin complexed with anti-DEC205 antibodies by WASp KO CD8 þ DCs and changes in presentation of MHC class I—ovalbumin peptides by WASp KO CD8 þ DCs13. In the present study we show that both WASp KO CD8 þ and CD8  DCs had reduced capacity to take up IgG-ovalbumin immune complexes via Fc receptors, whereas uptake of soluble ovalbumin was similar to wild-type DCs. In vivo, serial brief contacts (min) between antigen-presenting DCs and CD8 þ T cells induce early CD8 þ T-cell activation, proliferation, and differentiation into effector cytotoxic T cells40. However, long-lasting contacts (hours) are needed to form CD8 þ memory cells in response to antigen40. Discussion d f WASp KO CD8 þ DCs form immune synapses with CD8 þ T cells that are short-lived in vivo13, and we show in the present study that WASp KO CD8 þ and CD8  DCs formed immune synapses in vitro leading to increased activation and proliferation of CD8 þ T cells. Upon L. major infection, we showed that despite lower number of CD8 þ DCs with capacity to cross-present antigen in the dLNs, WASp KO mice had increased activation of dLNs CD8 þ T cells. Moreover, we detected a consistent increase in CD8 þ T cells over CD4 þ T cells in secondary lymphoid organs of WASp KO mice irrespective of inflammatory challenge. This finding suggested that the milieu in WASp KO mice favors CD8 þ T-cell homoeostasis. However, WASp KO CD8 þ T cells fail to respond efficiently when specific antigens are presented during viral infections as shown by other groups25,26,41. An explanation for these seemingly contradictory findings is that polyclonal activation of CD8 þ T cells in WASp KO mice impedes the expansion of antigen-specific CD8 þ T cells during viral infection. On the basis of the previous studies and the findings in the present study, we propose that downregulation of cross- presentation by WASp may be an active process that is essential to prevent over-activation of CD8 þ T cells. p Our data from re-expression of WASp in WASp KO BM-derived DCs suggests that WASp directly reduced cross-presentation in CD8  DCs in a process independent of the WASp-VCA domain. The WASp-VCA domain is also dispensable for WASp activity in T-cell receptor-mediated transcriptional activation44,45. Future studies are needed to address what part of WASp that is important for antigen processing and presentation. The WASp family of proteins have redundant and unique activities within the cell. N-WASp activity can compensate for critical functions of WASp during lymphocyte development since deletion of both WASp and N-WASp in B or T cells leads to severely compromised development and function46,47. Another example of such compensatory mechanism comes from studies of WASp KO NK cells. Treatment with IL-2 restores normal cytotoxicity of WASp KO NK cells by increased activation of the WASp-family protein WAVE2 (ref. 48). We have now identified another such compensatory mechanism in which WASp KO CD8  DCs adapt to a CD8 þ DCs phenotype by increasing Rac2 expression and localizing Rac2 to phagosomal membranes. Mice. All mice were bred and maintained in the same room at the animal facility at the Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, under specific pathogen-free conditions. WASp KO mice on C57Bl/6 and Balb/c background were backcrossed for at least nine generations. Balb/c, C57Bl/6, WASp KO Balb/c, WASp KO C57BL/6, DC/cWKO C57BL/6, OT-I Rag1  /  C57BL/6, OT-II C57BL/6, C57BL/6 J-Ncf1m1J/m1J (referred as Ncf1* in the text), WASp KO C57BL/6  C57BL/6 J-Ncf1m1J/m1J mice (referred as WASp KO  Ncf1* in the text) were used at 6–13 weeks of age. The DC/cWKO colony was maintained by breeding WASpfl/flCD11cwt/wt females to WASpfl/yCD11ccre/wt males or WASpfl/flCD11ccre/wt females to WASpfl/yCD11cwt/wt males. C57BL/6J-Ncf1m1J/m1J Discussion d f Importantly, CD8  DCs constitutes up to 40% of DCs in the spleen, as compared with 10% CD8 þ DCs, and in the setting of wild-type T cells in DC/cWKO mice, WASp KO DCs induced marked expansion of CD8 þ T cells. While WASp-deficient CD4 þ T cells are inherently hyporesponsive49–52, recent studies raises an emerging view in which WASp deficiency affects specific cells differently. In fact, WASp deficiency in plasmacytoid DCs53, B cells35,54–56 and DC-mediated activation of CD8 þ T cells (this study) induce hyper-responsive cells that become subjected to cellular exhaustion. Our data has implications for the treatment of WAS patients and raises concerns for those patients that have limited myeloid reconstitution and normal T-cell reconstitution after BM transplantation and gene therapy57–59. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 GFP  WASp KO BM DCs (Fig. 8e,f). Together, this data suggests that WASp directly reduced cross-presentation in CD8  DCs in a process independent of the WASp-VCA domain. During contact eczema, activated effector CD8 þ T cells are recruited into the skin where they initiate the inflammatory cascade by inducing apoptosis of keratinocytes42. Allergen- carrying Langerhans cells induce tolerance upon migration to the dLNs where allergen-specific CD8 þ T cells are deleted and regulatory T cells activated43. In the present study, WASp KO mice challenged with Der p 2 had increased accumulation of DCs in the dermis including the CD11c þEpCAM Langerin  DCs that can cross-present antigens. Using L. major infection, we detected fewer migratory MHC class IIhigh and CD103 þ DCs in the dLNs in WASp KO mice. Moreover, we found increased expansion of WASp KO CD8 þ T cells that were prone to produce IFNg both in vivo and in vitro. A unifying hypothesis is that skin pathology in WASp deficiency may result from decreased egress of Langerhans cells and dermal DCs from the skin, increased activation of CD8 þ T cells with high capacity to producing IFNg in the skin and dLNs, and decreased suppressive function of WASp KO regulatory T cells5–8. Local accumulation of DCs in the dermis, as we show in the present study, with capacity to cross-present antigen and activate CD8 þ T cells would worsen this vicious cycle of CD8 þ T-cell activation in WASp KO mice. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 Moreover, reduced NADPH activity in WASp KO  Ncf1* CD8  DCs restored the capacity to acidify phagosomes to pH 3–5 as measured by beads coated with pH rodo-ovalbumin (Supplementary Fig. 10). Together, our data shows that WASp KO CD8  DCs adapt to a CD8 þ DC phenotype with increased expression of Rac2 that co-localizes with the phagosome and induces elevated ROS production. 11 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE Single cell suspensions for dLN and spleen were prepared by incubating SPL and dLNs in 1% serum complete medium plus 0.5 mg ml  1 collagenase D for 30–45 min or plus 1.5 mg ml  1 collagenase D for 1.5 h. Ears were further processed using a BD Medimachine system (BD Biosciences) for tissue grinding. Single cell suspensions from spleen were cultured in complete RPMI medium with Der p 2 or PMA and ionomicin (Sigma), and incubated at 37 C for 4–48 h. Golgiplug (BD Biosciences) was added to the last 4 h of incubation to all cell cultures. For in vivo migration, Aldara 5% cream (imiquimod; Meda) was applied epicutaneously on ears of anesthetized mice and let dry to induce dendritic cell maturation and migration to the dLNs. Draining LNs were treated with col- lagenase D as described above for single cell suspension and analyzed by FACS. Immune synapse and cross-presentation assay. For immune synapse experi- ments, the splenic DC population was expanded by subcutaneous injection of 1  106 Flt3-ligand B16 melanoma cells and DCs purified after 7–10 days with the CD8 þ DC isolation kit followed by CD11c positive selection (Miltenyi Biotec). CD8 þ and CD8  DCs were pulsed with ovalbumin overnight and incubated with OT-I CD8 þ T cells for 2 h. Cells were transferred to fibronectin-coated slides, fixed and stained with g -ubulin (green) and actin (red). For cross-presentation assays, CD8 þ DCs and CD8  DCs from wild-type and WASp KO C57Bl/6 mice were isolated by incubating spleens with complete medium plus 1% serum and 0.5 mg ml  1 collagenase D followed by enrichment with the Dynabeads mouse DC enrichment kit (Invitrogen) and thereafter FACS sorted based on CD8 and CD11c using FACS Vantage. FACS-sorted DCs were pulsed with 16, 50 and 150 mg ml  1 soluble ovalbumin, with 3 mm ovalbumin-coated latex beads (Life Technologies), or with 2 mg ml  1 SIINFEKL peptide overnight. To examine cross-presentation of immune complexes (ICs), sorted DCs were incubated with pre-formed ICs of anti- TNP-IgG1 and TNP(5)-OVA at 0.25, 1.25 and 6.25 mg ml  1 final concentrations. OT-I CD8 þ T cells from spleen were sorted by negative selection using the CD8a T-cell isolation kit II (Miltenyi Biotec), with a purity of more than 95%, and labelled with 2 mM CFSE (Invitrogen). ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 immunohistochemistry46. Images were acquired using a Leica DM IRBE confocal laser scanning microscope (Leica Microsystems) equipped with 1 argon and 2 HeNe lasers, using an HC PL APO lens at 20  /0.70 CS and 63  /1.32 IMM CORR oil and 90% glycerol (MP Biomedicals). Images were processed with Adobe Photoshop CS4 Version 11.0.2 (Adobe Systems) and ImageJ software. Histological examination and analysis of skin sections were performed blindly where the identity of the section was unknown to the observer. The epidermis thickness values represent the picture with the highest measured thickness of each mouse. When counting cells on images, co-localization of two fluorochromes was based on colour intensity between 150–255 pixel co-localization of red, green, and/or blue using the ImageJ software. To quantify the cells per area on epidermis or in the dermis of whole-skin sections, the cells were counted and areas measured using ImageJ. For flow cytometry analysis, single cell suspensions from organs were prepared and erythrocytes lysed with ACK buffer. Cells were immunolabelled and acquired using a FACS Aria or LSR Fortessa (Becton Dickenson). Analyses were made using FlowJo software (version 7.2.5 TreeStar Inc.). mice were purchased from Jackson and backcrossed to WASp KO C57BL/6 mice to generate WASp KO  Ncf1* homozygous mice. All wild-type mice were littermate controls from heterozygous breedings with the respective gene-targeted allele and genotyped before use. Animal experiments were performed after approval from the local ethical committee (the north Stockholm district court). Der p 2 challenge protocol and in vivo migration. Wild-type and WASp KO Balb/c mice were challenged with Der p 2 (ref. 29). Briefly, mice were shaved on the back and patched 3  4 days with 50 mg of Der p 2 in 100 ml of PBS on 1 cm2 of shaved skin and sacrificed at day 50. We have used two different preparations of Der p 2 with o35 ng of lipopolysaccharide per mg protein and obtained similar results for the in vivo and in vitro experiments using both batches. Unchallanged mice were not patched or shaved. For epidermal sheets, full-thickness skin from back was incubated for 90 min at 37 C in a PBS solution plus 2.5 mg ml  1 dispase (Invitrogen). ARTICLE Sorted DCs were co-cultured with OT-I CD8 þ T cells at 1:10 DC:T-cell ratio and analyzed by FACS after 72 h. For the 3H-thymidine incorporation assay, OT-I T cells were co-cultured with CD8 þ or CD8  DCs for 72 h and pulsed with 1 mCi of 3H-thymidine for the last 12 h, collected and scintillation measured. L. major infection. L. major strain Fv1 (MHOM/IL/80/FN) was grown at 25 C to stationary phase in complete M199 supplemented with 20% foetal calf serum (parasite growth medium). Parasites were enriched for infectious metacyclic promastigotes by Ficoll 400 gradient separation33. Estimation of parasite number was determined by serial 1:2 dilutions of tissue homogenates in parasite growth medium. The number of viable parasites in each sample was calculated based on the highest dilution at which promastigotes could be grown out after 4–5 days of incubation at 25 C. The limitation of the assay was 10 parasites per ear and 20 parasites per dLN. Wild-type and WASp KO Balb/c mice were infected intradermally in both ears with 4  104 metacyclic L. major promastigotes in 10 ml DMEM. Control mice received sham injections of DMEM. Progression of lesion development was measured weekly using a digital veiner calliper and indicated as diameter of the lesion. After 2 or 6 weeks mice were euthanized, and ears and retromaxillar dLNs were removed. Single cell suspensions were prepared by incubating ears and dLNs in 1% serum complete medium plus 1.5 mg ml  1 collagenase D for 1.5 h or plus 0.5 mg ml  1 collagenase D for 30–45 min, respectively. Ears were further processed using a BD Medimachine system (BD Biosciences) for tissue grinding. To measure cytokine production, single cell suspensions from retromaxillar dLNs were cultured with PMA and ionomicin (Sigma) and golgiplug (Becton Dickenson) for 4 h. Antigen processing and ROS production assay. The splenic DC population was expanded by subcutaneous injection of 1  106 Flt3-ligand B16 melanoma cells and DCs purified after 7–10 days with the CD8 þ DC isolation kit followed by CD11c positive selection (Miltenyi Biotec). Enriched CD8 þ DCs and CD8  DCs were incubated with 16, 50 and 150 mg ml  1 ovalbumin-Alexa594 or DQ-ovalbumin (Invitrogen) for up to 6 h for ovalbumin uptake and degradation, respectively. ARTICLE To assess acidification capacity, ovalbumin was linked to pH-rodo according to the manufacturer’s instructions (Invitrogen) and DCs were incubated with 50 mg ml  1 ovalbumin-pH rodo for 6 h. To measure the production of ROS, the probe DHR (Life Technologies) was used. To control the amount of ovalbumin-pH rodo or DHR taken up over time by each cell, a particulate antigen assay was performed by coating 3 mm latex beads (Life Technologies) to pH rodo-ovalbumin or DHR. DCs that took up only one bead were gated on a flow cytometer according to FSC versus SSC parameters for acidification analysis. As negative controls for acidification and for ROS production, DCs were either pre-treated with 20 mM of NH4Cl before addition of the pH rodo-ovalbumin-coated beads, or with 20 mM DPI before addition of the DHR-coated beads. Upon incubation, cells were stained with CD11c, MHCII and CD8 to identify CD11c þMHCII þCD8 þ DC and CD11c þ MHCII þCD8  DCs and other cells including plasmacytoid DCs, B and T cells were excluded in a dump gate using B220 and CD3. Dead cells were excluded based on positive staining for DAPI. The pH was measured with a slightly modified protocol described in Savina et al.36 In short, DCs were enriched using dynabeads mouse DC enrichment (Life Technologies) and incubated with NHS-ester polyamino beads coupled with 1 mg ml  1 FITC and 1 mg ml  1 Alexa647. A standard curve was performed where enriched DCs were allowed to phagocytose the beads for 30 min. Cells were fixed with 4% paraformaldehyde, and incubated with 0.1% Triton X-100 in PBS with pH kept at 7. Afterwards DCs were submitted to PBS solutions with pH ranging from 5 to 8 and let incubate for a couple of minutes. Samples were immediately analysed by FACS and MFI of FITC and Alexa647 measured on CD11c þ-gated cells. The standard curve was obtained by calculating the ratio of ‘MFI FITC/MFI Alexa647’ at the corresponding pH. WT and WASp KO DCs were incubated with the same beads used for the standard curve, the MFI for FITC and Alexa647 was obtained on gated CD8 þCD11c þ or CD8 CD11c þ cells (dump gate used with B220, CD3 and live/dead antibodies) and the pH calculated using the formula obtained by the standard curve. Methods Mice. All mice were bred and maintained in the same room at the animal facility at the Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, under specific pathogen-free conditions. WASp KO mice on C57Bl/6 and Balb/c background were backcrossed for at least nine generations. Balb/c, C57Bl/6, WASp KO Balb/c, WASp KO C57BL/6, DC/cWKO C57BL/6, OT-I Rag1  /  C57BL/6, OT-II C57BL/6, C57BL/6 J-Ncf1m1J/m1J (referred as Ncf1* in the text), WASp KO C57BL/6  C57BL/6 J-Ncf1m1J/m1J mice (referred as WASp KO  Ncf1* in the text) were used at 6–13 weeks of age. The DC/cWKO colony was maintained by breeding WASpfl/flCD11cwt/wt females to WASpfl/yCD11ccre/wt males or WASpfl/flCD11ccre/wt females to WASpfl/yCD11cwt/wt males. C57BL/6J-Ncf1m1J/m1J 12 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications ARTICLE et al. Expression of Wiskott–Aldrich syndrome protein in dendritic cells regulates synapse formation and activation of naive CD8 þ T cells. J. Immunol. 181, 1135–1142 (2008). 14. Kim, A. S., Kakalis, L. T., Abdul-Manan, N., Liu, G. A. & Rosen, M. K. Autoinhibition and activation mechanisms of the Wiskott–Aldrich syndrome protein. Nature 404, 151–158 (2000). 15. Torres, E. & Rosen, M. K. Contingent phosphorylation/dephosphorylation provides a mechanism of molecular memory in WASP. Mol. Cell 11, 1215–1227 (2003). 16. Machesky, L. M. & Insall, R. H. Scar1 and the related Wiskott–Aldrich syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex. Curr. Biol. 8, 1347–1356 (1998). 17. Miki, H., Suetsugu, S. & Takenawa, T. WAVE, a novel WASP-family protein involved in actin reorganization induced by Rac. EMBO J. 17, 6932–6941 (1998). 18. Heath, W. R. & Carbone, F. R. Dendritic cell subsets in primary and secondary T cell responses at body surfaces. Nat. Immunol. 10, 1237–1244 (2009). 18. Heath, W. R. & Carbone, F. R. Dendritic cell subsets in primary and secondary T cell responses at body surfaces. Nat. Immunol. 10, 1237–1244 (2009). 19. Bedoui, S. et al. Cross-presentation of viral and self antigens by skin-derived CD103 þ dendritic cells. Nat. Immunol. 10, 488–495 (2009). p y 19. Bedoui, S. et al. Cross-presentation of viral and self antigens by skin-derived CD103 þ dendritic cells. Nat. Immunol. 10, 488–495 (2009). 20. Joffre, O. P., Segura, E., Savina, A. & Amigorena, S. Cross-presentation by dendritic cells. Nat. Rev. Immunol. 12, 557–569 (2012). 21. Savina, A. et al. The small GTPase Rac2 controls phagosomal alkalinization and antigen crosspresentation selectively in CD8( þ) dendritic cells. Immunity 30, 544–555 (2009). 22. den Haan, J. M. & Bevan, M. J. Constitutive versus activation-dependent cross-presentation of immune complexes by CD8( þ) and CD8(  ) dendritic cells in vivo. J. Exp. Med. 196, 817–827 (2002). Transfection of BM DCs. BM cells from femur and tibia were cultured in 20 ng ml  1 GMCSF (Peprotech) for 6 days. For confocal microscopy, BM DCs were incubated with 0.5 mg ml  1 soluble ovalbumin-pH rodo for 1 h and analyzed by confocal microscopy after staining with anti-CD11c. For re-expression of WASp, WASp KO BM DCs were transfected with eGFP-WASpWT or eGFP- WASpDVCA constructs60 using Amaxa transfection (Primary cell 4D nucleofector kit, Lonza). After 6 h, GFP þ and GFP  cells were FACS sorted using FACS Jazz. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 DCs in the presence of ICs was determined by flow cytometry and staining of DCs with the antibody towards H-2Kb-SIINFEKL (clone 25-D1.16, Biolegend). 11. Bouma, G., Burns, S. & Thrasher, A. J. Impaired T-cell priming in vivo resulting from dysfunction of WASp-deficient dendritic cells. Blood 110, 4278–4284 (2007). 12. Burns, S., Thrasher, A. J., Blundell, M. P., Machesky, L. & Jones, G. E. Configuration of human dendritic cell cytoskeleton by Rho GTPases, the WAS protein, and differentiation. Blood 98, 1142–1149 (2001). Rac1/2 assays. The splenic DC population was expanded by subcutaneous injection of 1  106 Flt3-ligand B16 melanoma cells and DCs enriched after 7–10 days with the CD8 þ DC isolation kit followed by CD11c positive selection (Miltenyi Biotec). To determine Rac1/2 and gp91phox expression and co-locali- zation to the phagosome, enriched CD8 þ DCs and CD8  DCs were incubated with 3 mm ovalbumin-coated latex beads for 2 h at 37 C, transferred to glass coverslips coated with 50 mg ml  1 fibronectin (Gibco) and incubated for 1 h at 37 C. Afterwards, cells were fixed with 4% paraformaldehyde, quenched with 0.1 M glycine, permeabilized and stained intracellularly with primary antibodies for Rac1, Rac2 and gp91phox. Cells were analysed using a Zeiss LSM 780 confocal microscope equipped with a diode laser with wavelength 405 nm, an Argon laser with wavelengths 458/488/514 nm DPSS 561 nm and a HeNe 633 nm laser. Pictures were acquired using a Plan-Apo 63  /1.4 oil-immersion lens and the ZEN Black 2011 software. Rac2 co-localization with the phagosome was calculated as: [(beads with Rac2)/(cells with beads)]  100. Five to 15 images per mouse were analyzed. Rac2 co-localization was further assessed by z-stack analysis from pictures taken with a 63  objective and a step-size of 1.3 mm per picture. Pictures were com- pressed into one single image using ImageJ. For Rac1/2 activity, protein lysates were prepared from DCs incubated with ovalbumin-coated latex beads for 2 h and protein content measured and diluted to equal concentration. Active GTP-bound Rac1/2 was quantified using the G-LISA Rac1,2,3 Activation Assay Biochem Kit (Cytoskeleton). To quantify GTP-bound Rac2 a modified version of the G-LISA kit was used with an anti-Rac2 (Santa Cruz Biotechnology) detection antibody, anti-rabbit-HRP (Santa Cruz Biotechnology) secondary, and anti-goat-HRP (Santa Cruz Biotechnology) tertiary antibodies. g y y protein, and differentiation. Blood 98, 1142–1149 (2001). 13. Pulecio, J. ARTICLE To assess acidification capacity, BM DCs after Amaxa transfection were incubated with ovalbumin-pH rodo beads for 30 min and pH rodo fluorescence determined in cells that had taken up one bead. To determine proliferation of OT-I CD8 þ T cells, BM DCs after Amaxa transfection were incubated with 0.5 mg ml  1 ovalbumin and LPS overnight. BM DCs were co-cultured with OT-I CD8 þ T cells at 1:10 DC:T-cell ratio and analyzed by FACS after 48 h. 23. Regnault, A. et al. Fcgamma receptor-mediated induction of dendritic cell maturation and major histocompatibility complex class I-restricted antigen presentation after immune complex internalization. J. Exp. Med. 189, 371–380 (1999). 24. Baker, K. et al. Neonatal Fc receptor for IgG (FcRn) regulates cross- presentation of IgG immune complexes by CD8-CD11b þ dendritic cells. Proc. Natl Acad. Sci. USA 108, 9927–9932 (2011). 25. Lang, P. A. et al. Reduced type I interferon production by dendritic cells and weakened antiviral immunity in patients with Wiskott–Aldrich syndrome protein deficiency. J. Allergy Clin. Immunol. 131, 815–824 (2013). g ovalbumin and LPS overnight. BM DCs were co-cultured with OT-I CD8 þ T cells at 1:10 DC:T-cell ratio and analyzed by FACS after 48 h. 26. Andreansky, S. et al. WASP- mice exhibit defective immune responses to influenza A virus, Streptococcus pneumoniae, and Mycobacterium bovis BCG. Exp. Hematol. 33, 443–451 (2005). Data and statistical analysis. For comparison between wild-type and WASp KO mice, data passed the normality test and was analyzed by the unpaired Student’s t-test using GraphPad Prism 5.0 software and a two-tailed P value with 95% confidence interval was acquired. Data is shown as mean±s.d. and Po0.05 was considered significant. 27. Strom, T. S., Gabbard, W., Kelly, P. F., Cunningham, J. M. & Nienhuis, A. W. Functional correction of T cells derived from patients with the Wiskott–Aldrich syndrome (WAS) by transduction with an oncoretroviral vector encoding the WAS protein. Gene Ther. 10, 803–809 (2003). g p confidence interval was acquired. Data is shown as mean±s.d. and Po0.05 was considered significant. 28. Darsow, U. et al. The prevalence of positive reactions in the atopy patch test with aeroallergens and food allergens in subjects with atopic eczema: a European multicenter study. Allergy 59, 1318–1325 (2004). Data availability. The data that support the findings of this study are available from the corresponding authors on request. Data availability. References is down-regulated in eczema. Allergy 64, 304–311 (2009). 30. Huang, C. H., Kuo, I. C., Xu, H., Lee, Y. S. & Chua, K. Y. Mite allergen induces allergic dermatitis with concomitant neurogenic inflammation in mouse. J. Invest. Dermatol. 121, 289–293 (2003). 1. Thrasher, A. J. & Burns, S. O. WASP: a key immunological multitasker. Nat. Rev. Immunol. 10, 182–192 (2010). 2. Bosticardo, M., Marangoni, F., Aiuti, A., Villa, A. & Roncarolo, M. G. Recent advances in understanding the pathophysiology of Wiskott–Aldrich syndrome. Blood 113, 6288–6295 (2009). 31. Henri, S. et al. Disentangling the complexity of the skin dendritic cell network. Immunol. Cell Biol. 88, 366–375 (2010). 32. Heath, W. R. & Carbone, F. R. The skin-resident and migratory immune system in steady state and memory: innate lymphocytes, dendritic cells and T cells. Nat. Immunol. 14, 978–985 (2013). 3. Massaad, M. J., Ramesh, N. & Geha, R. S. Wiskott–Aldrich syndrome: a comprehensive review. Ann. N. Y. Acad. Sci. 1285, 26–43 (2013). 4. Ozcan, E., Notarangelo, L. D. & Geha, R. S. Primary immune deficiencies with aberrant IgE production. J. Allergy Clin. Immunol. 122, 1054–1062 (2008). 33. Nylen, S. et al. Splenic accumulation of IL-10 mRNA in T cells distinct from CD4 þCD25 þ (Foxp3) regulatory T cells in human visceral leishmaniasis. J. Exp. Med. 204, 805–817 (2007). 5. Maillard, M. H. et al. The Wiskott–Aldrich syndrome protein is required for the function of CD4( þ)CD25( þ)Foxp3( þ) regulatory T cells. J. Exp. Med. 204, 381–391 (2007). p 34. Nylen, S. & Eidsmo, L. Tissue damage and immunity in cutaneous leishmaniasis. Parasite Immunol. 34, 551–561 (2012). leishmaniasis. Parasite Immunol. 34, 551–561 (2012) 6. Marangoni, F. et al. WASP regulates suppressor activity of human and murine CD4( þ)CD25( þ)FOXP3( þ) natural regulatory T cells. J. Exp. Med. 204, 369– 380 (2007). 35. Recher, M. et al. B cell-intrinsic deficiency of the Wiskott–Aldrich syndrome protein (WASp) causes severe abnormalities of the peripheral B-cell compartment in mice. Blood 119, 2819–2828 (2012). 7. Humblet-Baron, S. et al. Wiskott–Aldrich syndrome protein is required for regulatory T cell homeostasis. J. Clin. Invest. 117, 407–418 (2007). 36. Savina, A. et al. NOX2 controls phagosomal pH to regulate antigen processing during crosspresentation by dendritic cells. Cell 126, 205–218 (2006). 8. Adriani, M. et al. Impaired in vitro regulatory T cell function associated with Wiskott–Aldrich syndrome. Clin. Immunol. 124, 41–48 (2007). 37. Hultqvist, M. et al. ARTICLE The data that support the findings of this study are available from the corresponding authors on request. European multicenter study. Allergy 59, 1318–1325 (2004). 29. Lieden, A. et al. Cornulin, a marker of late epidermal differentiation, is down regulated in eczema Allergy 64 304 311 (2009) 29. Lieden, A. et al. Cornulin, a marker of late epidermal diffe 29. Lieden, A. et al. Cornulin, a marker of late epidermal differentiation, is down-regulated in eczema. Allergy 64, 304–311 (2009). ARTICLE To assess uptake of immunocomplexes (ICs) DCs were enriched as above described and incubated with a 1:20 dilution of pre-formed ICs of anti-TNP-IgG1 and TNP(5)- OVA at 0.25, 1.25 and 6.25 mg ml  1 final concentrations. Antigen presentation of Antibodies. The following antibodies were used for flow cytometry and/or immunohistochemistry: CD3-PE/Cy7 1:100 (557851), CD8a-PerCP/Cy5.5 1:100 (100734), CD8a-Alexa647 1:200 (557682), CD11c-biotin 1:200 (553800), CD45-APC/Cy7 1:200 (557659), CD45-PerCP/Cy5.5 1:200 (103236), CD62L-FITC 1:100 (553150), IgG1-FITC 1:100 (553443), IgE-biotin 1:200 (553419), PDL1-PE 1:200 (558091), B220-V500 1:200 (561227), CD3-V500 1:200 (560773), Va2-FITC 1:100 (553288), Vb5.1/5.2-PE 1:200 (553190), CD86-FITC 1:200 (553691), CD69-PE 1:200 (553237; BD Biosciences), CD4-eFluor450 1:100 (48-0042-82), CD4-APC 1:200 (100516), CD8b-biotin 1:200 (13-0083-81), CD103-FITC 1:200 (121420), F4/80-eFluor450 1:100 (48-4801-82), CD11c-PE/Cy7 1:100 (117318), MHCI-eFluor450 (H2-Kb) 1:100 (48-5958-80), IFNg-PerCP/Cy5.5 1:100 (505822), IL-17A-FITC 1:100 (11-7177-81; eBioscience), Strepavidin-alexa555 1:1000 (S32355) (ThermoFisher), 7AAD 7 ml per 200 ml of cell volume (51-2359KC) (BD Pharmingen), Live/Dead Fixable Vivid-nearIR 1:500 (L10119), live/dead-AmCyan 1:400 (L34966) (Invitrogen), B220-APC 1:400 (103212), B220-Pacific Blue 1:200 (103227), EpCAM-APC 1:200 (118214), CD11b-PerCP/Cy5.5 1:100 (101228), CD4-FITC 1:100 (100510), MHCII-APC (I-A/I-E) 1:400 (107614), TCRb-APC 1:200 (109212), B220-APC/Cy7 1:200 (103224), FcgRI 1:100 (139303), FcgRII/III 1:100 (101308), gp91phox 1:200 (650102), CD44-PE 1:400 (103008) (BioLegend), CD19-APC 1:200 (LS-C148489) (LSBio), CD207(langerin)-FITC 1:400 (DDX0362A488-100) (Dendritics), rac1-FITC 1:50 (bs-4186R-FITC), rac2-Alexa555 1:50 (bs-6153R-A555) (BIOSS), biotin-XX phalloidin 1:1000 (B7474) (Invitrogen), g-tubulin 1:200 (072M4808) (Sigma), FcRn 1:100 (AF6775) (R&D Systems), and cyclin A2 1:200 (ab38) (Abcam), aWASp 1:200 (sc-13139), aGAPDH 1:200 (sc-25778) (Santa Cruz). Histology and flow cytometry. For Der p 2-challenged mice, the skin within the 1 cm2 square shaved and treated with Der p2 was macroscopically examined, and 4 mm2 punch biopsies were taken from areas of red thickened skin for sections and image analysis. Skin tissues were snap frozen in Tissue-Tek OCT (Bio-Optica). Sections (8–10 mm) were cut, air dried overnight and fixed in cold acetone. Hematoxylin and eosin staining was performed using standard protocol and 13 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications Acknowledgements We acknowledge the valuable comments and technical help from Sebastian Amigorena (Institut Curie), Abdelhadi Saoudi (University of Toulouse), Benedict Chambers, John Anderson, Pontus Aspenstro¨m, Mari-Anne Hedblad, Sven Petersen, Jakob Michaelsson, Martin Ivarsson (all at Karolinska Institutet), and Eva Severinson (Stockholm University). This work was supported by a PhD fellowship from Fundac¸a˜o para a Ciaˆncia e a Tecnologia #SFRH/BD/47926/2008 and the Queen Silvia foundation to M.A.P.B., a postdoctoral fellowship from the Swedish Cancer Society to M.K., PhD fellowships from Karolinska Institutet to M.O. and C.I.M.D., an amfAR Mathilde Krim Fellowship in Basic Biomedical Research #108213-51-RKVA to M.N.E.F., grants from Wellcome Trust (104807/Z/14/Z) and by National Institute for Health Research Biomedical Research Centre at Great Ormond Street Hospital for Children NHS Foundation Trust and University College London to A.J.T., as well as the Swedish Research Council, Cancer Society, Childhood Cancer Society, Karolinska Institutet including Center for Allergy Research, the European Commission 7th framework program Marie Curie reintegration grant (#249177), Åke Olsson foundation, Jeansson foundation, Groschinsky Foundation, Åke Wiberg Foundation, Bergvall Foundation, and the Swedish Medical Society to L.S.W. L.S.W. is a Ragnar So¨derberg fellow in Medicine. 42. Akiba, H. et al. Skin inflammation during contact hypersensitivity is mediated by early recruitment of CD8 þ T cytotoxic 1 cells inducing keratinocyte apoptosis. J. Immunol. 168, 3079–3087 (2002). p p 43. Gomez de Agu¨ero, M. et al. Langerhans cells protect from allergic contact dermatitis in mice by tolerizing CD8 þ T cells and activating Foxp3 þ regulatory T cells. J. Clin. Invest. 122, 1700–1711 (2012). g y 44. Silvin, C., Belisle, B. & Abo, A. A role for Wiskott–Aldrich syndrome protein in T-cell receptor-mediated transcriptional activation independent of actin polymerization. J. Biol. Chem. 276, 21450–21457 (2001). p y 45. Sadhukhan, S., Sarkar, K., Taylor, M., Candotti, F. & Vyas, Y. M. Nuclear role of WASp in gene transcription is uncoupled from its ARP2/3-dependent cytoplasmic role in actin polymerization. J. Immunol. 193, 150–160 (2014). W b L S l Wi k Ald i h d i (WASP) d 46. Westerberg, L. S. et al. Wiskott–Aldrich syndrome protein (WASP) and N-WASP are critical for peripheral B-cell development and function. Blood 119, 3966–3974 (2012). 47. Cotta-de-Almeida, V. et al. Wiskott–Aldrich syndrome protein (WASP) and N-WASP are critical for T cell development. Proc. Natl Acad. Sci. USA 104, 15424–15429 (2007). ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms12175 39. Bouma, G. et al. Cytoskeletal remodeling mediated by WASp in dendritic cells is necessary for normal immune synapse formation and T cell priming. Blood 118, 2492–2501 (2011). 59. Griffith, L. M. et al. Primary Immune Deficiency Treatment Consortium (PIDTC) report. J. Allergy Clin. Immunol. 133, 335–347 ð2014Þ: 60. Worth, A. J. et al. Disease-associated missense mutations in the EVH1 domain disrupt intrinsic WASp function causing dysregulated actin dynamics and impaired dendritic cell migration. Blood 121, 72–84 ð2013Þ: 40. Scholer, A., Hugues, S., Boissonnas, A., Fetler, L. & Amigorena, S. Intercellular adhesion molecule-1-dependent stable interactions between T cells and dendritic cells determine CD8 þ T cell memory. Immunity 28, 258–270 (2008). 60. Worth, A. J. et al. Disease-associated missense mutations in the EVH1 domain disrupt intrinsic WASp function causing dysregulated 41. Strom, T. S. et al. Defects in T-cell-mediated immunity to influenza virus in murine Wiskott–Aldrich syndrome are corrected by oncoretroviral vector- mediated gene transfer into repopulating hematopoietic cells. Blood 102, 3108–3116 (2003). Author contributions 48. Orange, J. S. et al. IL-2 induces a WAVE2-dependent pathway for actin reorganization that enables WASp-independent human NK cell function. J. Clin. Invest. 121, 1535–1548 (2011). M.A.P.B. and L.S.W. designed research; M.A.P.B., M.K., M.O., K.K.S.S., J.A., C.I.M.D., A.J.W., A.L., I-C.K., R.P.A.W., S.N., L.S.W. performed research; S.B.S., L.E., A.S., M.C.I.K., G.B., S.O.B., M.N.E.F., A.J.T., S.N. contributed with ideas, new reagents and analytic tools; M.A.P.B., M.K., M.O., K.K.S.S., C.I.M.D., S.N., L.S.W. analyzed the data; M.A.P.B. and L.S.W. wrote the manuscripts; and all authors edited the manuscript. 49. Snapper, S. B. et al. Wiskott–Aldrich syndrome protein-deficient mice reveal a role for WASP in T but not B cell activation. Immunity 9, 81–91 (1998). y 50. Zhang, J. et al. Antigen receptor-induced activation and cytoskeletal rearrangement are impaired in Wiskott–Aldrich syndrome protein-deficient lymphocytes. J. Exp. Med. 190, 1329–1342 (1999). rearrangement are impaired in Wiskott–Aldrich syndrome protein-deficient lymphocytes. J. Exp. Med. 190, 1329–1342 (1999). References Enhanced autoimmunity, arthritis, and encephalomyelitis in mice with a reduced oxidative burst due to a mutation in the Ncf1 gene. Proc. Natl Acad. Sci. USA 101, 12646–12651 (2004). 9. Pivniouk, V. I. et al. Impaired signaling via the high-affinity IgE receptor in Wiskott–Aldrich syndrome protein-deficient mast cells. Int. Immunol. 15, 1431–1440 (2003). 38. Sims, T. N. et al. Opposing effects of PKCtheta and WASp on symmetry breaking and relocation of the immunological synapse. Cell 129, 773–785 (2007). 10. de Noronha, S. et al. Impaired dendritic-cell homing in vivo in the absence of Wiskott–Aldrich syndrome protein. Blood 105, 1590–1597 (2005). 14 Additional information 51. Morales-Tirado, V. et al. Cutting edge: selective requirement for the Wiskott–Aldrich syndrome protein in cytokine, but not chemokine, secretion by CD4 þ T cells. J. Immunol. 173, 726–730 (2004). Supplementary Information accompanies this paper at http://www.nature.com/ naturecommunications Competing financial interests: The authors declare no competing financial interests. Competing financial interests: The authors declare no competing financial interests. 52. Trifari, S. et al. Defective Th1 cytokine gene transcription in CD4 þ and CD8 þ T cells from Wiskott–Aldrich syndrome patients. J. Immunol. 177, 7451–7461 (2006). Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ 53. Prete, F. et al. Wiskott–Aldrich syndrome protein-mediated actin dynamics control type-I interferon production in plasmacytoid dendritic cells. J. Exp. Med. 210, 355–374 (2013). How to cite this article: Baptista, M. A. P. et al. Deletion of Wiskott–Aldrich syndrome protein triggers Rac2 activity and increased cross-presentation by dendritic cells. Nat. Commun. 7:12175 doi: 10.1038/ncomms12175 (2016). 54. Becker-Herman, S. et al. WASp-deficient B cells play a critical, cell-intrinsic role in triggering autoimmunity. J. Exp. Med. 208, 2033–2042 (2011). 54. Becker-Herman, S. et al. WASp-deficient B cells play a critical, cell-intrinsic role in triggering autoimmunity. J. Exp. Med. 208, 2033–2042 (2011). 55. Dahlberg, C. I. et al. Deletion of WASp and N-WASp in B cells cripples the germinal center response and results in production of IgM autoantibodies. J. Autoimmun. 62, 81–92 (2015). This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ 55. Dahlberg, C. I. et al. Deletion of WASp and N-WASp in B cells cripples the germinal center response and results in production of IgM autoantibodies. J. Autoimmun. 62, 81–92 (2015). 56. Volpi, S. et al. N-WASP is required for B-cell-mediated autoimmunity in Wiskott–Aldrich syndrome. Blood 127, 216–220 (2016). 57. Aiuti, A. et al. Lentiviral hematopoietic stem cell gene therapy in patients with Wiskott–Aldrich syndrome. Science 341, 1233151 (2013). 58. Boztug, K. et al. Stem-cell gene therapy for the Wiskott–Aldrich syndrome. N. Engl. J. Med. 363, 1918–1927 (2010). r The Author(s) 2016 15 NATURE COMMUNICATIONS | 7:12175 | DOI: 10.1038/ncomms12175 | www.nature.com/naturecommunications